BrainSCL: Subtype-Guided Contrastive Learning for Brain Disorder Diagnosis
This work addresses heterogeneity in mental disorder populations for improved diagnostic accuracy, representing an incremental advance through novel integration of subtype modeling with contrastive learning.
The authors tackled the challenge of patient heterogeneity in brain disorder diagnosis by proposing a subtype-guided contrastive learning framework that models latent subtypes to guide representation learning, achieving state-of-the-art performance on Major Depressive Disorder, Bipolar Disorder, and Autism Spectrum Disorders.
Mental disorder populations exhibit pronounced heterogeneity -- that is, the significant differences between samples -- poses a significant challenge to the definition of positive pairs in contrastive learning. To address this, we propose a subtype-guided contrastive learning framework that models patient heterogeneity as latent subtypes and incorporates them as structural priors to guide discriminative representation learning. Specifically, we construct multi-view representations by combining patients' clinical text with graph structure adaptively learned from BOLD signals, to uncover latent subtypes via unsupervised spectral clustering. A dual-level attention mechanism is proposed to construct prototypes for capturing stable subtype-specific connectivity patterns. We further propose a subtype-guided contrastive learning strategy that pulls samples toward their subtype prototype graph, reinforcing intra-subtype consistency for providing effective supervisory signals to improve model performance. We evaluate our method on Major Depressive Disorder (MDD), Bipolar Disorder (BD), and Autism Spectrum Disorders (ASD). Experimental results confirm the effectiveness of subtype prototype graphs in guiding contrastive learning and demonstrate that the proposed approach outperforms state-of-the-art approaches. Our code is available at https://anonymous.4open.science/r/BrainSCL-06D7.