A Classification Algorithm by Combination of Feature Decomposition and Kernel Discriminant Analysis (KDA) for Automatic MR Brain Image Classification and AD Diagnosis

Table 3

Results comparison of the feature decomposition, KDA, and the proposed method for classification of AD vs. NC.

Methods	ACC (%)	SEN (%)	SPE (%)	AUC (%)
Feature decomposition	88.08	80.87	94.31	92.73
KDA	88.07	80.84	94.31	93.42
Proposed method	90.41	84.37	95.63	93.89