Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

BrainSCL: Subtype-Guided Contrastive Learning for Brain Disorder Diagnosis

Xiaolong Li, Guiliang Guo, Guangqi Wen, Peng Cao, Jinzhu Yang, Honglin Wu, Xiaoli Liu, Fei Wang, Osmar R. Zaiane

Abstract

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.

BrainSCL: Subtype-Guided Contrastive Learning for Brain Disorder Diagnosis

Abstract

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.
Paper Structure (14 sections, 2 equations, 4 figures, 1 table)

This paper contains 14 sections, 2 equations, 4 figures, 1 table.

Table of Contents

  1. Introduction
  2. Method
  3. Multi-view Similarity Estimation
  4. Subtype Prototype Discovery
  5. Subtype-Guided Contrastive Learning
  6. Experiments and Results
  7. Datasets and Experimental Details
  8. Classification Results
  9. Comparison with SOTA
  10. Ablation Study
  11. Interpretability Analysis
  12. Validity of Latent Subtype Discovery
  13. Subtype Prototype Graph Interpretability Analysis
  14. Conclusion

Figures (4)

  • Figure 1: Comparison of conventional supervised contrastive learning (CL) and our subtype-guided contrastive learning.
  • Figure 2: Schematic of the proposed BrainSCL framework, which consists of three main modules: multi-view (text and graph structure) similarity estimation for generating a fused multi-view similarity matrix; subtype discovery for identifying latent subtypes and constructing subtype prototypes; subtype-guided contrastive learning for reinforcing intra-subtype consistency and enhancing discrimination between patients and healthy controls.
  • Figure 3: SNF-based patient subtypes visualized on the three disorder diagnosis scenarios with UMAP McInnes2018.
  • Figure 4: Top 10 brain regions in subtype prototype graphs for three subtypes.