Skip to main content
. 2021 Feb 5;15:605115. doi: 10.3389/fnins.2021.605115

TABLE 7.

10-fold cross-validation binary mean classification performance for HC against MCI using LSVM classifier using different feature selection methods.

Feature selection method Performance metrics Mean (%) Standard deviation Max (%) Min (%)
LASSO Accuracy 98.91 0.456 99.25 95.82
Sensitivity 99.68 0.56 100.0 95.48
Specificity 98.11 0.46 98.51 96.00
F-measure 0.9856
FSASL Accuracy 81.28 1.010 83.01 77.64
Sensitivity 84.61 1.389 86.62 79.81
Specificity 77.92 1.121 79.81 73.038
F-measure 0.833
LLCFS Accuracy 76.27 0.631 78.31 74.70
Sensitivity 71.08 1.388 75.37 68.23
Specificity 81.40 1.005 82.69 76.80
F-measure 0.800
CFS Accuracy 86.16 2.25 88.96 80.47
Sensitivity 92.28 2.33 95.11 86.14
Specificity 79.72 2.375 82.75 73.88
F-measure 0.8517
SVMRFE Accuracy 71.92 0.832 74.43 69.93
Sensitivity 66.90 1.493 71.14 63.53
Specificity 76.68 1.187 79.61 72.88
F-measure 0.7762