Frequency Feature Fusion Graph Network For Depression Diagnosis Via fNIRS

2025-04-29

AI Summary

Graph neural networks (GNNs) struggle to effectively model dynamic functional connectivity among fNIRS channels for depression diagnosis. Method: We propose a temporal graph convolutional network (TGCN) that integrates discrete Fourier transform (DFT)-derived frequency-domain features—introduced here for the first time in fNIRS-based depression recognition—as novel temporal biomarkers with structural graph representations. Contribution/Results: We construct the largest publicly available, propensity score matching (PSM)-balanced fNIRS dataset to date (1,086 subjects), enabling robust evaluation. The TGCN jointly models multi-channel temporal functional connectivity in the frequency domain and incorporates SHAP for clinical interpretability. Experiments demonstrate statistically significant improvements in F1-score on both the original and PSM-balanced subsets. This work establishes a new paradigm for objective, interpretable, AI-assisted depression diagnosis grounded in neurophysiologically meaningful spectral features and graph-structured temporal modeling.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Computational social science
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Data-driven approaches for depression diagnosis have emerged as a significant research focus in neuromedicine, driven by the development of relevant datasets. Recently, graph neural network (GNN)-based models have gained widespread adoption due to their ability to capture brain channel functional connectivity from both spatial and temporal perspectives. However, their effectiveness is hindered by the absence of a robust temporal biomarker. In this paper, we introduce a novel and effective biomarker for depression diagnosis by leveraging the discrete Fourier transform (DFT) and propose a customized graph network architecture based on Temporal Graph Convolutional Network (TGCN). Our model was trained on a dataset comprising 1,086 subjects, which is over 10 times larger than previous datasets in the field of depression diagnosis. Furthermore, to align with medical requirements, we performed propensity score matching (PSM) to create a refined subset, referred to as the PSM dataset. Experimental results demonstrate that incorporating our newly designed biomarker enhances the representation of temporal characteristics in brain channels, leading to improved F1 scores in both the real-world dataset and the PSM dataset. This advancement has the potential to contribute to the development of more effective depression diagnostic tools. In addition, we used SHapley Additive exPlaination (SHAP) to validate the interpretability of our model, ensuring its practical applicability in medical settings.

Problem

• Develops a novel biomarker for depression diagnosis using DFT
• Proposes a TGCN-based model for brain channel connectivity analysis
• Validates model interpretability via SHAP for medical applicability

Innovation

• Uses DFT for novel depression diagnostic biomarker
• Custom TGCN architecture enhances brain connectivity analysis
• Leverages SHAP for model interpretability in medicine