Automatic Diagnosis of Alzheimer’s Disease and Mild Cognitive Impairment Based on CNN + SVM Networks with End-to-End Training

Table 7

Ablations studies of the proposed 3DCNN + SVM model applied to binary classification of AD vs. NC (%).

Method	Training set				Testing set
	ACC	SEN	SPE	AUC	ACC	SEN	SPE	AUC
3DCNN	99.50	99.72	99.39	99.97	89.83	90.94	89.26	92.68
3DCNN + SVM	98.62	99.01	98.40	99.95	90.20	90.34	90.19	93.36
3DCNN + SVM + E2E	99.19	99.39	99.54	99.88	90.82	91.29	90.59	93.75