Table 1 Results from the 1D-CNN model for thyroid tissue classification.

From: A comparative analysis of deep learning architectures for thyroid tissue classification with hyperspectral imaging

Fold

Accuracy (%)

Precision (%)

Sensitivity (%)

Specificity (%)

AUC (%)

0

96.05

96.10

96.05

96.05

99.52

1

99.54

99.54

99.54

99.54

99.94

2

98.10

98.11

98.10

98.10

99.64

3

98.87

98.90

98.87

98.87

99.66

4

99.88

99.88

99.88

99.88

100.00

5

97.77

97.88

97.77

97.77

99.97

6

96.72

96.79

96.72

96.72

99.60

7

98.99

99.00

98.99

98.99

99.95

8

90.73

91.29

90.73

90.73

97.86

9

99.41

99.42

99.41

99.41

99.98

Mean

97.60

97.69

97.60

97.60

99.61

Std. dev.

± 2.57

± 2.42

± 2.57

± 2.57

± 0.60

  1. Bold value is used only to highlight the mean values and the std devs for the set of folds