Brain-JEPA: Brain Dynamics Foundation Model with Gradient Positioning and Spatiotemporal Masking

TL;DR

Brain-JEPA is a novel foundation model for brain activity analysis that leverages innovative techniques to achieve state-of-the-art results in demographic, disease, and trait prediction, while enhancing generalizability and interpretability.

Contribution

It introduces Brain Gradient Positioning and Spatiotemporal Masking, advancing brain functional parcellation and handling heterogeneous fMRI data for improved modeling.

Findings

State-of-the-art performance in demographic prediction

Superior generalizability across ethnic groups

Enhanced understanding of neural circuits

Abstract

We introduce Brain-JEPA, a brain dynamics foundation model with the Joint-Embedding Predictive Architecture (JEPA). This pioneering model achieves state-of-the-art performance in demographic prediction, disease diagnosis/prognosis, and trait prediction through fine-tuning. Furthermore, it excels in off-the-shelf evaluations (e.g., linear probing) and demonstrates superior generalizability across different ethnic groups, surpassing the previous large model for brain activity significantly. Brain-JEPA incorporates two innovative techniques: Brain Gradient Positioning and Spatiotemporal Masking. Brain Gradient Positioning introduces a functional coordinate system for brain functional parcellation, enhancing the positional encoding of different Regions of Interest (ROIs). Spatiotemporal Masking, tailored to the unique characteristics of fMRI data, addresses the challenge of heterogeneous time-series patches. These methodologies enhance model performance and advance our understanding of the neural circuits underlying cognition. Overall, Brain-JEPA is paving the way to address pivotal questions of building brain functional coordinate system and masking brain activity at the AI-neuroscience interface, and setting a potentially new paradigm in brain activity analysis through downstream adaptation.