Fig. 6.

ROC curves of the LR, SVM, and RF classifiers based on discriminative GM connectome features (i.e., significant topological metrics and connections) for distinguishing individuals with VM from HCs. The AUCs for the LR, SVM and RF models were 0.790 (P < 0.001), 0.831 (P < 0.001) and 0.801 (P < 0.001), respectively. ROC = receiver operating characteristic; LR = logistic regression; SVM = support vector machine; RF = random forest; GM = gray matter; VM = vestibular migraine; HC = healthy control; AUC = area under the curve