Skip to main content

Table 1 Comparison of deep learning auto-segmentation and manual contours of organs-at-risk and target volumes using test dataset #1

From: Clinical feasibility of deep learning-based auto-segmentation of target volumes and organs-at-risk in breast cancer patients after breast-conserving surgery

 

Total (n = 19)

Contrast (n = 10)

Non-contrast (n = 9)

DSC

STD

95% HD (mm)

STD (mm)

DSC

STD

95% HD (mm)

STD (mm)

DSC

STD

95% HD (mm)

STD (mm)

Organs-at-risk

 Heart

0.95

0.02

4.56

2.33

0.96

0.01

3.83

2.80

0.94

0.02

5.36

1.27

 Rt Lung

0.98

0.01

3.61

2.15

0.98

0.00

4.64

2.46

0.97

0.01

2.46

0.69

 Lt Lung

0.97

0.01

2.82

0.71

0.97

0.01

3.04

0.76

0.97

0.02

2.59

0.55

 Thyroid

0.89

0.05

1.88

0.90

0.90

0.04

1.55

0.65

0.88

0.05

2.25

0.99

 Esophagus

0.84

0.06

2.87

1.49

0.85

0.05

2.47

0.91

0.83

0.07

3.31

1.85

 Spinal cord

0.82

0.10

2.98

3.10

0.87

0.07

1.58

0.74

0.76

0.10

4.54

3.89

Target

 CTVp_breast

0.94

0.04

5.50

3.17

0.94

0.04

5.13

2.74

0.94

0.04

5.91

3.55

 CTVn_L1

0.74

0.08

10.93

6.27

0.71

0.09

13.51

7.10

0.78

0.05

8.07

3.40

 CTVn_L2

0.80

0.07

6.36

2.52

0.79

0.07

6.71

2.40

0.81

0.06

5.98

2.60

 CTVn_L3

0.64

0.13

7.99

3.81

0.66

0.10

6.97

2.87

0.62

0.16

9.11

4.37

 CTVn_IMN

0.72

0.09

5.75

3.36

0.67

0.09

7.53

3.71

0.77

0.07

3.77

1.00

 CTVn_L4

0.74

0.12

6.04

6.12

0.67

0.12

8.37

7.61

0.80

0.09

3.45

1.41

 CTVn_SCL RTOG

0.78

0.08

6.95

2.89

0.76

0.08

7.85

3.20

0.80

0.08

5.95

2.09

Back to article page

Radiation Oncology

ISSN: 1748-717X

Contact us