mlsanityData to 14 Oct 2026

MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI

arXiv 2604.11762 ↗First posted 13 Apr 2026cs.CV cs.LG eess.SPcited by 0

Not on the map. Why?

Abstract (arXiv)

Deep learning underpins a wide range of applications in MRI, including reconstruction, artifact removal, and segmentation. However, progress has been driven largely by public datasets focused on brain and knee imaging, shaping how models are trained and evaluated. As a result, careful studies of the reliability of these models across diverse anatomical settings remain limited. In this work, we introduce MosaicMRI, a large and diverse collection of fully sampled raw musculoskeletal (MSK) MR measurements designed for training and evaluating machine-learning--based methods. MosaicMRI is the largest open-source raw MSK MRI dataset to date, comprising 3,150 volumes and 89,830 slices, including a 3,096-volume 1.5~T core dataset and a 54-volume 0.55~T low-field set. The dataset offers substantial diversity in volume orientation (e.g., axial, sagittal), imaging contrasts (e.g., PD, T1, T2), anatomies (e.g., spine, knee, hip, ankle, and others), numbers of acquisition coils, and field strengths. Using accelerated reconstruction as a testbed, we study scaling, robustness, and data selection under realistic MSK distribution shifts. Across E2E-VarNet, U-Net, and ViT baselines, mixed-anatomy training improves over anatomy-specific training, showing that the benefit of anatomical diversity is not architecture-specific. Zero-shot experiments on the 0.55~T low-field set highlight that using more training data improves average reconstruction under field-strength shift. Controlled subset experiments further suggest that these gains come from transferable cross-anatomy structure, rather than anatomy or contrast matching alone.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 21 of its 40 references that are in the map. First posted on arXiv 13 Apr 2026.

Builds on

21 of 40 references.

References · 21

  1. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  2. 2014ImageNet Large Scale Visual Recognition Challengecited by 8,498
  3. 2018UNet++: A Nested U-Net Architecture for Medical Image Segmentationcited by 1,525
  4. 20163D U-Net: Learning Dense Volumetric Segmentation from Sparse Annotationcited by 1,476
  5. 2017The Perception-Distortion Tradeoffcited by 587
  6. 2018fastMRI: An Open Dataset and Benchmarks for Accelerated MRIcited by 470
  7. 2017Learning a Variational Network for Reconstruction of Accelerated MRI Datacited by 342
  8. 2021Score-based diffusion models for accelerated MRIcited by 311
  9. 2020End-to-End Variational Networks for Accelerated MRI Reconstructioncited by 202
  10. 2021ResViT: Residual vision transformers for multi-modal medical image synthesiscited by 177
  11. 2018Image Synthesis in Multi-Contrast MRI with Conditional Generative Adversarial Networkscited by 132
  12. 2017Deep learning for undersampled MRI reconstructioncited by 91
9 more
  1. 2017Framing U-Net via Deep Convolutional Framelets: Application to Sparse-view CTcited by 88
  2. 2021Measuring Robustness in Deep Learning Based Compressive Sensingcited by 53
  3. 2020OCMR (v1.0)--Open-Access Multi-Coil k-Space Dataset for Cardiovascular Magnetic Resonance Imagingcited by 36
  4. 2022HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstructioncited by 30
  5. 2019i-RIM applied to the fastMRI challengecited by 16
  6. 2023Robustness of Deep Learning for Accelerated MRI: Benefits of Diverse Training Datacited by 10
  7. 2019Sigma-net: Systematic Evaluation of Iterative Deep Neural Networks for Fast Parallel MR Image Reconstructioncited by 7
  8. 2024fastMRI Breast: A publicly available radial k-space dataset of breast dynamic contrast-enhanced MRIcited by 7
  9. 2025Improving Deep Learning for Accelerated MRI With Data Filteringcited by 2

Built on by

Not cited yet.