Robustness of explainable AI algorithms for disease biomarker discovery from functional connectivity datasets

Deepank Girish, Yi Hao Chan, Sukrit Gupta, Jing Xia, Jagath Rajapakse

Keywords: Biomarker discovery, Graph Neural Networks, Functional Connectivity, Model Explainability

TL;DR: We introduce objective quantitative metrics for evaluating the robustness of salient features in ML models applied to functional neuroimaging, particularly focusing on brain connectomes.

Abstract: Machine learning (ML) models have been used in functional neuroimaging for wide-ranging tasks, ranging from disease diagnosis to disease prognosis. There have been successive functional connectivity-based ML studies focused on improving model performances for disease detection. An increasing number of such studies use their trained models to detect and evaluate salient features that could be potential biomarkers of these neurological conditions. The evaluation of these salient features is often qualitative and limited to cross-referencing existing literature for similar findings. In this study, we present objective quantitative metrics to evaluate the robustness of these salient features. Building upon existing generic evaluation metrics, we propose metrics that capture topological properties known to be characteristic of brain functional connectomes. Using existing and newly proposed measures on a set of baselines and state-of-the-art GNN models, we found that when GNNExplainer is used with models that incorporate attention, the scores produced are relatively more robust than other combinations. On datasets of patients with ASD or ADHD, our proposed metrics highlighted that salient features identified in both disorders are highly involved in functional specialization, while salient ASD features expressed stronger functional integration than ADHD. We package these existing and novel metrics together in the RE-CONFIRM framework that holds promise to set the foundations for the quantitative evaluation of salient features detected by future studies.

Track: 6. AI for biomarker discovery and drug design

Registration Id: CXNBD6K8GYZ

Submission Number: 181