Table 4.

Models with highest performance for each task, based on rest, stress, delta, and combined feature sets.

Task	Feature set	Model	AUC	ACC	SEN	SPE
1: Normal/abnormal classification	Rest	RFE_KNN	0.56	0.65	0.71	0.41
	Stress	MRMR_KNN	0.61	0.63	0.64	0.6
	Delta	Boruta_RF	0.62	0.68	0.72	0.54
	Combined	RFE_NB	0.6	0.58	0.57	0.6
2: CAD low-risk/high-risk classification	Rest	MRMR_GB	0.7	0.63	0.61	0.64
	Stress	Boruta_GB	0.79	0.76	0.75	0.76
	Delta	MRMR_GB	0.69	0.64	0.58	0.69
	Combined	Boruta_LR	0.73	0.62	0.58	0.65