Skip to main content

Table 1 Experimental results for the transfer learning model

From: Artificial intelligence (AI) diagnostic tools: utilizing a convolutional neural network (CNN) to assess periodontal bone level radiographically—a retrospective study

Class

Correct validation

Accuracy of validation (%)

Correct test

Accuracy of test (%)

Experimental results per each class of the learning model – multi-classification

Normal

114

69.93

63

76.82

Mild

46

45.09

23

45.09

Moderate

35

60.34

19

65.51

Severe

10

45.45

2

18.18

Experimental results per each class of the learning model – binary classification

Normal

140

85.88

59

71.95

Abnormal

112

61.53

72

79.12