Automatic Diagnosis of Mild Cognitive Impairment Based on Spectral, Functional Connectivity, and Nonlinear EEG-Based Features

Algorithm 1

The SBFS algorithm.

Input: The entire feature set,

Output: The selected feature subset, where

1. Start with the entire set, .

2. Eliminate the worst feature, , where .

3. Update .

4. If , go to step 6.

5. Else go to the step 2.

6. Stop.