Table 6. Testing results when models run.

1 vs 1	Models	Accuracy	Precision	Recall	Auc	F1 score	MCC
CN/AD	EfficientNetB0	0.9894	0.9895	0.9895	0.9890	0.9895	0.9790
	DenseNet121	0.9716	0.9718	0.9716	0.9720	0.9716	0.9433
	AlexNet	0.9774	0.9783	0.9776	0.9780	0.9775	0.9558
MCI/AD	EfficientNetB0	0.9795	0.9795	0.9800	0.9800	0.9795	0.9594
	DenseNet121	0.9075	0.9077	0.9072	0.9070	0.9074	0.8149
	AlexNet	0.9750	0.9749	0.9751	0.9750	0.9750	0.9500
CN/MCI	EfficientNetB0	0.9731	0.9730	0.9734	0.9730	0.9731	0.9664
	DenseNet121	0.9842	0.9843	0.9842	0.9840	0.9843	0.9668
	AlexNet	0.9655	0.9661	0.9661	0.9660	0.9656	0.9311