BPI-GNN: Interpretable brain network-based psychiatric diagnosis and subtyping

• DOI: 10.1016/j.neuroimage.2024.120594
• Publication Date: 2024-04-01T15:48:32Z
• Authors: Kaizhong Zheng; Shujian Yu; Liangjun Chen; Lujuan Dang; Badong Chen
• Keywords: Subtyping

Converging evidence increasingly suggests that psychiatric disorders, such as major depressive disorder (MDD) and autism spectrum (ASD), are not unitary diseases, but rather heterogeneous syndromes involve diverse, co-occurring symptoms divergent responses to treatment. This clinical heterogeneity has hindered the progress of precision diagnosis treatment effectiveness in disorders. In this study, we propose BPI-GNN, a new interpretable graph neural network (GNN) framework for analyzing functional magnetic resonance images (fMRI), by leveraging famed prototype learning. addition, introduce novel generation process subgraph discover essential edges distinct prototypes employ total correlation (TC) ensure independence patterns. BPI-GNN can effectively discriminate patients healthy controls (HC), identify biological meaningful subtypes We evaluate performance against 11 popular brain classification methods on three datasets observe our always achieves highest accuracy. More importantly, examine differences symptom profiles gene expression among identified brain-based have relevance. It also discovers subtype biomarkers align with current neuro-scientific knowledge.