Table 1 Performance comparisons with the state-of-the-art in Dice \((\uparrow ,\%)\), \(HD95(\downarrow ,mm)\) and \(ASSD(\downarrow ,mm)\). We compute the mean ± std over 3 trials.

From: Prior knowledge of anatomical relationships supports automatic delineation of clinical target volume for cervical cancer

Model

Dice

HD95

ASSD

3D-UNet37

69.45%±10.80%

14.45±9.69

3.14±1.60

nnFormer38

76.38%±6.25%

14.03±9.23

2.97±1.18

UNETR39

75.77%±7.29%

11.39±8.36

2.55±1.21

3D-UXNet40

74.58%±9.18%

13.21±8.73

2.85±1.43

nnU-Net34

78.16%±6.10%

13.40±7.28

2.49±1.12

KDPN (Ours)

81.33%±6.36%

9.39±7.12

2.02±0.98