Rethinking Functional Brain Connectome Analysis: Do Graph Deep Learning Models Help

TL;DR

This study critically evaluates graph deep learning models for brain connectome analysis, revealing that their message aggregation mechanism does not improve predictive performance and may even degrade it, advocating for simpler, interpretable models.

Contribution

The paper introduces a hybrid model combining linear and graph attention networks, demonstrating improved robustness and interpretability over traditional graph deep learning approaches.

Findings

Message aggregation does not enhance predictive accuracy.

Hybrid model achieves more robust and interpretable predictions.

Complex deep learning models may not be necessary for connectome analysis.

Abstract

Graph deep learning models, a class of AI-driven approaches employing a message aggregation mechanism, have gained popularity for analyzing the functional brain connectome in neuroimaging. However, their actual effectiveness remains unclear. In this study, we re-examine graph deep learning versus classical machine learning models based on four large-scale neuroimaging studies. Surprisingly, we find that the message aggregation mechanism, a hallmark of graph deep learning models, does not help with predictive performance as typically assumed, but rather consistently degrades it. To address this issue, we propose a hybrid model combining a linear model with a graph attention network through dual pathways, achieving robust predictions and enhanced interpretability by revealing both localized and global neural connectivity patterns. Our findings urge caution in adopting complex deep learning models for functional brain connectome analysis, emphasizing the need for rigorous experimental designs to establish tangible performance gains and perhaps more importantly, to pursue improvements in model interpretability.