Skip to main content

Table 4 Performance of BiLSTM-Highway-CRF model for each entity type(unit: %)

From: Construction of a knowledge graph for breast cancer diagnosis based on Chinese electronic medical records: development and usability study

Entity Type

Precision

Recall

F1

Calcifications

99.71

100.00

99.86

Density

91.59

93.33

92.45

Distribute

93.02

93.02

93.02

Location

97.42

97.89

97.65

Mass

98.71

99.57

99.13

Lymph Node

98.30

98.72

98.51

Margin

89.57

94.81

92.11

Merge

94.14

96.98

95.54

Number

97.96

97.56

97.76

Shape

93.57

96.04

94.79

Size

97.56

98.77

98.16

Special

97.47

97.88

97.67

Structure

98.16

98.42

98.29

Category descriptions

98.28

99.57

98.92

Negation

99.84

100.00

99.92