A Structured Review and Quantitative Profiling of Public Brain MRI Datasets for Foundation Model Development

TL;DR

This paper systematically reviews 54 public brain MRI datasets, analyzing their variability and heterogeneity to inform the development of robust foundation models, highlighting the importance of preprocessing and domain adaptation.

Contribution

It provides a comprehensive, multi-level characterization of public brain MRI datasets and evaluates preprocessing effects, emphasizing the need for harmonization in foundation model development.

Findings

Significant heterogeneity in voxel spacing, orientation, and intensity across datasets.

Preprocessing improves within-dataset consistency but residual inter-dataset differences remain.

Residual covariate shift persists after standard preprocessing, affecting model generalization.

Abstract

The development of foundation models for brain MRI depends critically on the scale, diversity, and consistency of available data, yet systematic assessments of these factors remain scarce. In this study, we analyze 54 publicly accessible brain MRI datasets encompassing over 538,031 to provide a structured, multi-level overview tailored to foundation model development. At the dataset level, we characterize modality composition, disease coverage, and dataset scale, revealing strong imbalances between large healthy cohorts and smaller clinical populations. At the image level, we quantify voxel spacing, orientation, and intensity distributions across 15 representative datasets, demonstrating substantial heterogeneity that can influence representation learning. We then perform a quantitative evaluation of preprocessing variability, examining how intensity normalization, bias field correction, skull stripping, spatial registration, and interpolation alter voxel statistics and geometry. While these steps improve within-dataset consistency, residual differences persist between datasets. Finally, feature-space case study using a 3D DenseNet121 shows measurable residual covariate shift after standardized preprocessing, confirming that harmonization alone cannot eliminate inter-dataset bias. Together, these analyses provide a unified characterization of variability in public brain MRI resources and emphasize the need for preprocessing-aware and domain-adaptive strategies in the design of generalizable brain MRI foundation models.