Progressive Self-Supervised Learning with Individualized Community Assignment for Brain Network Analysis
2026-06-29 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors present BrainPICM, a new method to study brain networks that focuses on the unique patterns in each person's brain. Instead of using generic data masking, their method identifies and gradually learns from uncertain or abnormal brain regions, which helps the model understand both common structures and individual differences. They also created a way to measure changes in brain function compared to a typical brain. Testing their method on three brain scan datasets showed it works better than existing techniques at identifying neurological conditions. This suggests that considering brain community patterns helps create clearer and more useful brain network models.
brain networksmodular community structureself-supervised learningfMRIoptimal transportmasking strategyfunctional reorganizationdiagnostic accuracyROI (region of interest)curriculum learning
Authors
Hairui Chen, Yanwu Yang, Jianfeng Cao, Hanyang Peng, Chenfei Ye, Ting Ma
Abstract
Brain networks exhibit a modular community structure that varies across individuals and neurological conditions. However, existing self-supervised learning (SSL) methods often overlook this heterogeneity, relying on generic masking strategies that fail to capture subject-specific functional organization. We propose BrainPICM, a self-supervised framework for brain network analysis via progressive individualized community aware masking. BrainPICM formulates ROI-to-community mapping as a progressive unbalanced optimal transport process, yielding soft assignments and per-ROI confidence scores. Guided by these confidence estimates, a curriculum-style masking strategy gradually incorporates low-confidence, potentially pathological regions into training, enabling the model to learn both stable modular structures and individual variations. Additionally, a deviation-aware aggregation module quantifies functional reorganization by measuring mass redistribution relative to a population template, enhancing interpretability and downstream prediction. Experiments on three fMRI datasets (ABIDE-I, ADHD-200, ADNI) show that BrainPICM consistently outperforms state-of-the-art supervised and SSL methods in diagnostic accuracy, indicating that explicitly injecting modular community structure into masked modeling yields more functionally consistent and generalizable representations. The source code for this approach will be released at https://github.com/Hrychen7/BrainPICM.