Combining Deep Learning and Graph-Theoretic Brain Features to Detect Posttraumatic Stress Disorder at the Individual Level

Abstract

Previous studies using resting-state functional MRI (rs-fMRI) have revealed alterations in graphical metrics in groups of individuals with posttraumatic stress disorder (PTSD). To explore the ability of graph measures to diagnose PTSD and capture its essential features in individual patients, we used a deep learning (DL) model based on a graph-theoretic approach to discriminate PTSD from trauma-exposed non-PTSD at the individual level and to identify its most discriminant features. Our study was performed on rs-fMRI data from 91 individuals with PTSD and 126 trauma-exposed non-PTSD patients. To evaluate our DL method, we used the traditional support vector machine (SVM) classifier as a reference. Our results showed that the proposed DL model allowed single-subject discrimination of PTSD and trauma-exposed non-PTSD individuals with higher accuracy (average: 80%) than the traditional SVM (average: 57.7%). The top 10 DL features were identified within the default mode, central executive, and salience networks; the first two of these networks were also identified in the SVM classification. We also found that nodal efficiency in the left fusiform gyrus was negatively correlated with the Clinician Administered PTSD Scale score. These findings demonstrate that DL based on graphical features is a promising method for assisting in the diagnosis of PTSD.

Application of machine learning techniques in the diagnostic approach of PTSD using MRI neuroimaging data: A systematic review

“…Two studies by Zhu, Z et al [ 19 ]and Yang et al [ 20 ], combined deep learning and classical support vector machine (SVM) techniques. They employed graphic topological measures based on fMRI data to classify PTSD in adults vs. HC and PTSD in children vs. TEHC, achieving accuracies of 80% and 71.2%, respectively.…”

Systematic review of machine learning in PTSD studies for automated diagnosis evaluation

“…Zhu et al applied a graph-theoretic approach based on DL to discriminate PTSD from TE. An accuracy of 80% was achieved achieved with 0.81 sensitivity and 0.79 specificity by utilizing informative sets of brain graph measures of the central executive network, salience network, and default mode network 34 .…”

Brain morphometric features predict medication response in youth with bipolar disorder: a prospective randomized clinical trial

“…We implemented a two-stage prediction pipeline to differentiate medication responders from non-responders as described in previous prospective studies (Hazlett et al, 2017 ; Zhu et al, 2021 ; Yang et al, 2021 ). A feedforward multi-layer neural network was adopted as the initial stage for dimensionality reduction (Hinton & Salakhutdinov, 2006 ), and SVM was included as the second stage to individually discriminate responders from non-responders (Cortes & Vapnik, 1995 ).…”