Table 6 Grade-1 severity detection using performance metrics of various DL methods.

From: Enhanced swin transformer with dual attention for knee osteoarthritis severity grading from X-ray images

Algorithms

Performance metrics

Accuracy

Precision

Recall

Specificity

F1-score

MCC

CNN

82.3

81.2

80.3

79.5

81.2

0.81

VGG-19

77.2

76.2

74.1

70.4

70.5

0.72

DenseNETS

92.4

92.4

90.3

89.5

90.4

0.88

RESNETS-34

92.2

92.1

90.4

90.2

90.7

0.90

Ensemble methods

96.1

95.4

94.2

94

94.3

0.94

Resnets + CBAM

93.5

92.9

92.4

93.0

92.9

0.92

Proposed model

99.67

99.0

98.9

98.7

98.9

0.98