MosaicMRI: A Diverse Dataset and Benchmark for Raw Musculoskeletal MRI

TL;DR

MosaicMRI introduces the largest diverse raw musculoskeletal MRI dataset to advance deep learning applications, enabling comprehensive evaluation of model robustness, generalization, and the impact of anatomical diversity.

Contribution

This work provides the first large-scale, diverse open-source MSK MRI dataset and systematically studies the effects of anatomy, dataset size, and model capacity on reconstruction performance.

Findings

Models trained on combined anatomies outperform anatomy-specific models in low-sample regimes.

Cross-anatomy training improves generalization, with certain body parts showing strong transferability.

Performance under domain shifts depends on training set size, anatomy, and protocol-specific factors.

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 2,671 volumes and 80,156 slices. 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), and numbers of acquisition coils. Using VarNet as a baseline for accelerated reconstruction task, we perform a comprehensive set of experiments to study scaling behavior with respect to both model capacity and dataset size. Interestingly, models trained on the combined anatomies significantly outperform anatomy-specific models in low-sample regimes, highlighting the benefits of anatomical diversity and the presence of exploitable cross-anatomical correlations. We further evaluate robustness and cross-anatomy generalization by training models on one anatomy (e.g., spine) and testing them on another (e.g., knee). Notably, we identify groups of body parts (e.g., foot and elbow) that generalize well with each other, and highlight that performance under domain shifts depends on both training set size, anatomy, and protocol-specific factors.