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

Table 5

Evaluation of the proposed 3DCNN + SVM with E2E applied to binary classification of AD vs. MCI samples (%).

Method	Training set				Testing set
	ACC	SEN	SPE	AUC	ACC	SEN	SPE	AUC
Gray [6]	—	—	—	—	68.2	58.3	73.0	70.0
Lu [10]	—	—	—	—	—	—	—	—
Silveira [8]	—	—	—	—	70.0	—	—	—
Ding et al. [12]	92.39	97.50	90.29	98.59	71.19	68.52	72.36	77.28
Liu et al. [14]	96.10	99.93	94.52	99.18	73.79	75.00	73.28	79.16
Huang et al. [15]	96.09	99.66	94.53	99.39	73.83	74.93	73.42	78.53
Proposed	98.45	99.24	97.31	99.91	74.29	70.78	75.48	80.11