| MAPSHIM | Analysis | Used to improve the magnetic field homogeneity over the VOI for MRS acquisitions. | Not Listed | #12 |
| Brain Connectivity Toolbox | Analysis | Assess modularity of functional connectivity matrices using the Louvain algorithm. | Not Listed | #16 |
| Ginkgo toolbox | Analysis | Process diffusion MRI data. | Not Listed | #18 |
| RABIES | Analysis | Preprocesses MRI data for rat brain imaging, including registration to the Waxholm template. | Not Listed | #25 |
| REGAIN | Reconstruction | Resolution enhancement of low-resolution magnitude images after denoising and coil combination, using a generative AI-based model. | Not Listed | #32 |
| K-CC-MoCo | Reconstruction | Motion correction in k-space for first-pass perfusion cardiac MRI reconstruction, effective even at high acceleration factors. | Not Listed | #33 |
| InSil algorithm | Analysis | Non-rigid registration across inversion times and doses prior to T1 fitting for MOLLI data. | Not Listed | #36 |
| Generative Multitasking framework | Reconstruction | Learns an implicit neural representation of the temporal subspace of multidimensional images for improved motion characterization and separation of respiration from DCE in dynamic contrast enhanced (DCE) MRI. | Not Listed | #38 |
| MIJER | Reconstruction | Synergistically estimates motion and reconstructs images directly from highly accelerated SN-CMRA data using a joint implicit neural representation and a patch-based denoising prior. | Not Listed | #40 |
| ANTs toolkit | Analysis | Spatial normalization of individual metabolic maps to the MNI space via two-step registrations using first-echo water signal and T1-weighted MPRAGE as intermediate registration references with affine transformation. | http://stnava.github.io/ANTs/ | #44 |
| DL-Radiomics Model | Analysis | Automated segmentation and classification of parotid tumors to distinguish between benign and malignant tumors using MRI images. | Not Listed | #47 |
| Quantiphyse | Analysis | Analyze ASL images and quantify CBF. | Not Listed | #55 |
| MiniMORPH | Analysis | To estimate total and regional brain volumes from MRI scans. | Not Listed | #57 |
| SFNet | Reconstruction | Enhance the resolution and contrast of ultra-low-field (uLF) MRI scans to generate “Super-Field” (SF) scans suitable for volumetric measurements. | Not Listed | #59 |
| MaRCoS | Control System | Operates amplifiers and other circuitry in an MRI system, runs pulse sequences, and detects and processes MR signals. | Not Listed | #60 |
| MaRCoS | Pulse Sequence Design | Open-source MRI control system for pulse sequencing and data acquisition. | Not Listed | #61 |
| MaRCoS | Data Management | Open-source MRI console for controlling and acquiring data from the low-field MRI scanner. | Not Listed | #63 |
| KomaMRI.jl | Simulation | Implemented Magnus-based methods for solving Bloch equations to improve accuracy and speed of MRI simulations, particularly for RF excitation. | https://github.com/JuliaHealth/KomaMRI.jl/pull/612 | #68 |
| 6D-Bloch | Reconstruction | Performs Bloch-based modeling for the magnetization vector at each voxel in 3D space evolving under any 3D magnetic field for MR image reconstruction. | Not Listed | #73 |
| Real-time PMC pipeline | Reconstruction | Calculates motion parameters from raw inertial sensor data and transmits them to the scanner for real-time prospective motion correction. | Not Listed | #75 |
| streaming-MoCo | Reconstruction | Online motion correction for MRI data streamed from Siemens IDEA. | Not Listed | #77 |
| Temporal Manifold–Conditioned Diffusion Model | Reconstruction | An unsupervised dynamic MRI reconstruction framework that integrates the general spatial prior of a pre-trained diffusion model with a motion-aware latent manifold learned by a temporal Deep Image Prior (DIP). | Not Listed | #78 |
| rf_shimming | Pulse Sequence Design | Enables user-friendly control of individual transmit elements with minor code modifications in Pulseq for RF shimming and parallel transmission applications. | github.com/HarmonizedMRI/rf_shimming | #81 |
| PyRadiomics | Analysis | Extracts radiomics features from medical images, specifically T2WI and DWI sequences in this study. | Not Listed | #101 |
| DL fusion network | Analysis | Predict lymph node metastasis (LNM) in locally advanced rectal cancer (LARC) patients after neoadjuvant chemoradiotherapy (nCRT) by integrating clinical indicators, radiomics, and deep learning features from MRI images. | Not Listed | #105 |
| Custom preprocessing pipeline | Reconstruction | This pipeline registers NM-MRI images to a brainstem template via T1w-MRI, warps a brainstem mask to the native NM-MRI space, crops the image, and performs signal clipping using percentiles from the brainstem-masked region. | Not Listed | #110 |
| UDN | Analysis | Deep learning model for contrast-agnostic deep-brain localization and segmentation across the human lifespan. | Not Listed | #112 |
| nnU-Net | Reconstruction | Automatic segmentation of the parasagittal dura (PSD) in children with autism spectrum disorder using 3D T2-FLAIR images. | Not Listed | #113 |
| THOMAS | Segmentation | Generate thalamic masks for each hemisphere. | Not Listed | #118 |
| 2D U-Net DBS Lead Segmentation Tool | Reconstruction | An automated, open-source tool for accurate deep brain stimulation (DBS) lead segmentation from postoperative CT and MRI images using a 2D U-Net model. | Not Listed | #119 |
| nnU-Net | Segmentation | Automated segmentation of cerebellar peduncles from diffusion MRI. | Not Listed | #121 |
| nnU-Net | Reconstruction | To develop and evaluate a fully automated deep learning-based meningioma segmentation model using a large, multi-institutional dataset and assess its clinical readiness. | Not Listed | #122 |
| nnU-Net | Analysis | Automatically segment the vocal tract over the 3D volume through time. | Not Listed | #124 |
| LSTtoolbox | Analysis | Segment T1 images into CSF, GM, and WM, combine with coregistered FLAIR intensities to calculate lesion belief maps, and generate a lesion probability map for white matter lesion quantification. | [www.statistical- modelling.de/lst.html](www.statistical- modelling.de/lst.html) | #125 |
| Web-based viewer | Analysis | A Spanish-language web-based tool was developed for interactive visualization of neuroanatomy, including gross neuroanatomy, brainstem nuclei, cerebellum, and brainstem cranial nerves. | https://3dviewer.lavis.unam.mx/projects/neuroinmersa/ | #126 |
| DeepLabCut | Analysis | Label body parts in behavioral videos to train a geometric deep neural network for classifying mice across different groups. | Not Listed | #131 |
| SAM (Segment Anything Model) | Analysis | Automated segmentation of subcutaneous implants in Dixon water images for volumetric analysis. | Not Listed | #133 |
| In-house Python pipeline | Analysis | Reproduce NIST reference T2 data from the ISMRM/NIST phantom and analyze the influence of echo-time range and spatial resolution on T2 accuracy at 3T. | Not Listed | #150 |
| CRAFT (Complete Reduction to Amplitude-Frequency Table) | Analysis | Quantifies fatty acid composition from NMR and MRS data by analyzing FID signals as the sum of decaying sinusoids using Bayesian probability theory, calculating resonance frequencies and amplitudes. | Not Listed | #153 |
| DeepBM | Simulation | Generates two-pool z-spectra from tissue parameters using a four-layer fully connected network to accelerate the training process of a deep learning framework for NEMR fitting. | Not Listed | #178 |
| PyRadiomics | Analysis | To extract histogram features (mean, percentiles, entropy, skewness, kurtosis, uniformity, etc.) from medical images. | Not Listed | #181 |
| RealKeyMorph (RKM) | Reconstruction | A keypoint-based, resolution-agnostic image registration framework that aligns MR images with varying resolutions without resampling, designed to serve as a preprocessing tool for multi-stack reconstruction and longitudinal analysis. | Not Listed | #189 |
| MR-MOTUS (modified) | Reconstruction | Adapts MR-MOTUS to map MIMO B$^+$PT signals directly to non-rigid motion fields by calibrating spatial modes and inferring motion from B$^+$PT signals alone. | Not Listed | #192 |
| LMBS-INR | Reconstruction | A registration framework combining multimodal landmark matching with B-spline parameterized implicit neural representations for DWI distortion correction. | Not Listed | #193 |
| SynthSeg$^+$ | Segmentation | Automated tissue and subcortical segmentation on each age-group template. | Not Listed | #194 |
| Joint Synthesis and Registration Framework | Reconstruction | To synthesize T1w-like images from dMRI data and register them to real T1w images, jointly optimizing both synthesis and registration processes for improved cross-modal registration. | Not Listed | #195 |
| Bio-Pose-Net | Analysis | Predicts 17 anatomical landmarks to verify patient orientation and posture. | Not Listed | #196 |
| PATHmaps | Analysis | Visualize and quantify deviations of aortic anatomy in thoracic aorta disease (TAD) from normative aorta atlases. | Not Listed | #197 |
| Rep2Rep-Shuffle | Analysis | A noise-adaptive self-supervised denoising strategy that improves SNR by introducing shuffled repetition pairing and 3D spatial modeling within a noise-adaptive framework to improve training diversity and denoising robustness. | Not Listed | #199 |
| NUNet | Reconstruction | Denoise high-resolution 3D UTE-MR angiograms while preserving vessel details. | Not Listed | #202 |
| KLEAN | Analysis | A generalized acquisition- and reconstruction-agnostic denoising framework that operates directly on complex-valued k-space data before image reconstruction. | Not Listed | #203 |
| PCMRI | Simulation | PCMRI is a Prompt-Controlled MRI synthesis model that leverages latent diffusion models for text-to-image MRI synthesis, enabling parameter-conditioned generation of anatomically consistent and physically plausible multi-contrast MRI data. | Not Listed | #204 |
| Residual U-Net | Analysis | Suppress MRI-induced interference in EEG signals acquired during simultaneous sodium MRI–EEG acquisition using a deep encoder-decoder network with residual and skip connections. | Not Listed | #205 |
| SACRED | Reconstruction | Corrects geometric distortions in EPI images using deep learning, requiring only T1 and forward phase-encoding BOLD images. | Not Listed | #208 |
| Custom implementations (for zipper and Rician noise artifacts) | Simulation | To simulate zipper and Rician noise artifacts for MR image data augmentation. | Not Listed | #210 |
| FieldNet | Reconstruction | Estimates blade-specific field maps for distortion correction in DW-PROPELLER-EPI using a differentiable off-resonance model and blade-consistency training. | Not Listed | #213 |
| Multi-average joint Diff-PnP reconstruction | Reconstruction | Reconstructs DWI images from multi-average k-space data within a single reverse diffusion process, accelerating the reconstruction process and improving image quality. | Not Listed | #214 |
| LOFT BBB toolbox | Analysis | Calculate cerebral blood flow (CBF), BBB water-exchange rate (kw) and arterial transit time (ATT) maps from diffusion-prepared (DP) pCASL sequence. | Not Listed | #217 |
| STI Suite | Analysis | Process complex images from the second echo to generate both QSM and corresponding phase masks for venous imaging. | Not Listed | #220 |
| physics-informed neural network | Analysis | Synthesizes contrast agent-free DSC parameters and novel microstructural biomarkers from IVIM and conventional MR sequences using a physics-informed neural network. | Not Listed | #228 |
| GTF-Net | Analysis | Parametrically identifies magnetic resonance gradient systems by combining a rational-function linear dynamic model with a neural network to capture static nonlinearities, enabling accurate gradient prediction and pre-emphasis for real-time non-Cartesian MRI. | Not Listed | #233 |
| PyPulseq | Pulse Sequence Design | Generate MRI pulse sequences with a focus on finite raster times and interpolation of arbitrary waveforms for GIRF estimation. | Not Listed | #235 |
| BART toolbox | Reconstruction | Compressed sensing reconstruction with wavelet regularization. | Not Listed | #237 |
| matMRI toolbox | Reconstruction | Reconstructs images using an iterative model-based pipeline, including static off-resonance and coil sensitivity maps. | Not Listed | #238 |
| bfieldtools | Simulation | Framework used to compute the surface current distribution and optimize the current density pattern on the 3D head surface for the prefrontal shim coil design. | Not Listed | #240 |
| 3D Polynomial Phase Fitting | Reconstruction | Corrects for odd-even phase errors in simultaneous multi-slice EPI by estimating phase information via 3D polynomial fitting, leading to artifact reduction. | Not Listed | #243 |
| FEM tool | Simulation | A FEM-based simulation tool developed to guide the redesign of metallic components in low-field MRI systems, minimizing eddy currents while preserving RF shielding and structural integrity. | Not Listed | #244 |
| MRIReco.jl | Reconstruction | Image reconstruction with static and up to 3rd order dynamic field correction for 3D EPI data, formulated as a regularized least-squares optimization problem. | Not Listed | #247 |
| Transformer Signature / Combined Signature | Analysis | Predict pathological complete response (pCR) in breast cancer patients by fusing radiomics and clinical features using a Transformer-based model. | Not Listed | #248 |
| SDUM | Reconstruction | A universal deep-unrolled model (SDUM) for MRI reconstruction designed to operate across diverse clinical protocols by combining a Restormer reconstructor, learned coil-sensitivity estimation, and sampling-aware data consistency with universal conditioning and progressive cascade expansion. | Not Listed | #249 |
| Deep-DSP | Analysis | EMI elimination is performed after each scan | Not Listed | #251 |
| SpectrumMAE | Reconstruction | A self-supervised masked autoencoder model to reconstruct high-resolution 1H-MRSI data from low-resolution acquisitions for super-resolution of gliomas. | Not Listed | #252 |
| EGR-Net | Analysis | Predicts future brain connectivity from previously observed connections by combining a dual-branch graph encoder and a GRU-based recurrent mapper to capture spatial and temporal dependencies in brain networks. | Not Listed | #254 |
| Integrated Model | Analysis | To predict response to neoadjuvant endocrine therapy (NET) in ER+/HER2- breast cancer by integrating volumetric and kinetic features extracted from DCE-MRI tumor subregions using a deep learning model with a volume-proportion-weighted pooling mechanism. | Not Listed | #259 |
| SuStaIn | Analysis | Infer trajectories of cerebellar subregion volume reduction in schizophrenia using structural MRI data. | Not Listed | #292 |
| PLINK2 | Analysis | Quality control and polygenic score calculation for genetic data. | Not Listed | #293 |
| QALAS | Analysis | Used to derive R₂ and R₂* maps from ME-SE data, which were then used for χ-separation, resulting in paramagnetic (χpara) and diamagnetic ( | χdia | ) maps. |
| SpinDoctor | Simulation | To calculate the evolution of nuclear magnetization inside digitalized cells with assumed impermeable membranes. | Not Listed | #324 |
| in-house developed “flattening” approach | Analysis | To enhance visualisation of the TRN by flattening the TRN curvature slice-by-slice, creating a rectangular image of the “flattened” thalamic periphery. | Not Listed | #325 |
| TotalSegmentator | Reconstruction | Fine-tunes TotalSegmentator for automatic myocardium segmentation in dark-blood T2* MR images to enhance diagnostic accuracy for iron overload assessment. | Not Listed | #328 |
| nnU-Net | Analysis | To segment the aorta on a diastolic two-point Dixon MR angiography (MRA) scan. | Not Listed | #331 |
| Anatomy-aware segmentation framework | Reconstruction | Learns a shared vascular shape space and adapts to target-site images using unlabeled data for robust cross-center vessel lumen and wall segmentation on intracranial vessel wall MRI. | Not Listed | #334 |
| nnU-Net | Segmentation | Deep learning framework used for myocardial and scar segmentation from LGE images, leveraging a 2D U-Net architecture. | Not Listed | #335 |
| nnU-Net | Analysis | To train a neural network based on the nnU-Net framework for automated segmentation of multi-region 4D flow MRI data, including cardiovascular, cerebrovascular, and portal vein regions. | Not Listed | #337 |
| nnU-net | Segmentation | Automated myocardial segmentation in cardiac cine MRI images. | Not Listed | #338 |
| Image2Flow | Analysis | Generates a 3D pulmonary artery mesh with time-varying flow fields from a single 3D cardiac MRI input. | Not Listed | #346 |
| 3D q-aMRI pipeline | Analysis | Process displacement data for quantitative amplified MRI (q-aMRI) by amplifying data within 0.1–4 Hz, applying rigid registration and SynthSeg segmentation to assess region of interest-based displacement differences. | Not Listed | #348 |
| FlowMRI-Net | Reconstruction | Reconstructs 5D flow MRI k-space data from respiratory-resolved acquisitions using a deep learning approach. | Not Listed | #355 |
| UniConnNet | Analysis | Classifies carotid plaques as stable or unstable based on multi-contrast MRI data by leveraging multiple UNet architectures connected through Uniform Connections and a Multi Hypergraph Dynamic Node Framework. | Not Listed | #361 |
| Prototype Application | Analysis | Application for real-time, in-line monitoring of cine cardiac MRI image quality using a trained ResNet-18 CNN model to identify poor-quality slices and facilitate rescanning. | Not Listed | #365 |
| Automated framework | Analysis | Automated water-fat classification in musculoskeletal MRI by integrating adaptive multi-joint region-of-interest (ROI) extraction with 3D consistency voting. | Not Listed | #369 |
| Random Forest Model | Analysis | To classify renal interstitial fibrosis (RIF) severity using multiparametric MRI and clinical biomarkers. | Not Listed | #372 |
| AutoMAC-MRI | Analysis | An automated, explainable AI method to detect and grade motion severity in MR images by using supervised contrastive learning to generate distinct motion-grade embeddings and computing motion grade severity scores by measuring similarity between image embeddings and grade-specific templates for interpretable assessment. | Not Listed | #376 |
| Bi-Directional xLSTM-UNet | Analysis | A foundation model with a Bi-Directional xLSTM-UNet backbone was pretrained using self-supervised masked autoencoding on a large dataset of MRI scans for neuroimaging tasks such as neurodegenerative disease classification, brain-age prediction, and whole-brain segmentation. | Not Listed | #382 |
| 3D U-Net model | Segmentation | Segment PitNET lesions in MRI images, providing a region of interest (ROI) encompassing the entire tumor volume. | Not Listed | #383 |
| Chi-separation toolbox | Reconstruction | Separate paramagnetic (iron) and diamagnetic (myelin) susceptibility components, generating both paramagnetic and diamagnetic susceptibility maps. | Not Listed | #385 |
| Curvanato | Analysis | Compute morphometric scores, including volume, for caudate and putamen segmentations. | https://github.com/stnava/curvanato | #386 |
| PINGU | Reconstruction | To segment perivascular spaces (PVS) on 3D-T1 images. | Not Listed | #387 |
| Smoldyn MR module | Simulation | Fits time series of 13C-metabolites from HP 13C MRI brain scans using particle-based numerical modelling of enzyme reactions in multiple compartments with differing redox conditions. | Not Listed | #390 |
| U-Net | Analysis | To estimate regional pyruvate-to-lactate (kPL) and pyruvate-to-bicarbonate (kPB) conversion rate maps from dynamic HP 13C cardiac data using a 2D U-Net model trained on digital phantoms. | Not Listed | #395 |
| SVRTK | Reconstruction | Reconstructs isotropic 3D volumes from multiple 2D slice stacks by estimating motion between slices and solving a regularized inverse problem. | Not Listed | #407 |
| Dynamic-Causal ViT | Analysis | Model causal dependencies across timepoints using attention-based temporal encoding and quantify alignment between saliency evolution and actual tumor regression to predict chemoradiotherapy response. | Not Listed | #412 |
| End-to-end zero-shot SSL framework | Reconstruction / Analysis | Jointly reconstructs DW-images and estimates DTI-biomarkers directly from under-sampled 2D/3D-MUSE-DTI data using self-supervised learning. | Not Listed | #424 |
| dMRI-Lab toolbox | Analysis | Estimate free water (FW)-DTI metrics and FW volume fraction (FWVF) with spherical means. | Not Listed | #425 |
| mcPVS-Net | Analysis | Detect perivascular spaces (PVS) based on 3D T1-weighted structural images using a fully automated deep learning algorithm. | Not Listed | #426 |
| INDI | Analysis | Generate parametric maps for mean diffusivity (MD), fractional anisotropy (FA), helix angle (HA), and absolute secondary eigenvector angle (E2A) from cardiac diffusion tensor imaging (cDTI) data. | https://github.com/ImperialCollegeLondon/INDI | #428 |
| Dual-ADC gated U-Net | Reconstruction | Segment cytotoxic and combined edema lesions on DWI using ADC thresholds. | Not Listed | #429 |
| MRpro | Reconstruction | Implements operators and algorithms used for MR image reconstruction, including the proposed method for magnitude-phase reconstruction in MR Elastography. | Not Listed | #437 |
| UNITS | Reconstruction | A unified theoretical framework for self-supervised MRI reconstruction that proves the expected performance equivalence between self-supervised and supervised MRI reconstruction, and introduces sampling stochasticity and flexible data utilization to improve robustness and generalization. | Not Listed | #438 |
| ReconHMT | Reconstruction | Reconstructs MR images within the hierarchical latent space of a foundational autoencoder model by leveraging adapters to enforce token-level coherence and data-consistency across hierarchical stages. | Not Listed | #441 |
| spiral-DEFT | Reconstruction | A hybrid 2D/3D model-based reconstruction framework that jointly corrects off-resonance and chemical-shift blurring with Fat/Water separation while suppressing undersampling noise and artifacts. | Not Listed | #442 |
| HiFi-QSM | Reconstruction | Reconstructs a high-resolution QSM map by integrating high-frequency information from a high-resolution field map with low-frequency information from a low-resolution QSM map using a deep learning framework. | Not Listed | #443 |
| PyGrog | Reconstruction | Performs GROG interpolation using fractional GRAPPA operators to grid non-Cartesian k-space data onto a Cartesian grid, enabling efficient FFT-based reconstructions. | Not Listed | #444 |
| MoCo-SToRM | Reconstruction | Learns a continuous motion manifold and patient-specific latent vectors from stack-of-stars data for real-time 3D motion estimation in MRI-guided radiotherapy. | Not Listed | #446 |
| Joint binning, reconstruction, and fitting framework | Reconstruction | Simultaneously models multi-dose data using a joint Bloch signal model and shared spatial regularization for improved quantitative accuracy in multi-dose contrast-enhanced MRI. | Not Listed | #447 |
| DeepGuidedGS | Reconstruction | A reconstruction framework incorporating prior-scan information with robustness to imperfect or unreliable guidance by comparing latent-coefficients from a reconstruction target with those from guide images. | Not Listed | #448 |
| MoDL framework | Reconstruction | Accelerates the reconstruction of self-navigation in simultaneous multi-slab 3D DWI by replacing the time-consuming structured low-rank (SLR) reconstruction with a U-Net regularized model-based deep learning framework. | Not Listed | #449 |
| Diffusion model-based deep learning reconstruction | Reconstruction | Enhance image quality and evaluate the impact of radial undersampling on image quality and scan efficiency in UTE MRI. | Not Listed | #450 |
| iQSM+ | Reconstruction | Reconstructs QSM images from MRI phase data using a deep learning architecture that incorporates orientation-adaptive latent feature editing for improved robustness and image quality. | Not Listed | #452 |
| DOTCOPS-based gradient scheme optimization | Pulse Sequence Design | Optimizes crusher gradient design for edited MRS by integrating a volume-based likelihood model and b-value regularization to suppress out-of-voxel artifacts while maintaining signal integrity. | Not Listed | #457 |
| FIRE Pipeline | Reconstruction | Comprehensive online spectroscopy pipeline for the acquisition, processing, and tissue-corrected metabolite quantification of localized MEGA-PRESS and SPECIAL MRS acquisitions, performed entirely on the scanner. | https://github.com/aTrotier/ismrmrd_to_nifti | #460 |
| LIPNET | Analysis | Deep learning network for predicting and removing the lipid component from each voxel’s spectrum in MRSI. | Not Listed | #462 |
| Spinal Cord Toolbox (SCT) | Analysis | Process spinal cord MRI data for cord segmentation and vertebral labeling to extract morphometric measurements like CSA, AP, and RL widths. | Not Listed | #474 |
| BCT | Analysis | Perform graph theory analysis to evaluate global and nodal metrics of brain networks. | Not Listed | #485 |
| DMPLF | Analysis | Improves quantification accuracy of GuanCEST by first fitting strong background signals (MT, rNOE and direct water saturation) followed by fitting APT and GuanCEST peaks. | Not Listed | #490 |
| TWE-NITI | Reconstruction | Reconstruct anisotropic biomechanical parameters in human myocardium from multifrequency MRE data using a traveling wave expansion (TWE) model integrated with nearly incompressible transversely isotropic (NITI) inversion. | Not Listed | #496 |
| QSMnet | Reconstruction | Reconstruct quantitative susceptibility mapping (QSM) maps using deep learning. | Not Listed | #497 |
| In-house postprocessing tool | Analysis | Place planes orthogonal to vessels of interest for flow measurements. | Not Listed | #502 |
| OSPREY | Analysis | Processed MRS data for metabolite quantification. | Not Listed | #514 |
| FID-A software | Analysis | Process MRS data, calculate SNR as the ratio of the 2.0 ppm NAA peak amplitude to the standard deviation of the noise ([0, –2] ppm), and measure LW as the full-width at half-maximum of the NAA peak. | Not Listed | #517 |
| NVnmrLib | Analysis | Process MRI data and compute pyruvate, lactate, and bicarbonate AUCs over time. | Not Listed | #530 |
| BART-based simulation framework | Simulation | Simulates dynamic 13-C metabolic imaging for arbitrary phase encoding schemes, enabling realistic pyruvate-lactate exchange studies. | Not Listed | #536 |
| Automated biometry pipeline | Analysis | Automatically extract maternal pelvis and fetal head metrics from 3D T2-weighted fetal MRI using two MONAI-based 3D UNet models. | Not Listed | #540 |
| Custom time- and location-clustering method | Analysis | Collect catheter trajectories and provide dose-maps at 5 fps for real-time MRT catheter dosimetry. | Not Listed | #542 |
| CervFiT-AG3DNet | Analysis | To segment cervical cancer from time-series dynamic contrast enhanced MRI (DCE-MRI) using a deep learning model based on a 3D U-Net architecture with Res2Net block, Time-Intensity Curve (TIC), Feature-Wise Linear Modulation (FiLM), and attention gates (AGs). | Not Listed | #544 |
| RABIES | Preprocessing | Preprocessing fMRI data with frame-to-frame displacement scrubbing. | Not Listed | #557 |
| NORDIC | Analysis | Preprocessing of high-resolution BOLD-fMRI data. | Not Listed | #558 |
| NBS toolbox | Analysis | Implemented structural covariance analysis to identify brain regions with correlated volume changes across subjects. | Not Listed | #561 |
| SMART-Risk | Analysis | A radiomics model that integrates spatial organization, local heterogeneity, second-order texture, and morphology features to differentiate tumor recurrence vs pseudoprogression in brain tumors. | Not Listed | #571 |
| NeuroMamba | Analysis | Predicts behavior scores from resting-state fMRI data by leveraging temporal dynamics using a deep state space model based on the Mamba block. | Not Listed | #574 |
| Transformer-based multimodal framework | Analysis | Fuses disease trajectory embeddings with imaging-derived phenotypes using a Transformer architecture for improved disease risk prediction. | Not Listed | #577 |
| Bloch simulator | Simulation | Simulates Bloch equations using a 4th-order Runge-Kutta solver to investigate the behavior of the AM_HS1 pulse. | Not Listed | #587 |
| dynamic compressed sensing (CS) reconstruction | Reconstruction | Reconstruct high spatiotemporal resolution images with motion-resolved self-navigation from dynamic frames of spiral shots acquired using a rotated golden-angle stack-of-spirals sequence. | Not Listed | #595 |
| NORDIC | Analysis | Thermal denoising of fMRI data. | github.com/SteenMoeller/NORDIC_Raw | #596 |
| BrainMLSR | Reconstruction | Reconstructs multi-signal-layer cortical surfaces from 5.0T T2-weighted FLAIR MRI using an energy function-based framework. | Not Listed | #598 |
| LayNii | Analysis | To divide the cortex into 10 equidistant layers from WM to CSF for voxel-based depth-dependent profiling. | Not Listed | #601 |
| POPE | Pulse Sequence Design | Modifies EPI pulse shapes to selectively reduce slew rate at periods of high predicted PNS spikes, allowing for higher average slew rate and larger gradient moments, enabling robust high-resolution fMRI. | https://github.com/layerfMRI/Sequence_Github/blob/master/POPE/FRISGO_20251008_AZM_AA.pdf | #604 |
| DSpiral-RIM | Reconstruction | Accelerated spiral fMRI reconstruction by learning temporal dependencies across volumes to improve reconstruction fidelity at high acceleration and reduce processing time. | Not Listed | #605 |
| ACID toolbox | Analysis | Preprocesses and analyzes MRI data, including registration of b=0 and diffusion-weighted volumes, detrending BOLD fMRI data, fitting eddy current distortion parameters, and calculating DTI parameters and SNR. | Not Listed | #606 |
| RetroMocoBox | Reconstruction | Motion correction was applied on raw data using the RetroMocoBox toolbox to improve image sharpness. | Not Listed | #609 |
| UKRIN Kidney Analysis Toolbox (UKAT) | Analysis | Calculate quantitative MRI maps of the pancreas. | Not Listed | #612 |
| MRI-PancAge | Analysis | Estimates biological pancreas age from MRI-derived volumetric pancreas segmentation masks and predicts type 2 diabetes. | Not Listed | #615 |
| 3D U-Net-based deep learning model | Analysis | Vessel segmentation in liver MRI images to enable pure parenchyma extraction for liver function computation. | Not Listed | #622 |
| Spinal Cord Toolbox | Analysis | Segment the spinal cord, vertebral labeling, and lesions in spinal cord MRI images. | Not Listed | #630 |
| VFA-PETRA | Reconstruction | Enables efficient T1 mapping, even for tissues with extremely short T2, by combining PETRA with a variable flip angle strategy and prior knowledge of B0 and B1. | Not Listed | #632 |
| SimpleITK | Analysis | Rigid affine registration to align each individual average and diffusion-encoding direction to the T2-weighted HASTE images. | Not Listed | #633 |
| ITK toolkit | Analysis | Motion correct and register 3D images with different TEs to one reference set. | Not Listed | #638 |
| STRIFE | Analysis | STRIFE models how fixel-wise effects arise from underlying bundle-level changes, enabling bundle-wise effects estimates. | Not Listed | #643 |
| Custom triple-echo SMS DWI acquisition | Reconstruction | Enables dynamic, slice-wise field estimation for robust recovery of diffusion data under severe head motion and rapidly varying distortion fields. | Not Listed | #644 |
| Index of Tumor Invasiveness (ITI) | Analysis | Predicts patient-specific directional microscopic infiltration of glioblastoma to support personalized treatment by quantifying infiltration likelihood along fiber pathways using diffusion MRI tractography and kurtosis integration. | Not Listed | #651 |
| 1D-ResNet Autoencoder | Reconstruction | Suppresses residual water, ghosting, lipid contamination, and gradient modulation sidebands in in-vivo MRSI data using a dual-channel residual convolutional deep-autoencoder. | Not Listed | #661 |
| GIFT toolbox | Analysis | Computes independent component analysis (ICA) to generate maps of network connectivity from resting-state fMRI data. | mialab.mrn.org/software/gift/ | #670 |
| FuncTool | Reconstruction | Process CBF data based on the general kinetic model to obtain arterial transit time-corrected CBF maps. | Not Listed | #676 |
| nnU-Net | Segmentation | Automatically segment whole tumor volumes in MRI images. | Not Listed | #690 |
| uAI Research Portal | Analysis | Automated whole-tumor segmentation and feature extraction for radiomics analysis. | Not Listed | #692 |
| RADIATE-Net | Analysis | Predicts glioma treatment response using a radiomics-guided self-supervised 3D CNN with soft-label distillation from handcrafted radiomic features to leverage unlabeled data for knowledge transfer. | Not Listed | #693 |
| EP-vpMT | Pulse Sequence Design | To accelerate magnetization transfer (MT) imaging using a 3D segmented echo-planar variable-power MT sequence that reduces SAR and accelerates acquisition without compromising MT contrast. | Not Listed | #713 |
| Swiss Knife toolbox | Analysis | Fit biophysical models (ROS, ROC, and S+ROS) to dMRS data to extract microstructural parameters. | Not Listed | #753 |
| FlowINR | Reconstruction | Learns a probability distribution over a field of continuous functions mapping voxel coordinates to spherical harmonic representations of the dMRI signal, characterizing aleatoric uncertainty due to noise. | Not Listed | #754 |
| Regularized Iterative Reconstruction | Reconstruction | Estimates ADC and T2 maps from Phase-Based Diffusion MRI data by iteratively minimizing a composite cost function with TV and L2 regularization. | Not Listed | #755 |
| HARP | Analysis | A deep learning-based framework for harmonizing diffusion MRI data across different scanners using a 1D neural network trained solely on phantom data. | Not Listed | #756 |
| Surface-informed CSF correction framework | Reconstruction | Corrects for CSF contamination in cortical diffusion MRI by modeling and removing CSF spill-in to cortical voxels using surface information and an acquisition-derived point-spread function (PSF). | Not Listed | #757 |
| GAIA | Analysis | Uses knowledge distillation to compress large neural networks into smaller models for predicting high b-value diffusion MRI signals from lower b-values, reducing computational and energy costs. | Not Listed | #758 |
| DESIGNER | Pre-processing | Used for pre-processing of dMRI data with Maxwell-compensated free-gradient waveforms. | Not Listed | #759 |
| FlowVN | Reconstruction | Accurately reconstruct highly undersampled 4D Flow MRI phase-based velocity data and TKE data, enabling substantial scan-time reduction while maintaining accurate assessment of turbulence-sensitive hemodynamic parameters. | Not Listed | #765 |
| APT 4D FLOW | Analysis | A user-friendly and high-speed toolbox for 4D FLOW volumetric analysis, providing wall stress and peak velocity measurements comparable to cvi42 while enabling volumetric analysis across the entire vessel. | Not Listed | #767 |
| Dedicated MATLAB toolbox | Analysis | Standardized post-processing pipeline for background phase correction, semi-automatic segmentation of iliac arteries and veins, and 3D finite-element mesh generation for robust quantification of hemodynamic parameters. | Not Listed | #775 |
| HippUnfold | Analysis | Hippocampal subfield segmentation and surface-based hippocampal reconstruction. | Not Listed | #790 |
| AutoMetric | Analysis | Generate structured, quantitative MRI reports by integrating images, segmentation-derived metrics, and clinical data using LLaVA-Med. | Not Listed | #805 |
| MMRQA | Analysis | MMRQA integrates acquisition-aware signal metrics with multimodal large language models (MLLMs) to deliver clinically interpretable MRI quality reports with state-of-the-art performance and zero-shot generalization. | Not Listed | #806 |
| ScarNet-DPO | Analysis | A semi-supervised foundation model to achieve clinically deployable accuracy in left ventricular scar quantification while reducing annotation burden. | Not Listed | #807 |
| BboxHarmony | Analysis | Harmonizes MRI images to improve the performance of black-box deep learning models on data from unseen domains by leveraging disentanglement-based generator and Bayesian optimization. | Not Listed | #822 |
| Attention-based adaptive lag model | Analysis | Train a model to predict brain activity patterns from time-lagged visual stimuli by adaptively estimating optimal temporal lags between stimuli and fMRI responses using attention scores. | Not Listed | #823 |
| GDAR | Analysis | Capture and analyze the dynamic, directional information flow across brain regions from rs-fMRI data by employing a graph diffusion autoregressive model and integrating spatial information derived from diffusion tractography. | Not Listed | #824 |
| Multi-task diffusion framework | Reconstruction | Simultaneously generates virtual LGE images and automatically segments key structures directly from routine Cine and T2 STIR sequences using a diffusion model with feature-disentangled attention and cross-fusion modules. | Not Listed | #825 |
| CRaLDM | Simulation | Generates high-fidelity synthetic CT images from MRI by aligning MRI latents with radiomic embeddings using a contrastive radiomics-aligned latent diffusion model. | Not Listed | #826 |
| Denoising Diffusion Probabilistic Model (DDPM) | Reconstruction | To map weighted image series from a 3D Silent Multiparametric Mapping with Zero Echo Time (MuPa-ZTE) acquisition to quantitative MRI (qMRI) maps while incorporating a data consistency constraint during inference. | Not Listed | #827 |
| CineGen | Reconstruction | A conditional flow-matching (FM) model integrated within the Siemens FIRE framework to enable inline real-time cine MRI super-resolution inference. | Not Listed | #828 |
| Flowdiff | Reconstruction | A diffusion-based frame interpolation framework that integrates image prior information into the generation process to generate anatomically consistent, high-fidelity intermediate medical frames at arbitrary time positions. | Not Listed | #829 |
| ShapeWorks | Analysis | Establish spatial correspondences between cortical and cranial surfaces across subjects using particle optimization. | Not Listed | #836 |
| CPAC | Analysis | Preprocessing of human fMRI data using the CPAC pipeline, including alignment to the Brainnetome atlas. | Not Listed | #850 |
| RABIES | Preprocessing | Data preprocessing for fMRI data. | Not Listed | #851 |
| MaRCoS firmware | Pulse Sequence Design | Modified firmware to enable synchronized multichannel operation across multiple Red Pitaya boards for scalable, low-noise, multichannel MRI, while maintaining compatibility with the existing pulse-sequence framework. | Not Listed | #867 |
| Rev. D | Pulse Sequence Design | An open-source FPGA-based control and amplifier system capable of driving 4A/ch currents with scalability up to 64-ch for driving multi-coil arrays in MRI. | https://github.com/ | #870 |
| Custom Quadrature Digitizer | Data Management | A custom quadrature digitizer chain employing high dynamic range ADCs, with WiFi /Bluetooth interfacing, and the option to employ mm-wave modules or quadrature demodulators for arbitrary frequency sensing, with the aim to rapidly prototype radar sensing for MRI. | Not Listed | #873 |
| YOLOv11 | Analysis | A customized pupil and eye corner detection model was trained using YOLOv11 to ensure reliable tracking under varying conditions of pupil appearance due to IR illumination. | Not Listed | #874 |
| Walsh-Hadamard encoding reconstruction framework | Reconstruction | Enables high-accuracy MPI 3D reconstruction by using a 3D Walsh-Hadamard orthogonal coding and multi-harmonic fusion framework to decouple spatial information. | Not Listed | #880 |
| None | Analysis | To reconstruct diagnostic 12-lead ECG signals from compact wireless sensors operating inside the MRI bore using a calibrated regression model. | Not Listed | #884 |
| FM-driven multiomics model | Analysis | To predict platinum resistance in HGSOC patients by integrating clinical, MRI, and pathological data using pre-trained foundation models. | Not Listed | #889 |
| EPEN | Reconstruction | Regularized slab-combination reconstruction that incorporates learned priors, implemented using convolutional neural networks (CNNs) within a maximum-a-posteriori (MAP) framework, to suppress noise-like slab-boundary artifacts in 3D multi-slab dMRI. | Not Listed | #896 |
| RBFNet | Reconstruction | Reconstructs undersampled radial diffusion spectrum imaging (RDSI) data by learning the q-space representation from acquired samples using a Radial Basis Function Network. | Not Listed | #898 |
| V-NET with adaptive multi-scale attention | Reconstruction | Reconstructs voxel-level fiber orientation distribution functions (fODFs) from single-shell diffusion MRI data using a lightweight V-NET architecture with adaptive multi-scale attention. | Not Listed | #902 |
| FORCE | Analysis | Estimates microstructure and fiber orientations directly from standard dMRI data by matching measured diffusion signals with simulated signals in a forward modeling framework. | Not Listed | #903 |
| Noise-aware dMRI denoiser | Reconstruction | Denoise diffusion MRI data at the coil level, adapting to scanner- and coil-specific noise properties by conditioning on noise statistics estimated from a rapid noise scan. | Not Listed | #907 |
| INR framework | Reconstruction | Denoise pancreatic DWI by embedding a physically constrained diffusion model to suppress noise while preserving diffusion contrast and signal fidelity. | Not Listed | #909 |
| CoSimMAT | Simulation | A custom-built MATLAB toolbox for RF circuit co-simulation, optimizing lumped element values for tuning, matching, and decoupling in RF coil design. | Not Listed | #923 |
| UKRIN-Kidney-Analysis-Toolbox (UKAT) | Analysis | To provide a set of Python scripts for automated analysis of renal MRI data, including computation of B0 and B1 maps, kidney segmentation, T1 and T2 mapping, and DWI analysis. | https://github.com/UKRIN-MAPS/ukat | #928 |
| None | Reconstruction | Jointly estimates characteristic component spectra and their voxelwise abundances by exploiting signal correlations across many voxels in multidimensional contrast-encoded MRI (mdMRI). | Not Listed | #936 |
| T1D solver | Analysis | Estimates T1D map from Rdosl using nonlinear least-squares fitting (Levenberg-Marquardt algorithm) or the dictionary method, with MPF maps as input. | Not Listed | #937 |
| JuSpace toolbox | Analysis | Conducts spatial correlation analysis between neuroimaging data and spatial maps (e.g., neurotransmitter receptor densities). | Not Listed | #944 |
| DESIGNER | Reconstruction | Preprocesses diffusion MRI images. | Not Listed | #956 |
| hMRI toolbox | Analysis | To compute quantitative maps of proton density (PD), longitudinal relaxation rate (R1), and magnetization transfer saturation (MTsat) and correct for B1 inhomogeneities. | Not Listed | #964 |
| off-resonance-EPG-MT | Simulation | Extends the EPG framework to model off-resonance RF pulses in a two-pool magnetization transfer system, enabling rapid exploration of sequence and physiological parameters. | https://github.com/June-phy/off-resonance-EPG-MT | #968 |
| DL-MorphoBox | Analysis | Derives a brain segmentation mask from MP2RAGE images and co-registers it to the T2* maps for T2* analysis. | Not Listed | #972 |
| MOCHA | Analysis | Manually segment vessel lumen and outer wall boundary in multi-contrast, multi-planar framework for quantitative intracranial vessel wall (IVW) analysis. | Not Listed | #980 |
| VWI-RAP | Reconstruction | Accelerates intracranial vessel wall imaging while maintaining diagnostic quality by integrating k-space–guided recovery and anatomically weighted refinement. | Not Listed | #986 |
| MERIT | Analysis | Segments hyperintensity in the substantia nigra (SN) using a vision transformer model after training and validation on ground truth expert manual labels. | Not Listed | #992 |
| DIPY | Analysis | Generates diffusion maps of Fractional anisotropy (FA) and mean diffusivity (MD). | Not Listed | #997 |
| Osprey | Analysis | Process MRS data for quantitative assessment of brain metabolites, including eddy-current correction, frequency-and-phase alignment, water removal, frequency referencing, and fitting to the LC model basis set, as well as segmenting tissue compartments for metabolic concentration extraction. | Not Listed | #998 |
| U-Net | Reconstruction | Segment choroid plexus volume (CPV) from MRI images. | Not Listed | #999 |
| MG-CL | Analysis | Fuses modality-specific task-relevant embeddings via parallel GCN branches for FC and SC, while learning adaptive coupling weights to capture complementary relationships for Parkinson’s disease diagnosis. | Not Listed | #1001 |
| MPM2QSM | Analysis | Predict quantitative susceptibility maps (QSM) directly from MPM-magnitude-derived parameter maps (R1, R2*, proton density, and magnetisation transfer maps) using a U-Net model. | Not Listed | #1006 |
| TMT-net | Reconstruction | MRI translation network that leverages disentanglement architecture and metadata describing image contrast and scan parameters, enabling translation of input scan image to image that corresponds to target scan parameters. | Not Listed | #1007 |
| Graph-Attention Fusion Framework | Analysis | This framework integrates a graph-based GNN with flexible degree-adaptive aggregation and an image restoration Transformer leveraging pixel-level self-attention for global context to achieve high-quality brain MRI synthesis, specifically for motion artifact removal and cross-modal translation. | Not Listed | #1008 |
| MSSC-cGAN (Multiscale Subtraction Consistency-based conditional GAN) | Simulation | Synthesizes multi-phase DCE-MRI images directly from non-contrast MRI inputs (T1w, DWI, and ADC maps). | Not Listed | #1009 |
| PIPE | Analysis | Incorporates spatial smoothness constraints into a parametric mapping framework based on phase-cycled bSSFP to improve robustness against undersampling noise and yield more accurate quantitative maps. | Not Listed | #1013 |
| Physics-guided Neural Network | Analysis | Estimates T1, T2, Meff, and ∆f parameters from bSSFP data using a physics-guided neural network trained on simulated data. | Not Listed | #1016 |
| REFINE-MORE | Reconstruction | A scan-specific framework that integrates implicit neural representation with MR physics-based reinforcement for accurate, reference-free multiparametric qMRI reconstruction. | Not Listed | #1017 |
| CV-MRF | Reconstruction | Simultaneously estimates reversible transverse relaxation rate R2*, susceptibility-induced magnetic field perturbations Δf, and transceive phase offset ϕ0 from complex multi-echo gradient-echo data using a dictionary matching technique. | Not Listed | #1018 |
| MC-ZSSSL | Reconstruction | Reconstructs multiple contrast images jointly to achieve higher acceleration factors in T1 mapping while preserving spatial resolution and quantitative accuracy. | Not Listed | #1019 |
| SSL-MIMOSA | Reconstruction | Estimates T1, T2, T2*, proton density (PD), and inversion efficiency (IE) maps from multi-contrast MIMOSA images via self-supervised learning. | Not Listed | #1020 |
| BART sequence framework | Pulse Sequence Design | Implements MRI sequences in BART with dynamic parameter adjustment online on the scanner and sequence export to pulseq-format, enabling joint design of sequences and reconstruction. | Not Listed | #1021 |
| Synthetic MRI | Simulation | A pipeline to synthesize physics-consistent images and matching multi-coil k-space from multi-parameter maps (MPM), enabling controlled parameter sweeps plus artifact/pathology insertion for broad research. | https://github.com/borankilic/synthetic_mri | #1023 |
| eduMRIsim | Simulation | To extend an open-source educational MRI simulator with realistically simulated artefacts and noise, allowing students to interactively explore how scan parameters affect image contrast, artefacts, and SNR. | eduMRIsim.github.io | #1024 |
| E2E-VN-MC (Multi-Channel) and E2E-VN-3D | Reconstruction | These variational networks reconstruct accelerated phase-cycled bSSFP MRI data by incorporating the phase-cycling dimension within their internal representation to improve image quality and T1/T2 mapping accuracy. | Not Listed | #1025 |
| Custom fused Triton GPU kernels | Reconstruction | Compute the MRF signal and its analytical gradient concurrently, bypassing PyTorch autodiff for faster gradient-based optimization. | Not Listed | #1026 |
| Certis Solutions | Reconstruction | Reconstruct temperature and thermal dose maps from segmented 2D gradient‑echo echoplanar imaging (EPI) data. | Not Listed | #1033 |
| RAFT framework (extended for feature- and contrast-aware motion estimation) | Analysis | Estimates motion fields from 2D Cine MR images for real-time abdominal landmark tracking using a self-supervised learning approach. | Not Listed | #1040 |
| DESIGNER | Data Management | Process diffusion-weighted MRI data. | Not Listed | #1044 |
| MaxEnt-DTD | Reconstruction | Estimate diffusion tensor distribution (DTD) to simultaneously estimate fiber orientation and microstructure using a maximum-entropy based convex optimization algorithm. | Not Listed | #1049 |
| DMIPY | Simulation | Simulates diffusion weighted signals for DTI, DKI and NODDI models in Python. | Not Listed | #1050 |
| PulSeq | Pulse Sequence Design | Used to implement 3D MP-RAGE and 3D MP-FISP sequences. | Not Listed | #1054 |
| neuroComBat | Analysis | Harmonizes functional and structural connectomes in multi-site MRI studies to mitigate scanner-related variability, modelling scanner vendor as a batch variable and adjusting for covariates. | Not Listed | #1061 |
| Matlab software package | Analysis | Adapted for CSF analysis from a cardiovascular 4D Flow quantification software package. | Not Listed | #1067 |
| SYRMEP Tomo Project (STP) | Reconstruction | Image reconstruction for X-ray phase contrast tomography using a phase retrieval algorithm. | Not Listed | #1074 |
| iQ-Cog | Analysis | Derives precise volumetric measures from T1-weighted and FLAIR MRI scans to improve differentiation of AD from cognitively normal (CN), and mild cognitive impairment (MCI) from CN. | Not Listed | #1075 |
| 3D conditional rectified flow model | Simulation | Generate subject-specific 3D MR images reflecting brain atrophy to predict Alzheimer’s disease progression. | Not Listed | #1076 |
| DL-MorphoBox | Analysis | Segments brain volumes into 55 separate regions and returns their absolute and relative volumes (normalized by the total intracranial volume), and the corresponding Z-score values derived from age- and sex-specific reference ranges. | Not Listed | #1077 |
| PDDF-Net | Analysis | Diagnoses Parkinson’s Disease using QSM and T1w images by training a deep classification network on 2D slices extracted from 3D volumes and refining the network using a multiple-instance-learning-based strategy. | Not Listed | #1078 |
| MIND | Analysis | Calculates structural divergence between brain regions based on statistical comparison of multiple morphometric features to construct structural similarity networks. | https://github.com/isebenius/MIND | #1079 |
| DAF-Net | Analysis | An end-to-end deep learning framework featuring anatomically-consistent cross-modal fusion for early Parkinson’s Disease diagnosis using dual-modal diffusion MRI. | Not Listed | #1081 |
| Zero-MIRID | Reconstruction | Deep learning reconstruction for multi-shot EPIK that suppresses undersampling artifacts and corrects geometric distortions by incorporating field-map information directly into the reconstruction process within a zero-shot, self-supervised dual-domain network incorporating field map–guided data consistency. | Not Listed | #1082 |
| SCoRe | Reconstruction | Sparsity-adaptive Composite Recovery (SCoRe) reconstruction is used for fat-water cine MRI with pseudo-random sampling to improve image quality. | Not Listed | #1083 |
| 3D Residual-UNet (ResUNet) | Reconstruction | Denoises and sharpens low-quality low-field MRI images using a two-stage framework trained with a hybrid loss. | Not Listed | #1085 |
| RobFuse | Reconstruction | A robust data-fusion-based slice-GRAPPA method for SMS reconstruction aiming to balance in-plane artifacts and cross-plane leakage in the presence of compromised consistency between calibration and MB data. | Not Listed | #1086 |
| SONIC-MRE | Reconstruction | Reconstructs undersampled MRE images based on transform-domain sparsity and incoherent sampling patterns. | Not Listed | #1090 |
| Segment sharing | Reconstruction | A novel method for increasing temporal resolution in compressed sensing 4D flow by grouping segments of consecutive cardiac bins to create an extra, in-between timeframe. | Not Listed | #1093 |
| BART | Reconstruction | Performs L1-regularized CS reconstruction and combined L1+LLR regularization for Snapshot GRE-CEST data from retrospectively undersampled datasets. | Not Listed | #1094 |
| GPU-accelerated EPG-based simulator | Simulation | Simulates MRF signal trajectories using a GPU-accelerated EPG implementation within a Gym-style environment for reinforcement learning-based acquisition parameter optimization. | Not Listed | #1095 |
| OSPREY | Analysis | MRS data analysis with advanced fitting tools using sequence- and scanner-specific basis sets. | Not Listed | #1097 |
| Multilayer Perceptron (MLP) | Analysis | Quantify unsaturated (UFA), polyunsaturated (PUFA), monounsaturated (MUFA), and saturated (SFA) fatty acids from MRS data using a deep learning framework trained on augmented synthetic spectra. | Not Listed | #1099 |
| WALINET+ | Analysis | Removes water peak, water sidebands, and lipid signals from water-unsuppressed MRSI data to improve metabolite quantification. | Not Listed | #1102 |
| CloudBrain-VisualAI | Analysis | Provides a web-based, code-free visual programming environment for deep learning model creation, training, and evaluation for MRI and MRS analysis. | Not Listed | #1105 |
| In-house developed pulse sequence | Pulse Sequence Design | Spectrally-selective CSI pulse sequence to excite targeted downfield metabolites while preserving the water signal for in vivo NAD+ and tryptophan CSI in the human brain at 7T and 3T. | Not Listed | #1106 |
| Osprey | Analysis | An open-source MRS processing and analysis software for exploring metabolic differences. | Not Listed | #1109 |
| SCREENot | Analysis | Determines the optimal number of retained principal components in PCA-based denoising by ensuring only physiologically plausible spectral variance is preserved. | Not Listed | #1110 |
| MagTetris | Simulation | Performs magnetic field calculations for optimization of permanent magnet arrays. | Not Listed | #1112 |
| CoilGen | Pulse Sequence Design | Generate layouts for gradient channels on a cylindrical carrier for MRI gradient coil design. | Not Listed | #1113 |
| magpylib | Simulation | Simulation to determine the locations and orientations of magnet pieces for passive shimming. | Not Listed | #1114 |
| DTDM Coil Designer | Pulse Sequence Design | A MATLAB-based GUI tool for designing and impedance-tuning double-tuned and double-matched RF coils or baluns to 50 Ω at two distinct Larmor frequencies. | https://github.com/CHSun2022/Double_Tuned_Double_Matched_Circuit_Design_Helper | #1117 |
| Custom-coded differentiable simulator | Simulation | Infers the tuning voltages of all degrees of freedom for a desired field profile using gradient-descent based optimization. | Not Listed | #1119 |
| LoMINA-SC | Reconstruction | A Mamba-based 3D deep learning framework designed for accurate segmentation of subcortical brain structures in T2-weighted (T2w) uLF pediatric MRI. | Not Listed | #1125 |
| SegResNetVAE | Analysis | Performs automated skull-stripping of pediatric low-field T2-weighted brain MR images using a deep learning model based on the SegResNetVAE architecture. | Not Listed | #1132 |
| PREFUL App | Analysis | Post-processes free-breathing balanced steady state free precession sequence data to yield spatial maps and aggregate statistics of ventilation defect (VDP), perfusion defect (QDP), ventilation-perfusion mismatch (VQMD), and ventilation perfusion match (VQM). | Not Listed | #1134 |
| Custom Python scripts | Analysis | Compute voxel-wise ADC maps using subject-specific bval/bvec inputs. | Not Listed | #1136 |
| DWImodMR-TRG | Analysis | Improves the accuracy of MRI in detecting residual disease after immunotherapy-based total neoadjuvant therapy for rectal cancer by adjusting the MR-TRG score based on diffusion-weighted imaging. | Not Listed | #1153 |
| MATI | Data Management | Data fitting platform used for calculating microstructural parameters. | Not Listed | #1154 |
| MCMRSimulator | Simulation | To predict the dMRI signal across a range of discrete values of target microstructural and sequence parameters using Monte-Carlo simulations. | Not Listed | #1160 |
| CATERPillar | Simulation | Generation of numerical gray matter-like substrates with beading, undulation, orientation dispersion, and somas. | Not Listed | #1161 |
| Modified Yeong-Torquato simulated annealing (SA) algorithm | Reconstruction | To invert dMRI signals into microstructure by modifying the Yeong-Torquato simulated annealing algorithm in q-space with a regularization term to promote phase connectivity and reduce pixel swaps. | Not Listed | #1162 |
| None | Analysis | Interpret low b-value DTI by inferring flow-velocity vector fields under oscillatory flow. | Not Listed | #1164 |
| Green-dMRIPrep | Reconstruction | A GUI-driven pipeline that streamlines diffusion MRI preprocessing, embeds carbon auditing and automated quality control, and promotes reproducible, energy-aware neuroimaging practices. | Not Listed | #1166 |
| Realistic Microstructure Simulator | Simulation | To simulate diffusion signals in tissue voxels containing impermeable spherical somata with cytoplasm and concentric nuclei surrounded by cerebrospinal fluid, enabling the evaluation of multishell DWI schemes and their impact on RSI sensitivity to tumor cellularity. | Not Listed | #1173 |
| Shimming Toolbox | Analysis | Compute B0 field maps and slice-wise shimming coefficients. | Not Listed | #1179 |
| DenseNet-based slice-wise regressor | Reconstruction | Estimates slice-wise rigid translation parameters for motion correction in spinal cord dMRI and fMRI. | https://github.com/NTucksinapinunchai/moco_dmri_fmri | #1181 |
| sct_deepseg graymatter | Reconstruction | Automatic gray matter segmentation of the spinal cord across multiple MRI contrasts, field strengths, SC regions, and pathologies using a deep learning model integrated into the Spinal Cord Toolbox (SCT). | Not Listed | #1183 |
| PCNtoolkit | Analysis | Train a normative model for spinal cord cross-sectional area using hierarchical Bayesian regression to quantify patient-level morphometric deviations relative to a healthy cohort. | Not Listed | #1184 |
| Rep2Rep | Reconstruction | Performs noise-adaptive self-supervised denoising on MRI images by learning from paired clinical repetitions with shared anatomical content and independent noise realizations. | Not Listed | #1185 |
| ESRGAN-CEST | Reconstruction | A deep learning framework that enables CEST imaging with enhanced spectral and spatial resolution using an Enhanced Super-Resolution Generative Adversarial Network (ESRGAN). | Not Listed | #1198 |
| Physics-driven deep learning method | Reconstruction | Suppresses Gibbs-ringing artifacts by embedding the k-space truncation model directly into the training process, enabling the network to invert the artifact formation process. | Not Listed | #1202 |
| SynthSeg | Segmentation | Generates pseudo tissue maps from T1w images to guide image translation. | Not Listed | #1203 |
| Hybrid Attention Transformer (HAT) | Analysis | Enhance 0.05T 3D cardiac cine MRI by adapting a vision transformer pre-trained on natural images to improve noise suppression, restore cardiac structures, and improve temporal fidelity. | Not Listed | #1204 |
| 4D U-Net | Reconstruction | Denoise and superresolve 4DFlow-MRI by integrating spatial and temporal flow constraints using a physics-informed 4D U-Net. | Not Listed | #1205 |
| GAN model | Reconstruction | Enhances the contrast of 64mT neonatal T2-weighted MRI images by transferring contrast from corresponding 3T images using a generative adversarial network (GAN). | Not Listed | #1207 |
| DDPM-based framework | Reconstruction | Denoise single-acquisition breast DWI images using a denoising diffusion probabilistic model (DDPM) with domain-specific adaptations for Rician noise and automated timestep selection. | Not Listed | #1209 |
| 3D-DLIE | Reconstruction | Improves UTE MRI image quality by reducing noise and artifacts while preserving anatomical structures using deep learning-based adaptive filtering within an iterative reconstruction framework. | Not Listed | #1210 |
| 4DFlowNet | Reconstruction | Super-resolve 4D flow MRI data using a deep learning approach with publicly available pre-trained weights. | Not Listed | #1211 |
| SNRDiff | Reconstruction | A lightweight diffusion-based MRI denoiser optimized for training with limited data, allowing efficient region-adaptive noise reduction while preserving intricate anatomical details. | Not Listed | #1212 |
| Two-stage Refinement Framework | Reconstruction | Enhances mid-field (0.6T) T2-weighted prostate MRI by denoising and super-resolution in two stages: a U-Net for initial enhancement followed by a flow matching pipeline for detail recovery. | Not Listed | #1213 |
| SIINR | Reconstruction | Reconstructs high-resolution diffusion MRI images from low-resolution, highly anisotropic clinical data by combining a supervised learning module for structural priors with a self-supervised implicit neural representation for signal consistency. | Not Listed | #1214 |
| 3D conditional diffusion model | Reconstruction | Simultaneously perform partial Fourier reconstruction, denoising, and super-resolution of 0.05T 3D knee images using a 3D conditional diffusion model. | Not Listed | #1215 |
| S²-MoCo | Reconstruction | A fully self-supervised k-space framework that seamlessly integrates motion detection, correction, and parameter estimation within a unified pipeline for physically consistent motion-aware k-space reconstruction. | Not Listed | #1217 |
| CoNAD | Analysis | A generative AI framework that integrates Feature-wise Linear Modulation (FiLM) to condition the network on user-defined denoising levels for MRI images. | Not Listed | #1218 |
| χ-separation toolbox | Reconstruction | Process QSM and χ-separation maps. | https://github.com/SNU-LIST/chi-separation | #1222 |
| Forsberg toolbox | Reconstruction | Register respiratory resolved images for each measurement towards end-inspiration image. | Not Listed | #1234 |
| Forsberg toolbox | Analysis | Provides registration of reconstructed images to the end-inspiration image using a group-oriented approach. | Not Listed | #1235 |
| Gadgetron | Reconstruction | Low-latency dual-image reconstruction pipeline for cardiac and pulmonary MRI. | Not Listed | #1242 |
| OpenRecon | Reconstruction | An inline post-processing framework used for non-rigid image registration, deep learning-based lung segmentation, and field map computation. | Not Listed | #1243 |
| OpenRecon | Reconstruction | Reconstruct MR-ARFI focal point maps from raw data using a ‘raw to complex image’ feature to output a mean phase difference map. | Not Listed | #1251 |
| mrftools | Reconstruction | Dictionary-based reconstruction of T1, T2, and M0 maps plus synthetic MPRAGE images from MRF data. | Not Listed | #1259 |
| FTL-MAPLE | Reconstruction | Rapidly reconstructs T1, T2*, frequency, and proton density maps from multi-echo, multi-flip angle GRE acquisitions using a scan-specific, unsupervised, hybrid-regularized network. | https://github.com/autmr-git/ftl_maple | #1262 |
| DL-SAMER+B₀ | Reconstruction | Integrates deep learning into the SAMER image reconstruction, alternating between classical conjugate-gradient SENSE+motion+B₀ optimization and deep-learning-based updates of the image prior. | Not Listed | #1263 |
| vNav-QALAS | Pulse Sequence Design | Integrates volumetric EPI navigators (vNavs) into the 3D-QALAS sequence to provide real-time prospective motion correction for 3D multiparametric mapping without external tracking hardware. | Not Listed | #1265 |
| Multitasking (MT) framework | Reconstruction | Reconstructs undersampled 3D LGE MRI data with low-rank tensor subspace modeling, motion correction, and dictionary-based dynamic DCE (dDCE) modeling to improve SNR and preserve contrast dynamics for simultaneous RV fibrosis and cine imaging. | Not Listed | #1267 |
| T2_KneeActive | Pulse Sequence Design | Optimize Multi-Spin Echo (MSE) T2 mapping sequences for knee articular cartilage using the Cramer-Rao Lower Bound formalism to improve repeatability. | https://github.com/ZemaTimoteo/T2_KneeActive | #1269 |
| Physics-guided deep learning (DL) framework | Reconstruction | Reconstruct accelerated carotid T1/T2 mapping data by combining model-based data consistency with learned image priors. | Not Listed | #1270 |
| ORACLE | Analysis | Extends an analytical framework for T1 and T2 mapping to include water-fat quantification, thereby overcoming T1 signal bias as well as limits in SNR efficiency. | Not Listed | #1272 |
| pTV toolbox | Analysis | Register acquired images. | Not Listed | #1273 |
| DENSE-DDPM | Simulation | Synthesize realistic DENSE images from simulated DENSE data using a conditional denoising diffusion probabilistic model. | Not Listed | #1275 |
| CONN toolbox | Analysis | Preprocesses fMRI data, extracts ROI-to-ROI functional connectivity matrices, performs graph theoretical analysis and network visualization. | Not Listed | #1280 |
| GRETNA toolbox | Analysis | Brain functional network construction and analysis of global and nodal topological properties from resting-state fMRI data. | Not Listed | #1285 |
| e2D-FAM | Pulse Sequence Design | Implements an extended 2D-FAM sequence in the vendor-agnostic Pulseq framework to enable combined PDFF/R2* mapping with high-resolution T1-weighted imaging during free-breathing. | Not Listed | #1291 |
| DL-based reconstruction algorithm | Reconstruction | Reconstructs accelerated MRI data acquired with CAIPIRINHA acceleration, finetuned for high-resolution 7T data. | Not Listed | #1292 |
| oNLI | Reconstruction | Learns a mapping between MRE displacement fields and complex shear stiffness to enable real-time nonlinear inversion of magnetic resonance elastography data. | Not Listed | #1294 |
| QMRITools | Analysis | Used to obtain the segmentation of the thigh muscles into three groups (posterior, medial, and anterior) for region-wise stiffness reconstruction. | Not Listed | #1296 |
| Transformer-based neural network | Analysis | Predict AIFs from noisy voxel time courses using a transformer-based neural network trained on synthetic DCE-MRI data. | Not Listed | #1299 |
| QCMRI | Analysis | Predict ADC map quality from T2 images acquired earlier in the scanning process using a deep learning model. | https://github.com/jbrender/QCMRI | #1307 |
| TWE-NITI | Reconstruction | Invert anisotropic mechanical parameters from MRE data in vivo using a traveling-wave-expansion approach for near-incompressible transversely isotropic tissue. | Not Listed | #1313 |
| SUIT toolbox | Analysis | Cerebellar segmentation and volumetric analysis from T1-weighted data. | Not Listed | #1315 |
| MIND | Analysis | Computes inter-parcel similarity based on morphometric features using a symmetric KL-divergence approach to capture cortical organization. | Not Listed | #1316 |
| OpenPIV | Analysis | Compute voxel-wise velocity fields from reconstructed ²³Na signal fluctuations. | Not Listed | #1317 |
| IntelliSpace Medicina Scientia (ISMS) | Analysis | An in-house software used to generate multiple diffusion parameter maps from DICOM images using DIPY, including DTI, DKI, NODDI, and MAP parameters. | Not Listed | #1318 |
| Spinal Cord Toolbox (SCT) | Analysis | To perform spinal cord morphometry, including segmentation using DeepSeg, manual quality control, and vertebral labeling. | Not Listed | #1325 |
| Gannet | Analysis | Quantification of GABA-edited MRS data. | Not Listed | #1326 |
| Tractometry | Analysis | Quantify microstructural changes in myelin water fraction (MWF) and fractional anisotropy (FA) over time along white matter tracts and apply it to characterize the propagation of damage along WM tracts that pass through MS lesions. | Not Listed | #1328 |
| AMICO | Analysis | Performs NODDI modeling to generate intra-, extra-neurite, and isotropic fraction maps from diffusion-weighted MRI data. | Not Listed | #1329 |
| CLEAR-SWI | Reconstruction | Calculate susceptibility weighted imaging (SWI) offline. | https://github.com/korbinian90/ClearSwi.jl | #1331 |
| DEcomposition and Component Analysis of Exponential Signals (DECAES) | Analysis | Generate MWF maps by computing the voxel-wise ratio of short T2 signal to total water signal from multi-echo spin-echo MRI data. | Not Listed | #1332 |
| Chi-SepNet | Analysis | Decomposes total susceptibility into paramagnetic and diamagnetic components using a convolutional neural network. | Not Listed | #1336 |
| CONN toolbox | Analysis | Analyze genotype–connectivity associations in resting-state functional MRI (rs-fMRI). | Not Listed | #1342 |
| JuSpace | Analysis | To investigate the spatial correspondence between FC changes and neurotransmitter/cellular/metabolic atlas maps, and to examine their associations with cognitive impairment. | Not Listed | #1344 |
| dynamic network-attack framework | Simulation | Simulates sequential node removals in healthy functional connectomes to model neurodegeneration as a cascading failure process and identify selectively vulnerable brain regions. | Not Listed | #1345 |
| Yazdan-Panah convolutional neural network model | Segmentation | Generates choroid plexus masks from T1-weighted MRI images. | Not Listed | #1347 |
| CONN Functional Toolbox | Analysis | Analyze functional MRI (fMRI) data to perform ROI-to-ROI functional connectivity analysis. | Not Listed | #1349 |
| TbCAPs toolbox | Analysis | Extract dynamic Co-Activation Patterns (CAPs) from resting-state fMRI data using the posterior cingulate cortex as a seed region. | Not Listed | #1352 |
| ROMEO | Reconstruction | Unwrapping the phase to produce a smooth phase map for EPT reconstruction. | Not Listed | #1353 |
| NIFTI_NORDIC | Analysis | Denoising fMRI data by thermal noise reduction. | github.com/SteenMoeller/NORDIC_Raw | #1369 |
| Diffusion model-inspired framework | Reconstruction | Compensates for T2*-induced blurring in ultrashort TE MRI by integrating a physics-informed tissue-dependent signal decay simulation into the forward process of a diffusion model. | Not Listed | #1371 |
| U-Net diffusion model | Simulation | To generate post-contrast T1 mappings from pre-contrast T1 mappings and MOLLI images using a conditional diffusion model with a U-Net architecture. | Not Listed | #1372 |
| In-house scripts | Reconstruction | Reconstruct k-space data offline, including 2D-CAIPIRINHA undersampling using the SENSE algorithm and PF undersampling via projections onto convex sets (POCS) reconstruction. | Not Listed | #1374 |
| BART toolbox | Reconstruction | Reconstruct images from under-sampled k-space data using NUFFT and solve the optimization problem with LLR regularization. | Not Listed | #1383 |
| PSO-optimized VQVAE | Reconstruction | Reconstructs high-quality cerebral blood flow (CBF) maps from fewer Arterial Spin Labeling (ASL) acquisitions by integrating a Vector Quantized Variational Autoencoder (VQVAE) with a Particle Swarm Optimization (PSO) algorithm for latent space refinement. | Not Listed | #1388 |
| gammaSTAR | Data Management | Acquire multi-TE ASL scans with a 3D GRASE readout and two FOCI pulses for background suppression. | Not Listed | #1389 |
| Pulseq-based PCASL sequence | Pulse Sequence Design | Implements a vendor-agnostic pseudo-continuous arterial spin labeling (PCASL) sequence using the Pulseq framework to reduce inter-scanner variability in cerebral blood flow (CBF) quantification. | Not Listed | #1390 |
| LOFT BBB Toolbox | Analysis | Generate Kw, ATT, and CBF maps incorporating motion correction, co-registration, skull stripping, quality assurance procedures as well as quantification steps. | Not Listed | #1392 |
| HippoUnfold toolbox | Analysis | Segment the hippocampus and its subfields in MP2RAGE images and compute voxel-wise gradients. | Not Listed | #1394 |
| MVPANI toolbox | Analysis | Conduct pattern analysis for multivariate pattern analysis (MVPA). | Not Listed | #1395 |
| U-Net | Segmentation | To automatically segment the hand motor cortex (HMC) based on gyrus anatomy from structural MRI, trained on expert-annotated HMC regions. | Not Listed | #1398 |
| HBDN | Analysis | Integrates static, dynamic, and topological fMRI representations with a noise-robust semi-supervised learning strategy to improve diagnostic reliability in disorders of consciousness (DOC). | Not Listed | #1399 |
| Cortex MAE | Analysis | Learns fine-grained cortical features via deformation-based masked autoencoding to generate generalizable representations for downstream cortical analysis tasks. | Not Listed | #1401 |
| PyRadiomics | Analysis | Extract radiomic features from post-T1 and ECV mappings for disease classification. | Not Listed | #1402 |
| OXSA toolbox | Analysis | Process 7T 31P-MRSI data in MATLAB to analyze cardiac energy metabolism and correlate it with functional cardiac parameters. | Not Listed | #1430 |
| ASLtoolkit | Analysis | Simplifies multi-echo ASL processing and lowers the barrier for adoption by providing a standardized, accessible, and open-source tool for generating quantitative perfusion and BBB maps from raw multiTE-ASL data. | Not Listed | #1435 |
| microTorch | Analysis | A software framework enabling flexible, user-friendly self-supervised fitting of dMRI models. | https://github.com/snigdha-sen/microtorch | #1436 |
| VASAL | Simulation | VASAL is an algorithm for generating brain microvascular phantoms with flexible network morphology, together with a flow simulation toolbox. | Not Listed | #1437 |
| CEDQ | Reconstruction | Fast and accurate estimation of QTI parameters that are physics consistent and robust to a range of protocols using a hierarchical encoder-decoder network. | Not Listed | #1438 |
| IVIM-INR | Analysis | A two-stage framework for IVIM parameter estimation that uses implicit neural representation (INR) to denoise the signal and fit IVIM parameters. | Not Listed | #1439 |
| Data-sharing-guided DL framework | Reconstruction / Analysis | Jointly reconstruct high-quality multi-shot DW-images and generate IVIM-biomarker maps from highly accelerated MS-DW-EPI data using a data-sharing-guided deep learning framework. | Not Listed | #1440 |
| SCNN* | Analysis | Extends a spherical convolutional neural network (SCNN) with a hypernetwork to achieve b-value generalisation in brain tissue microstructure estimation from diffusion-weighted MRI data. | Not Listed | #1441 |
| OpenRecon–Python container | Analysis | Automates kidney segmentation and performs Twelve-Layers Concentric objects (TLCO) analysis for renal T1 mapping on the MRI scanner. | Not Listed | #1442 |
| find_variance_lines.tcsh | Analysis | Detects variance line artifacts in FMRI data by estimating variance in each voxel time series, scaling the variance, and identifying columns with high mean scaled variance. | Not Listed | #1444 |
| RF Coil Standard | Data Management | Defines a standard data format and transmission protocol for identifying an RF coil’s capabilities and operating parameters to the scanner, storing the data in BSON format on an EEPROM. | https://github.com/ | #1445 |
| OSIPI IVIM code repository | Analysis | Evaluates and compares different IVIM model fitting algorithms to assess the impact of algorithm design choices on the reproducibility of IVIM parameters. | https://github.com/OSIPI/TF2.4_IVIM-MRI_CodeCollection | #1447 |
| Hyperwheel | Data Management | Automates the transfer of DICOM data from Hyperfine Swoop MRI scanners to a Flywheel server using a Raspberry Pi and custom Lua script. | Not Listed | #1448 |
| VoxLoc | Analysis | An open-source, vendor-independent automated tool for Magnetic Resonance Spectroscopy (MRS) voxel localization. | Not Listed | #1449 |
| D-SAME-IVIM | Analysis | Predicts the weighting factors of 40 basis diffusion coefficients from 0.0007 mm2/s to 0.3 mm2/s (output) from 16 b-value DWI data (input) at different SNR levels. | Not Listed | #1453 |
| DL-reconstruction algorithm | Reconstruction | To reconstruct diffusion-weighted images from an accelerated IVIM-DWI protocol, reducing the number of averages needed at high b-values. | Not Listed | #1454 |
| DL rs-EPI | Reconstruction | Improves diffusion image quality in breast DWI with deep learning-based reconstruction, enabling faster acquisition times. | Not Listed | #1455 |
| Spinal Cord Toolbox (SCT) | Analysis | Motion correction and segmentation of spinal cord MRI data. | Not Listed | #1457 |
| IVIM-NET | Reconstruction | Deep learning framework for IVIM parameter estimation from diffusion-weighted MRI data, aiming to improve fitting stability and reproducibility compared to conventional methods. | Not Listed | #1460 |
| AIM-IVIM (Anatomy-Informed Model IVIM) | Analysis | Improves tri-compartmental IVIM parameter estimation by incorporating anatomical priors derived from T1-weighted and FLAIR-weighted MRI using a fusion network with CNN and DNN branches. | Not Listed | #1462 |
| INDI | Analysis | Post-processing of cardiac diffusion tensor imaging (cDTI) data, including exclusion of frames with signal voids and misregistration for tensor calculation. | https://github.com/ImperialCollegeLondon/INDI | #1463 |
| AI-THiNRV1.0 | Analysis | A user-friendly AI-based toolbox designed for clinical use, comprising two novel toolkits: (i) MRI-Deep Feature Analysis (MRI-DFA) was tested on papillary thyroid carcinoma (PTC), (ii) MRI-Physics-Based Deep Learning (MRI-PBDL) analysis was tested on HN Cancer (HNC). | Not Listed | #1464 |
| JuSpace toolbox | Analysis | To evaluate neurotransmitter system associations via correlations with public PET-derived receptor/transporter distribution maps. | Not Listed | #1468 |
| GRETNA | Analysis | Quantify global, nodal, and modular properties of brain networks using graph-theoretical metrics. | Not Listed | #1473 |
| AdjPerm | Analysis | Performs adjacency-restricted permutations to test spatially embedded data while preserving local structure, maintaining local topology and spatial autocorrelation. | Not Listed | #1474 |
| HONeD | Analysis | Infers directed structural connectivity from dMRI and fMRI data using a physics-informed generative model. | Not Listed | #1475 |
| Pulseq | Pulse Sequence Design | Implementation of custom multi-shot spiral and EPI diffusion-weighted sequences. | Not Listed | #1476 |
| IPA | Pulse Sequence Design | A flexible, fully self-gated framework for robust physiological gating independent of imaging plane orientation. | Not Listed | #1477 |
| FastSafeDesign | Pulse Sequence Design | Designs Peripheral Nerve Stimulation (PNS) and mechanical resonance safe spiral trajectories on-the-fly by adjusting gradient and slew rate limits. | https://github.com/mriphysics/FastSafeDesign | #1478 |
| Multiband RF Toolbox | Pulse Sequence Design | Designed variable-rate selective excitation multi-slice prepulses for a simultaneous multi-slice UTE sequence. | Not Listed | #1481 |
| BART toolbox | Reconstruction | Image reconstruction was performed offline using compressed sensing with wavelet regularization. | Not Listed | #1482 |
| None | Reconstruction | Reconstruct dynamic volumes at high spatiotemporal resolutions from incoherently-sampled Zero-TE imaging data using a low-rank formulation and self-estimated temporal subspace. | Not Listed | #1483 |
| ANTs | Analysis | Registration of MP2RAGE images to MNI152 and Harvard-Oxford atlas. | Not Listed | #1484 |
| FixelBayes | Analysis | A Bayesian extension to fixel-based analysis (FBA) that estimates a Bayes Factor (BF) at each fixel to provide continuous evidence measures for group differences in fixel metrics. | Not Listed | #1486 |
| GenTract | Reconstruction | Generates entire streamlines at once, conditioned on global anatomical representations for more robust white-matter tractography. | Not Listed | #1487 |
| TractCloud3 | Analysis | Performs prediction of white matter fiber tracts into 4 classes based on geometric features by sampling 25 points per fiber and 20 neighboring fibers. | Not Listed | #1488 |
| dipy | Analysis | Generate parametric profiles for seven metrics (NAA/Cr, Cho/Cr, NDI, ODI, FWF, MD, FA) using automated fiber-tract quantification. | Not Listed | #1491 |
| Graph-regularized spherical deconvolution | Reconstruction | Generate asymmetric Fiber Orientation Distributions (aFODs) for improved white matter reconstruction by incorporating a graph-based regularization term into damped Richardson-Lucy spherical deconvolution. | Not Listed | #1493 |
| Bash-based pipeline | Reconstruction | To automatically perform arcuate fasciculus segmentation from raw diffusion images using MRtrix3, FSL, and ANTs. | Not Listed | #1494 |
| ExploreASL | Analysis | Quantifies CBF maps from ASL data. | Not Listed | #1497 |
| Slicer-DTI-ALPS software | Analysis | Computes DTI and DTI-ALPS indices from diffusion-weighted images. | Not Listed | #1500 |
| Track DFC | Analysis | Parcellates white matter streamlines based on dynamic functional connectivity derived from resting-state fMRI data at streamline endpoints, and analyzes the relationship with age and cognition. | Not Listed | #1504 |
| Custom atlas generation script | Analysis | Generates a custom atlas by merging regions from the Automated Anatomical Labeling Atlas version 3 (AAL3) within SPM12 for ROIs to match those used in visual analysis. | Not Listed | #1531 |
| THOMAS | Reconstruction | Parcellates thalamic subregions into 12 bilateral nuclei. | Not Listed | #1533 |
| HypperMapp3r | Segmentation | Perform automated white matter hyperintensity (WMH) segmentation using a convolutional neural network (CNN). | Not Listed | #1535 |
| CAT12 | Analysis | Performs volumetric analysis of 3D T1-weighted images. | http://www.neuro.unijena.de/cat/ | #1539 |
| United Imaging Artificial Intelligence (uAI) system | Analysis | Performs fully automated segmentation to calculate the volumes of hyperintense signals on Cube FLAIR in juxtaventricular, periventricular, deep, and juxtacortical WMH regions. | Not Listed | #1540 |
| PCNtoolkit | Analysis | Used for Hierarchical Bayesian Regression (HBR) to compute normative lifespan trajectories and corresponding centile curves for 22 JHU-WM regions using DTI metrics. | Not Listed | #1544 |
| Gacelle toolbox | Analysis | Provides a GPU-accelerated, minimization-based optimization framework, including the askAdam optimizer, used to estimate parametric maps from dMRI data. | Not Listed | #1547 |
| JuSpace toolbox | Analysis | Computes spatial correlations between free water changes and molecular maps of neurotransmitter receptor densities. | Not Listed | #1549 |
| SANDI-Matlab-Toolbox-Latest-Release | Analysis | SANDI modelling was performed using the toolbox, yielding maps of soma and neurite density, soma radius, extracellular fraction, and extracellular diffusivity. | https://github.com/palombom/SANDI-Matlab-Toolbox-Latest-Release | #1550 |
| Motion-corrected TR-MRF | Reconstruction | Mitigate respiratory-induced artifacts and improve anatomical fidelity in both quantitative maps and synthetic contrast-weighted images by integrating rigid motion correction within the subspace reconstruction framework of Time-Resolved MR Fingerprinting (TR-MRF). | Not Listed | #1554 |
| PETALUTE | Pulse Sequence Design | Designed a 3D rosette UTE pulse sequence for dynamic 19F gas imaging with ultra-short echo time to quantify concentration changes. | Not Listed | #1561 |
| Generative MR Multitasking framework | Reconstruction | Unified reconstruction paradigm that produces both variability-robust gated-like images and beat-to-beat time-resolved images using a neural implicit representation with learned complex-harmonic cardiac motion. | Not Listed | #1566 |
| CMR-MOTUS | Reconstruction | Jointly reconstructs motion fields and motion-corrected images from free-running cardiac MRI data through alternating optimization and low-rank decomposition. | Not Listed | #1573 |
| 3D U-Net | Reconstruction | Generate peak systolic flow MRI angiographies from TOF MR angiographies using a 3D U-Net implemented in PyTorch. | Not Listed | #1575 |
| Deep Unrolled 2D PC-MRI Reconstruction | Reconstruction | Reconstruct 2D phase-contrast MRI images from highly undersampled k-space data using a deep unrolling framework with a Cross Attention Transformer model and learnable complex-difference sparsity. | Not Listed | #1579 |
| OXSA | Analysis | Spectral fitting of spectroscopy data. | Not Listed | #1580 |
| MambaRoll | Reconstruction | MambaRoll is a physics-driven autoregressive state-space model for accelerated MRI reconstruction that progressively predicts finer-scale feature maps conditioned on coarser representations through multi-scale autoregressive inference. | https://github.com/icon-lab/MambaRoll | #1584 |
| SSCU | Reconstruction | A self-supervised learning strategy via coil undersampling (SSCU) is proposed, which randomly selects a subset of coils to predict the remaining coils, and is integrated into both image-domain and k-space model-based reconstruction frameworks. | Not Listed | #1585 |
| Water Content-Guided Physics Informed Neural Network (WCG-PINN) | Reconstruction | Reconstructs electrical properties (conductivity and permittivity) by incorporating water content information and physical laws into a neural network to improve accuracy and reduce the need for large training datasets. | Not Listed | #1586 |
| UMPIRE-Net | Reconstruction | MRI reconstruction with separate magnitude and phase regularizers in algorithm unrolling to improve reconstruction quality in scenarios with large phase variations. | Not Listed | #1587 |
| AutoSamp | Pulse Sequence Design | Jointly optimizes k-space sampling patterns and image reconstruction networks for accelerated MRI. | Not Listed | #1588 |
| Uncertainty Module | Analysis | Estimates statistically rigorous uncertainty intervals for accelerated MRI reconstructions by combining quantile regression and conformal prediction with a pre-trained end-to-end variational network. | Not Listed | #1589 |
| Implicit ESPIRiT | Reconstruction | To develop a compact, implicit neural representation of ESPIRiT coil sensitivities by training an MLP to learn spatially-smooth basis vectors for ESPIRiT subspaces in a memory-efficient manner. | Not Listed | #1590 |
| None | Reconstruction | Reconstructs MRI images from k-space data using a continuous representation based on adaptive 2D Gaussian basis functions (“blobs”) to avoid pixelation artifacts and enable resolution enhancement. | Not Listed | #1591 |
| mte-ms-DAGER | Reconstruction | Extends the DAGER framework to jointly exploit shared information across q-space and multiple TEs in multi-TE diffusion MRI, enabling higher acceleration. | Not Listed | #1592 |
| DMIWizard | Data Management | Processing of the 3D Deuterium Metabolic Imaging (DMI) datasets. | Not Listed | #1595 |
| INSPECTOR | Analysis | Optimized preprocessing, analysis, and quantification of pancreas spectra in MATLAB for ¹H-MRS data. | Not Listed | #1598 |
| ART-Net | Analysis | To identify the dominant artifact source in ASL CBF maps among motion, arterial transit-time effects, fat-shift, and z-blurring using a deep learning model. | Not Listed | #1612 |
| BART | Reconstruction | Reconstruct multi-TE images from partition-randomized simulation data with subspace bases and locally low-rank regularization. | Not Listed | #1615 |
| AI Model / Label Plane Extractor / Stack Plane Optimizer | Analysis | Predicts segmentation masks for carotid arteries and labeling planes, extracts labeling plane geometry, and computes imaging stack parameters for ASL planning. | Not Listed | #1616 |
| Robust temporal decomposition model | Reconstruction | Decomposes 4D ASL data into a static angiographic prior and a sparse, dynamic vascular signal for denoising. | Not Listed | #1617 |
| GigaBlochs | Simulation | A Bloch simulation framework designed for flexible investigation of large parameter spaces, enabling comprehensive characterization of PCASL under pulsatile flow conditions. | Not Listed | #1618 |
| TransUNet | Reconstruction | Automated tumor segmentation using a combination of convolutional neural networks (CNNs) and vision transformers to achieve voxel-level tumor segmentation. | Not Listed | #1624 |
| Hyperparameter-optimized CNN | Analysis | Brain tumor classification from MRI images using a hyperparameter-optimized convolutional neural network. | Not Listed | #1626 |
| Temporal Gradient Phase Unwrapping (TGPU) | Reconstruction | Voxel-wise algorithm operating in the time domain that preserves local phase continuity across vibration offsets for robust reconstruction of long-wavelength, low-frequency wave data in MRE. | Not Listed | #1632 |
| 3D Slicer | Analysis | Registration of CT-based planning target volumes (PTVs) to b500 diffusion-weighted images. | Not Listed | #1637 |
| SANDI toolbox | Analysis | Generates five Soma and Neurite Density Imaging (SANDI) parameter maps from diffusion weighted images. | https://github.com/palombom/SANDI-Matlab-Toolbox-v1.0 | #1641 |
| SEPIA toolbox | Reconstruction | Reconstruct QSM maps using FSL BET, Optimum weights Laplacian (MEDI), VSHARP, and MEDI. | Not Listed | #1646 |
| MegaTrack | Analysis | A framework enabling manual dissection of large-scale tractography datasets when no tract templates or automatic methods exist. | Not Listed | #1647 |
| DESIGNER | Analysis | Preprocesses dMRI data. | Not Listed | #1651 |
| ROCKETSHIP | Analysis | Generates arterial input function (AIF) for DCE processing. | Not Listed | #1652 |
| Self-gated motion compensation method | Reconstruction | Motion compensation for breath-held liver MRI using CASPR trajectory and anterior-to-posterior frequency encoding by extracting respiratory signals from k-space data and applying soft/hard gating. | Not Listed | #1672 |
| 4D-iReKAM | Reconstruction | Reconstructs 4D motion-resolved abdominal MRI images from free-breathing FSE data by incorporating respiratory information and minimizing motion artifacts using a sliding-windowing method based on gradient entropy and POCSMUSE. | Not Listed | #1673 |
| Pulseq | Pulse Sequence Design | An open-source pulse sequence programming platform that was used to implement GIRF sequences and a radial imaging sequence for GIRF-based image correction. | Not Listed | #1682 |
| PyPulseqRemoteSequence | Pulse Sequence Design | Implements a Pulseq interpreter in Python that communicates with the scanner via a standardized network interface to stream sequence instructions and react to user input. | Not Listed | #1685 |
| ScanHub | Acquisition | Provides an open, cloud-native platform for scanner control, reconstruction workflow orchestration, and visualization of MRI data, enabling efficient remote operation of point-of-care low-field systems. | Not Listed | #1686 |
| Pulseq | Pulse Sequence Design | Extends the FOV positioning algorithm in Pulseq by adding gradient delay correction to enable accurate FOV positioning without additional artifacts in the presence of gradient delays. | Not Listed | #1687 |
| MR-STAT Pulseq Implementation | Pulse Sequence Design | Implement a vendor-agnostic 3D MR-STAT pulse sequence in Pulseq for rapid, high-resolution T1, T2, and proton density mapping across different scanner platforms. | Not Listed | #1689 |
| OpenMRF | Reconstruction | Provides an end-to-end solution for cardiac Magnetic Resonance Fingerprinting (cMRF). | Not Listed | #1690 |
| PROPELLER-EPI sequence | Pulse Sequence Design | Implement an open-source PROPELLER-EPI sequence for cardiac diffusion imaging and accelerate acquisitions using the RoSA method. | Not Listed | #1691 |
| Pulseq | Pulse Sequence Design | Open-source framework used for developing the diffusion-weighted MREG sequence. | Not Listed | #1692 |
| Enzyme.jl | Pulse Sequence Design | Enables highly efficient RF pulse optimization using GPU-accelerated Bloch simulations with compiler-level reverse-mode automatic differentiation. | www.juliahealth.org/KomaMRI.jl/enzyme-ad/ | #1694 |
| jMRUI | Analysis | To quantify myocardial PCr/ATP and PCr/γATP ratios from cardiac ³¹P-MRS data. | Not Listed | #1696 |
| WSVIE solver and co-simulator | Simulation | Rapid full-wave wire-surface-volume integral equation (WVSIE) solver and a fully automatic circuit co-simulator for coil tuning, matching, decoupling, preamplifier decoupling, and detuning. | https://github.com/cloudmrhub/marie | #1697 |
| DL Reconstruction Pipeline | Reconstruction | Reconstructs 3D-cine images from multiple real-time 2D acquisitions using a deep learning framework, performing motion correction, super-resolution, and automatic segmentation of cardiac structures. | Not Listed | #1704 |
| nnUnet | Analysis | Semiautomatically segment the lung contour of end-inspiratory image with vessels excluded. | Not Listed | #1710 |
| GRASP | Reconstruction | Reconstruct multi-phase 3D volumes using compressed sensing with a total variation constraint forcing sparsity in the motion state dimension. | Not Listed | #1711 |
| LayNii | Analysis | Parcellates the cortex into cortical columns and samples layer profiles as a function of geodesic distance. | Not Listed | #1713 |
| NORDIC | Analysis | Denoising of time series data using NORDIC prior to motion correction. | Not Listed | #1718 |
| CycleGAN | Reconstruction | Corrects MRI artifacts based on selection by a Vision-Language Model (VLM). | Not Listed | #1722 |
| MeshD²R | Reconstruction | Reconstructs viewed images from fMRI data using a surface-based convolutional network, achieving high semantic fidelity and fine-grained detail by decoding cortex-wide brain activity. | Not Listed | #1723 |
| TurboFIRE | Analysis | Real-time rsfMRI analysis pipeline for performing seed-based correlation analysis and Fisher Z-transformation of correlation coefficients. | Not Listed | #1724 |
| U-Net | Analysis | Enhance resting-state cerebrovascular reactivity (rs-CVR) maps to improve their similarity and accuracy relative to standard CVR maps. | Not Listed | #1734 |
| BrainSpace toolbox | Analysis | Performs gradient analysis on structural covariance networks (SCNs) using diffusion embedding and normalized angle kernel. | Not Listed | #1735 |
| automated tortuosity quantification | Analysis | Automatically evaluates centerline displacement and centerline distance to quantify tortuosity index (TI) from 4D flow MRI data. | Not Listed | #1740 |
| MRpro | Analysis | Derives relaxation times from quantitative imaging in phantom liquid and spherical inclusions. | Not Listed | #1750 |
| KOMA | Simulation | Simulation pipeline used for sequence search based on the KOMA framework. | Not Listed | #1752 |
| NUFFT-SENSE | Reconstruction | Reconstructs images from ultrasonically encoded MR-STAT data, accounting for the ultrasonic trajectory and coil sensitivity profiles. | Not Listed | #1753 |
| MRF BLADE Framework | Pulse Sequence Design | Implemented a research MRF sequence using a BLADE trajectory and a novel framework allowing a generic description of MRF experiments and dictionary generation, including a mode for trajectory measurement. | Not Listed | #1754 |
| WB-MRF | Pulse Sequence Design | Developed a wideband cardiac MRF (WB-MRF) technique for robust T1 and T2 mapping near CIEDs. | Not Listed | #1757 |
| In-house software | Analysis | Manually segment parotid regions of interest (ROIs) to derive mean T1/T2 values from MRF images. | Not Listed | #1759 |
| DL-MorphoBox | Analysis | Research application software used to obtain brain segmentation masks from MP2RAGE datasets and to compare contrast ratio (CR) and contrast-to-noise ratio (CNR) between gray and white matter in FLAWSmin images. It was also investigated to achieve automatic brain segmentation directly from FLAWS-derived images. | Not Listed | #1769 |
| Multiscale registration pipeline | Reconstruction | Registers XPCT to high-resolution MRI data, then inversely applies this registration to map the ultra-high-resolution DESY data onto the MRI space. | Not Listed | #1772 |
| Custom in-house reconstruction pipeline | Reconstruction | Process and fit diffusion data, correct for distortion, eddy currents, bulk motion and non-linearity of diffusion-encoding gradients, and estimate the intra-axonal radial diffusivity and radii. | Not Listed | #1776 |
| MATRIX-DP | Pulse Sequence Design | Integrates diffusion preparation (DP) into a modified fast spin-echo (FSE) sequence with variable refocusing flip angles (MATRIX) to suppress vascular signal and improve nerve-tissue contrast for 3D brachial plexus MR neurography. | Not Listed | #1779 |
| DLPC | Reconstruction | Generates high-quality phase maps for each shot in multishot DTI acquisitions using a residual neural network, enabling robust complex signal averaging and improved image quality. | Not Listed | #1781 |
| SMT Toolbox | Reconstruction | Fitting the SMT model to generate Vax maps. | Not Listed | #1783 |
| ROVER-dMRI | Reconstruction | Super-resolution reconstruction of diffusion MRI images from thick-slice rotating-view acquisitions using implicit neural representation to achieve high SNR and submillimeter resolution. | Not Listed | #1793 |
| FESTIVE | Reconstruction | Efficiently learns an ensemble of GRAPPA-kernels that maps clean k-t data to field-corrupted spiral k-t data using a compact implicit neural representation (INR); from which spatio-temporal field-imperfections are extracted via an ESPRIT-like operation on the GRAPPA-kernels. | Not Listed | #1794 |
| Implicit Neural Representation-based framework | Reconstruction | Jointly reconstructs diffusion MRI data across k-space, q-space, and echo times (TEs) using a self-supervised implicit neural representation, leveraging redundancy and continuity in these domains. | Not Listed | #1795 |
| System-aware EPTI model | Reconstruction | Stabilize the phase across orientations/directions in EPTI reconstruction by modeling orientation-dependent concomitant fields and direction-dependent eddy-current fields, particularly for ultra-high performance gradients. | Not Listed | #1798 |
| Open-Source RF Coil Mixed Contact Connector | Hardware Design | Design an inexpensive RF coil connector using off-the-shelf components and 3D printing to replace costly proprietary connectors in MRI systems. | https://github.com/ | #1812 |
| sigpy | Pulse Sequence Design | Designs Shinnar-LeRoux pulses with specified time-bandwidth product and in-band ripple. | https://sigpy.readthedocs.io/en/latest/guide_basic.html | #1818 |
| OSI² | Analysis | Analyzes B₀ homogeneity in terms of spherical harmonics and generates optimized magnet positions to improve field uniformity. | https://gitlab.com/osi-square/osi-square | #1820 |
| Modular Robotic Mapper Software | Control | Controls the movement of the robot along a defined path, initiates data acquisition, performs data queries, saves the data, and displays the data graphically in intensity color maps. | https://github.com/ | #1822 |
| Magpylib | Simulation | Calculate the magnetic field of each coil at 1A to create a sensitivity matrix. | https://github.com/mri4all/active_shimming | #1823 |
| FANSI | Reconstruction | Reconstruct QSM maps from phase data. | Not Listed | #1825 |
| Multi-resolution graph-cut algorithm | Reconstruction | Enables more reliable water–fat separation, reducing swaps and improves the overall image quality of water- and fat-separated images under low SNR conditions. | Not Listed | #1827 |
| DEBRA | Pulse Sequence Design | A multi-echo acquisition sequence with optimized echo spacing to minimize fat shift in diffusion-weighted imaging, enabling robust water-fat separation. | Not Listed | #1828 |
| mDIXON-Quant | Analysis | Generate proton-density fat-fraction (PDFF) maps from multi-echo gradient-echo data to quantify intramyocardial fat. | Not Listed | #1837 |
| U-Net | Reconstruction | To generate T1rho maps directly from PD- and T2-weighted images using a 2D U-Net architecture. | Not Listed | #1839 |
| Nila | Reconstruction | Enhance low-field MRI image quality using a high-field trained diffusion prior and a noise-level-adaptive data-consistency term, without paired data or retraining. | https://github.com/Solor-pikachu/Nila | #1840 |
| DL-SENSE and DL-GRAPPA | Reconstruction | Enhance MRI image quality and diagnostic performance for FAI by integrating deep learning with SENSE and GRAPPA parallel imaging, while also reducing scan time. | Not Listed | #1842 |
| MCVSR | Reconstruction | Reconstructs a high-resolution 3D MRI volume from a fast-acquired low-resolution 2D scan, guided by an auxiliary high-resolution 3D scan of another contrast. | Not Listed | #1844 |
| UniCMR | Reconstruction | Enhances cross-center generalization in MRI reconstruction by integrating a cascaded network with domain feature unifying (DFU) modules to address domain shifts from different scanners or protocols. | Not Listed | #1847 |
| SLOMOCO | Reconstruction | Correct inter-/intra-volume motion using Slice-Oriented Motion Correction with volume- and slice-wise motion and voxel-wise partial volume nuisance regression. | https://github.com/wanyongshinccf/SLOMOCO.git | #1852 |
| Adjusted open-source QA code | Analysis | Performs data analysis for quality assurance measurements, including calculating GSR, SNR, SFNR, signal drift, fluctuation, and radius of decorrelation. | Not Listed | #1856 |
| NORDIC_RAW | Analysis | Reduce thermal noise in fMRI data using magnitude and phase data, and noise-only volumes. | Not Listed | #1857 |
| Physics-Integrated Neural Network | Reconstruction | Mitigates B0 field inhomogeneity artifacts in GRE-EPI images by integrating physics-based modeling of intravoxel dephasing with a neural network to restore brain signals in regions near air/tissue interfaces. | Not Listed | #1858 |
| Pulseq sequence | Pulse Sequence Design | Implemented a custom 3D radial-spiral-Phyllotaxis GRE sequence optimized for T2*-weighting for high-resolution and fast-sampled retinal imaging. | Not Listed | #1861 |
| Semi-automatic pipeline | Analysis | Quantifies PHVS-based hypoperfusion on SWI-EPI to estimate penumbra in acute ischemic stroke. | Not Listed | #1872 |
| Virtualized vascular network | Simulation | A simple vascular network was developed to simulate collateral flow patterns capable of compensating perfusion when flow through a simulated MCA occlusion was restricted. | Not Listed | #1876 |
| PyASLReport | Analysis | Extracts ASL parameters from DICOM or BIDS datasets, performs consistency checks, and generates parameter tables, text summaries, and error/warning reports. | Not Listed | #1894 |
| hurahura | Data Management | Provides an examination-level framework for imaging studies, standardizing directory structures and supporting logging, tag management, and configuration-based operation. | https://github.com/fraser29/ | #1895 |
| QA toolkit | Analysis | Automatically generates QA maps and ROI-based graphs after each scan for immediate console review and longitudinal tracking of MRI system stability. | https://github.com/byronphy/QA-7T-fMRI | #1897 |
| dicompare | Analysis | A browser-based DICOM validator that validates MRI scanning sessions with pre-defined schema files, which can be generated from a reference session or selected from a pre-existing library of schema for landmark studies and other specific imaging applications. | https://dicompare-web.vercel.app/ | #1898 |
| dsv2pulseq | Pulse Sequence Design | A Python-based converter to transform Siemens simulation (“.dsv”) files into the Pulseq format while preserving the sequence timing. | Not Listed | #1899 |
| FIRE | Reconstruction | Provides an interface to the MRD format, converts scanner’s raw readouts into the MRD format, and sends raw data to a reconstruction container, receiving reconstructed images/spectra back for inline display. | Not Listed | #1900 |
| tissue-to-MRproperty | Data Management | Assigns literature-derived 3T MR tissue properties (T1, T2, T2*), proton density (PD) and χ values to each tissue label in the digital phantom. | https://github.com/shimming-toolbox/tissue-to-MRproperty | #1901 |
| Hierarchy Flow model | Reconstruction | Generate realistic digital MRI phantoms from weighted brain MR images using a self-supervised learning approach with hierarchical coupling layers and Adaptive Instance Normalization. | Not Listed | #1902 |
| mtrk | Pulse Sequence Design | mtrk was improved by incorporating readout building blocks, implementing time management, and making its GUI accessible online through CloudMR, enabling streamlined pulse sequence design and integration into automated simulation pipelines. | Not Listed | #1903 |
| QWIST | Analysis | A modular MATLAB pipeline for reproducible MRI processing that defines workflows by connecting nodes, each declaring input-output relationships and defining execution logic. | https://github.com/markus-nilsson/dpp | #1904 |
| UIH Pulseq interpreter | Pulse Sequence Design | To interpret Pulseq sequences, including a novel binary file format (.bseq), on United Imaging Healthcare (UIH) MRI systems. | Not Listed | #1905 |
| MRSeqStudio | Pulse Sequence Design | A web-based tool for MR sequence design and simulation, allowing users to design, simulate, and visualize pulse sequences directly in the web browser. | https://github.com/pvillacorta/MRSeqStudio | #1906 |
| MR-zero | Simulation | Simulate and optimize NCE lung MRI sequences using a moving virtual lung phantom. | Not Listed | #1907 |
| ndslice | Analysis | Interactive tool for visualizing and analyzing N-dimensional, potentially complex arrays, supporting a broad range of MRI data formats. | https://github.com/henricryden/ndslice | #1908 |
| Multimodal age regression framework | Analysis | Integrates MRIs, imaging-derived features, blood-based measures, and lifestyle factors to predict biological age (BA) in the brain and heart. | Not Listed | #1911 |
| MDEV | Reconstruction | Invert mechanical properties from MRE data. | Not Listed | #1934 |
| None | Reconstruction | Tumours were segmented using an in-house image processing pipeline into non-enhancing, enhancing and oedema regions, according to BRATs criteria. | Not Listed | #1936 |
| Knowledge distillation framework | Analysis | Accurate brain tumor segmentation using incomplete MRI inputs by transferring tumor features from a full-modality teacher network to an incomplete-modality student network via Kullback–Leibler divergence loss. | Not Listed | #1942 |
| TGV-regularized iterative methods | Reconstruction | Reconstruct 31P spectra from MRSI data. | Not Listed | #1953 |
| MRtrix3Tissue | Reconstruction | Preprocesses diffusion data and derives fibre orientation distributions (FODs). | https://3Tissue.github.io | #1959 |
| Customized MATLAB script | Reconstruction | Generates synthetic T1-weighted, T1 map, synthetic double inversion recovery (DIR), synthetic Fluid and White Matter Suppression (FLAWS) and synthetic lesion attenuated contrasts from a single MP2RAGE scan. | Not Listed | #1965 |
| Deep learning–based segmentation model | Reconstruction | Automated delineation of the prostate gland on axial T2-weighted MRI using a deep learning model previously trained and validated on publicly available prostate MRI datasets. | Not Listed | #1972 |
| nnU-Net | Analysis | Automated lesion segmentation and detection were performed using the 3D nnU-Net framework. | Not Listed | #1973 |
| nnU-Net | Segmentation | Automated breast tumor segmentation on DCE and DWI MRI data. | Not Listed | #1974 |
| ExVivo-SCI-SCT | Reconstruction | To accurately segment injured ex-vivo spinal cord tissue in 7T MRI images by adapting a Spinal Cord Toolbox (SCT)-derived deep-learning architecture. | Not Listed | #1975 |
| RectoMap | Analysis | Automatic segmentation of rectal tumor and mesorectum on T2-weighted MRI using an ensemble of deep learning architectures. | https://gitlab.dei.unipd.it/fair/RectoMap | #1976 |
| probabilistic soft-edge post-processing pipeline | Analysis | Enables accurate and computationally efficient total cyst volume (TCV) quantification in ADPKD MRI images by refining 2.5D segmentation outputs. | Not Listed | #1977 |
| None | Analysis | Descriptive analyses and data visualization were performed to analyze the data of the participants of the webinar. | Not Listed | #2008 |
| Spinal Cord Toolbox SCT | Analysis | Automated segmentation techniques for spinal cord imaging. | www.spinalcordtoolbox | #2021 |
| physics-informed, self-supervised deep learning model | Pulse Sequence Design | Generates 2DRF pulses in real time that account for patient head shape, as well as B0 and B1+ inhomogeneities, with multi-shot excitation and readout to suppress off-resonance effects. | Not Listed | #2023 |
| ROS-JRESI-MC | Pulse Sequence Design | Implemented a metabolite-cycled 4D J-resolved MRSI acquisition using a rosette k-space trajectory for accelerated and motion-robust spectroscopic imaging. | Not Listed | #2024 |
| MRS4Brain toolbox | Analysis | Processed MRSI datasets with non-Cartesian reconstruction using NUFFT from the theoretical CRT. | Not Listed | #2025 |
| Total Field Inversion (TFI) | Reconstruction | Reconstructs QSM images directly on the MR console via GPU acceleration from multi-echo data. | Not Listed | #2043 |
| Automated pipeline for U-fiber segmentation | Analysis | To automatically segment U-fibers from T1-weighted images and QSM data for reproducible quantification of U-fiber susceptibility. | Not Listed | #2047 |
| Pixelwise Quantile Regression | Analysis | Quantifies uncertainty in image-to-image regression problems by generating an interval around each pixel that is guaranteed to contain the true value with a user-specified high probability. | Not Listed | #2049 |
| None | Analysis | The study trains simple voxel-wise multilayer perceptrons on synthetic data to predict M₀, R₂*, and total field maps from multi-echo GRE signals for quantitative susceptibility mapping (QSM). | Not Listed | #2052 |
| ViewMotionQSM toolbox | Simulation | Simulates motion-corrupted QSM datasets, allowing users to evaluate the robustness of different QSM reconstruction methods to motion artifacts. | Not Listed | #2055 |
| PRIDE tool | Analysis | A Python application built within Philips research development environment that enables direct fiducial placement and automated grid overlay on the scanner console for MRI-guided trans-perineal prostate procedures. | Not Listed | #2061 |
| ST-GNNs | Analysis | Predict tumor evolution and prognosis in prostate cancer by modeling tumors as dynamic graphs and capturing spatio-temporal dependencies in longitudinal mpMRI data. | Not Listed | #2068 |
| ResUMamba3D | Reconstruction | Synthesizes CT-like images directly from UTE-MRI using a 3D deep learning model combining residual U-Net encoding with Mamba state-space modeling. | Not Listed | #2072 |
| AI-enabled executable bundle | Analysis | Automatically removes bright background from FRACTURE MRI images to improve image quality and enable better downstream processing and analysis. | Not Listed | #2076 |
| Artificial Intelligent Kit (A.K.) | Analysis | Extracts radiomic features from volumes of interest (VOIs) delineated in ITK-SNAP. | Not Listed | #2084 |
| BASS | Pulse Sequence Design | Optimizes k-space sampling patterns for accelerated T1ρ mapping by adaptively reallocating measurements to information-rich regions. | Not Listed | #2085 |
| VSASL sequence | Pulse Sequence Design | Implemented a velocity-selective ASL (VSASL) pulse sequence with a 3D stack-of-spirals FLASH readout for quantitative measurement of foot perfusion. | Not Listed | #2086 |
| nnU-Net | Reconstruction | To segment the quadriceps across three MRI vendors. | Not Listed | #2093 |
| DualEnc-Rad-HL | Reconstruction | Automatic thigh muscle segmentation in MRI by integrating MRI and radiomics feature maps through self-/cross-attention fusion and hierarchical multi-stage supervision. | Not Listed | #2096 |
| SynthStrip | Analysis | Skull-stripping of brain images, particularly when spinal anatomy is present, for generating brain masks used in combination with spinal cord masks for B0 shimming optimization. | Not Listed | #2107 |
| Workflow Simulation | Simulation | Simulates MRI suite operations including patient pathways, staff activities, and resource utilization to evaluate the impact of different scheduling and protocol scenarios on patient throughput and environmental sustainability. | Not Listed | #2108 |
| MARIE | Simulation | Calculate electromagnetic fields of coils, with improvements to exploit GPU acceleration and increase accuracy. | Not Listed | #2112 |
| MaRGE GUI | Analysis | Controls the MRgFUS system, including image co-registration, calculation of required displacements for transducer positioning, and launching the US treatment. | Not Listed | #2113 |
| Nomogram model | Analysis | Diagnostic prediction model for myocardial fibrosis in HCM based on T1max, T1ρmax, and GLS, visualized through a nomogram. | Not Listed | #2115 |
| LRMC-DIP | Reconstruction | Incorporates non-rigid cardiac motion correction into a Deep Image Prior (DIP) reconstruction framework for cardiac MRF to improve motion robustness. | Not Listed | #2116 |
| Low-rank tensor (LRT) multitasking framework | Reconstruction | Reconstructs a motion-resolved T1/T2-weighted 5D dataset from a single acquisition of coronary sinus oximetry data. | Not Listed | #2124 |
| Automated baseline-referenced alignment algorithm | Analysis | Corrects frequency, phase, and baseline offsets in MRS spectra by referencing only stable metabolite regions to align spectra from separate scans while preserving dynamic spectral changes. | Not Listed | #2133 |
| U-Net model | Analysis | Automatically identify anatomical landmarks (MV leaflet attachments, TV leaflet attachments, left ventricular apex, and RV apex) in four-chamber cardiac MRI images. | Not Listed | #2146 |
| 3D U-Net model | Analysis | Automatic fiducial segmentation during image reconstruction on T1-weighted GRE prostate images. | Not Listed | #2149 |
| AI-driven pipeline | Analysis | To automatically localize femoral endpoints, prescribe in-plane EPI acquisitions, and extract femur length in real-time from fetal MRI datasets. | Not Listed | #2150 |
| None | Analysis | A geometric deep learning model was trained to predict 6-DoF motion parameters from in vivo k-space data. | Not Listed | #2152 |
| Transformer | Analysis | Estimating myelin water fraction (MWF) from multi-echo gradient-echo (mGRE) imaging data using a self-supervised learning approach. | Not Listed | #2153 |
| CNN-Transformer | Reconstruction | Robust CBF and ATT quantification from multi-delay pCASL with reduced number of repetitions, maintaining high accuracy and noise resilience. | Not Listed | #2154 |
| Deep Blind AIF | Analysis | Estimates the unsaturated arterial input function (AIF) directly from the saturated AIF and myocardial tissue signals using a convolutional neural network (CNN). | Not Listed | #2155 |
| None | Reconstruction | Implicit neural representation (INR) is optimized directly from undersampled k-space, regularized by a population-pretrained autoencoder prior, to achieve high accuracy and generalize across different sampling patterns. | Not Listed | #2156 |
| IR-Sync-Gate & Depth-Sync-Gate & Resp-Net | Analysis | The software extracts r-PPG signals from facial microvascular reflectance, segments the upper chest and abdomen in depth maps, and predicts respiratory and PPG gating signals using a fusion of these features and CNN embeddings. | Not Listed | #2159 |
| BasisREMY | Simulation | Automates MRS basis set generation directly from MRS data files, simplifying the process of creating study-specific basis sets for spectral fitting. | Not Listed | #2160 |
| jMRUI | Analysis | To process spectral data in the time domain, particularly for model-based fitting of phosphorus spectra acquired via ³¹P-MRS, including blood contamination and partial-saturation correction to quantify PCr/ATP ratios. | https://www.mrui.uab.es | #2165 |
| multidimensional diffusion MRI toolbox | Analysis | To compute DIVIDE parameters (total kurtosis (MKt), anisotropic and isotropic components (MKa, MKi), and microscopic fractional anisotropy (μFA)) from diffusion MRI data. | https://github.com/markus-nilsson/md-dmri | #2167 |
| Osprey | Analysis | Fully automated MRSI analysis including preprocessing, tissue composition mapping, and simultaneous modeling of both sides of the echo using a generalized LCM algorithm. | Not Listed | #2168 |
| PETPVC Toolbox | Analysis | Correct for partial volume effects in 23Na and 23Na_IR images. | Not Listed | #2176 |
| 3D-FAM | Pulse Sequence Design | Implemented a 3D spoiled gradient-echo (SGRE) pulse sequence with flip-angle modulation (FAM) for improved PDFF quantification in MRI. | Not Listed | #2195 |
| Free-Running Framework (FRF) | Reconstruction | Reconstructs free-running 3D radial MRI data into respiratory phases using compressed sensing for motion-resolved liver imaging. | Not Listed | #2211 |
| Pulseq | Pulse Sequence Design | Enables flexible and independent multiple sub-volume positioning within a single MRI sequence by applying unique RF and ADC phase offsets for each sub-volume without duplicating sequence blocks. | Not Listed | #2218 |
| Python | Analysis | Developed a machine learning pipeline using Python to predict PASAT scores from subcortical U-fiber FA features, incorporating feature selection, model training, and explainability analysis. | Not Listed | #2220 |
| LFACM | Reconstruction | To achieve segmentation of nine whole-body tissues/organs under few-shot learning scenario, enabling robust BCA from PDFF images with minimal manual annotation by integrating convolution into the active contour model’s analytical framework, replacing its traditional similarity function with learnable operators. | Not Listed | #2222 |
| PlaneCNN | Reconstruction | Directly prescribes eleven CMR planes without relying on intermediate landmark detection by formulating plane prescription as a semantic segmentation problem. | Not Listed | #2223 |
| BioTTA | Analysis | A two-stage unsupervised test-time adaptation framework that trains on source data and at inference minimizes entropy, contour length, and boundary gradient losses on target data while integrating atlas informed anatomical priors via registration for robust automatic fetal brain biometry on out-of-distribution target domains. | Not Listed | #2224 |
| UNet3+ | Segmentation | Automated left ventricular blood pool and myocardium segmentation in mouse cardiac MRI images. | https://github.com/mrphys/Open-Source_Pre-Clinical_Segmentation | #2225 |
| None | Segmentation | Automated neonatal choroid plexus segmentation using a 3D U-Net with conditional instance normalization layers modulated by sinusoidal age encoding, with a rule-based controller to identify age-sensitive layers. | Not Listed | #2226 |
| Anatomy-Aware Dynamic Causal Augmentation (AADCA) | Analysis | A framework for single-source domain generalization in medical image segmentation that integrates anatomical semantic priors with dynamic causal intervention to achieve precise, adaptive, and clinically compatible segmentation. | Not Listed | #2231 |
| uAI Research Portal (uRP) | Reconstruction | Automatically segment bilateral hippocampus and amygdala based on a VB-Net deep learning model. | Not Listed | #2238 |
| pyradiomics | Analysis | To perform feature extraction from CE-T1 MRI images of meningioma tumors. | Not Listed | #2242 |
| OpenMRF | Reconstruction | OpenMRF is an open-source framework that enables reproducible magnetic resonance fingerprinting by unifying sequence components, automated simulation and dictionary generation, state-of-the-art reconstruction, and robust metadata handling. | Not Listed | #2249 |
| CRB-MyelinQalas | Pulse Sequence Design | Optimizes TEs and FAs in the QALAS pulse sequence using a self-supervised multilayer perceptron (MLP) trained to minimize a Cramér–Rao bound-based loss, to improve the accuracy of multiparametric estimation, including MWF, and reduce scan time. | https://github.com/shizhuo-li/CRB-MyelinQalas | #2251 |
| QQ-S | Analysis | To estimate oxygen extraction fraction (OEF) by modeling the temporal evolution of magnitude signals using a deep learning model incorporating Convolutional Long Short-Term Memory (ConvLSTM) modules. | Not Listed | #2252 |
| AI-based eddy current correction (ECC) module | Reconstruction | Correct eddy current-induced distortions in self-gating signals of TrueFISP Multitasking sequence using a conditional variational autoencoder (CVAE). | Not Listed | #2254 |
| Bloch simulation | Simulation | To compare signal evolution under two sequence configurations (T2 Prep→BB→ACQ and BB→T2 Prep→ACQ) for different tissue compartments with varying T1/T2 values. | Not Listed | #2283 |
| SQM brain atlas | Analysis | To provide an improved brain atlas for more accurate delineation of structures in the squirrel monkey brain, enabling unbiased quantitation of regional brain volumes from T2*-w images. | Not Listed | #2300 |
| Perinatal Lamb Brain MRI Template and Atlas | Analysis | Provides a standardized reference space for automated spatial normalization, tissue segmentation, and ROI-based analyses in ovine perinatal neurodevelopmental studies. | Not Listed | #2303 |
| FastHenry | Simulation | Simulating inductance and resistance of the 4-layer spiral coil. | Not Listed | #2306 |
| ADAS 3D | Analysis | Assess fibrosis using 3D high resolution LGE sequence. | Not Listed | #2307 |
| MCR-MWI-Chisep | Reconstruction | An improved myelin water imaging (MWI) framework that enhances myelin water fraction estimation while enabling simultaneous iron quantification. | Not Listed | #2313 |
| χ-sepnet | Analysis | Decomposes susceptibility into positive (χ⁺) and negative (χ⁻) components using reconstructed local field, QSM, and R₂* maps as inputs. | Not Listed | #2315 |
| QSim | Reconstruction | QSim is a novel OEF mapping approach that integrates QSM of phase signal modeling and Monte-Carlo simulation of magnitude signal evolution to improve OEF accuracy by incorporating diffusion effects and realistic vascular structure. | Not Listed | #2316 |
| Dipole-lets | Reconstruction / Analysis | An undecimated multiscale transform that decomposes phase images by radial scale and proximity to the magic cone for QSM reconstruction and analysis, separating dipolar and non-dipolar content to reduce streaking artifacts. | http:/gitlab.com/cmilovic/FANSI-toolbox | #2319 |
| VesselBoost | Reconstruction | VesselBoost is a deep-learning pipeline used for vessel segmentation of post-mortem midbrain MRI, incorporating intensity-based augmentation to handle T2* contrast variability. | Not Listed | #2321 |
| VesselMass | Analysis | Measure vessel wall thickness from MRI images. | Not Listed | #2328 |
| Sub-space low-rank reconstruction framework | Reconstruction | Reconstruct 4D MRA datasets with 10 radial spokes per frame, corresponding to a temporal resolution of 76 ms/frame. | Not Listed | #2329 |
| LSRDN | Reconstruction | Enhancing the spatial resolution of rapidly acquired low-resolution 3D TOF-MRA images using a ladder-shaped residual dense network. | Not Listed | #2330 |
| Hybrid SVM-GCN Framework | Analysis | Predict cerebrovascular age from cerebral artery features by using a two-stage hybrid Support Vector Machine-Graph Convolutional Network model. | Not Listed | #2339 |
| ITK-SNAP | Analysis | Delineate the whole tumor’s volume based on a 55Hz-OGSE images. | http://www.itksnap.org/ | #2356 |
| Longitudinal-Aware CLIP framework | Analysis | Predict pathological complete response (pCR) to neoadjuvant chemotherapy (NAC) in breast cancer patients using longitudinal DCE-MRI data and a biomedical foundation model (biomedCLIP), incorporating a temporal encoding mechanism and a longitudinal-aware module to capture lesion changes across time points. | Not Listed | #2360 |
| nnDetection | Analysis | Predicts bounding boxes and confidence scores for breast lesions with restricted diffusion in whole-body DWI. | Not Listed | #2363 |
| SPINet | Reconstruction | Reconstructs complete MRF trajectories from highly compressed acquisitions by discovering signals under novel acquisition settings. | Not Listed | #2382 |
| CS Reconstruction Algorithm | Reconstruction | Reconstructs 4D image data from undersampled k-space data using a wavelet-based compressed sensing (CS) algorithm with multi-dimensional regularization. | Not Listed | #2403 |
| LayNii toolbox | Analysis | Create inner and outer cortical layer masks using the equi-volume laminar approach. | Not Listed | #2408 |
| SVRTK | Reconstruction | Solve a minimization problem to integrate images from different slices and orientations into a high-resolution isotropic volume, accounting for motion and encoding. | Not Listed | #2409 |
| FIRE | Data Management | Real-time export and processing of MRI acquisitions on a high-performance computer for an AI-driven fetal MRI pipeline. | Not Listed | #2410 |
| NeSVoR | Reconstruction | Reconstructs 3D fetal MRI volumes from stacks of slices, correcting for fetal motion using a neural network approach. | Not Listed | #2419 |
| Pulseq-based implementation of EPTI | Pulse Sequence Design | Implements the Echo Planar Time-Resolved Imaging (EPTI) sequence using the Pulseq framework for flexible adaptation and use in neonatal brain imaging. | Not Listed | #2421 |
| auto SVRTK toolbox | Reconstruction | Reconstruct 3D fetal lung images from motion-corrupted 2D T2-weighted image stacks and perform automated lung volumetry. | https://github.com/SVRTK/auto-proc-svrtk | #2424 |
| BOUNTI | Analysis | Automatic Segmentation of cortical grey matter (GM), deep GM and white matter (WM) from T2-weighted images. | Not Listed | #2426 |
| DL-MorphoBox | Analysis | Segment T1w MRI volumes into 55 separate brain regions and return structure volumes compared to age-/sex-matched reference ranges by means of Z-Scores. | Not Listed | #2428 |
| SE(3)-ResNet34 | Analysis | Brain age prediction using white matter fiber orientation distributions (WM-FODs) with a rotation-equivariant convolutional neural network. | Not Listed | #2436 |
| LPCNN | Reconstruction | Dipole inversion in QSM processing pipeline. | Not Listed | #2437 |
| SegMambaBET | Reconstruction | Extracts the brain from T1-weighted MRI images using a patch-sampled 3D residual encoder–decoder with Mamba state-space blocks and skip connections. | Not Listed | #2439 |
| OXSA pipeline | Analysis | Process spectral data from deuterium metabolic imaging (DMI) in MATLAB, including receive channel combination, spectral fitting with AMARES, and quality control based on CRLB and SNR. | Not Listed | #2441 |
| µTVB | Simulation | Integrates microstructure-informed connectomes and subject-specific conduction velocities into The Virtual Brain (TVB) framework to improve the physiological consistency of brain dynamic simulations. | Not Listed | #2444 |
| nnU-Net | Segmentation | Trains a ventricle segmentation network and reuses its encoder as an MRI feature extractor. | Not Listed | #2446 |
| Custom 2D U-Net ensemble | Segmentation | Combines a 2D U-Net ensemble with WMH-informed negative masking to harmonize T1- and FLAIR-weighted MRI to distinguish infarcts from vascular hyperintensities. | Not Listed | #2449 |
| Rectified Flow model | Reconstruction | Synthesizes T1ce and FLAIR images from T1- and T2-weighted MRI inputs using a rectified flow model to reduce scanning time and eliminate the need for contrast agent injections. | Not Listed | #2450 |
| nnU-Net | Reconstruction | To automatically segment white matter in dSIR UHC brain images. | Not Listed | #2453 |
| Nigranet | Reconstruction | Automatic segmentation of the substantia nigra pars compacta (SNpc) in neuromelanin MRI images using a U-Net model. | Not Listed | #2457 |
| BM4D | Analysis | Denoises 23Na-MRI images to improve SNR while preserving image quality and accuracy of total sodium concentration. | https://webpages.tuni.fi/foi/GCF-BM3D/ | #2474 |
| BART toolbox | Reconstruction | Reconstruction used a subspace-constrained pipeline. | Not Listed | #2476 |
| None | Reconstruction | Anomaly detection in female pelvic MRI using a residual variational autoencoder for real-time reconstruction and deviation mapping. | Not Listed | #2482 |
| tMPPCA | Analysis | Denoise multi-band fMRI data using tensor-based Marchenko–Pastur Principal Component Analysis (tMPPCA) to suppress noise while preserving low-frequency fluctuations. | Not Listed | #2485 |
| STASIS_Control | Analysis | Provides a graphical user interface written in Python for easy control and adaptation of the open-source RF exposure system, including pulse design, system calibration, and settings adjustment. | https://github.com/sOrzada/STASIS_Control | #2490 |
| WALINET & PHIVE (integrated within Siemens FIRE framework) | Reconstruction | Lipid/water removal (WALINET) and spectral fitting (PHIVE) in an online 3D MRSI reconstruction pipeline. | Not Listed | #2505 |
| ECCENTRIC 3D FID-MRSI | Pulse Sequence Design | Optimized for 7T fMRSI, the sequence enables high-resolution, whole-brain mapping of task-induced metabolic changes with near-isotropic 3 mm resolution. | Not Listed | #2507 |
| TriPeak optimal combination | Reconstruction | A bias-free, optimal-SNR coil combination method for X-nuclear MRSI that performs two iterative combinations per voxel based on Roemer’s optimal SNR framework in the spectral domain. | Not Listed | #2509 |
| OmniShim | Analysis | Provides better B0 shimming compared to vendor methods, particularly in challenging brain regions. | Not Listed | #2512 |
| Csar-EPSI | Pulse Sequence Design | Implemented a chemical-shift-selective adiabatic refocusing EPSI(Csar-EPSI) sequence with variable TR, TE and FA for simultaneous and rapid mapping of metabolite T₁, T₂ and absolute concentration with MRSI. | Not Listed | #2514 |
| MC BTS (Multi-Component BTS) | Reconstruction | Three-pool MC BTS parameter estimation from multi-echo magnitude and phase images, combined with flip-angle maps, using a three-pool Bloch-McConnell equation and two-pool MT parameters as priors. | Not Listed | #2524 |
| 3D SPARCS | Reconstruction | Reconstructs cine images from steady-state spiral data using L1-SENSE reconstruction. | Not Listed | #2526 |
| CMRperf | Simulation | Simulates cardiac perfusion MRI data with realistic motion and undersampling artifacts for sequence, reconstruction, and quantification validation and optimization. | Not Listed | #2528 |
| MST | Analysis | Predict hematocrit levels from cardiac MRI data using a multi-stage deep-learning model to derive synthetic extracellular volume without blood sampling. | Not Listed | #2530 |
| DeepSENC | Analysis | To quantify liver strain from cardiac-induced motion in SENC-MRI images for liver fibrosis assessment using a supervised deep learning framework. | Not Listed | #2534 |
| PUDIP-Flow | Reconstruction | An unsupervised and segmentation-free phase unwrapping method for aortic and cerebrovascular 4D flow MRI. | https://github.com/AssociatedPrimeIdeal/PUDIP-Flow | #2536 |
| DeepStrain | Analysis | Automated computation of global radial and circumferential strain in single-ventricle hearts from short-axis CMR using a combination of segmentation and motion estimation models. | Not Listed | #2537 |
| In-house developed python scripts | Reconstruction | Reconstruct parametric maps from MRI images. | Not Listed | #2541 |
| Spectral-spatial fusion network | Analysis | Accurately estimates GAG concentration in lumbar intervertebral discs from gagCEST data by fusing spectral and spatial features. | Not Listed | #2542 |
| Frequency Feature Extractor (FFE) | Analysis | Extracts inter-offset dependencies and enhances frequency-aware representations by employing a 1×1 convolution, a squeeze-and-excitation (SE) attention block, and a channel-wise MLP to improve amide contrast mapping. | Not Listed | #2544 |
| CRISTINA-3D-Pulseq | Pulse Sequence Design | Implements the 3D CRISTINA sequence for simultaneous single and triple quantum sodium imaging in the Pulseq open-sequence framework. | https://github.com/Computer-Assisted-Clinical-Medicine/CRISTINA-3D-Pulseq | #2549 |
| B1-Mapping-Seq | Pulse Sequence Design | Provides Pulseq sequences for Bloch-Siegert and double flip angle B1 mapping for sodium MRI, with optional TV-accelerated reconstruction. | https://github.com/Computer-Assisted-Clinical-Medicine/B1-Mapping-Seq | #2559 |
| Geometric Filter Optimization (GFO) | Pulse Sequence Design | Generates optimal sets of rotations for encoding objects with arbitrary shapes in dMRI, minimizing the rotation variance in the powder-averaged signal. | Not Listed | #2566 |
| PCHIP | Analysis | Interpolates sparse b-value diffusion MRI data using piecewise cubic Hermite interpolating polynomials to enforce monotonicity and smoothness, improving parameter fitting accuracy for IVIM. | Not Listed | #2567 |
| PyMR | Data Management | Operates and controls the CIERMag system, an 8Tx/8Rx Digital Magnetic Resonance Spectrometer (DMRS). | Not Listed | #2568 |
| Triton EPG | Simulation | A high-performance, Pytorch-compatible single-kernel EPG implementation for 3D MRF dictionary simulation, seamlessly accessible from Python. | Not Listed | #2569 |
| Bloch simulation package | Simulation | To model signal evolution in multi-echo spin echo (MESE) sequences and enable detailed slice profile analysis. | Not Listed | #2570 |
| CaPTk (Cancer Imaging Phenomics Toolkit) | Analysis | Calculate radiological tumor characteristics and quantitative brain reserve metrics. | Not Listed | #2579 |
| DL-CSDNet | Analysis | Predict voxel-wise cell volume fractions and cell-size distributions from normalized GESFIDE signals using a fully connected neural network trained on simulated 3D microstructures. | Not Listed | #2582 |
| pyfeats | Analysis | Extract Haralick texture features from 3D gray-level co-occurrence matrices (GLCMs) alongside volume measures. | Not Listed | #2596 |
| nnU-Net v2 | Segmentation | Automatically segments enhancing tumor parenchyma and peritumoral edema in MRI images of primary central nervous system lymphoma (PCNSL). | Not Listed | #2600 |
| AI-based pharmacokinetic parametric mapping | Analysis | Estimates quantitative kinetic parameters (Ktrans, Kep, and Ve) with the extended Tofts model to generate voxel-wise quantitative maps from DCE-MRI data. | Not Listed | #2601 |
| PyRadiomics | Analysis | Extracts quantitative features from medical images, following IBSI compliance guidelines, including T1 relaxation metrics, first-order statistics, texture features, and reference tissue metrics. | Not Listed | #2604 |
| SPURS | Reconstruction | Increase the resolution of zerofilled data via SPURS from the MEDI Toolbox. | Not Listed | #2607 |
| ARMA framework | Analysis | To estimate spectral frequencies and amplitudes from multi-gradient echo (mGRE) MRI signals for spectral decomposition and profiling fatty acid compositions. | Not Listed | #2608 |
| Random Forest | Analysis | Predict NASH severity using MRI features (T1, T2, FA, MD, AD, RD). | Not Listed | #2610 |
| MR AutoProcessor Body Composition Analysis v1.0 | Analysis | Automated analysis of whole-body MRI scans to quantify body composition metrics such as fat and muscle distribution and organ volumetry. | Not Listed | #2619 |
| QQ-NET | Reconstruction | Facilitates OEF (oxygen extraction fraction) calculations from QSM and qBOLD imaging using a deep learning model, reducing computation time and preserving accuracy. | Not Listed | #2622 |
| cuDIMOT | Reconstruction | Fits NODDI data to obtain NDI, ODI, ISOVF maps. | Not Listed | #2623 |
| 3D U-Net | Reconstruction | Performs fast MRI segmentation and extracts cognition-sensitive features to predict disease stage, tauopathy seeding, and StaND parameters from baseline MRI, demographics, and plasma biomarkers. | Not Listed | #2635 |
| NODDI MATLAB Toolbox | Analysis | Estimates NODDI (Neurite Orientation Dispersion and Density Imaging) metrics from diffusion MRI data, separating intra- and extra-cellular water compartments for microstructural characterization. | Not Listed | #2636 |
| pyAMARES and CSIgui toolbox | Reconstruction | Process spectra using manual phasing, baseline correction, and advanced AMARES fitting. | Not Listed | #2640 |
| APART-QSM | Reconstruction | Separates paramagnetic and diamagnetic sources at a sub-voxel level from QSM images to derive paramagnetic (iron-related) and diamagnetic (protein-related) components. | Not Listed | #2641 |
| FreeSurfer | Analysis | Automatically segment thalamic volume from T1-weighted MRI images. | Not Listed | #2644 |
| MRtrix3 | Reconstruction | To obtain ADC maps from diffusion-weighted images. | Not Listed | #2650 |
| OREX | Pulse Sequence Design | A novel omni-directional REX sequence (OREX) acquires all REX-induced components while maintaining an improved signal-to-noise ratio and enables accelerated acquisition via magnetization recycling. | Not Listed | #2656 |
| CoilGen | Pulse Sequence Design | Design x-gradient coil for phase-encoding. | Not Listed | #2657 |
| DINOv2-attention | Analysis | Automated headache subtyping from 3D brain MRI using a slice-level attention aggregation network to extract patient-level embeddings from T1-weighted MRI, followed by four binary classifiers integrated via a hierarchical decision-tree. | Not Listed | #2662 |
| Random Forest Classifier | Analysis | Predicts manual QC ratings of ultra-low-field MRI scans based on MRIQC-derived image quality metrics. | Not Listed | #2664 |
| Bespoke platform | Guidance | Guiding cryoprobe placement intraprocedurally to achieve the probe placement necessary to cover a prescribed ablation region. | Not Listed | #2673 |
| automated MRI-based segmentation algorithm | Analysis | To quantify visceral adipose tissue volume from MRI images. | Not Listed | #2676 |
| RUNet | Reconstruction | Super-resolution of prostate ADC maps using a UNet architecture, optionally incorporating feature injection from high-resolution T2W images. | Not Listed | #2684 |
| NeuroCombat | Analysis | Harmonize data from different cohorts to minimize site and scanner-related variance. | https://github.com/Jfortin1/neuroCombat | #2688 |
| AgeGate Dual-Pathway Model | Analysis | Predicts pediatric brain age by simultaneously modeling myelination- and gyrification-sensitive brain age, using an age-dependent weight gate to capture the evolving balance between these processes. | Not Listed | #2690 |
| ECA-LDM | Reconstruction | Integrate a latent diffusion model with efficient channel attention mechanism for MRI super-resolution, achieving comparable performance to state-of-the-art models with reduced computational resource requirements. | Not Listed | #2699 |
| Motion-resolved Gaussian splatting | Reconstruction | A novel framework for dynamic MRI reconstruction using 2D Gaussian splatting that learns an explicit spatiotemporal representation of the underlying anatomy and models anatomical motion for fast and high-quality cardiac MRI reconstruction from undersampled data. | Not Listed | #2701 |
| MRI-ESRGAN | Reconstruction | Enhances through-plane resolution in multi-slice MRI reformatting by refining a pre-trained ESRGAN network through transfer learning. | Not Listed | #2702 |
| None | Reconstruction | Reconstructs multi-planar 2D MRI images into submillimeter isotropic 3D volumes using a super-resolution algorithm. | Not Listed | #2703 |
| CONSIS-Net | Reconstruction | A lightweight unrolled neural network that combines a transformer-based super-resolution module with iterative data-consistency steps operating in both k-space and image domains for low-field knee MRI reconstruction. | Not Listed | #2704 |
| None | Reconstruction | Quantify in vivo cartilage stiffness and stiffness heterogeneity using DENSE MRI. | Not Listed | #2718 |
| Deep Kernel Learning Gaussian Process (DKL-GP) | Analysis | Predict cartilage proteoglycan (PG) content from MR Fingerprinting (MRF) data by combining convolutional neural networks (CNNs) for feature extraction with a sparse variational Gaussian Process Regression (GPR) layer. | Not Listed | #2719 |
| EPG dictionary-based fitting | Reconstruction | Generate T₂ maps using dictionary-based Extended Phase Graph (EPG) fitting accounting for slice profiles and B1+ inhomogeneities. | Not Listed | #2728 |
| Open-MOLLI | Pulse Sequence Design | An open-source myocardial T1-mapping sequence designed to provide consistent and reproducible cardiac T1 values across different MRI vendors. | https://github.com/asgaspar/OpenMOLLI | #2732 |
| Temporal Convolutional Network (TCN) | Simulation | Simulate gradient waveforms in low-field MRI systems to enable their efficient optimization and improve image quality. | Not Listed | #2738 |
| MR-IQ | Analysis | A reference-free MR image-quality metric that integrates multiple normalized sub-metrics into a single composite score for reproducible, interpretable, and quantitative evaluation of MRI quality. | Not Listed | #2740 |
| Sepia toolbox | Reconstruction | Reconstruct QSM images using phase unwrapping, background field removal, and dipole inversion. | https://github.com/kschan0214/sepia | #2744 |
| MRS4Brain toolbox | Analysis | Process 1H-FID-MRSI data for neurometabolic mapping. | Not Listed | #2745 |
| Spinal cord toolbox | Analysis | Used to obtain a spinal cord mask for post-processing of MRI data. | Not Listed | #2746 |
| GRETNA | Analysis | Compute graph-theoretical metrics, including small-worldness (σ), clustering coefficient (C), characteristic path length (L), and global/local efficiency, to assess network topology. | Not Listed | #2747 |
| QSM-PLAQUE Aβ Detector | Analysis | Decomposes a high-resolution QSM MRI image into background, signal, and noise components to detect Aβ plaques. | Not Listed | #2748 |
| AlignedSENSE | Reconstruction | Joint iterative image reconstruction and motion estimation across temporally grouped k-space shots. | Not Listed | #2749 |
| Morphometric Inverse Divergence | Analysis | Calculates structural difference between brain regions based on statistical comparison of multiple morphometric features. | https://github.com/isebenius/MIND | #2750 |
| pediatric-data-analysis | Analysis | Analyze pediatric spinal cord MRI data to establish normative morphometric values and investigate age-related structural variations. | https://github.com/sct-pipeline/pediatric-data-analysis | #2751 |
| APART-QSM | Reconstruction | Separates paramagnetic and diamagnetic susceptibility on a voxel-wise basis by incorporating transverse relaxation constraints. | Not Listed | #2752 |
| Osprey | Analysis | Process HERMES spectra to generate GSH-edited, GABA-edited, and sum (with glutamate, Glu, co-edited) spectra, and generate voxel masks for tissue fraction calculation. | Not Listed | #2753 |
| Rank-1 Spherical Harmonics representation | Reconstruction | To reconstruct diffusion MRI data from sparse multidimensional acquisitions by representing the signal at each Spherical Harmonics order as a rank-1 matrix, enabling clinically feasible microstructure mapping. | Not Listed | #2760 |
| RMS toolbox | Simulation | Performs Monte Carlo simulations of diffusion in extracellular space within micro-geometries composed of randomly oriented cylinders and packed spheres. | Not Listed | #2762 |
| PyTorch | Analysis | Compute Jacobians for QTI parameter estimation using automatic differentiation to enable gradient-based optimization of b-tensor sampling schemes. | Not Listed | #2763 |
| MRtrix3 | Analysis | DTI preprocessing (denoising, Gibbs ringing removal, eddy current correction) and calculation of quantitative maps. | Not Listed | #2765 |
| qMRLab | Analysis | Fitting T1 maps from MP2RAGE images. | Not Listed | #2766 |
| FeAture Explorer | Analysis | Extract radiomics features from the brain-to-tumor interface region in accordance with the IBSI guidelines. | Not Listed | #2787 |
| Custom Attention module | Analysis | Enhance structurally informative regions in the concatenated encoded features using a module based on CBAM. | Not Listed | #2789 |
| Super-Resolution Net | Reconstruction | Remove ringing artifacts, replace the traditional zero-filling strategy used to increase the matrix size, and enhance image sharpness in 2D Cartesian acquisitions. | Not Listed | #2792 |
| PyRadiomics | Analysis | Extract radiomics features from medical images to quantify tumor heterogeneity. | Not Listed | #2795 |
| pediatric-data-analysis | Analysis | Provides a processing pipeline for extracting DTI metrics from pediatric cervical spinal cord MRI data and includes an open-source database of normative DTI values. | https://github.com/sct-pipeline/pediatric-data-analysis | #2801 |
| Magpylib | Simulation | Simulating and optimizing the magnetic field and homogeneity of Halbach array based MRI magnets. | Not Listed | #2802 |
| Custom simulated annealing optimizer | Optimization | Optimizes the design of permanent magnet based low-field MRI magnets and shim arrays by integrating GPU-accelerated field pre-computation with a custom simulated annealing algorithm. | Not Listed | #2803 |
| Low-Field MRI Readiness Assessment Framework | Analysis | Evaluate site preparedness and inform future hardware adaptation for resilient LF-MRI deployment in low-resource settings. | Not Listed | #2805 |
| Custom GA | Simulation | Optimizes Halbach magnet configurations by using a genetic algorithm to determine the optimal ring combination from a predefined dataset of Halbach array rings, minimizing field inhomogeneity and targeting a specific magnetic field strength. | Not Listed | #2806 |
| KiCad 9.0 | Pulse Sequence Design | Used to design flexible PCBs for a slotted Faraday shield to be placed inside an RF coil for EMI mitigation. | Not Listed | #2813 |
| MagTetris+ | Simulation | A hybrid simulator for both permanent magnets and ferromagnetic materials that has high memory efficiency and rapid computation for applications such as passive shimming and iron yoke designs for low-field portable MRI. | Not Listed | #2815 |
| Pymoo | Analysis | Employed to optimize the geometric parameters of the magnet using the trained deep learning simulator. | Not Listed | #2827 |
| MacroQA | Analysis | Automates the complete ACR MRI phantom quality assurance test suite, providing a transparent, time-efficient, and accessible alternative for standardized MRI quality monitoring. | Not Listed | #2829 |
| FIRE | Reconstruction | Packages and deploys inline the reconstruction and processing of 3D non-Cartesian high-resolution whole-brain MRSI acquisitions to generate spectroscopy and metabolic maps image DICOMs. | Not Listed | #2830 |
| MaRCoS | Pulse Sequence Design | Updated control software to improve system calibration and usability. | Not Listed | #2831 |
| QuIDBBIDS | Analysis | A BIDS-app that automatically computes quantitative MRI biomarkers from diverse datasets, adhering to the BIDS standard to ensure reproducibility and simplify large population studies. | github.com/Donders-Institute/quidbbids | #2832 |
| seeVieweR | Visualization | A MATLAB App Designer tool for interactive 3D/4D visualization of volumetric MRI data, enabling real-time adjustment of transparency, masking, colormaps, and rendering modes across multiple overlays, with automatic alignment of overlay datasets and reproducible figure generation. | https://github.com/abhogal-lab/seeVieweR | #2833 |
| LN2_FRISGO | Reconstruction | Corrects 7T EPI artifacts on reconstructed NIfTI time-series data using dual polarity EPI acquisitions. | [github.com/layerfMRI/LAYNII (devel branch)](github.com/layerfMRI/LAYNII (devel branch)) | #2834 |
| Bloch simulator | Simulation | Converts Pulseq (.seq) files into executable operator lists to accurately model arbitrary RF pulses, including spin-lock states and rotating frame relaxation (T1ρ/T2ρ). | Not Listed | #2835 |
| MRpro | Reconstruction | Reconstructs data acquired using the Nexus console. | Not Listed | #2836 |
| MR RawDeface | Data Management | Anonymize/de-identify raw, 3D multi-channel brain k-space datasets for privacy preserving sharing without fundamentally altering the raw k-space signal characteristics. | Not Listed | #2838 |
| MR-MOTUS | Reconstruction | Reconstructs time-resolved 3D motion fields and motion-corrected images from free-running MRI data, enabling high temporal resolution 3D speech imaging. | Not Listed | #2839 |
| QMRITools | Analysis | Create muscle segmentation masks from 3D DIXON scans. | Not Listed | #2843 |
| FireVoxel | Analysis | Manually segment layer-by-layer bilateral hippocampal regions as regions of interest (ROIs) in the axial images of three MRI sequences (T2WI/FLAIR/ADC). | https://www.firevoxel.org/ | #2854 |
| NN | Reconstruction | Correct B1 inhomogeneities in CEST MRI data using a feed-forward neural network trained to emulate a multi-power reference method. | Not Listed | #2865 |
| MTRsigm | Analysis | Selectively suppresses fluid signal in APTw MRI while preserving ssMT contrast using a sigmoid-based weighting function. | Not Listed | #2867 |
| None | Reconstruction | Calibration of gradient coil imperfections, including eddy currents or gradient coupling induced imperfections, using paired no-wave and wave calibration data acquired across a few central k-space lines. | Not Listed | #2879 |
| T2DL | Reconstruction | Volumetric deep-learning super-resolution reconstruction of 3D T2-weighted breast MRI images from low-resolution acquisitions, reducing scan time while improving image quality. | Not Listed | #2891 |
| Self-Supervised Learning Model | Reconstruction | Directly reconstructs prostate DWI from raw undersampled k-space data while reducing noise using a self-supervised learning approach. | Not Listed | #2893 |
| pGA-Movienet | Reconstruction | Reconstructs motion-resolved 4D MRI from aliased images by exploiting time-coil-space information and using both pre- and post-contrast data. | Not Listed | #2895 |
| CNN-based deep learning reconstruction model | Reconstruction | Reconstructs 3D abdominal MR images from radially under-sampled MR data, mitigating streaking artifacts and improving spatial resolution. | Not Listed | #2896 |
| LoSP-Prompt | Reconstruction | Reconstructs high-resolution multi-organ diffusion-weighted images (DWI) by decoupling high-order locally smooth phase of specific organ and the low-order locally smooth phase of other organs, using a 1D low-rank optimization method and a prompt learning technique to improve robustness. | https://arxiv.org/abs/2510.15400 | #2903 |
| Habbita Cluster model | Analysis | A clustering framework that leverages radiomic superpixels and spatial interaction matrices to decode spatial heterogeneity in acute ischemic stroke (AIS) using multimodal MRI. | Not Listed | #2910 |
| PyRadiomics | Analysis | To extract radiomics features from regions of interest in diffusion weighted imaging (DWI) data. | Not Listed | #2913 |
| Prior-Guided Denoising Diffusion Model (PGDDM) | Reconstruction | Synthesizes high-fidelity time-of-flight magnetic resonance angiography (TOF-MRA) and T1-weighted VWI from a single contrast-enhanced T1-VWI input by incorporating anatomical priors into the diffusion process. | Not Listed | #2914 |
| 3DViT | Analysis | Process MRI data to extract relevant latent features for predicting cardiovascular disease risk. | Not Listed | #2915 |
| CESTQuant | Analysis | A multi-pool quantitative CEST analysis platform that integrates Lorentzian decomposition, Rex analysis, and physical modeling for absolute quantification of five metabolites (APT, NOE-lipid/protein, NOE-fatty acid, Cr, MT) from CEST-MRI data. | Not Listed | #2918 |
| RapidTide | Analysis | Generate hemodynamic lag (LAG) maps from BOLD images. | Not Listed | #2919 |
| DifSim | Simulation | A monte carlo dMRI simulation platform used to simulate dMRI experiments for each model using spin echo, stimulated echo, and dPFG pulse sequences. | Not Listed | #2922 |
| MRtrix3 | Analysis | Compute DTI metrics (Fractional Anisotropy (FA), Mean Diffusivity (MD), Radial Diffusivity (RD), Axial Diffusivity (AD)) | Not Listed | #2926 |
| Deep Anatomical Federated Network (DAFNE) | Analysis | Automatically segments Dixon and MESE sequences into 11 regions of interest (ROIs) per each thigh. | Not Listed | #2928 |
| pulseq | Pulse Sequence Design | Development of a CINE 4D flow sequence for time-resolved phase-contrast MRI to capture forearm muscle motion during voluntary and NMES-evoked grip. | Not Listed | #2929 |
| RPBM | Analysis | Fits radial diffusivities per diffusion time to estimate per-muscle membrane permeabilities and fibre diameters from diffusion-tensor-MRI data. | https://github.com/NYU-DiffusionMRI/RPBM | #2934 |
| TSG-DM | Analysis | Predicts structural connectivity (SC) from functional connectivity (FC) while preserving topological consistency and handling both positive and negative FC. | Not Listed | #2952 |
| 3D-CNN | Analysis | Rapidly predict patient-specific in-situ peak 1 g-averaged SAR for pedicle screw systems at 1.5 T using a multi-channel 3D convolutional neural network. | Not Listed | #2957 |
| Transformer Network | Reconstruction | Predicts EPG signal evolutions from variable timing inputs to generate T1-T2 mapping dictionaries for cardiac MRI. | Not Listed | #2959 |
| Learnable Contrast Generation Module (LCM) | Reconstruction | The LCM receives quantitative SynMRI maps (T1, T2, PD) and synthesizes standard MRI contrasts using differentiable MRI contrast equations and learnable contrast parameters. | Not Listed | #2964 |
| SAMER | Reconstruction | Retrospective motion correction using a rapid low-resolution scout and repeated acquisition of motion-guidance lines to decouple motion estimation from motion-mitigated reconstruction. | Not Listed | #2991 |
| FIRE | Reconstruction | Prototype Siemens Framework for Image Reconstruction Environments. | Not Listed | #2992 |
| retroMoCoBox | Reconstruction | Corrects for motion artefacts using a non-uniform FFT (nuFFT) operator with corrected k-space trajectories after applying the GRAPPA method. | Not Listed | #2993 |
| ZS-SSL (Zero-Shot Self-Supervised Learning) | Reconstruction | Reconstructs motion-resolved abdominal MRI from highly undersampled radial data without external training data using a network with channel attention and joint motion-state training. | Not Listed | #2997 |
| ADMM | Reconstruction | Jointly reconstructs a motion-corrected image and motion fields by incorporating motion vector field estimates into the forward model to map the reference gate back to the gated k-space data, enforcing data-consistency. | Not Listed | #2998 |
| PulPy | Pulse Sequence Design | To program a 10 ms Shinnar-Le Roux pulse with time bandwidth 8 for refocusing in the BURST-MRF pulse sequence. | Not Listed | #3007 |
| TorchKbNufft | Reconstruction | Implements the Toeplitz-NUFFT linearization for Non-Uniform Fast Fourier Transform (NUFFT) using interpolation weights. | Not Listed | #3008 |
| SPUR-iG | Reconstruction | A fully 3D unrolled reconstruction framework for Magnetic Resonance Fingerprinting (MRF) that combines implicit-gridding-based data-consistency with a three-stage training scheme to enable efficient training of such models. | Not Listed | #3009 |
| mrftools | Reconstruction | Executes pattern matching with dynamically-calculated dictionaries and B1+ correction to generate T1 and T2 maps from raw 4D MRF data. | Not Listed | #3010 |
| DIP-MRF | Reconstruction | Reconstructs T1 and T2 maps from MRF data using a Deep Image Prior approach that combines low-rank subspace modeling with zero-shot deep learning. | Not Listed | #3011 |
| OpenMRF | Pulse Sequence Design | A vendor-agnostic framework for sharing and testing cMRF sequences across different MRI systems. | Not Listed | #3012 |
| fidall | Pulse Sequence Design | Implemented the customized spectrally-selective RF pulses and 3D stack-of-spiral bSSFP architecture for HP-MRF sequence design. | Not Listed | #3013 |
| UBNAno | Simulation | Simulates pathological tissue on healthy brains by altering T1/T2 relaxation values and modeling realistic lesion shapes for brain tumor detection. | Not Listed | #3016 |
| Open-SPEN | Pulse Sequence Design | Implemented and validated an open-source, multi-slice diffusion-weighted SPEN-SE-EPI sequence in the Pulseq framework, accompanied by educational code illustrating the SPEN forward operator derivation and reconstruction process. | https://github.com/andih98/Open-SPEN.git | #3017 |
| MRS4Brain toolbox | Reconstruction / Analysis | Streamline reconstruction and processing of dMRSI data, including reconstruction with SPIRiT, k-space spatial filtering, phase correction, water residual removal, LCModel fitting, and quality control. | Not Listed | #3023 |
| GrOpt | Pulse Sequence Design | Solves a constraint-optimized diffusion encoding gradient waveform design problem that leverages full gradient system capabilities. | Not Listed | #3024 |
| INDI | Analysis | Implements motion correction, outlier rejection, segmentation, and tensor fitting for diffusion data to generate parametric maps for FA, MD, HA, E2A, and TA. | https://github.com/ImperialCollegeLondon/INDI | #3025 |
| STAIR-SE-EPI | Pulse Sequence Design | Implemented a short TR adiabatic inversion recovery-prepared spin echo EPI sequence to isolate myelin water diffusion with whole-brain coverage. | Not Listed | #3028 |
| Extended reconstruction framework | Reconstruction | Mitigates in-plane B₁⁺ variations and through-plane slab-boundary artifacts in gSlider diffusion-weighted imaging at 7T by incorporating B₁⁺ inhomogeneity and T₁ recovery corrections. | Not Listed | #3029 |
| MatMRI toolbox | Reconstruction | Reconstructs images using an expanded encoding model with k-space coefficients up to 3rd spatial order and l1-wavelet regularization, incorporating coil sensitivity maps and B0 field maps estimated from a GRE calibration prescan. | Not Listed | #3030 |
| OXSA | Analysis | Overlay the reconstructed 3D CSI grid onto the 1H anatomical images for voxel localization and perform AMARES fitting of the area under the curve of the spectra. | Not Listed | #3043 |
| In-house MATLAB scripts | Analysis | Process 31P-MRS spectra, including filtering, phasing, Fourier transformation, baseline correction, peak fitting, and quantification of ATP flux rates. | Not Listed | #3048 |
| SGDR-Net | Reconstruction | A self-guide deformable registration network for eliminating motion artifacts in DCE-MRI breast subtraction images. | Not Listed | #3060 |
| RIM-augmented Delta-MRI | Reconstruction | Estimates anatomical changes directly from a reference image and undersampled follow-up scans using a physics-informed iterative inverse-problem-solving network. | Not Listed | #3061 |
| Deformetrica | Reconstruction | Non-rigid registration and landmark tracking to construct temporally consistent 4D cardiac atlases. | Not Listed | #3063 |
| U-Net based unsupervised deformable registration model | Reconstruction | Predicts 2-D displacements between moving and fixed frames in cine CMR images using a U-Net architecture and unsupervised learning. | Not Listed | #3068 |
| None | Reconstruction | The framework performs real-time volume-level prospective motion correction (PMC) using rigid-body volume-to-volume registration (VVR) and slice-to-volume registration (SVR) to maintain the imaging field-of-view (FoV) during acquisition while concurrently monitoring slice-level motion to quantify the remaining intra-volume motion. | Not Listed | #3069 |
| vSHARP | Reconstruction | Reconstructs high-fidelity images from sampled k-space data using an unrolled optimization 3D reconstruction network. | Not Listed | #3070 |
| Microstructure.jl | Analysis | Estimate soma- and neurite-volume fractions from diffusion MRI data. | Not Listed | #3071 |
| hierarchical registration network | Reconstruction | Integrates MR signal models into a contrast-agnostic deep learning registration pipeline to correct intra-scan motion at the weighted-image level and inter-scan misalignment between derived maps. | Not Listed | #3072 |
| Hybrid AI–Physics Framework | Reconstruction | Reconstructs 3D heart motion from differential scattering parameters (ΔS) by integrating AI-based estimation with physics-based refinement. | Not Listed | #3074 |
| 4 Quadrant Linear Flow Decomposition (4QLiFD) | Analysis | Decompose renal IVIM components from multiple cardiac cycle and gradient waveform renal DWI data into steady and pulsatile aspects of fast and slow flow components, attributed to renal vascular and tubular spaces. | Not Listed | #3079 |
| AI-enhanced FOD | Reconstruction | Enhance fibre orientation distribution estimation from single-shell diffusion MRI data using a pre-trained deep learning model. | Not Listed | #3080 |
| ON_sampling | Analysis | Automates region-of-interest sampling along the optic nerves to provide a simple and reproducible mechanism for dMRI measurement of optic nerve microstructure. | https://github.com/filipp02/on_sampling | #3085 |
| LOFT BBB Toolbox | Analysis | Process DW-pCASL data to obtain kw, CBF, and ATT maps. | Not Listed | #3086 |
| Stim-CODE | Pulse Sequence Design | Designs M₁M₂-compensated diffusion gradient waveforms with PNS and CNS constraints for cardiac DTI acquisitions to achieve shorter TEs at higher b-values. | Not Listed | #3089 |
| Dmipy | Analysis | To fit diffusion data with the DKI model for obtaining diffusion metrics. | Not Listed | #3090 |
| 3D deep-learning framework | Reconstruction | Automated segmentation and quantitative assessment of the internal capsule (IC) in 7T neonatal T2-weighted MRI, enabling volumetric and signal-based evaluation of regional maturation in preterm and term neonates. | Not Listed | #3093 |
| nnU-Net based framework | Segmentation | Automated subthalamic nucleus segmentation from 7T MRI scans, trained on manual segmentations performed for DBS surgery. | Not Listed | #3094 |
| MIST | Analysis | Lightweight segmentation models for glioma segmentation in resource-constrained settings, achieving significant reductions in training time and model size with architecture-dependent accuracy tradeoffs. | https://github.com/mist-medical/MIST | #3095 |
| UMamba | Reconstruction | Automated segmentation of esophageal squamous cell carcinoma (ESCC) on contrast-enhanced, high-resolution, free-breathing 3D-GRE MRI. | Not Listed | #3098 |
| PVS segmentation repository | Analysis | To create and maintain an open-source repository for perivascular space (PVS) segmentation code, providing a benchmark for future development and comparison of segmentation methods. | https://github.com/PVS-segmentation-repository | #3099 |
| SpineLabelNet | Analysis | Automate cervical and lumbar vertebral labelling on 2D 3-plane localizers using a deep-learning model and a post-processing correction algorithm to refine and propagate labels. | Not Listed | #3100 |
| provenance-aware active learning framework | Analysis | Reduces annotation burden in fetal MRI segmentation by combining diversity-driven cold-start selection, hybrid uncertainty fusion, confidence-gated auto-label acceptance, and provenance-weighted retraining. | Not Listed | #3102 |
| BrainSegNet | Reconstruction | A deep learning framework for whole-brain multi-label segmentation using MRI, integrating SAM with U-Net and incorporating ASPP, CSAM, and Boundary Refinement for improved accuracy and boundary fidelity. | Not Listed | #3103 |
| UMM-Net | Segmentation | An uncertainty-guided multi-sequence fusion network designed to simultaneously tackle effective cross-sequence fusion and accurate boundary localization for placenta accreta spectrum segmentation. | Not Listed | #3104 |
| PRIME (Philips Reconstruction with Injector, Modulator, and Emitter) | Reconstruction | Enables inline execution of vendor-neutral custom data-processing algorithms within the Philips reconstruction environment. | Not Listed | #3105 |
| nnU-Net | Analysis | Train a 3D nnU-Net with multi-channel MRI data (CBF, FLAIR, MPRAGE, WMH segmentation) for spatial prediction of WMH after ~1 year from baseline. | Not Listed | #3107 |
| AI model | Analysis | Detect and segment cerebral aneurysms automatically from TOF-MRA scans using a 3D deep learning framework. | Not Listed | #3108 |
| CaPTk | Analysis | Cancer Imaging Phenomics Toolkit (CaPTk) for skull stripping, bias correction, and automated segmentation of enhancing, necrotic, and edema regions in MRI data. | Not Listed | #3109 |
| 3D U-Net | Reconstruction | To automatically segment fetal body organs from 3D T2-weighted MRI images. | Not Listed | #3110 |
| Gannet toolbox | Analysis | To process MRS data and obtain CSF-corrected metabolite estimates. | Not Listed | #3118 |
| TrufiStrain | Analysis | Fully automated atrial volumetric and strain analysis from CMR using a deep learning model for LA segmentation and deformable registration for contour propagation. | Not Listed | #3124 |
| ST-DIP | Reconstruction | Reconstructs multi-slice cine MRI data using a deep image prior self-supervised deep learning approach parameterized by time and slice. | Not Listed | #3126 |
| Wave-Driven Simulator | Simulation | A physics-based simulator that generates structural and functional cine MRI data with pixel-wise labeled displacement and strain for benchmarking deep learning algorithms. | Not Listed | #3134 |
| Pulseq | Pulse Sequence Design | Implement the mpDYCI sequence with a stack-of-spirals sampling trajectory using a vendor-neutral open-source framework. | Not Listed | #3135 |
| BART | Reconstruction / Pulse Sequence Design | To provide an open-source computational MRI framework for advanced image reconstruction and pulse sequence design, specifically for radial ASL MRI, enabling sequence generation, data acquisition, and reconstruction with ASL-specific regularization. | Not Listed | #3136 |
| Pulseq-LIBRE | Pulse Sequence Design | Implements a vendor-agnostic 3D radial GRE sequence with LIBRE water excitation pulses for fat suppression in ophthalmic MRI. | Not Listed | #3139 |
| Pulseq-4DFlow | Pulse Sequence Design | Provides a 4-point 4D Flow sequence implemented in PyPulseq. | https://github.com/BAMMri/Pulseq-4DFlow | #3140 |
| Gadgetron | Reconstruction | Provides vendor-independent image reconstruction and post-processing for MRI data. | Not Listed | #3141 |
| Modular MT sequence | Pulse Sequence Design | Implements a modular magnetization transfer sequence for 31P MRS within the Pulseq framework, allowing for vendor-agnostic application and modification for different experiments. | Not Listed | #3143 |
| Python-based JupyterLab interface | Pulse Sequence Design | Enables sequence design, parameter tuning, and real-time results visualization for an FPGA-based MRI console. | Not Listed | #3144 |
| INR | Reconstruction | Estimate IVIM parameters with spatial regularization in a self-supervised manner using implicit neural representations. | Not Listed | #3155 |
| PANDA | Analysis | Detect fetal brain anomalies by predicting gestational age from 3D MRI patches and using the maximum absolute age difference (MaxAAD) as a biomarker. | Not Listed | #3161 |
| nnUNet | Analysis | Segment brain injuries in infants using multi-center T1WI and T2WI MRI data. | Not Listed | #3163 |
| TA-GAN | Reconstruction | Synthesizes gadolinium-free CBV maps from non-contrast MRI sequences for perfusion imaging in primary intracranial tumors. | Not Listed | #3165 |
| SRNR (Super-Resolution using Noisy Reference) | Reconstruction | Super-resolves low-resolution MRI images using a neural network trained on accelerated, noisy high-resolution scans, suitable for pediatric MRI. | Not Listed | #3168 |
| Sonic DL | Reconstruction | Accelerate MRI acquisition using sparse random sampling with an unrolled CNN-regularized reconstruction to reconstruct high-quality images in a shorter time. | Not Listed | #3169 |
| Custom Python scripts | Analysis | Automated data processing of temperature-time curve data to calculate power output at the MWA antenna, with and without correction for heat loss, and to determine the average available heating power depending on the cable length. | Not Listed | #3174 |
| TGRAPPA-initialized unrolled network | Reconstruction | Achieves high-quality, low-latency cine reconstruction by using a TGRAPPA-initialized unrolled network with a pretrained coil sensitivity estimation module, removing dependence on ESPIRiT and supporting robust real-time imaging for MRI-guided cardiovascular interventions. | Not Listed | #3178 |
| In-house deep-learning toolkit | Analysis | Automated body composition segmentation from MR images to quantify muscle and fat tissue areas. | Not Listed | #3183 |
| WARP-SPACE | Pulse Sequence Design | Modified SPACE sequence incorporating an optimized acquisition scheme with view-angle-tilting (VAT), slab-selective pulses, and matched RF bandwidths to reduce susceptibility artifacts in head-and-neck MRI. | Not Listed | #3193 |
| nnU-Net model | Segmentation | Automated, precise multi-structure orbital segmentation in Thyroid Eye Disease (TED) patients from MRI images. | Not Listed | #3200 |
| gtx | Analysis | Generate whole-genome and cell-type-specific polygenic risk scores (PRS) by integrating AD summary statistics with cell-type-enriched transcriptomic profiles. | Not Listed | #3211 |
| nnU-Net-based framework | Reconstruction | Automated segmentation of vascular perfusion territories using super-selective pseudo-continuous arterial spin labelling (ss-pCASL)-derived perfusion maps, complemented by structural information from MP-RAGE and FLAIR MRI. | Not Listed | #3213 |
| Brain Connectivity Toolbox | Analysis | Computes graph theory metrics such as characteristic path length (CPL), clustering coefficient (CC), global efficiency (GE), and small-worldness (SWN) from functional connectivity matrices. | Not Listed | #3214 |
| DESIGNER | Reconstruction | Post-processing of dMRI data. | Not Listed | #3237 |
| U-Net | Reconstruction | Synthesize HBP images from pre-contrast T1-weighted liver MRI using a deep learning model with a combination of L1 loss, structural similarity (SSIM) loss, and a feature-wise perceptual loss. | Not Listed | #3246 |
| FAE | Analysis | To extract radiomic features, including shape and first-order histogram metrics, from the whole tumor and each habitat. | Not Listed | #3248 |
| CHORUS-RARE | Pulse Sequence Design | Extends the CHORUS sequence for imaging with a RARE scheme, adding one RF pulse to improve robustness to B0/B1 inhomogeneities in low-field MRI. | Not Listed | #3253 |
| ERNIE | Education | To provide a low-cost/open-source/hands-on educational toolkit for practical learning of Halbach-based MRI scanner assembly and field mapping. | https://github.com/IMAGINE | #3255 |
| Tyger–MaRGE framework | Reconstruction | Integrates cloud reconstruction directly into the acquisition workflow of low-field MRI, enabling parallel execution and application-specific pipelines for denoising, distortion correction, and non-Cartesian imaging. | Not Listed | #3257 |
| QC-framework | Analysis | An open-source QC-framework tailored to low-field MRI systems, which allows monitoring spatial and temporal system instabilities which may infer image quality. | Not Listed | #3258 |
| EEG-driven optimization framework | Analysis | Optimizes neuronal model parameters in The Virtual Brain (TVB) using EEG data to personalize brain digital twins, particularly in low-resource settings where fMRI data is unavailable. | Not Listed | #3260 |
| HalbachMRIDesigner | Design | Generates 3D geometry of Halbach array magnets as an OpenSCAD (.scad) file based on parameters. | Not Listed | #3261 |
| MARGE | Pulse Sequence Design | An open-source platform used to write and debug a PETRA sequence, including the possibility to pre-emphasize RF pulses. | Not Listed | #3262 |
| Deep-DSP2 | Analysis | Directly predicts EMI-free MR signals from experimental MRI receive coil signals and EMI sensing coil signals via active EMI sensing and cross-domain deep learning. | Not Listed | #3264 |
| Automated pipeline for spatial co-registration of 4D flow MRI with CE-MRA | Analysis | Spatially co-register 4D flow MRI with CE-MRA to enable integrated anatomy-hemodynamics analysis, specifically for assessing WSS-dilatation pattern relationships in BAV aortopathy. | Not Listed | #3268 |
| 3D J-Net | Reconstruction | Enhances the quality of rapidly acquired, low-resolution whole-heart scans using a deep learning-based super-resolution network. | Not Listed | #3272 |
| Uncertainty-guided active learning pipeline | Analysis | An active learning pipeline for accurate segmentation and diameter quantification of the aorta and iliac arteries in post-contrast CMR, designed to support TAVI access route planning. | Not Listed | #3274 |
| DESIGNER pipeline | Analysis | Denoise dMRI data using MPPCA before preprocessing. | Not Listed | #3283 |
| Surface-Normal Diffusivity (SN-D) Analysis Framework | Analysis | To compute and analyze surface-normal diffusivity (SN-D) from diffusion tensor imaging (DTI) data by projecting diffusion tensors onto local vectors orthogonal to the cortical surface, aiming to provide a more sensitive marker of glymphatic function compared to mean diffusivity (MD). | Not Listed | #3284 |
| Microstructure.jl | Analysis | Fitting q-space data to the SANDI model. | Not Listed | #3285 |
| End-to-end quantitative pipeline for TOF MRA | Analysis | Derives a robust vessel mask, reconstructs vascular surfaces and centerline skeletons, and computes per-vertex and global graph features for cerebrovascular morphometry from TOF MRA images. | Not Listed | #3291 |
| Unnamed Deep Learning Framework | Reconstruction | Enables rapid, high-quality reconstruction of accelerated 4D-ASL-MRA using limited acquired data by leveraging open 3D-TOF-MRA data through simulation-driven deep learning. | Not Listed | #3296 |
| Eight neural network families | Analysis | Benchmark different neural network architectures for fitting the VERDICT model to diffusion MRI data in brain tumours, including a 3-layer MLP (with a ResNet variant), an LSTM-based RNN, a 1D convolutional network, a Transformer encoder, a variational autoencoder (VAE) regressor, a Mixture-of-Experts (MoE) model, and a TabNet. | Not Listed | #3303 |
| Swin Transformer | Analysis | Classify brain MRI images into glioma, meningioma, pituitary, and non-tumor classes. | Not Listed | #3308 |
| MFEM | Simulation | Solves linear elasticity equations using a preconditioned conjugate gradient method to simulate patient-specific deformation, stress, and stiffness based on MRI-derived meshes. | Not Listed | #3310 |
| Patch2Self | Analysis | Denoising multi-shell diffusion MRI data. | Not Listed | #3314 |
| CAT12 | Analysis | Volumetric analysis of 3D T1-weighted images. | http://www.neuro.unijena.de/cat/ | #3328 |
| Dipy | Analysis | Calculating extracellular free water fraction using parameter fitting, model computation, and spatial normalization of DTI data. | Not Listed | #3334 |
| Standard Model Imaging (SMI) toolbox | Analysis | Estimate free-water fraction (FWF) from diffusion MRI data using the diffusion-MRI Standard Model. | Not Listed | #3335 |
| Dmipy | Analysis | Computes diffusion tensor and diffusion kurtosis (DTI and DKI) metrics. | Not Listed | #3341 |
| cuDIMOT | Reconstruction | Computes NODDI parameter maps, including orientation dispersion index (ODI), neurite density index (NDI) and free water fraction (FWF) using the Watson distribution. | Not Listed | #3342 |
| TRACED | Analysis | TRACED is a biophysical model to estimate intracellular volume fraction, cell size distribution, extracellular intrinsic diffusivity, and tortuosity in tumors by modeling time-dependent diffusion in tissue microstructure with cell size distributions. | Not Listed | #3345 |
| ME-AxCaliberSMT | Reconstruction | A combined diffusion-relaxometry MRI framework that extends the dynamic range of axon calibre estimation across the full physiological range by integrating an R2-based model into AxCaliberSMT. | Not Listed | #3352 |
| TractSeg | Reconstruction | Automatically reconstructs 72 tracts from diffusion MRI data. | Not Listed | #3353 |
| TumorCLIP | Analysis | A CLIP-inspired vision–language fusion framework that enhances both brain tumor classification accuracy and explainability by combining image and text embeddings. | Not Listed | #3357 |
| OmniMRI | Reconstruction / Segmentation / Detection / Diagnosis / Report Generation | A unified vision-language foundation model trained on large-scale heterogeneous MRI data, performing reconstruction, segmentation, detection, diagnosis, and report generation in one system. | Not Listed | #3358 |
| BrainMR Specialist | Analysis | A brain MRI foundation model is developed and fine-tuned for diverse downstream tasks including accelerated MRI reconstruction, image translation, classification, segmentation, and report generation. | Not Listed | #3359 |
| BrainDFMAE | Analysis | Learns a unified representation capturing feature consistency for multi-task Alzheimer’s MRI analysis, including segmentation, classification, and generation, through deformation-aware pretraining. | Not Listed | #3360 |
| Cardiac Perfusion MRI Foundation Model | Analysis | Pretrain a foundation model using unlabeled cardiac perfusion MRI images and fine-tune it for myocardial segmentation, achieving state-of-the-art performance with significantly fewer labeled images. | Not Listed | #3362 |
| KIMRA (K-space–Image Multimodal Representation Alignment) | Analysis | An end-to-end framework that unifies representation learning across k-space and image domains, enabling comprehensive cardiac analysis directly from undersampled k-space measurements. | Not Listed | #3363 |
| CardiVLSM | Segmentation | A fully automated, language-guided segmentation pipeline that uses a vision-language model (VLM) to convert text queries into semantic features that guide SAM2 to generate segmentation masks for cardiac cine MRI. | Not Listed | #3364 |
| Matcher | Analysis | Automates the evaluation of AI-generated medical images by matching them against exemplar template images using Vision Transformer-based feature extraction and one-shot segmentation. | Not Listed | #3365 |
| MuscleMap Toolbox (MMT) | Analysis | Streamlines 3D processing, segmentation, and quantification of muscle health from multimodal MRI and CT images, providing automated, contrast-agnostic muscle segmentation and quantification of muscle size and intramuscular fat. | https://github.com/MuscleMap/MuscleMap | #3371 |
| nnU-Net / FatSegNet frameworks | Analysis | Automatically segment bilateral psoas and erector spinae muscles in MRI images. | Not Listed | #3372 |
| MRS4Brain Toolbox | Analysis | Process spectroscopic data to estimate metabolite concentrations in each voxel. | Not Listed | #3380 |
| ivim | Analysis | To generate IVIM maps for Hypoxia Scoring calculation. | https://github.com/oscarjalnefjord/ivim | #3386 |
| Python scripts | Reconstruction | Image reconstruction (non-uniform fast Fourier transform) of MRI data. | Not Listed | #3387 |
| Auto-TI framework | Analysis | Predicts optimal inversion time (TI) for myocardium and blood pool in LGE imaging and provides actionable feedback to the user regarding the confidence of the prediction. | Not Listed | #3391 |
| Multitasking (MT) T1/T1ρ sequence | Pulse Sequence Design | A modified 2D MT T1/T1ρ sequence was developed for non-contrast myocardial tissue characterization, enabling simultaneous quantification of T1 and T1ρ relaxation times. | Not Listed | #3397 |
| Sonic DL | Reconstruction | Enhances spatial resolution for MOLLI T1 mapping by integrating variable-density undersampling with a pretrained deep learning model, while preserving temporal resolution and quantification accuracy. | Not Listed | #3400 |
| lesion-aware mathematical simulator | Simulation | Generates physiologically plausible multi-class label maps by synthesizing LV cavity and myocardial rings and inserting infarct and no-reflow lesions with realistic shapes, sizes, and locations. | Not Listed | #3402 |
| AUTOSEQ-based MRF | Reconstruction | Implements an AUTOSEQ-based magnetic resonance fingerprinting (AUTOSEQ-based MRF) framework for relaxation spectroscopy at 6.5 mT scanner to validate T1/T2 estimation in tissue-mimicking phantoms. | Not Listed | #3427 |
| PF-DiffSR | Reconstruction | A novel 3D progressive deep learning image formation framework that integrates a PF-SR model for image noise and artifacts suppression with a conditional residual diffusion model to recover fine anatomical details in ultra-low-field lumbar spine MRI. | Not Listed | #3428 |
| CT-GNN | Analysis | Predicts longitudinal, patient-level language outcomes from MRI by hard-thresholding structural connectivity and jointly encoding structural and functional connectomes using a graph neural network. | Not Listed | #3432 |
| HICAG-U-Net | Reconstruction | Segment acute ischemic lesions from DWI and ADC images using a deep learning U-Net architecture with specific ADC thresholds. | Not Listed | #3433 |
| parallel GRU+LSTM network | Simulation | Applies a random ADC value to each signal in a pre-existing MR signal dictionary for simulating diffusion effects in MR vascular fingerprinting. | Not Listed | #3435 |
| TorchIO | Analysis | Image augmentations were performed using the TorchIO package for creating a more robust training set. | Not Listed | #3437 |
| Multidimensional-diffusion-MRI toolbox | Reconstruction | Reconstruct FA and μFA maps from diffusion-weighted images obtained from tensor-encoded diffusion MRI. | https://github.com/markus-nilsson/md-dmri | #3439 |
| DenseResU-Net++ | Segmentation | Accurately estimates stroke lesions using dense connections and attention mechanisms. | Not Listed | #3440 |
| alignedSENSE+JointT/C+LLR | Reconstruction | Jointly reconstruct tag/control ASL images with alignedSENSE motion correction and locally low-rank (LLR) regularization directly on the perfusion images to improve SNR and motion robustness. | Not Listed | #3443 |
| DISTI | Reconstruction | Reconstruct accurate susceptibility tensors from minimal orientations using a deep learning framework that integrates physics-based modeling with DTI and QSM priors. | Not Listed | #3444 |
| dynamic joint static SMS imaging framework / cine-referenced dynamic SMS method | Reconstruction | Integrates cine imaging with quantitative T₁ and T₂ mapping in a single acquisition by deriving calibration information from cine reference data, eliminating the need for separate reference line acquisitions. | Not Listed | #3460 |
| Custom Python gadgets | Reconstruction | Implement four different methods for ML-based image reconstruction (Single-frame, Sliding-window, Key-hole, Variable-radius) within the Gadgetron framework. | Not Listed | #3463 |
| Time-resolved low-rank subspace model | Reconstruction | Reconstructs blipped-CAIPI time-resolved images from multiple scans by exploiting a shared time-resolved subspace derived from concatenated central slice images to synchronize the reconstruction across all slices and time points. | Not Listed | #3465 |
| PFT-based CS with spatially variable temporal-TV | Reconstruction | Reconstructs 2D radial cardiac MRI data using a polar Fourier transform-based compressed sensing method with spatially variable temporal total variation regularization to improve image sharpness in the cardiac region. | Not Listed | #3466 |
| Radial DV 4D-Flow MRI sequence | Pulse Sequence Design | Development of a radial dual VENC 4D-Flow MRI sequence with modified trajectories to reduce stimulated echo artifacts when using RF-only spoiling. | Not Listed | #3469 |
| ZTE-STAR | Reconstruction | A framework integrating a 3D GA spiral-phyllotaxis trajectory, algebraic dead-time filling, and Temporal Total Variation (TTV) reconstruction for dynamic ZTE imaging. | Not Listed | #3471 |
| Poblano | Reconstruction | Limited-memory BFGS was used for the iterative conjugate gradient reconstruction modeling coil sensitivities, B0 inhomogeneity, and R2* decay with total variation regularization. | Not Listed | #3472 |
| BART (Berkeley Advanced Reconstruction Toolbox) | Reconstruction | Performs iterative SENSE reconstruction for real-time MRI of speech production using radial cine sequences. | Not Listed | #3475 |
| Retrospective Simulation Tool | Simulation | Models signal evolution across the slice thickness for a spin echo to assess the discrepancy between spin dynamics at the slice center and the slice profile for large-tip-angle pulses in 2D-TSE. | Not Listed | #3484 |
| 3D Slicer | Analysis | Semiautomated segmentation of the superior sagittal sinus and tumor volumes (including enhancing and necrotic regions) was performed on postcontrast-precontrast subtraction images. | Not Listed | #3505 |
| Dynamic inference U-Net model | Reconstruction | Reconstructs a prospective dense Z-spectrum (DZS) for each candidate offset by “simulating” the effect of acquiring that offset using the current DZS. | Not Listed | #3509 |
| In-house script | Analysis | Quantify the tumor subcomponents geometry using 3D fractal dimension (FD3D) and lacunarity (Lac3D). | Not Listed | #3515 |
| CINDERELLA | Data Management | Automatically label DICOM sequences using a minimal number of sequence parameters for classification. | Not Listed | #3516 |
| STISuite | Reconstruction | Reconstruct quantitative susceptibility maps (QSM) using Laplacian unwrapping, V-SHARP filtering, and STAR-QSM inversion. | Not Listed | #3521 |
| χ-sepnet | Reconstruction | Deep learning-based reconstruction to generate an artifact-resistant χpara map for iron mapping. | Not Listed | #3522 |
| SPICE | Reconstruction | Simultaneously map the whole-brain oxidative, glial, and neuronal alterations in vivo. | Not Listed | #3527 |
| 3D CNN pipeline | Reconstruction | Automates tumor segmentation and classification of breast tumors from multi-modal MRI data, enabling streamlined clinical workflows. | Not Listed | #3542 |
| DCE-HI map | Analysis | Quantifies intratumoral heterogeneity dynamics during therapy by combining temporal entropy and necrotic fraction derived from DCE-MRI data. | Not Listed | #3545 |
| ResNet34 | Analysis | Developed a ResNet34 model with feature and decision fusions based on D and K maps from DKI for classification of cervical cancer histological subtypes. | Not Listed | #3548 |
| SeeIt | Analysis | Extracting radiomic features (morphological, first-order, textural) from MRI images. | Not Listed | #3553 |
| LRMC-DP | Reconstruction | Reconstructs MRF data with low-rank motion correction and a dictionary-patch regularizer to generate volumetric T1/T2/T2*/PDFF maps. | Not Listed | #3556 |
| Spiral-QALAS | Pulse Sequence Design | A stack-of-spirals pulse sequence for accelerated, motion-corrected, water-fat separated T1 and T2 mapping. | Not Listed | #3562 |
| SEPIA | Analysis | Process complex multi-echo data for Quantitative Susceptibility Mapping (QSM). | Not Listed | #3565 |
| MATI package | Analysis | Calculate microstructural parameters (cell diameter d, intracellular volume fraction vin, extracellular diffusivity Dex, and transytolemmal water exchange rate constant Kin) from the JOINT model using diffusion MRI data. | Not Listed | #3571 |
| Pyradiomics | Analysis | To compute histogram parameters (mean, median, 10th percentile, 90th percentile, skewness, and kurtosis values) from fractional anisotropy (FA), mean diffusivity (MD), and cerebral blood flow (CBF) maps. | Not Listed | #3574 |
| IMPULSED (Imaging Microstructural Parameters Using Limited Spectrally Edited Diffusion) | Analysis | A modeling framework for quantifying cellular microstructure from time-dependent diffusion MRI (td-dMRI) by probing water diffusion at multiple time scales. | Not Listed | #3575 |
| MrTrix v3.0.3 | Analysis | To process and analyze diffusion MRI data using fixel-based analysis (FBA). | Not Listed | #3576 |
| AMURA | Analysis | Derive EAP-like sensitivity diffusion metrics from single-shell diffusion MRI data using clinically feasible single-shell data. | Not Listed | #3579 |
| Conditional Diffusion Model | Simulation | Generate synthetic breast implant images conditioned on implant masks to augment real data for improved implant classification. | Not Listed | #3581 |
| scPINN-MREPT | Reconstruction | Reconstructs conductivity maps for Magnetic Resonance Electrical Properties Tomography (MREPT) using a stepwise training approach with physics-informed neural networks and collocation points. | Not Listed | #3585 |
| IQMR | Reconstruction | Perform image denoising and enhance resolution of undersampled k-space data acquired with a reduced number of excitations. | Not Listed | #3586 |
| Reconstruction framework | Reconstruction | Reconstructs 2D brain tumor MRI scans and transforms thick slices into thin slices with flexible interpolation rates. | Not Listed | #3587 |
| Recurrent Inference Machine (RIM) | Reconstruction | Reconstructs accelerated DCE MRI data at high spatial and temporal resolution using a supervised physics-informed iterative deep learning model. | Not Listed | #3591 |
| DL-CS-CAIPI | Reconstruction | Combines coherent and incoherent undersampling with a deep learning-based reconstruction for accelerated MRI. | Not Listed | #3593 |
| CRUNet-MR | Reconstruction | An unrolled network designed for cardiac cine MRI reconstruction that enhances spatio-temporal feature utilization by integrating convolutional recurrent operations into the UNet architecture. | Not Listed | #3594 |
| DLSpeed | Reconstruction | Deep learning image reconstruction from undersampled k-space to accelerate MRI acquisition. | Not Listed | #3595 |
| Deep learning based reconstruction method | Reconstruction | Reconstruct high-quality METMOLED images from accelerated single-shot multi-echo-train multiple overlapping-echo detachment (METMOLED) data using a weighted k-space diffusion model based on score-based stochastic differential equations. | Not Listed | #3596 |
| RS-Recon | Reconstruction | A novel reconstruction method that uses high-SNR reference signals to aid the detection of low-SNR ASL signals. | Not Listed | #3599 |
| Unrolled SSDU network | Reconstruction | Reconstructs 3D volumes from undersampled k-space data using self-supervised physics-guided learning with alternating data-consistency and CNN-based regularization modules. | Not Listed | #3600 |
| SNR-guided non-Cartesian CS reconstruction | Reconstruction | Improves non-Cartesian compressed sensing reconstruction for ultra-high field MRI by incorporating spatially adaptive weights calculated from SNR maps into L1 regularization. | Not Listed | #3601 |
| Chunk-Wise Diffusion Reconstruction Framework | Reconstruction | A framework that partitions a 3D volume into overlapping chunks, applies a score-based diffusion model independently to each chunk, and then assembles the denoised chunks into a full-volume image in CPU memory to enable memory-efficient diffusion reconstruction. | Not Listed | #3602 |
| IAS | Reconstruction | Dynamic MRI reconstruction with reduced hyper-parameter sensitivity compared to compressed sensing. | Not Listed | #3603 |
| Jülich Advanced Reconstruction Toolbox (JUART) | Reconstruction | In-house developed ESPIRiT reconstruction toolbox for sensitivity map estimation that is OS-independent and has shorter computation times compared to BART. | Not Listed | #3606 |
| Accelerated ML-DIP | Reconstruction | Accelerates the training of ML-DIP for 3D cardiac cine MRI reconstruction by leveraging distributed training and optimization techniques, reducing training time while preserving image quality. | Not Listed | #3609 |
| DREME-GSMR | Reconstruction | DREME-GSMR enables dynamic MRI reconstruction and real-time MRI inference from limited k-space data by representing patient anatomy and motion fields as 3D Gaussians. | Not Listed | #3610 |
| MRE-FIn | Reconstruction | Open-source framework for non-linear inversion (NLI) in Magnetic Resonance Elastography (MRE) to estimate tissue material properties from measured displacement data. | Not Listed | #3612 |
| NiftyMIC | Reconstruction | Perform super-resolution reconstruction of fetal MRI images using rigid brain reconstruction. | https://github.com/NiftyMIC | #3613 |
| MNet | Pulse Sequence Design | Predicts scan-specific Cartesian MRI undersampling masks directly from low-frequency k-space, trained using pre-optimized scan-adaptive masks as supervision. | Not Listed | #3616 |
| PDDL | Reconstruction | Deep learning-accelerated reconstruction of isotropic PD-weighted SPACE MRI sequence for faster knee imaging while maintaining diagnostic quality. | Not Listed | #3617 |
| U-Net | Reconstruction | Synthesizes MT- and T1-weighted images with varying contrasts from limited acquisitions for accurate macromolecular proton fraction (MMF) mapping. | Not Listed | #3619 |
| Rep-CRNN | Reconstruction | Enhances a convolutional recurrent neural network with a re-parameterization mechanism to achieve low-latency, high-quality MRI reconstruction for intervention. | Not Listed | #3624 |
| Deep Resolve Boost (DRB) | Reconstruction | To accelerate MRI scans using deep learning reconstruction while maintaining or improving image quality, thereby increasing throughput on a 0.55T MRI system. | Not Listed | #3627 |
| TW-dFC | Analysis | Mapping functional connections along structural white matter pathways to identify treatment-related network disruptions. | Not Listed | #3630 |
| JuSpace toolbox | Analysis | Investigate atlas-based nuclear imaging-derived neurotransmitter maps. | Not Listed | #3632 |
| SimNIBS | Simulation | To estimate the electric field (E) and current density (J) for each individual participant’s head and brain, using the “CHARM” function. | Not Listed | #3634 |
| Bloch simulator | Simulation | Simulates Bloch equations using a 4th-order Runge-Kutta solver to investigate the properties of the AM_HS1 pulse. | Not Listed | #3644 |
| dynamic compressed sensing (CS) reconstruction | Reconstruction | Reconstruct ASL perfusion images with high spatiotemporal resolution and motion-resolved self-navigation using dynamic compressed sensing. | Not Listed | #3652 |
| presurfer | Analysis | Process images for brain extraction, segmentation, and cortical parcellation using the Desikan–Killiany atlas. | Not Listed | #3655 |
| NIFTI_NORDIC.m | Denoising | Applies NORDIC denoising to diffusion-weighted images within MATLAB. | [https://github.com/NORDIC github repository](https://github.com/NORDIC github repository) | #3656 |
| MATI | Simulation | Simulates the signal for a microstructure set with varying cell radii and intracellular volume fractions to optimize OGSE and PGSE waveforms for IMPULSED. | Not Listed | #3663 |
| iRICE-SDP | Reconstruction | Implements a constrained convex semidefinite program (SDP) to fit QTI data from the instant RICE (iRICE) minimal sampling scheme, incorporating established diffusion tensor constraints and bounds on isotropic spherical components to enhance robustness. | Not Listed | #3666 |
| INDI | Analysis | Implements a pipeline for cardiac Diffusion Tensor Imaging (cDTI) post-processing, including data conversion, outlier removal, image registration, cardiac segmentation, tensor fitting, and metric calculation. | https://github.com/ImperialCollegeLondon/INDI | #3667 |
| New fixed TQTPPI reconstruction | Reconstruction | Enables the simultaneous determination of relaxation times (T2s and T2f) and TQ/SQ ratio from fixed TQTPPI sequence data by fitting transfer functions along the acquisition time. | Not Listed | #3669 |
| MOCCO | Reconstruction | Reconstructs free-breathing liver DCE-MRI data acquired with golden-angle radial-sampling and adaptive self-gating, balancing spatial quality and temporal accuracy. | Not Listed | #3671 |
| FIRE (Framework for Image Reconstruction Environments) | Reconstruction | Provides a real-time streaming interface using the ISMRM raw data format to send reconstructed images to a Python program and send shim currents back to the sequence. | Not Listed | #3673 |
| Research sequence and reconstruction framework | Reconstruction | Reconstructs 3D cine CMR data with spatiotemporal wavelet regularization and integrated motion compensation from single- or multi-breathhold acquisitions with a 2-region sampling pattern. | Not Listed | #3680 |
| MRpro | Reconstruction | An open-source software framework used for developing the reconstruction algorithms. | Not Listed | #3681 |
| MRS4Brain toolbox | Analysis | Atlas-based 3D regional segmentation and 1H-FID-MRSI data quantification. | Not Listed | #3684 |
| LRT framework | Reconstruction | Reconstruct acquisitions using a low-rank tensor (LRT) framework incorporating blood-oxygenation signal modeling to generate pixelwise SbO2 maps. | Not Listed | #3687 |
| MRtrix3 | Reconstruction | Used for diffusion tensor imaging (DTI) processing to build weighted structural connectivity (SC) matrices. | Not Listed | #3712 |
| PyRadiomics | Analysis | Extract radiomic features from medical images. | Not Listed | #3713 |
| Pulseq | Pulse Sequence Design | Designed different saturation pulse types within the Pulseq framework for fat saturation in diffusion-weighted EPI sequences. | Not Listed | #3715 |
| Motion-Aware Fieldmap Estimation | Reconstruction | Jointly estimates the fieldmap and a single undistorted image from EPI volumes, accounting for inter-scan motion using a particle-in-cell forward model and optimization implemented in JAX. | Not Listed | #3719 |
| Pulseq | Pulse Sequence Design | Implemented a spiral-out readout for OSSI data acquisition on a 3T GE scanner. | Not Listed | #3724 |
| Flip Proof | Analysis | Ensures the orientation correctness of QSM DICOMs by integrating novel, custom software that leverages the compiler to enforce orientation-safe image operations. | Not Listed | #3726 |
| μSep | Reconstruction | Extends current χ-Separation by introducing the mesoscopic Larmor-frequency contribution in white matter to reduce orientation-dependent errors in susceptibility estimation. | Not Listed | #3729 |
| X-SepNet | Reconstruction | Separates diamagnetic and paramagnetic components from multi-echo GRE data for χ-dia processing. | Not Listed | #3732 |
| DeepAcq | Reconstruction | Synthesizes high-resolution whole-brain multi-contrast images and T1/T2 maps from ultra-fast acquisitions using a deep learning framework. | Not Listed | #3734 |
| Patch2Voxel (P2V) | Reconstruction | Denoise dynamic MRI data by learning voxel-wise temporal dynamics from spatially localized patches using a self-supervised learning framework. | Not Listed | #3735 |
| nn-UNet | Reconstruction | To predict non-enhancing/edema, necrosis, and enhancement from T1w, T2w, and T2-FLAIR images to delineate the tumor core. | Not Listed | #3736 |
| MIMOSA | Reconstruction | Enables rapid simultaneous mapping of T1, T2, T2*, and quantitative susceptibility mapping (QSM) using an optimized acquisition and multi-contrast/-slice, zero-shot SSL (MZS-SSL). | https://github.com/yutingchen11/MIMOSA | #3738 |
| PISCO | Analysis | Computes spatiotemporal maps (STMs) from autocalibration data for use in magnetic resonance fingerprinting (MRF) reconstruction. | Not Listed | #3739 |
| nnU-Net | Segmentation | Automated segmentation of paediatric optic pathway gliomas (OPGs) using a transfer learning approach with Channel Dropout for robustness to missing MRI modalities. | Not Listed | #3740 |
| DIME | Reconstruction | Directly estimates stiffness from MRE wavefields using a U-Net based CNN, bypassing the assumptions required by conventional analytical methods. | Not Listed | #3741 |
| VIBEsegmentator | Analysis | Segment the entire heart from CMR images, providing bounding boxes for cropping the heart from localizer images. | Not Listed | #3743 |
| VAE MoCo | Reconstruction | Compensates for respiratory motion in cardiac MRI by learning time-resolved respiratory displacements from self-gating data using a variational autoencoder (VAE) and applying this information for k-t space data compensation. | Not Listed | #3744 |
| spatial-temporal dealiasing network integrated into a FISTA-based PnP framework | Reconstruction | Rapidly reconstruct 3D radial free-running CMR data by integrating a deep learning regularizer within an iterative reconstruction approach. | Not Listed | #3749 |
| CGE framework | Simulation | Generates LGE-equivalent images from cine sequences via a diffusion model with a spatiotemporal encoder to extract motion features. | Not Listed | #3752 |
| hMRI-toolbox | Analysis | Process ME-PDw and T1w-images alongside B1+-maps to get qPD maps for water density (WD). | Not Listed | #3767 |
| MEDI (Morphology Enabled Dipole Inversion) toolbox | Analysis | Performs Quantitative Susceptibility Mapping (QSM) on multi-echo gradient recalled echo (GRE) datasets, incorporating phase unwrapping, background field removal, and dipole inversion regularized by magnitude priors. | Not Listed | #3769 |
| In-house software package | Analysis | Perform tissue segmentation for BPE quantification, including the delineation of breast outlines and segmentation of fibroglandular tissue. | Not Listed | #3814 |
| DIPY | Analysis | Calculate DTI metrics including FA, MD, axial diffusivity (AD), radial diffusivity (RD), and fiber density (FD) from diffusion MRI data. | Not Listed | #3817 |
| BENCH | Analysis | The Bayesian EstimatioN of CHange (BENCH) framework identifies the probability of each biophysical parameter explaining variation in dMRI data with respect to a continuous or categorical variable. | Not Listed | #3821 |
| HippUnfold | Reconstruction | Segment hippocampal subfields from high-resolution T1-weighted images. | Not Listed | #3823 |
| SpinDoctor | Simulation | Solves the Bloch–Torrey equation using the finite-element method to simulate water exchange in neurons. | Not Listed | #3824 |
| nnU-Net | Reconstruction | Automated segmentation of 3D T1-weighted images to quantify perivascular space volume fraction (PVS). | Not Listed | #3825 |
| IVIM toolbox | Reconstruction | Generate IVIM maps from diffusion-weighted MRI data using exponential fitting. | Not Listed | #3827 |
| nnUNet | Segmentation | Automated lesion segmentation across 3D T1W, FLAIR, and DIR sequences. | Not Listed | #3878 |
| In-house developed pipeline | Reconstruction | Processes MRSI data, including coil combination, k-space reconstruction, spatial filtering, lipid signal removal, and spectral fitting. | Not Listed | #3879 |
| Deep learning model (based on ResNet) | Analysis | To classify MS patients as cognitively preserved or impaired at follow-up using a pre-trained 10-layer Residual Network (ResNet) for extracting image features, combined with non-imaging data. | Not Listed | #3880 |
| Spinal Cord Toolbox (SCT) | Analysis | Co-register images, automatically segment the spinal cord and gray matter, and generate a white matter mask. | Not Listed | #3884 |
| NOW toolbox | Pulse Sequence Design | Design diffusion gradient waveforms with motion compensation. | Not Listed | #3891 |
| nnU-Nets | Analysis | To segment the lung boundary and remove vessels from images in the PREFUL post-processing pipeline. | Not Listed | #3896 |
| VOLVE | Analysis | Quantifies amplitude and timing of signal changes due to respiration in free-breathing 1H lung MRI to provide categorical ventilation metrics. | Not Listed | #3897 |
| RadiolGAN | Reconstruction | To develop a GAN-based network for high-fidelity UTE-to-CT image synthesis across various pulmonary diseases, enhancing image quality and diagnostic performance. | Not Listed | #3900 |
| FastICA | Analysis | Implements the FastICA algorithm for independent component analysis to separate ventilation and perfusion signals from cine-MRI data. | Not Listed | #3901 |
| Reiter k-MDEV inversion algorithm | Analysis | Process ASL and MRE images to calculate cortical and medullary shear-wave speed. | Not Listed | #3905 |
| BART | Reconstruction | Provides a framework for rapid prototyping of MRI acquisition and reconstruction methods with a focus on reproducibility, used here for CEST-MRI sequence definition, reconstruction, and quantification. | Not Listed | #3906 |
| None | Analysis | A data-driven, voxel-wise DCE-MRI framework for fibrosis assessment using histopathology as the reference, implemented in Python or MATLAB. | Not Listed | #3913 |
| VIBESegmentator | Analysis | Liver segmentation for each participant to create probabilistic label atlases. | Not Listed | #3914 |
| DONA | Pulse Sequence Design | Optimizes CEST sampling distributions by maximizing the combined coefficient of determination (R²) for key CEST contrasts using numerical optimization on simulated data. | Not Listed | #3919 |
| ARTEMIS | Reconstruction | Reconstructs high-fidelity breast diffusion-weighted images from multi-shot EPI data by integrating data consistency with dual task-optimized convolutional neural networks for joint aliasing suppression and k-space completion, enabling navigator-free reconstruction. | Not Listed | #3929 |
| ETICones | Pulse Sequence Design | Numerically optimize a dual-echo time cones trajectory for k-space coverage with low-magnetic field constraints. | Not Listed | #3935 |
| DESS-EPI | Pulse Sequence Design | Implemented a customized 3D DESS-EPI sequence using the PulSeq framework for multi-contrast water-only and fat-only imaging at both 3T and 0.55T. | Not Listed | #3939 |
| multi-model-direct-inversion reconstruction algorithm | Reconstruction | Post-process acquired phase images to generate stiffness maps. | Not Listed | #3943 |
| Koma simulator | Simulation | Used for Bloch simulations to optimize acquisition parameters. | Not Listed | #3949 |
| DL de-Gibbs | Reconstruction | Denoises and removes Gibbs artifacts from Cartesian ¹²⁹Xe ventilation imaging data using a deep learning model trained on natural images. | Not Listed | #3953 |
| open-source xenon image pipeline | Analysis | Process raw hyperpolarized xenon-129 MRI data to generate ventilation, diffusion, and gas exchange maps, including functional overlays with anatomical proton images. | Not Listed | #3955 |
| kMXE | Simulation | A kinetic model of xenon exchange (kMXE) is used to simulate tracer transport across tissue and blood compartments to estimate tissue and blood 129Xe dynamics within the Gas Exchange Zone (GEZ). | Not Listed | #3962 |
| In-house MATLAB-based pipeline | Analysis | To realign the EPI images, build a brain mask, apply voxelwise de-meaning, linear detrending, and Hamming-windowing to the realigned EPI time series, compute the FFT power spectrum per voxel, select band-limited power for HR, BR, and low-frequency (LF) ranges, extract per-voxel HR peak frequency and power within the HR band, derive narrowband maps at the four dominant peaks of the median spectrum, and quantize HR power into N bins and apply 3-D connected-component labeling to identify spatially coherent HR components. | Not Listed | #3969 |
| SEPIA | Analysis | Process phase data for field map fitting, phase unwrapping, background removal, and susceptibility inversion to obtain voxel-wise magnetic susceptibility. | Not Listed | #3972 |
| LRMCH-DP | Reconstruction | Respiratory motion correction and suppression of residual aliasing/noise artefacts in diffusion-weighted images using Low Rank Motion Correction (LRMC) combined with Virtual Coil Concept (VCC) and Dictionary-Patch based regularization (DP). | Not Listed | #3984 |
| OpenMRF | Pulse Sequence Design | Facilitates the prototyping and implementation of advanced MRF sequences, including non-standard magnetization preparations and rosette k-space trajectory calibration. | Not Listed | #3993 |
| nnU-Net | Reconstruction | Segment visceral fat, subcutaneous fat, and muscle volumes in whole-body MRI scans. | Not Listed | #3996 |
| ARTS software | Analysis | Computes a score linked to the likelihood a person suffers from arteriolosclerosis using MRI data. | www.nitrc.org/projects/arts | #4000 |
| mrGrad | Analysis | To segment bilateral nigrostriatal tracts (NST) along the nigra-to-striatum axis. | Not Listed | #4001 |
| SANDI toolbox | Analysis | Generate SANDI parameter maps, including neurite signal fraction (fneurite), soma signal fraction (fsoma), extracellular signal fraction (fextra), and extracellular diffusivity coefficient (De). | https://github.com/palombom/SANDI-Matlab-Toolbox-v1.0 | #4002 |
| GBSS script | Analysis | Analyzes gray matter microstructure in conjunction with NODDI to identify microstructural changes in subjective cognitive decline (SCD). | [GitHub script](GitHub script) | #4010 |
| In-house pipeline | Analysis | Computes three key distances: Frontal Horn (FH), Inter-Caudate (CC), and Inter-Table (IT) on standardised axial slices, extending classical linear ventricular indices into a 3D-distance framework. | Not Listed | #4012 |
| CSF-Flow-4D | Analysis | A semi-automatic analysis tool for sampling flow waveforms in the spinal canal and cerebral aqueduct from 4D flow MRI data. | https://github.com/tomasvikner/CSF-Flow-4D | #4013 |
| nnU-Net | Segmentation | Automated lesion segmentation of infant brain MRI using a trained 3D U-Net model. | Not Listed | #4019 |
| The Virtual Brain (TVB) | Simulation | Simulates network dynamics using subject-specific structural connectivity matrices and optimizes parameters by comparing simulated and empirical functional connectivity data. | Not Listed | #4020 |
| TractSeg | Analysis | Delineates corticospinal tracts (CSTs) from diffusion MRI data. | Not Listed | #4021 |
| DMIPY | Analysis | Implemented the three-compartment IVIM-FWI model to obtain FW, PF, and FAt metrics from diffusion MRI data. | Not Listed | #4023 |
| PT-GAN | Reconstruction | Synthesize fractional anisotropy (FA) maps from T1WI, T2WI, and T2-FLAIR images. | Not Listed | #4026 |
| HydrOptiFrame | Pulse Sequence Design | Uses a machine learning framework with numerical optimization and Bloch equation simulation to design spectrally selective water-excitation (WE) RF pulses for fat suppression in MRI. | Not Listed | #4031 |
| Deep learning based water-fat separation method | Reconstruction | Predict water-only and fat-only images from chemical shift encoded multi-echo images, robust to small water-fat phase shifts or fewer echoes. | Not Listed | #4034 |
| FERNET FW estimation | Analysis | Fits the free-water DTI model to DWI data and generates FW fraction, FA, MD, AD, and RD maps. | https://github.com/neuro-stivenr/fernet-freewater | #4046 |
| In-house software | Analysis | Derives the T1/T2 ratio and cortical thicknesses of the Desikan-Killiany atlas regions from MPRAGE and FSE images. | Not Listed | #4050 |
| Cycle-GAN | Analysis | Predict comprehensive E-field distributions, and subsequently 10g-averaged SAR, solely from B1+ maps. | Not Listed | #4053 |
| PointNet | Analysis | Estimate neurite orientation dispersion and density imaging (NODDI) parameters from diffusion MRI data with variable acquisition protocols using a permutation-invariant deep learning model. | Not Listed | #4054 |
| AlignPET | Reconstruction | Synthesizes PET-equivalent metabolic maps from MRI using a Variational Autoregressive Transformer with Structure-Aware Weighting and Linear Multistep Refinement. | Not Listed | #4056 |
| 1D Convolutional GAN | Reconstruction | Reconstruct complete cardiac flow cycles from control and stroke datasets at 1, 3, and 6 months from 4D Flow MRI data. | Not Listed | #4057 |
| PINN | Reconstruction | A two-stage Physics-Informed Neural Network (PINN) is developed for ASL quantification that embeds the ASL Buxton model as a physical constraint for robust estimation of cerebral blood flow (CBF) and arterial transit time (ATT) without relying on prior informations. | https://gitlab.dei.unipd.it/fair/pinn_for_asl | #4058 |
| UnA2LGENet | Segmentation | A generalizable, fully automatic LGE segmentation framework that unifies SAM-guided proposals with an adapter-enhanced transformer refiner for blood pool, myocardium, and infarct segmentation in multi-center LGE CMR images. | Not Listed | #4059 |
| UniDiff | Reconstruction | A unified diffusion-based framework for end-to-end brain image super-resolution, outpainting, skull-stripping, and segmentation. | Not Listed | #4060 |
| Augmentrum | Data Augmentation | A modular spectral augmentation suite for synthetic and in-vivo MRS/MRSI data that combines physical validity with flexible transformations to expand data diversity, bridge the simulation–in-vivo gap, and improve deep learning robustness. | Not Listed | #4061 |
| ITK-SNAP | Analysis | Manual segmentation of 16 organs per volunteer by 2 radiologists. | Not Listed | #4063 |
| PINN | Reconstruction | Estimate tissue-specific tissue residue function (TRF) from DSC-MRI data by embedding the DSC convolution model directly into the learning objective, enabling zero-shot learning without predefined functions or large pre-training datasets. | Not Listed | #4064 |
| Dynamic fusion foundation model | Analysis | Integrates multimodal MRI features using a self-attention mechanism to classify white matter lesions by etiology. | Not Listed | #4065 |
| JHU/KKI QSM toolbox | Reconstruction | Reconstruct quantitative susceptibility mapping (QSM) images. | https://github.com/xuli99/JHUKKI_QSM_Toolbox | #4078 |
| EPG-model-based reconstruction | Reconstruction | Reconstructs undersampled multi-echo spin-echo (MESE) data by incorporating extended phase graph (EPG) simulations to model signal evolution and account for B1+ inhomogeneities. | Not Listed | #4082 |
| interactive HTML dashboard | Analysis | Allows slice-wise inspection and subject-level comparisons of myelin volume fraction (MVF) data in the optic nerve. | https://optic-nerve-mvf.netlify.app/ | #4087 |
| qMT | Analysis | Performs joint single-point qMT (JSP-qMT) mappings to estimate MPF and apparent T1 from reduced datasets. | https://github.com/lsoustelle/qMT | #4088 |
| CAT12 toolbox | Analysis | Conduct surface-based morphometric analyses to extract cortical parameters. | Not Listed | #4094 |
| In-house MATLAB tools | Analysis | Pre-process 4D-Flow MRI data including eddy-current correction, noise-masking, and dual-venc-based anti-aliasing, and extract hemodynamic parameters from segmented vessels. | Not Listed | #4095 |
| BART toolbox | Reconstruction | Reconstruct cardiac-resolved sparse k-space datasets using a local low-rank (LLR) algorithm. | Not Listed | #4098 |
| peNLI | Reconstruction | Estimate spatial maps of shear modulus from aMRI displacements using a poroelastic nonlinear inversion finite-element framework with fluid-solid coupling. | Not Listed | #4102 |
| Python pipeline | Analysis | To estimate the geometric complexity of each subcomponent (enhancing, non-enhancing + necrotic, and edematous) using 3D fractal dimension (FD3D) and lacunarity (Lac3D) in brain metastasis and glioma patients. | Not Listed | #4111 |
| NeuDiLab | Analysis | Calculate NODDI parameters, including intracellular volume fraction (ICVF), orientation dispersion index (ODI), and isotropic volume fraction (ISOVF), from DWI data. | Not Listed | #4112 |
| 3D Slicer | Analysis | Manual segmentation of the ACL on the short-TE magnitude image. | https://www.slicer.org | #4121 |
| nnInteractive deep-learning server | Analysis | Used for interactive segmentation of regions of interest (ROIs) on 3D SPACE images. | Not Listed | #4125 |
| mr-image-tools | Analysis | Provides a CLI for single-dataset analysis of MR images. | https://github.com/gold-standard-phantoms/mr-image-tools | #4132 |
| PhySIC (Physics-informed Susceptibility Invariant Correction) | Reconstruction | A deep learning method that suppresses susceptibility-induced artifacts in MR thermometry by learning artifact characteristics and thermal field priors from simulations and applying this knowledge unsupervised to real MR temperature maps. | Not Listed | #4133 |
| BART | Reconstruction | Compressed sensing reconstruction pipeline. | Not Listed | #4139 |
| In-house MATLAB and Python tools based on the MEDI toolbox implementation of LBV | Reconstruction | Apply Laplacian Boundary Value (LBV) method to phase maps to remove susceptibility effects in MR thermometry data. | Not Listed | #4143 |
| BLAKJac | Pulse Sequence Design | Optimize flip-angle train for multi-echo multi-band MR-STAT sequence to ensure sufficient T1 and T2 encoding. | Not Listed | #4148 |
| MFEM | Simulation | Simulate E-fields using a quasistatic electromagnetic solver. | Not Listed | #4153 |
| open-birdcagebuilder toolbox | Pulse Sequence Design | Determines the capacitor values for a high-pass shielded whole-body birdcage RF transmit coil. | Not Listed | #4163 |
| UqEPT toolbox | Analysis | EPT-derived electrical properties (EPs) were obtained from in vivo data using the UqEPT toolbox. | Not Listed | #4164 |
| MFEM | Simulation | Simulates E-fields using an adaptive mesh strategy within a peripheral nerve stimulation (PNS) modeling framework to predict thresholds and stimulation sites. | Not Listed | #4165 |
| TEDANA | Analysis | Estimation of T2* and S0 maps from multi-echo fMRI data using multi-echo independent component analysis. | Not Listed | #4169 |
| TASIR | Reconstruction | A self-supervised implicit neural representation (INR) framework for fMRI reconstruction that incorporates two temporal-attention mechanisms (background/dynamics decomposition and separated spatial-temporal encodings) to improve SNR and activation fidelity. | Not Listed | #4174 |
| rDCM toolbox | Analysis | Estimates effective connectivity using dynamic causal modeling based on fMRI time-series data. | Not Listed | #4178 |
| YOLO v11 model | Analysis | Performs automatic detection and labeling of social features (human faces, animated characters, and interactive objects) in video frames. | Not Listed | #4179 |
| BioImage Suite | Analysis | Preprocess WF-Ca2+ data and compute correlation matrices using both WF-Ca2+ and oxy/deoxy-hemoglobin data. | https://bioimagesuiteweb.github.io | #4189 |
| Dedicated Phase-Reconstruction Pipeline | Reconstruction | A two-step phase-reconstruction and dictionary matching framework to remove both B0- and chemical-shift-induced off-resonance effects from fat-containing voxels, allowing to quantify T1/T2/off-resonance/tissue-fractions from a single scan. | Not Listed | #4209 |
| Voxel-wise physics-informed deep neural network (DNN) | Analysis | Estimates fatty acid composition (FAC) in mammary adipose tissue using a voxel-wise physics-informed deep neural network that models multi-echo complex signal evolution. | Not Listed | #4211 |
| VAE-GAN | Reconstruction | Generates high-fidelity abdominal synthetic CT images from a single T1-weighted MR sequence using a hybrid Variational Autoencoder-Generative Adversarial Network (VAE-GAN) with a 2.5D multi-slice architecture. | Not Listed | #4235 |
| OurGAN | Reconstruction | Synthesizes high-quality T2FLAIR images directly from T2 mapping data using a 2D end-to-end Generative Adversarial Network with a four-times upsampling strategy. | Not Listed | #4237 |
| 3D Conditional VAE with ViT-UNETR | Simulation | Generates realistic multi-sequence brain MRIs using a 3D conditional variational autoencoder with a Vision Transformer-based U-Net backbone. | Not Listed | #4238 |
| CycleGAN | Reconstruction | Synthesizes Hybrid Multidimensional MRI (HM-MRI) tissue maps from a reduced set of MRI channels using a CycleGAN architecture with wavelet feature fusion and physics-aware attention, reducing scan time while preserving diagnostic fidelity. | Not Listed | #4239 |
| TransUnet | Reconstruction | Synthesizes T2w, T2-FLAIR and T1-FLAIR images from high-resolution, distortion-free PSF-EPI DWI using a two-stage deep learning framework. | Not Listed | #4242 |
| CALAMITI | Analysis | Learns disentangled representations of anatomy and intensity from paired MRI contrasts to harmonize images across different scanners and sites. | Not Listed | #4243 |
| DeepOxyMap | Analysis | Extracts oxygenation patterns from OS-CMR images and highlights fibrosis patterns across different pathologies, using LGE as the reference ground truth. | Not Listed | #4245 |
| ANTS (Advanced Normalization Tools) | Reconstruction | Registration and template construction for lung ventilation MRI images. | http://stnava.github.io/ANTs/ | #4246 |
| T1w-GWR synthesis pipeline | Simulation | A pipeline to simulate T1-weighted gray-white matter ratio images from quantitative maps using MR-physics simulations, including B1+ correction and EPG modeling. | Not Listed | #4248 |
| Neuro GPT | Analysis | An AI-assisted tool leveraging large language models (LLM) for scientific discovery, enabling data analytics, targeted question-answering over curated documents, and reasoned literature synthesis. | Not Listed | #4250 |
| microSolenoidGeometryOptimizer | Simulation | Performs numerical simulations to maximize SNR of a micro-solenoid coil as a function of wire diameter and turn count, accounting for coil sensitivity, inductance, and effective resistance from coil, lead wires, capacitors, and sample losses. | https://github.com/BibekDhakal977/microSolenoidGeometryOptimizer | #4256 |
| PsychoPy | Simulation | To present auditory word stimuli (high-calorie, low-calorie, and non-food related) and record participant responses regarding their desire and restraint related to the stimuli. | Not Listed | #4272 |
| RESTplus | Analysis | Preprocesses rs-fMRI data, including format conversion, time point removal, slice timing correction, head motion correction, space standardization, smoothing, linear trend removal, and nuisance covariate regression. | Not Listed | #4277 |
| BrainSpace Toolbox | Analysis | Performs functional connectome gradient analysis on preprocessed fMRI data. | http://github.com/MICA-MNI/BrainSpace | #4280 |
| SCORE | Reconstruction | Mitigates the distortion-blurring tradeoff associated with VAT by providing the same correction error bound with the added capability of reconstructing both corrected and uncorrected images from a single acquisition. | Not Listed | #4287 |
| None | Reconstruction | Correct distortion artifacts in echo planar imaging (EPI) MR images by estimating field inhomogeneity using a CNN and then incorporating it into a model-based reconstruction framework. | Not Listed | #4289 |
| ECHO | Reconstruction | A new algorithm for motion and eddy-current correction in diffusion MRI that combines global, volume, and slice-level distortion models. | Not Listed | #4291 |
| SPINNED | Analysis | Accelerates deconvolution for perfusion map calculation, reducing computation time compared to conventional methods. | Not Listed | #4293 |
| NuFT-based reconstruction | Reconstruction | Implemented a NuFT-based reconstruction in Python within the Siemens Open Recon framework to correct for motion artifacts using motion parameters logged alongside k-space data. | Not Listed | #4294 |
| APART-QSM | Reconstruction | Sub-voxel quantitative susceptibility mapping to separate paramagnetic/positive, diamagnetic/negative, and total susceptibilities. | Not Listed | #4307 |
| SP-CNN | Reconstruction | Generate subject-specific vasculature models for lenticulostriate arteries (LSA) using a small-patch convolutional neural network. | Not Listed | #4320 |
| GIFT toolbox | Analysis | To perform spatial ICA to identify key resting-state networks (DMN, SN, CEN). | Not Listed | #4323 |
| Siemens Framework for Image Reconstruction Environments (FIRE) | Reconstruction | Framework used for real-time processing of calibration and navigator data, calculating field coefficients, and transferring them back to the sequence for prospective adjustment of the center frequency and linear shim gradients. | Not Listed | #4341 |
| MESS_pulseq | Pulse Sequence Design | Design and implement a multi-echo SSFP (MESS) pulse sequence in PyPulseq to acquire discrete dephasing orders and synthesize banding-free bSSFP contrast. | https://github.com/bughht/MESS_pulseq | #4345 |
| NLINV-PP | Reconstruction | Corrects phase poles in NLINV reconstructions to produce phase-pole-free images and reliably estimates coil sensitivity maps from small auto-calibration regions. | Not Listed | #4346 |
| 2PC-T1 | Reconstruction | Co-estimate T1 and B0 maps via dictionary matching using a subject-specific dictionary generated by simulating the pulse sequence. | Not Listed | #4347 |
| Physics-guided self-supervised DL framework | Reconstruction | Infers quantitative T1-/T2-/PD-maps directly from conventional T1w-, T2w-, and FLAIR-scans using a 2D U-Net-like convolutional neural network with an attention-based fusion module. | Not Listed | #4353 |
| Python scripts | Analysis | Analyze DGE-MRI data by co-registering scans, deriving masks, and obtaining R1ρ values through nonnegativity-enforcing dampened least squares fit. | Not Listed | #4358 |
| Trackvis | Analysis | Segmentation of the physis was performed manually. | http://www.trackvis.org | #4386 |
| hMRI toolbox | Reconstruction | Reconstruct quantitative parameter maps of longitudinal relaxation rate (R1), effective transverse relaxation rate (R2*), proton density, and magnetization transfer saturation (MTsat). | Not Listed | #4388 |
| 3D Slicer | Analysis | Segmentation of bone from ZTE datasets to generate 3D models for printing and surgical planning. | Not Listed | #4391 |
| Spinal Cord Toolbox | Analysis | Picture post-processing to calculate the Magnetisation Transfer Ratio (MTR) between the C2 and C5 vertebrae. | Not Listed | #4394 |
| Python tool | Analysis | Detect power-law behavior and quantify long-range temporal correlations in both simulated and real fMRI data by integrating simulation, preprocessing, spectral estimation, and statistical validation. | Not Listed | #4398 |
| FerroQuant | Analysis | A multimodal MRI framework that integrates T1, T2, T2*, and quantitative susceptibility mapping (QSM) to characterize iron in ex vivo human intracranial hematomas. | Not Listed | #4409 |
| MODEM (MOdel-free Diffusion-wEighted MRI) | Analysis | Maps raw DWI signals to pathological states using deep learning, bypassing biases and constraints of explicit model selection for tumor classification. | Not Listed | #4410 |
| IMPULSED-perfusion (IMPULSED-P) | Analysis | Incorporate perfusion into the IMPULSED framework to evaluate the perfusion effect on quantitative microstructural parameters derived from diffusion MRI. | Not Listed | #4412 |
| MATI package | Analysis | Calculate MR cytometry parameters from the IMPULSED model. | Not Listed | #4413 |
| Gadgetron | Reconstruction | Fully automated inline perfusion quantification incorporating motion correction and surface coil intensity correction. | Not Listed | #4429 |
| Gadgetron | Reconstruction | Fully automated inline perfusion quantification incorporating motion correction and surface coil intensity correction. | Not Listed | #4433 |
| PerfCine3D | Reconstruction | A self-supervised image reconstruction method that provides whole-heart coverage at high spatial and temporal resolution from a single steady-state 3D acquisition. | Not Listed | #4437 |
| v2 | Analysis | Retrain nnU-Net model v1 to improve robustness to severe pathologies and motion artifacts in fetal brain segmentation for artifact detection and selective reacquisition in fetal dMRI. | Not Listed | #4439 |
| MoCo | Reconstruction | A 3-step motion correction reconstruction was performed including breath-to-breath, intra-breath and beat-to-beat motion correction using elastix. | Not Listed | #4440 |
| δB₀-informed AlignedSENSE | Reconstruction | Jointly estimates motion, δB₀, and image reconstruction through interleaved iteration to improve motion-correction performance in neonatal 7T MP2RAGE imaging. | Not Listed | #4441 |
| RANGR | Reconstruction | Deep learning framework for motion auto-navigation using radial k-space acquisition, trained with pediatric-specific data for robust motion estimation and guiding XD-GRASP reconstruction in free-breathing pediatric MRI. | Not Listed | #4442 |
| SAMER | Reconstruction | Retrospective motion correction for high-resolution MPRAGE and MP2RAGE acquisitions at 7T by updating k-space trajectories and followed by deep learning reconstruction. | Not Listed | #4443 |
| None | Reconstruction | Correct respiratory motion in free-breathing supine breast MRI using a 3D cones trajectory with golden-angle sampling. | Not Listed | #4446 |
| CVI42 | Analysis | Measures bi-ventricular volumes from cardiac imaging data. | Not Listed | #4447 |
| Attention-U-Net | Analysis | Estimates 2D+t motion fields for cardiac motion correction in a subject-specific, unsupervised manner. | Not Listed | #4448 |
| MRTrix3 | Analysis | Diffusion data preprocessing, including denoising, eddy-current correction and motion correction. | Not Listed | #4449 |
| CONN toolbox | Analysis | Preprocesses and analyzes resting-state fMRI data, including realignment, slice-timing correction, co-registration, segmentation, normalization to MNI space, spatial smoothing, temporal filtering, and physiological noise removal. | Not Listed | #4450 |
| PVSSAS | Reconstruction | Segment perivascular spaces on T2w images within white matter masks. | Not Listed | #4454 |
| BEN toolbox 2 | Analysis | Generate voxel-wise brain entropy parametric maps from fMRI data. | Not Listed | #4455 |
| None | Pulse Sequence Design | Developed and evaluated a rapid, phase-based composite-pulse B1+ mapping method optimized for 23Na-MRI at 7T, generating a new magnetization trajectory optimized for SNR and phase accrual to achieve robust calibration for quantitative sodium imaging. | Not Listed | #4465 |
| SKINCOILS | Pulse Sequence Design | A Python script to automatically generate custom RF coil shapes within FreeCAD, enabling parametric design and export to simulation and fabrication software. | https://github.com/HFMRCrflab/SKINCOILS | #4472 |
| MPLUS | Reconstruction | Augments Mattes Mutual Information with the Normalized Gradient Field for improved boundary sensitivity in image registration. | Not Listed | #4480 |
| LLM-enhanced multi-modal network | Segmentation | To accurately segment tibiofemoral joint tissues in knee MRI by integrating visual features from MRI images with textual features generated by a Large Language Model (LLM) describing the tissues. | Not Listed | #4481 |
| Physics-Informed Detection Refinement Pipeline (PIDRP) | Reconstruction | Jointly reconstructs vascular territories and locations of corresponding feeding arteries in rVE-ASL with limited encoding steps. | Not Listed | #4482 |
| Hybrid prompted U-Net | Segmentation | To develop a deep learning framework for accurate cartilage and menisci segmentation on 7T T2* MRI with minimal manual input using a hybrid prompted U-Net. | Not Listed | #4488 |
| nnU-Net | Reconstruction | To perform robust and high-precision segmentation of thigh muscles in MRI images, even in cases with severe muscular degeneration. | Not Listed | #4489 |
| DL pipeline | Analysis | Automated tissue segmentation and volumetric quantification of the lower leg using a deep learning pipeline. | Not Listed | #4490 |
| nnDetection | Detection | An object detection framework based on RetinaU-Net architecture used for pancreatic cystic lesion detection in WB-MRI. | Not Listed | #4502 |
| Interventional MRI User Interface Extension Research Software Package | Analysis | To view and reslice online 3DRT images during interventional cardiovascular MR procedures. | Not Listed | #4510 |
| k-Wave | Simulation | Acoustic simulations of TUS transducers in realistic head models to generate ultrasound heating rate maps, which were applied in thermal simulation of a head model. | Not Listed | #4513 |
| CAFOP | Analysis | Predicts the next cine-MR image for all orthogonal imaging planes to anticipate organ motion in MRI-guided radiation therapy. | Not Listed | #4514 |
| MITK Diffusion | Analysis | Performs fiber tractography based on the Orientation Distribution Function (ODF) for Q-ball imaging (QBI). | Not Listed | #4515 |
| TenF-INR | Reconstruction | Reconstructs dynamic cardiac MRI images from highly undersampled k-space data using an unsupervised framework that integrates low-rank tensor modeling with implicit neural representation to capture temporal correlations and inter-patch similarities. | Not Listed | #4521 |
| MRH-INR | Reconstruction | Reconstructs accelerated cardiac-cine-MRI using a self-supervised implicit neural representation with multi-resolution hash encoding and ESPIRiT guidance, trained with randomly masked subsets of acquired k-space data to enforce data-consistency. | Not Listed | #4522 |
| S³-MUSIC | Reconstruction | A self-supervised spatiotemporal deep learning model for reconstructing high-quality 4D MUSIC cardiac images with very high acceleration factors (up to 32x) by integrating a complex bidirectional CRNN with low-rank sparse modelling, jointly enforcing temporal consistency and k-space fidelity. | Not Listed | #4524 |
| XD-GRASP | Reconstruction | Reconstructs 3D LGE images using temporal total variation (TTV) regularization. | Not Listed | #4525 |
| Subspace-MOCO | Reconstruction | Reconstructs motion-corrected high-resolution whole-heart spiral perfusion images using a navigator-guided subspace approach, incorporating motion estimation via multiscale registration and deformation fields into the subspace forward model. | Not Listed | #4529 |
| FlowMRI-Net | Reconstruction | Deep learning-based 4D flow MRI reconstruction that incorporates respiratory motion to improve velocity quantification. | Not Listed | #4531 |
| pTVreg | Analysis | Non-rigid registration toolbox used for extracting respiratory motion from in-vivo 5D Flow scans by registering end-expiratory and end-inspiratory states. | Not Listed | #4546 |
| Automated CMR postprocessing pipeline | Analysis | An automated pipeline for large-scale generation and analysis of personalized cardiac geometries from CMR data, including segmentation, alignment, shape model fitting, and analysis of shape mode amplitudes. | Not Listed | #4547 |
| U-Net model and ResNet-50 classifier | Analysis | A U-Net model segments transition regions in inter-frame distance maps, and a ResNet-50 classifier identifies tissue nulling types to predict myocardium and blood pool inversion times from cardiac TI-scout data. | Not Listed | #4548 |
| DENSE-specific deep learning denoiser | Reconstruction | Suppress noise in DENSE MRI images without compromising the information needed for accurate myocardial strain computation. | Not Listed | #4552 |
| PM-GAN (Physics and Mask-Guided GAN) | Reconstruction | Synthesizes high-quality, gadolinium-free CE-T1w images from non-contrast T1w and time-of-flight (TOF) inputs for improved assessment of carotid plaque vulnerability and detection of lipid-rich necrotic cores. | Not Listed | #4555 |
| ZS-SSL | Reconstruction | Reconstructs 3D images from undersampled ultra-low-field MRI data using a zero-shot self-supervised unrolled network without paired training datasets. | Not Listed | #4560 |
| DS3N | Reconstruction | Dual-domain self-supervised learning-based reconstruction to accelerate 3D non-Cartesian mcUTE acquisitions. | Not Listed | #4561 |
| INR model | Reconstruction | Reconstruct voxel-level tau-PET SUVR from regional SUVR and MRI-derived atrophy using an Implicit Neural Representation. | Not Listed | #4562 |
| Coil-STRAINER | Reconstruction | A subject-specific, self-supervised k-space implicit neural representation for ACS-free parallel MRI reconstruction that jointly learns global inter-coil correlations and local coil-dependent features. | Not Listed | #4563 |
| FEP-Net | Reconstruction | Reconstructs high-fidelity pseudo-high-field images from low-field MRI data using an unrolled network with frozen pretrained rectified flow models as expert priors and a coil-split self-supervision training paradigm. | Not Listed | #4564 |
| Implicit Neural Representation (INR) model | Reconstruction | Jointly reconstructs multiple MRI contrasts with mismatched resolutions using a single implicit neural representation (INR) model, enabling information sharing between contrasts. | Not Listed | #4565 |
| Implicit Neural Representation (INR) network | Reconstruction | Jointly reconstructs multi-inversion contrast images from undersampled k-space data by learning a continuous mapping from image coordinates to complex voxel-wise signal evolution across contrasts. | Not Listed | #4566 |
| UnrollINR | Reconstruction | A zero-shot self-supervised reconstruction method enabling scan-specific MRI reconstruction without external training data by combining a physics-guided unrolled iterative reconstruction architecture with implicit neural representation (INR) as a regularization prior. | Not Listed | #4567 |
| NSDI | Reconstruction | Jointly estimates the image and coil sensitivity maps directly from undersampled k-space data using a self-supervised reconstruction method. | Not Listed | #4568 |
| Temporal Annihilation Reconstruction Model | Reconstruction | Learns and imposes temporal annihilation relationships in x-t space for myocardial perfusion image reconstruction using zero-shot self-supervised learning. | Not Listed | #4569 |
| SS-4DMRF | Reconstruction | A self-supervised 4DMRF framework that enables accurate and efficient reconstruction of 4D tissue properties and respiratory motion from undersampled k-space signals. | Not Listed | #4570 |
| LSE-INR | Reconstruction | Reconstructs dense CEST Z-spectra from sparsely acquired data by integrating Lorentz Subspace Encoding (LSE) with Implicit Neural Representation (INR). | Not Listed | #4571 |
| ImMAP | Reconstruction | A diffusion-based method for MRI reconstruction that integrates the physics of acquisition with diffusion based on MAP estimation. | Not Listed | #4572 |
| INDI | Analysis | Process diffusion images using a consensus workflow for cardiovascular magnetic resonance. | https://github.com/ImperialCollegeLondon/INDI | #4578 |
| DESIGNER | Analysis | Processing of diffusion MRI data with linear/nonlinear b-tensor encodings. | Not Listed | #4579 |
| Federated Style-Transfer Learning Framework | Analysis | Harmonizes multi-site diffusion MRI data across sites using federated learning by extracting a compact style code representing each site’s imaging characteristics and guiding a generator network to transform source images into the target style. | Not Listed | #4583 |
| JuSpace toolbox | Analysis | To assess the structure-neurotransmitter spatial correlations. | Not Listed | #4596 |
| SEPIA toolbox | Reconstruction | QSM reconstruction was conducted using | Not Listed | #4597 |
| MIDAS | Reconstruction | Generate whole-brain metabolites maps from 3D EPSI data using Fourier transform reconstruction and automate spectral fittings (Lorentzian–Gaussian line-shape), with corrections for B0 shifts, gradient eddy currents, and signal intensity, and k-space extrapolations to reduce lipid contamination. | Not Listed | #4598 |
| LST-AI toolbox | Analysis | Deriving white matter hyperintensities (WMH) burden from 3D FLAIR images. | Not Listed | #4604 |
| GA-CS-tTV | Reconstruction | Reconstructs MR images from golden-angle radial data using compressed sensing with temporal total variation regularization to suppress artifacts caused by magnetic field fluctuations in MR-guided proton therapy. | Not Listed | #4607 |
| CRTP typed-parameter class | Pulse Sequence Design | Provides a C++ design that makes parameter math unit-safe, concise, and efficient for MRI sequence design. | Not Listed | #4610 |
| DW-RARE sequence implementation | Pulse Sequence Design | Implemented and validated single-shot and multi-shot diffusion-weighted RARE (DW-RARE) sequences within the open-source Pulseq framework, optimized for MRN, to deliver robust, high-resolution images with superior nerve visibility. | Not Listed | #4615 |
| FFA-PSF | Pulse Sequence Design | Jointly designs k-space filters and flip angles based on a point spread function to improve SNR in FSE sequences. | Not Listed | #4616 |
| UNETR | Reconstruction | Predicts B0 map at a new position from a pre-motion B0 map and a conditioning vector representing motion parameters using a UNETR architecture. | Not Listed | #4619 |
| Body Diffusion Toolbox | Analysis | Calculate mean diffusivity (MD) and mean kurtosis (MK) maps from diffusion kurtosis imaging data. | Not Listed | #4622 |
| MERA toolbox | Analysis | Calculate the myelin water fraction (MWF) map using a multicomponent T2 relaxation model. | Not Listed | #4636 |
| MIMRAT | Reconstruction | Performs skull stripping of rat brain MRI data using a U-Net trained on Wistar rat data. | Not Listed | #4641 |
| ANTS-based brain Extraction Pipeline | Reconstruction | Perform brain extraction through a series of ANTS and FSL modules using a non-linear registration approach where a template brain scan is warped into the subject’s native space. | Not Listed | #4643 |
| MCAST | Reconstruction | Uses a hybrid CNN-BiLSTM network with cross-attention to simultaneously model EMI’s spatial patterns and temporal dynamics for EMI suppression in unshielded MRI. | Not Listed | #4645 |
| System Control Unit (SCU) | Data Management | Centralizes supervision, safety management, and inter-module communication for low-field MRI systems using a modular, open-source architecture. | Not Listed | #4652 |
| MR diffusion toolbox | Analysis | Calculate parameter maps for T1, mean kurtosis (MK), mean diffusivity (MD), true diffusion coefficient (D), perfusion-related diffusion coefficient (Dp), perfusion fraction (Fp), and T2* from diffusion MRI data. | Not Listed | #4656 |
| QMRItools | Analysis | Semi-automatically stitch full-volume series from multi-station Dixon MRI acquisitions and perform image processing. | Not Listed | #4663 |
| MATI | Analysis | Derives microstructural parameters from time-dependent diffusion MRI data using the IMPULSED model. | https://github.com/jzxu0622/mati | #4668 |
| nnInteractive | Analysis | Rapidly segment surrounding tissues in 3D MR images with AI assistance to create multi-tissue segmentation maps for contrast-agnostic deep learning segmentation. | Not Listed | #4670 |
| nnU-Net | Analysis | To identify epicardial and pericardial fat on non-gated thoracic T2-weighted non-fat suppressed MR images for automated quantification of cardiac fat. | Not Listed | #4672 |
| Diffusion-based k-space inpainting framework | Reconstruction | Predicts missing k-space samples for implicit data consistency in 5D free-running cardiac MRI reconstruction, improving generalization and tolerance to undersampling compared to image-space diffusion. | Not Listed | #4675 |
| Differential Evolution global solver | Analysis | Compute a vector of shimming currents minimizing the mean residual B0-field over the targeted volume. | Not Listed | #4676 |
| AdaptCMR | Reconstruction | A parameter-efficient deep learning framework for cardiac MRI reconstruction that uses a spectrally guided mixture of experts to generalize across different views, contrasts, and acceleration factors. | Not Listed | #4680 |
| Residual U-Net | Analysis | To generate pixel-wise displacements and principal strains from cine SSFP images. | Not Listed | #4683 |
| pCSF mapping | Analysis | Quantifies the whole brain voxelwise fraction of CSF-like long T2 signal within brain parenchyma using multi-echo T2 MRI data. | Not Listed | #4686 |
| MaZda | Analysis | Texture analysis was conducted on periventricular white matter at the semi-oval center using MaZda, extracting five features—variance, skewness, kurtosis, difference entropy, and sum average—representing signal variability and complexity. | Not Listed | #4687 |
| LoRA-MCG framework | Reconstruction | A cross-field-strength framework for accelerated low-field MRI reconstruction that adapts a high-field diffusion prior to low-field contrast using LoRA updates and a k-space fidelity term, enabling high-fidelity low-field reconstruction without paired high-/low-field datasets or retraining. | Not Listed | #4699 |
| EDM-FastMRI | Reconstruction | Code for training multiscale Energy-Based Models (EBMs) via score distillation on MRI data for tasks like prior sampling, MAP estimation, MMSE estimation, and uncertainty estimation. | https://github.com/asad-aali/EDM-FastMRI.git | #4700 |
| Physics-Corrected Diffusion Model (PCDM) | Reconstruction | Reconstructs high-fidelity multi-contrast cardiac MRI from highly undersampled data by integrating a dictionary match driven by MRI physics and a gradient correction module into reverse diffusion sampling, mitigating domain shift. | Not Listed | #4701 |
| Quantization-Compensated Diffusion Refiner (QCDR) | Reconstruction | Predicts residuals between continuous and quantized representations in a VQ-VAE framework, restoring fine structures critical for clinical interpretation that are suppressed during vector quantization for T2 MRI reconstruction. | Not Listed | #4703 |
| DDfire | Reconstruction | DDfire is a diffusion-based MRI reconstruction method that improves image quality and convergence speed in accelerated multicoil brain MRI by whitening denoiser input error via colored “renoising” within a diffusion sampler. | Not Listed | #4705 |
| OptDiff | Reconstruction | A diffusion–optimization framework that adopts convex optimization and diffusion-based image priors for accelerated MRI reconstruction. | Not Listed | #4707 |
| Deep learning method based on diffusion bridge model | Reconstruction | Jointly denoises and reconstructs undersampled low-SNR images using a diffusion bridge model with a step-size-adjusted strategy. | Not Listed | #4708 |
| reg-BBrg | Reconstruction | Generates high temporal resolution (tRes) data from low tRes breast DCE-MRI data using a regularized Brownian Bridge (reg-BBrg) diffusion model for improved pharmacokinetic analysis. | Not Listed | #4710 |
| NC-PDNet | Reconstruction | Jointly reconstructs dual-contrast, undersampled multi-coil 3D MP2RAGE data and performs phase-sensitive combination for generating T1-weighted UNI images. | Not Listed | #4712 |
| All-in-One DeepGrasp | Reconstruction | A unified self-supervised reconstruction framework for highly-accelerated 4D golden-angle radial MRI that can handle different organs, resolutions, and temporal dynamics. | Not Listed | #4717 |
| FROSEN | Reconstruction | A framework for contrast-enhanced abdominal imaging that combines subspace-guided real-time image reconstruction using Multitasking, non-rigid motion correction, and compressed sensing to enable clear, motion-frozen imaging in free-breathing liver MRI. | Not Listed | #4718 |
| Diffusion bridge generative AI network | Reconstruction | Removes aliasing artifacts and motion blurring from highly accelerated free-breathing acquisitions without motion navigation by learning a distribution-to-distribution mapping. | Not Listed | #4720 |
| PSIRNet | Reconstruction | Reconstructs a phase-sensitive inversion recovery (PSIR) image from a single interleaved IR/PD acquisition using a physics-guided deep-learning method with surface coil correction. | Not Listed | #4721 |
| GLAPA (Gating based on Local All-Pass Alignment) | Reconstruction | A deep learning–based self-gating method for fetal cardiac MRI that leverages motion estimation to create a navigator signal for retrospective gating and reconstruction, enabling hardware-free tracking of the cardiac cycle. | Not Listed | #4722 |
| Real4DFlow | Reconstruction | A self-supervised deep learning-based real-time 4D flow reconstruction framework that integrates multi-dynamic manifold learning, low-rank representations, deformation-based motion modeling, and a deep image prior to achieve real-time whole-heart flow imaging. | Not Listed | #4723 |
| Spatiotemporal Transformer network | Reconstruction | Maps a pair of artifact-corrupted, phase-cycled cine inputs to an artifact-suppressed output by using a cross-attention mechanism to facilitate efficient spatiotemporal feature extraction and fusion of the two input movies. | Not Listed | #4726 |
| nnU-Net | Analysis | Predict contrast enhancement in brain tumors from quantitative MRI data (T1, T2*, QSM, and PD maps) using a deep learning model. | Not Listed | #4729 |
| DL-based OGSE fitting model | Reconstruction | Deep learning model for fast parameter estimation from oscillating gradient spin-echo (OGSE) MRI, specifically for cell diameter, intra-cellular volume fraction, extra-cellular diffusivity, and cell density. | Not Listed | #4730 |
| In-house MATLAB-based pipeline | Analysis | Process 7T MRSI data with spectral fitting by LCModel and analyze ratios of tCho, Glu, Gln, Glx, Gly, Ins, Ser and GSH to tNAA and tCr after spectral quality filtering. | Not Listed | #4734 |
| fmristroke | Analysis | Perform hemodynamic lag correction for resting-state fMRI data, specifically designed for stroke patients. | Not Listed | #4751 |
| χ-SepNet | Reconstruction | Deep neural network-based χ-separation algorithm enabling magnetic source decomposition without explicit R2 mapping. | Not Listed | #4755 |
| MSNet | Analysis | Individualized voxel-level mapping of eloquent regions and tumor areas in brain tumor patients using minimal rs-fMRI data. | Not Listed | #4765 |
| StaND | Simulation | Predicts patient-specific tau progression across entire disease course using baseline MRI, tau-PET, and clinical data. | Not Listed | #4774 |
| MR-guided PET reconstruction (MRg recon) | Reconstruction | To reconstruct PET images using a modified asymmetrical Bowsher’s algorithm, penalizing the PET image according to a similarity metric of each pixel to its neighbors, using a high-resolution MRI as a prior. | Not Listed | #4775 |
| ExploreASL | Analysis | Quantifies grey matter cerebral blood flow (CBF) using T1w, FLAIR, and ASL images. | Not Listed | #4776 |
| COMET (Correction Of Motion in Elastography via Transformation) | Reconstruction | A motion-compensation iterative reconstruction for multi-shot and joint-reconstruction acquisitions in MRE, and demonstrates the efficacy of post-reconstruction correction for single-shot MRE. | Not Listed | #4789 |
| Uncertainty-guided filter | Reconstruction | Improves EPT reconstructions by selectively weighted-averaging only the lowest-uncertainty neighboring voxels based on voxel-wise uncertainty maps derived from a bivariate error-propagation strategy. | Not Listed | #4795 |
| ANN-based GESSE | Analysis | Analyzes Gradient-Echo Sampling of the Spin Echo (GESSE) data using an artificial neural network to map muscle oxygen extraction fraction (MOEF). | Not Listed | #4798 |
| DL-ZTE | Reconstruction | Denoise ZTE4D images using a deep-learning denoising model. | Not Listed | #4803 |
| Dedicated in house-software | Analysis | Analyze MRE-data, derive metrics (in particular IG*I), and longitudinally monitor them in the tumor core and the corpus callosum. | Not Listed | #4804 |
| MoCo-MoDL | Reconstruction | Reconstructs 3D whole-heart joint T1/T2 maps from undersampled k-space data using a motion-corrected deep learning approach, incorporating motion estimation and dual-echo information for accelerated reconstruction. | Not Listed | #4811 |
| MoE-Unet | Reconstruction | An end-to-end MRI reconstruction model using Mixture-of-Experts (MoE) with unsupervised domain assignment for robust reconstruction across diverse datasets. | Not Listed | #4812 |
| Mamba-MRF | Reconstruction | Reconstructs T1 and T2 maps from highly undersampled magnetic resonance fingerprinting (MRF) data using a dual Mamba-based deep learning architecture. | Not Listed | #4815 |
| MoCo-MoDL | Reconstruction | An end-to-end non-rigid motion-corrected model-based deep learning reconstruction framework for the acceleration of 3D water/fat LGE imaging. | Not Listed | #4817 |
| Deep-learning model | Reconstruction | Enhance temporal resolution of undersampled cardiac cine MRI. | Not Listed | #4818 |
| SUNO | Pulse Sequence Design | Learns patient-specific Cartesian undersampling masks using a randomized batched iterative coordinate-descent (RB-ICD) algorithm and predicts test-time masks through a nearest neighbor search for scan-adaptive cardiac MRI. | Not Listed | #4819 |
| Deep-learning based tool | Reconstruction | Automates 3D segmentation of STIR images into muscle, fat, and edema tissue classes for radiomic feature extraction. | Not Listed | #4823 |
| EBGA | Reconstruction | Uncertainty-aware fusion mechanism that adaptively balances self-attention (anatomical features) and cross-attention (text priors) using pixel/voxel-wise evidential confidence for MRI reconstruction. | Not Listed | #4824 |
| Automated pipeline | Simulation | Generates ready-to-print CAD files for 3D-printed phantoms with appropriate materials based on high-field scans, material characterization, and segmentation of anatomical layers. | Not Listed | #4828 |
| CNN | Simulation | Predict peripheral nerve stimulation (PNS) thresholds for arbitrary gradient waveforms and axes combinations in real-time. | Not Listed | #4829 |
| Sim4Life | Simulation | Electromagnetic field simulations were performed to evaluate implant-induced SAR at 2 MHz (~0.05 T) and 127 MHz (3 T) using the finite-difference time-domain solver. | Not Listed | #4832 |
| Wrapped Gaussian Process (WGP) | Simulation | To replace computationally expensive Bloch simulations with a fast, high-fidelity surrogate model for rapid, subject-specific online LTA pulse design. | Not Listed | #4835 |
| pulsesDL (Deep Learning predicted pulses) | Pulse Sequence Design | Predict tailored pTx pulses for a 3D turbo spin echo sequence at 7T based on in vivo B1+ and B0 maps, achieving comparable excitation homogeneity to universal pulses. | Not Listed | #4836 |
| SelExNet | Pulse Sequence Design | Jointly optimizes parallel transmit RF pulses and gradient waveforms for high-fidelity, spatially selective excitation in ultra-high field MRI. | Not Listed | #4838 |
| CNN-UHP | Pulse Sequence Design | An end-to-end deep learning framework for parallel transmit design that incorporates hard constraints via an unrolled homeomorphic projection method, explicitly enforcing quadratic SAR limits. | Not Listed | #4839 |
| SsfpSym | Pulse Sequence Design | Design a symmetric Universal Pulse (UP) with null 0-th order gradient moments to maintain the “balanced” condition of bSSFP sequences, compensating for B1+ inhomogeneities at 7T. | Not Listed | #4844 |
| Spatiotemporal Maps (STM) | Reconstruction | A dynamic reconstruction framework that decomposes the dynamic image series into voxel-level temporal bases and static coefficient maps for multi-echo fMRI data, aiming to improve tSNR and functional contrast while preserving spatial resolution. | Not Listed | #4853 |
| qfBOLD | Analysis | Estimates oxygen extraction fraction (OEF) by analyzing the ratio of SE-BOLD to GE-BOLD fMRI signals acquired during a vasodilatory challenge to decouple OEF from dHb-sensitive cerebral blood volume (CBVdHb). | Not Listed | #4859 |
| Vibration-Optimized Waveform Design | Pulse Sequence Design | Design vibration-optimized oscillating gradient waveforms by solving a multi-constraint optimization problem to minimize predicted vibration by suppressing gradient spectral components that resonate with the system’s mechanical vibration. | Not Listed | #4870 |
| Optimization framework | Pulse Sequence Design | Minimize the A-weighted Sound Pressure Level (SPL) of OGSE waveforms by optimizing gradient waveforms based on the scanner’s Frequency Response Function (FRF) and various constraints. | Not Listed | #4871 |
| Pulseq | Pulse Sequence Design | Program pulse sequences which were executed on the MRI scanner and the additional gradient hardware. | Not Listed | #4873 |
| CoilGen | Pulse Sequence Design | Optimize gradient coil designs for Halbach magnets, exporting the results as .stl files for 3D printing. | Not Listed | #4874 |
| Modified Thin Wire Approximation | Pulse Sequence Design | Extended the open-source Thin Wire Approximation library to 2D thin wire and modified it to design actively shielded multi-coil gradient coils with overlapping and multi-turn coils using least-squares fitting. | Not Listed | #4878 |
| SynthSeg | Reconstruction | Segment brain MRI scans of varying resolution and contrast, including T2-weighted data, to perform brain volumetry. | Not Listed | #4879 |
| 5ttgen_neonatal | Analysis | To perform multi-atlas label fusion using the M-CRIB neonatal atlas to generate tissue segmentation suitable for Anatomically Constrained Tractography (ACT) on neonatal T1-weighted or T2-weighted images. | https://github.com/MRtrix3/mrtrix3/tree/5ttgen_neonatal | #4880 |
| Multi-BOUNTI | Reconstruction | Unified deep learning pipeline for multi-lobe segmentation of fetal and neonatal 3D T2-weighted MRI with 43 anatomically and clinically relevant ROIs. | Not Listed | #4882 |
| FreeSurfer | Analysis | To perform cortical and subcortical segmentation of T1-weighted MRI scans to extract left and right hippocampal volumes. | Not Listed | #4884 |
| SEVB-CAM-Net | Analysis | Extracts spatial structural features from T1WI, T2WI, key brain region masks, and white matter hyperintensity (WMH) masks using a 3D CNN for early CP prediction. | Not Listed | #4890 |
| Morphology Enabled Dipole Inversion (MEDI) toolbox | Reconstruction | Reconstruct QSMs from the RDF. | Not Listed | #4896 |
| 4D Flow MATLAB Toolbox | Analysis | Quantify hemodynamic parameters from 4D flow MR images using a previously validated finite-element scheme. | Not Listed | #4902 |
| HyperSLICE | Pulse Sequence Design | Joint optimization of spiral trajectories and training of a deep artefact suppression ML network for low-latency interactive cardiac imaging. | Not Listed | #4905 |
| Physics-guided self-supervised cGAN | Reconstruction | Estimates T1, T2, and proton-density maps from routine weighted images using a conditional GAN with Bloch-equation-based forward modeling for label-free training, and computes synthetic DIR images from the derived quantitative maps. | Not Listed | #4907 |
| LSE | Reconstruction | Removes banding artifacts in bSSFP imaging using a regularized least-square-error approach with three phase-cycled acquisitions, also providing B0 mapping. | Not Listed | #4908 |
| J-GRAPPA, JVC-GRAPPA, J-LORAKS | Reconstruction | Jointly reconstruct multiple contrasts (INV1 and INV2) from undersampled MP2RAGE data to improve image quality and T1 map accuracy compared to GRAPPA. | Not Listed | #4910 |
| PERK + L-BFGS framework | Reconstruction | Refines T₁, T₂, and exchange-rate maps by using PERK for initialization and L-BFGS for final refinement, enabling rapid and accurate characterization of white matter microstructure. | Not Listed | #4917 |
| APART-QSM | Reconstruction | Iteratively estimates a voxel-specific magnitude decay kernel to account for tissue heterogeneity in QSM reconstruction, improving upon the fixed relaxometric constant of $\chi$-separation. | Not Listed | #4918 |
| PCI | Analysis | Identifies the coherence transfer pathways (CTPs) responsible for out-of-voxel (OOV) artifacts in MRS data by inverting the phase cycle. | Not Listed | #4924 |
| BrainSpec | Analysis | Quantifying and visualizing multiple types of MRS data including chemical shift imaging (CSI) for monitoring disease activity and therapeutic response. | Not Listed | #4925 |
| Koopman autoencoder (KAE) | Analysis | Jointly learns a nonlinear embedding and a linear latent dynamics model for fMRI data, enabling reconstruction, prediction, and controllability analysis. | Not Listed | #4929 |
| Structural Generator | Reconstruction | Converts spatial-processing voxels from fMRI signals into latent diffusion variables for structural layout reconstruction. | Not Listed | #4930 |
| VAN model | Simulation | Simulate macrovascular intra- and extra-vascular BOLD components based on a subject-specific macroscopic vascular anatomical network (VAN) model. | Not Listed | #4933 |
| Brain Annotation Toolbox | Analysis | Extracts region-specific gene expression profiles from the Allen Human Brain Atlas. | Not Listed | #4934 |
| Pulseq | Pulse Sequence Design | Developed a multi-echo GRE sequence accelerated by wave encoding and Complementary Poisson Disk sampling, incorporating a B0 navigator and echo-shifting. | Not Listed | #4943 |
| BrainSpace toolbox | Analysis | Computes functional connectivity gradients (FCG) and aligns them via Procrustes rotation. | Not Listed | #4945 |
| meFD-T2Net | Reconstruction | Jointly performs distortion correction and direct T2 mapping from blip-reversed multi-echo EPI data using a physics-driven deep model. | Not Listed | #4954 |
| UWCL | Analysis | A semi-supervised segmentation method that leverages predicted uncertainty to distinguish high- and low-confidence regions, improving pseudo-label quality and segmentation accuracy. | Not Listed | #4959 |
| 3D SNC-PDNet | Reconstruction | To reconstruct 3D non-Cartesian ASL MRI images with improved artifact reduction and generalization by incorporating an adaptive density compensation factor (DCF) refinement module. | Not Listed | #4961 |
| Adaptive Correction Diffusion Bridges (ACDB) | Reconstruction | ACDB accelerates diffusion bridge sampling for MRI reconstruction by incorporating adaptive-correction terms into the reverse process, refining previously imputed frequencies and leveraging knowledge of the MRI measurement operator to enforce both data-fidelity and prior-consistency at every step. | Not Listed | #4964 |
| VHU-Net | Reconstruction | Bias field correction in body MRI by integrating convolutional Hadamard transform blocks with trainable semi-soft thresholding and ELBO-based variational regularization to enforce frequency-domain sparsity. | Not Listed | #4965 |
| CycleGAN | Analysis | Synthesizes X-rays from MRI and vice versa using a CycleGAN architecture with a novel combination of losses for knee joint imaging. | Not Listed | #4966 |
| DeepKAM | Reconstruction | To develop a deep learning framework using a generative adversarial network (GAN) trained on datasets with realistic motion artifacts to enable efficient and robust motion correction in free-breathing T2-FSE MRI. | Not Listed | #4969 |
| CSDS-binning | Reconstruction | Improves image quality in free-breathing liver DW-PROPELLER-EPI by sharing k-space blades from similar anatomical locations across adjacent respiratory bins to improve k-space uniformity. | Not Listed | #4975 |
| MRS4Brain toolbox | Analysis | Process MRS data, including segmentation for voxel selection. | Not Listed | #4995 |
| In-house software | Analysis | Analyze DENSE data by removing confounding phase components, estimating displacement over cardiac bins, computing the strain tensor, and defining volumetric strain. | Not Listed | #5004 |
| happy | Analysis | Derives cardiac waveforms from functional MRI acquisitions and corresponding motion alignment parameters. | Not Listed | #5009 |
| Customized MATLAB scripts | Analysis | Assess cerebrovascular reactivity (CVR) in combination with SPM12. | Not Listed | #5013 |
| QSMxT | Reconstruction | Reconstruct QSM, R2-star, and T2-star maps from multi-echo GRE MRI data. | Not Listed | #5014 |
| Gaussian PCA-Hidden Markov Model (HMM) | Analysis | Extracts rsfMRI network states, state transition matrix, and state time courses from resting-state fMRI (RS) and optogenetic fMRI (OG) data. | Not Listed | #5024 |
| MEYE | Analysis | Process eye-camera data to generate binary blink indices for arousal assessment. | Not Listed | #5026 |
| PINN-Based Tensor Diffusion EPT framework | Reconstruction | Reconstructs tissue conductivity from single-acquisition MR data with enhanced stability and boundary contrast by integrating the governing PDEs of EPT into a physics-informed neural network and introducing a spatially varying diffusion tensor guided by local electromagnetic field distributions. | Not Listed | #5029 |
| Spatially-resolved quality score (SQS) | Analysis | Derives a reference-free image quality score from learned artifact probability masks obtained through contrastive representation learning to assess image quality in brain MRI. | Not Listed | #5030 |
| Modified AutoSamp | Reconstruction | Enhanced AutoSamp framework with dual-echo cross-attention modules for joint optimization of k-space sampling and image reconstruction in accelerated dual-echo MRI, facilitating inter-echo information exchange. | Not Listed | #5032 |
| Self-supervised U-Net | Reconstruction | Denoise dynamic deuterium metabolic imaging (DMI) data in the frequency–repetition domain after an initial low-rank denoising step. | Not Listed | #5034 |
| Sequence Search | Pulse Sequence Design | Automatically designs MRI pulse sequences based on objectives and properties without prior sequence knowledge, iteratively optimizing sequence structure and parameters through interaction with a Bloch simulator and a loss function. | Not Listed | #5038 |
| GPS | Pulse Sequence Design | Design subject-specific RF pulses that adjust for subject-specific B0/B1+ inhomogeneity, improving flip-angle uniformity and image contrast at ultra-high-field MRI. | Not Listed | #5040 |
| MR-RT Seed CNN | Segmentation | Predict gold seed segmentation along with surrounding anatomical structures from MR in-phase image and the computed ΔB map. | Not Listed | #5042 |
| PINN-MRE | Analysis | Estimate the shear modulus in MR Elastography by jointly learning separate wave components and shear modulus using physics-informed neural networks. | Not Listed | #5043 |
| Tiramisu | Analysis | Performs cardiac MRI segmentation of CINE and Late Gadolinium Enhancement (LGE) images to assist in automated patient report generation. | Not Listed | #5044 |
| CTM (Cortex-to-Tract Mapping) | Reconstruction | Deep learning framework that generates bundle-specific streamlines directly from cortical surfaces, taking into account the brain anatomy, to infer cortical-to-cortical white matter connectivity. | Not Listed | #5045 |
| GCN (Graph Convolutional Network) | Analysis | Predict functional connectivity (FC) from structural connectivity (SC) using a graph convolutional network, comparing the influence of superficial white matter (SWM) and deep white matter (DWM). | Not Listed | #5048 |
| AFQ | Analysis | Automatically identifies major white matter fascicles and quantifies diffusion properties along their trajectories. | http://github.com/jyeatman/AFQ | #5050 |
| SHARPEN | Reconstruction | A self-supervised deep learning network to reconstruct and correct for the slab profile in 3D multi-slab diffusion MRI to reduce slab boundary artifacts. | Not Listed | #5060 |
| PreENCODE | Pulse Sequence Design | Numerically designs pre-compensation gradients to reduce eddy-current-induced signal loss in diffusion-prepared imaging. | https://github.com/KMoulin/DiffusionRecon-v2 | #5065 |
| None | Simulation | Simulates the influence of X-shim and Y-shim gradients and excitation frequency offset to model their effects on signal excitation for slice-specific shim optimization. | Not Listed | #5066 |
| CL+μGUIDE | Pulse Sequence Design | A machine learning framework for multidimensional MRI sequence optimization that selects a subset of measurements based on the posterior distributions of tissue parameters to reduce acquisition time while maintaining parameter estimation accuracy. | Not Listed | #5069 |
| DPMU | Pulse Sequence Design | Learns an optimal undersampling pattern through probabilistic sampling for Diffusion Spectrum Imaging, enabling accurate estimation of microstructural properties while significantly reducing scan time. | Not Listed | #5071 |
| BART (Berkeley Advanced Reconstruction Toolbox) | Reconstruction | Reconstruction of diffusion-weighted images from k-space data. | Not Listed | #5072 |
| TADRED | Reconstruction | Optimizes diffusion MRI acquisition protocols and SANDI parameter estimation by identifying the most informative measurements using a subsampling network combined with a task network that performs reconstruction or parameter estimations. | Not Listed | #5073 |
| BART | Reconstruction | Implementation of Toeplitz embedding for regularized iterative subspace reconstruction for memory-efficient iterative reconstructions and machine learning on GPUs. | Not Listed | #5075 |
| Tyger | Data Management | Provides coordinated ISMRMRD data streaming and container orchestration for real-time EPTI reconstruction on the Azure cloud. | Not Listed | #5076 |
| Phase-Guided Conditional Diffusion Framework | Reconstruction | Reconstruct high-fidelity cardiac cine images from highly undersampled real-time data by leveraging aggregated k-space priors and enforcing phase consistency. | Not Listed | #5077 |
| End-to-End Framework | Reconstruction | Integrates real-time MRI reconstruction, needle tracking, and tissue tracking for MRI-guided liver interventions with tunable reconstruction parameters optimized for latency and accuracy. | Not Listed | #5079 |
| Accelerated DIP-MRF | Reconstruction | Accelerates the Deep Image Prior (DIP) reconstruction for cardiac MR Fingerprinting (MRF) by using meta-learning to initialize the deep learning model, followed by scan-specific fine-tuning, to reduce reconstruction time. | Not Listed | #5080 |
| Clinically Significant Diagnosis (CSD) scale | Analysis | A novel internally developed scale that categorizes findings from screening whole-body MRI from 1 to 5, representing the screening-diagnostic severity and follow-up urgency. | Not Listed | #5084 |
| 3D U-Net | Reconstruction | Implemented a customised 3D U-Net with scSE modules, centre-cropped skip connections, and test-time augmentation for few-shot infant brain MRI segmentation. | Not Listed | #5089 |
| VQ-VAE | Analysis | Maps continuous visual features from a 3D Swin UNETR to discrete visual tokens using a learned codebook for alignment with a language model to generate narrative clinical reports. | Not Listed | #5100 |
| BERT (Bidirection Encoder Representations from Transformers) | Analysis | Summarizes the semantic content of prostate MRI reports into low-dimensional latent representations indicative of PI-RADS categories using contrastive learning. | Not Listed | #5102 |
| NeuroRAP | Analysis | A vision-language model trained to predict two-year prognosis in patients with multiple sclerosis (MS) and mild cognitive impairment (MCI) by integrating MRI data and automatically generated imaging reports and using a retrieval-assisted prediction mechanism. | Not Listed | #5103 |
| ITTR (Image-to-Image Translation with Transformers) | Reconstruction | Synthesizes mean diffusivity (MD) maps directly from 3D volumetric T1-weighted (T1w) MRI using a transformer-based architecture with hybrid perception blocks (HPB) and dual-pruned self-attention (DPSA). | Not Listed | #5105 |
| CGE-Diffusion | Reconstruction | Synthesizes LGE-equivalent images from contrast-free cine cardiac MR images using a diffusion model conditioned on a spatiotemporal cine encoder and multi-task learning to predict slice-level scar and MVO presence. | Not Listed | #5107 |
| Custom Python interface | Data Management | Visualisation and data collection of force measurements from the knee extensor dynamometer, as well as transmission of a synchronised trigger signal to the scanner and regulation of pulse train parameters for NMES. | Not Listed | #5116 |
| Open Sequence | Pulse Sequence Design | Open Sequence framework was used to implement and run a novel 3D TSE sequence with dynamically changing parameters across echo trains, allowing for flexible sequence specification. | Not Listed | #5118 |
| PF-SR | Reconstruction | Reconstructs high resolution knee images from low resolution data acquired at 0.05 T using a deep learning based superresolution framework. | Not Listed | #5122 |
| IPOPT | Pulse Sequence Design | Optimizes RF magnitude and phase for spectral pulses targeting multiple frequency bands and B1+ ranges to enhance fat saturation homogeneity. | Not Listed | #5123 |
| Python script | Reconstruction | Reconstruct water, fat, FF, and B0 maps from 3D-GRE-CEST-DIXON and 3D-GRE-DIXON acquisitions using a multi-peak fat model. | Not Listed | #5127 |
| SuperMAP | Reconstruction | Reconstructs quantitative T1ρ and T2 maps directly from undersampled images, reducing scan and reconstruction time compared to compressed sensing (CS). | Not Listed | #5137 |
| nnU-Net | Segmentation | Automatic segmentation model to localize solitary spinal lesions (SSLs) using sagittal CET1 and axial DWI images. | Not Listed | #5142 |
| BERC | Analysis | Classifies breast cancer ER status (negative, low positive, high positive) from DCE-MRI using deep learning. | Not Listed | #5143 |
| ITK-SNAP | Segmentation | Delineating breast lesions on peak-phase DCE-MRI images. | Not Listed | #5146 |
| Spatial-Temporal Encoding Methodology (STEM) | Analysis | Automated processing of DCE-MRI time-intensity curve data to differentiate benign and malignant breast tumors and provide tumor staging capabilities. | Not Listed | #5151 |
| QQ-F | Reconstruction | A deep learning framework for OEF mapping that generalizes across diverse TE schemes without the need for retraining, by extracting QQ-related features from MRI signals using a feature extraction unit (FEU) and U-NET. | Not Listed | #5161 |
| GenIQ | Analysis | To measure the quantitative parameter, Ktrans, of the DCE-MRI images. | Not Listed | #5166 |
| SSL-QALAS++ | Analysis | Estimates T1/T2/T2*/PD/IE maps from multi-echo QALAS images while accounting for imperfect inversion during the IR pulse by using a CNN with a self-consistency loss between acquired and Bloch-simulated images, and an IE optimization loss. | Not Listed | #5170 |
| fastGRAPE | Pulse Sequence Design | Optimizes RF and gradient waveforms for parallel transmission pulse design, addressing B₀ and B₁⁺ inhomogeneities in ultra-high-field MRI. | Not Listed | #5175 |
| nnUNet | Segmentation | Generates segmentations of the myocardium and blood pool from ²³Na/¹H images. | Not Listed | #5176 |
| PCNtoolkit | Analysis | Perform Hierarchical Bayesian Regression (HBR) for normative modeling of muscle volume and fat. | Not Listed | #5178 |
| 3D Slicer | Analysis | Segmentation of the paraspinal muscles (erector spinae and multifidus) at the L2 vertebral level was performed on PDFF maps at both time points. | Not Listed | #5184 |
| S2V-DREME | Reconstruction | Reconstructs time-resolved volumetric MRIs from orthogonal 2D MRI slice pairs by jointly solving a reference 3D MRI and a motion model. | Not Listed | #5193 |
| Shimmer | Simulation | A passive shimming toolbox to systematically improve the B0 field homogeneity in low-field magnets by computing the configuration of shims using optimization algorithms and generating 3D printable CAD files for installation. | https://gitlab.com/osii/shimming/shimmer | #5196 |
| Julia-based shimming tool | Simulation | This open-source tool creates a consistent and transparent B0 shimming workflow on multiple vendors’ MRI scanners by harmonizing the field map acquisition, shim volume selection, and shim calculation algorithm. | https://github.com/HarmonizedMRI/functional | #5199 |
| Shim Control | Analysis | A MATLAB application that integrates masking, fieldmap processing, optimization, and hardware communication for hybrid Siemens and MCA shims at 7 T and 11.7 T to improve B₀ shimming performance and reproducibility. | Not Listed | #5200 |
| MEIGO toolbox | Analysis | Global optimization for determining the optimal distribution of ferromagnetic material to reduce B0 variation. | Not Listed | #5202 |
| OmniShim Toolbox | Analysis | Calibrates and performs B0 shimming in user-defined regions of interest across human MRI scanners from different vendors, supporting various shim orders, using a 3D B0 map with complete cross-term calibration of all shim channels. | Not Listed | #5208 |
| Custom Framework (integrating Genetic Algorithm and 3D Finite Element Method) | Simulation | Optimizes the mechanical structure of permanent magnets for low-field and ultra-low field MRI, balancing uniformity, field strength, and system weight using a genetic algorithm and 3D finite element method. | Not Listed | #5210 |
| Longitudinal-QSM | Reconstruction | Simultaneously reconstructs QSM data from two time points, accounting for head rotation and background field variations while enforcing spatial sparsity on susceptibility changes. | Not Listed | #5242 |
| synthMWF | Analysis | Generate synthetic myelin water fraction (MWF) maps from T1 maps to improve cross-site comparability and reproducibility in multicenter studies. | Not Listed | #5246 |
| Fully convolutional neural network | Reconstruction | Automated segmentation of choroid plexus from 3D T1-weighted MR images. | Not Listed | #5258 |
| DPARSF | Analysis | Preprocesses resting-state fMRI data. | Not Listed | #5262 |
| SLOMOCO | Reconstruction | Implements the SLice-Oriented MOtion COrrection (SLOMOCO) method in the HCP pipeline to reduce motion effects on functional connectivity. | https://github.com/wanyongshinccf/HCPpipelines-CCF.git | #5266 |
| SDM-PSI | Analysis | Perform meta-analysis by coordinate-based method to identify consistent cerebral cortex volume changes. | http://www.sdmproject.com | #5281 |
| HYDRA | Analysis | Identifies distinct neuroanatomical subtypes of T2DM based on cortical morphometric similarity networks. | Not Listed | #5284 |
| QQ-NET | Reconstruction | Calculate oxygen extraction fraction (OEF) from 3D multi-echo GRE data. | https://github.com/jc2852/QQ-NET | #5286 |
| NBS-predict | Analysis | Predict subject-level connectomes associated with long-term cognitive worsening using network-based statistics. | Not Listed | #5291 |
Leave a Reply