Four-way classification of Alzheimer’s disease using deep Siamese convolutional neural network with triplet-loss function

Brain Informatics

Table 3 Results of proposed models on the ADNI dataset

Models	Sensitivity	Specificity	1-Sensitivity	1-Specificity	Accuracy
Siamese CNN	0.9179	0.9383	0.0821	0.0670	0.9183
Siamese CNN with VGG16	0.950	0.853	0.146	0.0490	0.8531
VGG16 (as classifier)	0.963	0.891	0.108	0.0390	0.8900