Pairing Explainable Deep Learning Classification with Clustering to Uncover Effects of Schizophrenia Upon Whole Brain Functional Network Connectivity Dynamics

Abstract

Many studies have analyzed resting state functional magnetic resonance imaging (rs-fMRI) dynamic functional network connectivity (dFNC) data to elucidate the effects of neurological and neuropsychiatric disorders upon the interactions of brain regions over time. Existing studies often use either machine learning classification or clustering algorithms. Additionally, several studies have used clustering algorithms to extract features related to brain states trajectories that can be used to train interpretable classifiers. However, the combination of explainable dFNC classifiers followed by clustering algorithms is highly underutilized. In this study, we show how such an approach can be used to study the effects of schizophrenia (SZ) upon brain activity. Specifically, we train an explainable deep learning model to classify between individuals with SZ and healthy controls. We then cluster the resulting explanations, identifying discriminatory states of dFNC. We lastly apply several novel measures to quantify aspects of the classifier explanations and obtain additional insights into the effects of SZ upon brain network dynamics. Specifically, we uncover effects of schizophrenia upon subcortical, sensory, and cerebellar network interactions. We also find that individuals with SZ likely have reduced variability in overall brain activity and that the effects of SZ may be temporally localized. In addition to uncovering effects of SZ upon brain network dynamics, our approach could provide novel insights into a variety of neurological and neuropsychiatric disorders in future dFNC studies.

A Framework for Systematically Evaluating the Representations Learned by A Deep Learning Classifier from Raw Multi-Channel Electroencephalogram Data

“…We applied the αβ-rule [90] of LRP [58], [91] for insight into the relative importance of each channel. LRP is a popular approach in the domain of explainability for image classification and has also been used extensively in the domain of neuroimaging and neurological time-series classification [5], [39], [40], [62], [63], [72], [73], [78], [92]–[97]. We implemented LRP using the Innvestigate library [98].…”

“…This is a tentative category because while the approach has been used to identify important time points [71], it has not yet been applied to identify key waveforms. As demonstrated in several fMRI classification studies, this approach can also provide global insight into patterns of importance distribution [72], [73].…”