Region-Aware Reconstruction Strategy for Pre-training fMRI Foundation Model

Ruthwik Reddy Doodipala, Pankaj Pandey, Carolina Torres Rojas, Manob Jyoti Saikia, Ranganatha Sitaram

Keywords: foundation models, self-supervised learning, fMRI, masked autoencoder, neuroimaging

TL;DR: We introduce an ROI-guided masking strategy for fMRI foundation model pre-training that improves interpretability and classification performance by integrating anatomical region masking into self-supervised masked autoencoding.

Abstract:

The emergence of foundation models in neuroimaging is driven by the increasing
availability of large-scale and heterogeneous brain imaging datasets. Recent ad-
vances in self-supervised learning, particularly reconstruction-based objectives,
have demonstrated strong potential for pretraining models that generalize effec-
tively across diverse downstream functional MRI (fMRI) tasks. In this study, we ex-
plore region-aware reconstruction strategies for a foundation model in resting-state
fMRI, moving beyond approaches that rely on random region masking. Specifically,
we introduce an ROI-guided masking strategy using the Automated Anatomical
Labelling Atlas (AAL3), applied directly to full 4D fMRI volumes to selectively
mask semantically coherent brain regions during self-supervised pretraining. Using
the ADHD-200 dataset comprising 973 subjects with resting-state fMRI scans,
we show that our method achieves a 4.23% improvement in classification accu-
racy compared to conventional random masking. Region-level attribution analysis
reveals that brain volumes within the limbic region and cerebellum contribute
most significantly to reconstruction fidelity and model representation. Our results
demonstrate that masking anatomical regions during model pretraining not only
enhances interpretability but also yields more robust and discriminative representa-
tions. In future work, we plan to extend this approach by evaluating it on additional
neuroimaging datasets, and developing new loss functions explicitly derived from
region-aware reconstruction objectives. These directions aim to further improve the
robustness and interpretability of foundation models for functional neuroimaging.

Submission Number: 25