MosaicMRI provides a diverse raw MSK MRI dataset that enables deep learning models to exploit cross-anatomical correlations, outperforming anatomy-specific training in low-sample regimes for accelerated reconstruction.
hub
fastMRI: An Open Dataset and Benchmarks for Accelerated MRI
31 Pith papers cite this work. Polarity classification is still indexing.
31
Pith papers citing it
abstract
Accelerating Magnetic Resonance Imaging (MRI) by taking fewer measurements has the potential to reduce medical costs, minimize stress to patients and make MRI possible in applications where it is currently prohibitively slow or expensive. We introduce the fastMRI dataset, a large-scale collection of both raw MR measurements and clinical MR images, that can be used for training and evaluation of machine-learning approaches to MR image reconstruction. By introducing standardized evaluation criteria and a freely-accessible dataset, our goal is to help the community make rapid advances in the state of the art for MR image reconstruction. We also provide a self-contained introduction to MRI for machine learning researchers with no medical imaging background.
hub tools
citation-role summary
dataset 2
background 1
method 1
citation-polarity summary
representative citing papers
AutoMedBench evaluates AI agents on long-horizon medical workflows across five stages and finds validation and submission as dominant failure points based on thousands of runs.
Hallucinations in inverse problem reconstructions are fundamental to ill-posedness, with necessary and sufficient conditions plus computable bounds depending only on the forward model.
UNITS framework proves self-supervised splitting risk in MRI reconstruction is a weighted supervised risk, yielding identical Bayes-optimal predictors and relating training residuals to prediction bias.
A temporal diffusion model enables adaptive selection of focused ultrasound transmits, outperforming random subsampling and diverging waves on EchoNet-Dynamic and in-house echocardiogram datasets while supporting real-time operation.
MSC is a black-box terminal operator based on classical range/null-space data consistency that reduces measured-subspace dispersion in diffusion MRI samplers while preserving unmeasured diversity and image quality metrics.
MoRE integrates a sparsely activated MoE module with unsupervised routing into a variational network for stable multimodal MRI reconstruction on fastMRI brain and knee data at 8x undersampling.
MeniOmni is a new structured multimodal benchmark dataset and evaluation framework for fine-grained Stoller grading and diagnostic report generation from knee MRI combined with clinical priors.
CoilDrop-MRI uses coil dropout in self-supervised training for parallel MRI reconstruction and outperforms prior self-supervised methods while matching supervised quality across multi-site, multi-field, and multi-modality data.
SO-Mamba introduces state-ownership routing in Mamba regularizers for unrolled MRI reconstruction to separate resident carrier content from non-resident evidence across stages.
NexOP jointly optimizes NEX-aware k-space sampling probabilities and multi-measurement reconstruction to raise effective SNR in low-field MRI under a fixed total sampling budget.
Conditional score-based diffusion models synthesize phase maps from magnitude-only MR images to generate k-space datasets that train superior deep learning models for accelerated MRI reconstruction compared to smooth-phase or GAN-based alternatives.
Natural-domain foundation models provide competitive and more robust priors than task-specific models for accelerated cardiac MRI reconstruction in cross-domain settings.
A cVAE plus flow-matching model generates realistic complex-valued brain MRI that preserves phase coherence above 0.997 and yields synthetic data that trains abnormality classifiers to 0.880 AUROC, beating the 0.842 real-data baseline on fastMRI.
An information-theoretic optimization framework for task-adapted CS-MRI enables adaptive sampling at arbitrary ratios and probabilistic inference for uncertainty while supporting joint reconstruction-task or privacy-focused scenarios.
PASS combines a vision-language model with physics-based deep unrolling to create personalized, anomaly-aware fast MRI that improves image quality and downstream diagnostic tasks.
Noise injection into plug-and-play algorithms using pretrained score-based diffusion denoisers optimizes a Gaussian-smoothed objective and yields better reconstructions for severely ill-posed imaging tasks.
Diffusion inverse solvers are assessed for posterior fidelity using a new score-based Kernel Stein Discrepancy metric that requires no ground-truth posterior, revealing that reconstruction accuracy alone is insufficient.
iR2D2 extends the R2D2 DNN series paradigm with an interlaced dual-series architecture and error-controlled updates to jointly reconstruct MR images and self-calibrate sensitivity maps from undersampled radial k-space data.
Bilevel-optimized implicit neural representation with Gaussian process hyperparameter tuning enables scan-specific accelerated MRI reconstruction without training data.
SUNO learns per-scan adaptive k-space undersampling patterns via ICD optimization and NN lookup from low-frequency data, showing better reconstruction quality than standard patterns at 4x and 8x acceleration on fastMRI knee and brain data.
A task-aware flow-based generative framework optimizes subsampling masks in compressed sensing, reporting SOTA PSNR of 25.17 dB at 5% rate on CelebA and 29.24 dB for 8x MRI on fastMRI.
Introduces a cascade-based structural decomposition of posterior uncertainty to isolate intrinsic ambiguity from estimation uncertainty in deep generative models for linear inverse problems.
MK-ResRecon predicts missing slices with a multi-kernel texture-aware loss while IdentityRefineNet3D refines the combined 3D volume to enable accurate reconstruction from highly sparse 2D inputs.
citing papers explorer
No citing papers match the current filters.