Fig. 3: The ROC curves of the GCNs models. | npj Digital Medicine

Fig. 3: The ROC curves of the GCNs models.

From: An interpretable model based on graph learning for diagnosis of Parkinson’s disease with voice-related EEG

Fig. 3

Panels a–d show the ROC curves for the PCC+GCNs, PLV+GCNs, PCC + GSP-GCNs, and PLV + GSP-GCNs models, respectively. The solid blue line represents the average performance across different test sets using 5-fold cross-validation, while the other colored lines show the performance of individual test set. The gray shaded area indicates the variance range of the average ROC curve.

Back to article page