Skip to main content

Table 2 Evaluation of whether the segment was present, including precision, sensitivity, accuracy, and F1 score

From: A deep learning-based automated algorithm for labeling coronary arteries in computed tomography angiography images

Segment

TP

TN

FP

FN

Precision

Recall

Accuracy

F1

pRCA

157

0

0

0

100.0%

100.0%

100.0%

100.0%

mRCA

157

0

0

0

100.0%

100.0%

100.0%

100.0%

dRCA

156

1

0

0

100.0%

100.0%

100.0%

100.0%

R-PDA

139

16

1

1

99.3%

99.3%

98.7%

99.3%

R-PLB

94

30

0

33

100.0%

74.0%

79.0%

85.1%

LM

157

0

0

0

100.0%

100.0%

100.0%

100.0%

pLAD

157

0

0

0

100.0%

100.0%

100.0%

100.0%

mLAD

157

0

0

0

100.0%

100.0%

100.0%

100.0%

dLAD

156

0

1

0

99.4%

100.0%

99.4%

99.7%

D1

152

3

0

2

100.0%

98.7%

98.7%

99.4%

D2

110

29

12

6

90.2%

94.8%

88.5%

92.4%

pCx

156

0

0

1

100.0%

99.4%

99.4%

99.7%

LCx

151

0

0

6

100.0%

96.2%

96.2%

98.1%

OM1

128

19

1

9

99.2%

93.4%

93.6%

96.2%

OM2

75

72

6

4

92.6%

94.9%

93.6%

93.8%

RI

46

87

11

13

80.7%

78.0%

84.7%

79.3%

In total

2148

257

32

75

98.5%

96.6%

95.7%

97.6%