MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI

Paula Arguello; Berk Tinaz; Mohammad Shahab Sepehri; Maryam Soltanolkotabi; Mahdi Soltanolkotabi

cs.CVApr 13, 2026

MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI

Authors: Paula Arguello, Berk Tinaz, Mohammad Shahab Sepehri, Maryam Soltanolkotabi, Mahdi Soltanolkotabi

Organizations: Department of Computer Science, University of Southern California · USC Center on AI Foundations for the Sciences · Department of Electrical and Computer Engineering, University of Southern California · Department of Radiology and Imaging Sciences, University of Utah

Abstract

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.

Figures & tables

Panel of Figure 1 — Figure 1: MosaicMRI overview. (left) Anatomy distribution for 3,096 fully sampled 1.5 T volumes, led by spine (50.9%, 1,576 volumes), shoulder (14.4%, 446), and hip (8.9%, 274), with all ten anatomy categories shown separately. (right) Representative slices spanning seven anatomy groups and three orientations.

Dataset	Anatomy	View	Contrast	Vendor	Magnet	Coils	# Scans / # Subjects
fastMRI brain [ 33 ]	brain	axial	T1, T1POST, T2, FLAIR	Siemens	1.5T, 3T	4-20	6.4k / 6.4k
OCMR [ 7 ]	heart	various	SSFP	Siemens	0.5T - 3T	15-38	165 / 165
AHEAD [ 3 ]	brain	various	MP2RAGE-ME	Philips	7T	32	105 / 105
M4Raw [ 23 ]	brain	axial	T1, T2, FLAIR	XGY	0.3T	4	1.3k / 183
CMRxRecon2023 [ 32 ]	heart	various	SSFP-Balanced	Siemens	3T	10	300 / 300
fastMRI prostate [ 29 ]	prostate	axial	T2, DWI	Siemens	3T	10-30	312 / 312

Table 1: Comparison of public MRI datasets with raw k-space data. Statistics are compiled from respective dataset papers and Table 1 of [ 22 ] .

Dataset split	# scans	# patients	# slices
train	2,166	598	61,903
val	465	128	13,265
test	465	128	13,248
0.55T low-field	54	8	1,414

Table 2: Dataset statistics for the

1.5

T splits and the

0.55

T out-of-distribution set.

	$4\times$ accel.		$8\times$ accel.
Method	PSNR (dB) $\uparrow$	SSIM $\uparrow$	PSNR (dB) $\uparrow$	SSIM $\uparrow$
ESPIRiT	$32.34\!\pm\!5.21$	$0.627\!\pm\!0.193$	$30.75\!\pm\!4.44$	$0.598\!\pm\!0.189$
E2E-VarNet	$\textbf{41.62}\!\pm\!4.25$	$\textbf{0.946}\!\pm\!0.046$	$\textbf{38.57}\!\pm\!4.01$	$\textbf{0.924}\!\pm\!0.054$
U-Net	–	–	$37.44\!\pm\!4.34$	$0.913\!\pm\!0.056$
ViT	–	–	$36.61\!\pm\!4.36$	$0.904\!\pm\!0.059$
LBM (1 step)	$33.12\!\pm\!3.13$	$0.844\!\pm\!0.076$	–	–

Table 3: Test-set reconstruction of various methods on MosaicMRI at

4\times

and

8\times

acceleration. Values are mean

\pm

standard deviation over 465 volumes.

Figure 2: PSNR versus training-set fraction for E2E-VarNet with 4, 8, and 12 cascades, stratified into extremity and non-extremity test volumes. All panels use the same y-axis scale.

Model	10%	20%	50%	100%
U-Net	35.78 / 0.898	36.36 / 0.903	36.99 / 0.909	37.44 / 0.913
ViT	32.73 / 0.853	32.04 / 0.837	35.35 / 0.890	36.61 / 0.904

Table 4: Architecture scaling with MosaicMRI training data for the held-out test split.

Model	Test anatomy	Target anatomy only	Full MosaicMRI
U-Net	Knee	37.35 / 0.894	37.94 / 0.898
U-Net	Shoulder	42.05 / 0.952	42.41 / 0.954
ViT	Knee	34.01 / 0.850	37.47 / 0.892
ViT	Shoulder	35.79 / 0.912	41.66 / 0.950

Table 5: PSNR (dB) / SSIM for reconstruction with mixed-anatomy training across architectures.

Figure 3: Protocol generalization in MosaicMRI with E2E-VarNet trained on each protocol (columns) and tested on each protocol (rows); Baseline is trained on all protocols. Boxes mark single-protocol models within 1 dB of the best per row.

Figure 4: Mean PSNR (dB) of E2E-VarNet for cross-anatomy transfer on MosaicMRI . Rows are test anatomies, and columns are the training anatomy; the final Baseline column corresponds to the model trained on all data. Anatomies are ordered from higher to lower volume counts (top-left to bottom-right), yielding three groups: high-data anchors (blue), distal extremities (brown), and low-data groups (pink). For each test anatomy (row), black outlines mark all anatomy models within 1 dB of the best anatomy result for that row (excluding the baseline).

	Model trained on
Test set	fastMRI	MosaicMRI	MosaicMRI (knee)
fastMRI	36.90 / 0.888	35.89 / 0.876	33.13 / 0.832
MosaicMRI	34.76 / 0.886	38.57 / 0.924	35.81 / 0.889
MosaicMRI (knee)	33.04 / 0.849	38.09 / 0.891	37.95 / 0.890

Table 6: Cross-dataset performance in PSNR (dB) and SSIM. The MosaicMRI (knee) model is trained only on the knee subset of MosaicMRI . Values in bold represent the best model by test set.

	Broad/full training			Matched slice budget
Test anatomy (contrast)	Anatomy-only	Contrast-specific	MosaicMRI	Slice-matched random	Anatomy+contrast	DreamSim
Knee (PD-FS)	34.41 / 0.802	34.54 / 0.803	34.61 / 0.806	32.95 / 0.781	33.40 / 0.787	33.75 / 0.792
Training slices	5,864	8,661	61,903	541	541	541
Shoulder (PD-FS)	44.64 / 0.974	44.39 / 0.973	44.63 / 0.974	42.96 / 0.968	42.81 / 0.968	43.63 / 0.971
Training slices	8,534	8,661	61,903	1,744	1,744	1,744

Table 7: Controlled-subset PSNR (dB) / SSIM and training-slice counts. The vertical rule separates the broad/full-training comparison from the slice-budget-matched comparison; bold denotes the best result within each three-model group.

		Test set
Model	Training data	Knee	Shoulder	Spine	Average
	MosaicMRI (10%)	29.285 / 0.753	28.794 / 0.676	28.385 / 0.708	28.821 / 0.712
E2E-VarNet	MosaicMRI (50%)	29.836 / 0.773	29.028 / 0.688	28.443 / 0.720	29.102 / 0.727
	MosaicMRI	30.184 / 0.788	29.382 / 0.711	29.202 / 0.757	29.590 / 0.752
	MosaicMRI (10%)	29.759 / 0.805	28.449 / 0.712	29.205 / 0.777	29.093 / 0.762
U-Net	MosaicMRI (50%)	29.855 / 0.798	28.858 / 0.715	29.564 / 0.780	29.396 / 0.762

Table 8: Zero-shot reconstruction on the 0.55 T low-field set at

8\times

acceleration. Each cell reports PSNR (dB) / SSIM; bold marks the best metric within each architecture and test-set column.

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 5: Distribution of the 0.55 T low-field set by patient, anatomy, and contrast.

Figure 6: Dataset diversity by anatomy. Violin/box plots summarize the distribution of receive-coil counts and slices per volume for each anatomy group. A contrast panel reports the presence of major contrast families (T1, T1-FS, T2, T2-FS, PD, PD-FS, STIR) per anatomy, reflecting protocol heterogeneity. A stacked bar chart shows the orientation mix (axial/sagittal/coronal) within each anatomy, highlighting anatomy-dependent acquisition geometry.

Metric	spine	knee	shoulder	hip	ankle	foot	wrist/hand	elbow	pelvis	tib/fib
FOV x (mm)	238.28	159.42	162.69	281.82	165.86	155.96	109.46	167.38	305.26	257.37
FOV y (mm)	237.46	156.31	162.43	278.46	166.59	150.62	99.13	166.36	294.00	243.40
Res x (mm)	0.7016	0.3674	0.3076	0.7881	0.4483	0.4364	0.3185	0.5200	0.8628	0.7128
Res y (mm)	0.7073	0.4138	0.4254	0.8983	0.4483	0.4527	0.3319	0.5620	0.9485	0.8700

Appendix

Table 9: Average in-plane field-of-view (FOV) and in-plane resolution by anatomy.

Panel of Figure 7 — Figure 7: U-Net cross-anatomy (left) and cross-contrast (right) PSNR. Columns are training subsets and rows are test subsets; All data is trained on the complete training split.

Panel of Figure 8 — Figure 8: Qualitative accelerated reconstruction examples across anatomies. For each panel, columns (left to right) show the masked k-space after applying the undersampling pattern, the zero-filled RSS reconstruction, the reconstruction produced by VarNet trained on full MosaicMRI , and the fully sampled target.

Explore similar work

OrthoDiffusion: A Generalizable Multi-Task Diffusion Foundation Model for Musculoskeletal MRI Interpretation
Feb 24, 2026Tian Lan, Lei Xu, Zimu Yuan +8Medical Imaging Foundation Models Diffusion Models
Center for Applied Statistics and School of Statistics, Renmin University of China, Beijing, China. · Department of Sports Medicine, Peking University Third Hospital, Institute of Sports Medicine of Peking University, Beijing, China. · Beijing Key Laboratory of Research and Translation for Drugs and Medical Devices in Precision Diagnosis and Treatment of Sports Injuries, Beijing, China. +4
A4-MR: Abnormality-aware Anatomical Alignment for MR
Oct 9, 2026Noel D'souza, Zhijian Yang, Nishank Singla +14Medical Imaging Foundation Models Radiology VLMs
GE Healthcare, Seattle, USA · GE Healthcare, Budapest, Hungary · Amazon, USA
KneePreM: Towards 3D Knee MRI Foundation Models via Large-Scale Unlabeled Pretraining and Label-Efficient Fine-Tuning
Sep 25, 2026Xinxin Wang, Liam Hazan, Jing Li +3Medical Imaging Foundation Models Self-Supervised Pre-Training
Program of Advanced Musculoskeletal Imaging, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, United States · Department of Biomedical Engineering, Lerner Research Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, United States · Department of Computer and Data Science, Case Western Reserve University, 10900 Euclid Avenue, Cleveland, OH 44106, United States +2

Feb 24, 2026J/K move · Enter open · S save

OrthoDiffusion: A Generalizable Multi-Task Diffusion Foundation Model for Musculoskeletal MRI Interpretation

Tian Lan, Lei Xu, Zimu Yuan, Shanggui Liu+7

Center for Applied Statistics and School of Statistics, Renmin University of China, Beijing, China. · Department of Sports Medicine, Peking University Third Hospital, Institute of Sports Medicine of Peking University, Beijing, China. · Beijing Key Laboratory of Research and Translation for Drugs and Medical Devices in Precision Diagnosis and Treatment of Sports Injuries, Beijing, China.+4

Medical Imaging Foundation Models Diffusion Models Magnetic Resonance Imaging Medical Image Segmentation Medical Image Classification

Open paper AskPDF arXiv

Musculoskeletal disorders represent a significant global health burden and are a leading cause of disability worldwide. While MRI is essential for accurate diagnosis, its interpretation remains exceptionally challenging. Radiologists must identify multiple potential abnormalities within complex anatomical structures across different imaging planes, a process that requires significant expertise and is prone to variability. We developed OrthoDiffusion, a unified diffusion-based foundation model designed for multi-task musculoskeletal MRI interpretation. The framework utilizes three orientation-specific 3D diffusion models, pre-trained in a self-supervised manner on 15,948 unlabeled knee MRI scans, to learn robust anatomical features from sagittal, coronal, and axial views. These view-specific representations are integrated to support diverse clinical tasks, including anatomical segmentation and multi-label diagnosis. Our evaluation demonstrates that OrthoDiffusion achieves excellent performance in the segmentation of 11 knee structures and the detection of 8 knee abnormalities. The model exhibited remarkable robustness across different clinical centers and MRI field strengths, consistently outperforming traditional supervised models. Notably, in settings where labeled data was scarce, OrthoDiffusion maintained high diagnostic precision using only 10% of training labels. Furthermore, the anatomical representations learned from knee imaging proved highly transferable to other joints, achieving strong diagnostic performance across 11 diseases of the ankle and shoulder. These findings suggest that diffusion-based foundation models can serve as a unified platform for multi-disease diagnosis and anatomical segmentation, potentially improving the efficiency and accuracy of musculoskeletal MRI interpretation in real-world clinical workflows.

Abstract

Explore similar work

OrthoDiffusion: A Generalizable Multi-Task Diffusion Foundation Model for Musculoskeletal MRI Interpretation

A4-MR: Abnormality-aware Anatomical Alignment for MR

KneePreM: Towards 3D Knee MRI Foundation Models via Large-Scale Unlabeled Pretraining and Label-Efficient Fine-Tuning