Table 2 Comparative evaluation of the proposed model versus state-of-the-art models using the OASIS dataset

From: Automated Alzheimer’s disease detection using active learning model with reinforcement learning and scope loss function

Model

Accuracy

F-measure

G-means

AUC

HBOA-MLP19

76.210 ± 0.084

79.736 ± 0.057

80.696 ± 0.031

0.594 ± 0.059

CNN20

77.226 ± 0.038

80.827 ± 0.023

81.799 ± 0.090

0.610 ± 0.046

VGG1621

78.787 ± 0.089

81.453 ± 0.076

82.438 ± 0.073

0.624 ± 0.071

CLADSI22

80.238 ± 0.068

82.787 ± 0.024

83.799 ± 0.071

0.640 ± 0.054

DQN29

81.029 ± 0.076

83.602 ± 0.073

84.593 ± 0.089

0.652 ± 0.075

4D-AlzNet50

82.266 ± 0.013

84.681 ± 0.096

85.645 ± 0.035

0.670 ± 0.090

XAI51

83.097 ± 0.027

85.916 ± 0.008

86.831 ± 0.023

0.683 ± 0.036

RESNET50 KNN52

84.179 ± 0.067

87.068 ± 0.017

88.020 ± 0.054

0.698 ± 0.079

Proposed w/o AL

84.906 ± 0.013

88.536 ± 0.098

89.474 ± 0.085

0.714 ± 0.045

Proposed w/o SLF

86.220 ± 0.058

89.909 ± 0.091

90.762 ± 0.053

0.730 ± 0.097

Proposed w/o HO

87.739 ± 0.099

90.847 ± 0.074

91.678 ± 0.068

0.744 ± 0.013

Proposed

88.806 ± 0.051

92.044 ± 0.048

92.923 ± 0.079

0.753 ± 0.045

Back to article page