Graph Convolutional Networks Reveal Network-Level Functional Dysconnectivity in Schizophrenia

Abstract

Background and Hypothesis Schizophrenia is increasingly understood as a disorder of brain dysconnectivity. Recently, graph-based approaches such as graph convolutional network (GCN) have been leveraged to explore complex pairwise similarities in imaging features among brain regions, which can reveal abstract and complex relationships within brain networks. Study Design We used GCN to investigate topological abnormalities of functional brain networks in schizophrenia. Resting-state functional magnetic resonance imaging data were acquired from 505 individuals with schizophrenia and 907 controls across 6 sites. Whole-brain functional connectivity matrix was extracted for each individual. We examined the performance of GCN relative to support vector machine (SVM), extracted the most salient regions contributing to both classification models, investigated the topological profiles of identified salient regions, and explored correlation between nodal topological properties of each salient region and severity of symptom. Study Results GCN enabled nominally higher classification accuracy (85.8%) compared with SVM (80.9%). Based on the saliency map, the most discriminative brain regions were located in a distributed network including striatal areas (ie, putamen, pallidum, and caudate) and the amygdala. Significant differences in the nodal efficiency of bilateral putamen and pallidum between patients and controls and its correlations with negative symptoms were detected in post hoc analysis. Conclusions The present study demonstrates that GCN allows classification of schizophrenia at the individual level with high accuracy, indicating a promising direction for detection of individual patients with schizophrenia. Functional topological deficits of striatal areas may represent a focal neural deficit of negative symptomatology in schizophrenia.

Discriminative analysis of schizophrenia patients using graph convolutional networks: A combined multimodal MRI and connectomics analysis

“…Deep learning ( Hatcher and Yu, 2018 ; Le et al, 2020 ), as a subfield of machine learning, can create a fully automated diagnostic process with no expert clinical intervention ( Qureshi et al, 2019 ) because of its powerful feature representation capability. However, most machine learning methods adopted in previous studies were typically based on independent neuroimaging features or connection features instead of the connectome itself ( Lei et al, 2022 ). Currently, graphs are the most commonly used representation of brain networks in neuropsychiatric disorder diagnosis.…”

Construction of Deep Learning Model using RESNET 50 for Schizophrenia Prediction from rsFMRI Images

“…In [11], the authors have used VGG16 netand transfer learningto predict schizophrenia and healthy controls. In [14], the researchers classi ed SZ patients from health controls from various sites with an accuracy of 85.8% using a Graph Convolutional model. The authors constructed BrainNet-GACNN and produced an accuracy of 83.13% in [15].…”

Acquired Pedophilia: international Delphi-method-based consensus guidelines

“…For this reason, the algorithms and online tools that have been created to support clinicians in the identification and quantification of brain abnormalities cannot be applied to psychiatric disorders [34]. Furthermore, while cutting-edge machine learning algorithms can be applied with the final aim of identifying potential biomarkers for psychiatric disorders [35][36][37][38][39], these new methodologies are not free from limitations [40]; thus, the results obtained are not reliable enough to be applied in the clinical context [41].…”