Table 2.

Performances comparison of our models with traditional machine learning methods. Our models were trained on the training dataset and evaluated on the test dataset. Our models were compared with traditional machine learning methods. The performance data of the traditional machine learning methods were from Rorbach, G., et al.²³. “—” means “data not provided in the original paper”.

Model	Sensitivity	Specificity	F1	MCC	Accuracy
CNN-filter3-128	0.846	0.945	0.890	0.795	0.895
CNN-filter4-128	0.786	0.980	0.871	0.781	0.883
CNN-filter5-128	0.861	0.955	0.903	0.819	0.908
CNN-filter6-128	0.871	0.970	0.916	0.845	0.920
CNN-concat-filters	0.846	0.975	0.904	0.827	0.910
Support Vector Machines	0.926	0.945	0.901	0.859	—
Random Forest	0.870	0.957	0.883	0.836	—
Linear Discriminant Analysis	0.935	0.919	0.881	0.830	—
Logistic Regression	0.875	0.941	0.867	0.816	—
Decision Tree	0.861	0.943	0.863	0.808	—
Naive Bayes	0.875	0.894	0.824	0.746	—