Skip to main content

Table 5 Comparison of the quantitative results of rdaunet and AMSMW on the test set of dataset A. fold0, fold1, fold2, fold3 and fold4 are the five test sets in the five-fold cross-validation experiment

From: Lesion segmentation in breast ultrasound images using the optimized marked watershed method

   \( {Loss\, (\%)}\) \( {Acc\, (\%)}\) \( {DC\, (\%)}\) \( {Sen\, (\%)}\) \( {Sp \,(\%)}\) \( {F1 \,(\%)}\) \( {Pc \,(\%)}\) \( {M-IOU \,(\%)}\)
fold0 RDAU 28.22 96.44 71.78 75.43 98.52 74.00 77.96 61.15
AMSMW 14.55 97.33 85.45 93.64 97.41 85.45 79.89 75.36
fold1 RDAU 25.93 97.17 74.07 73.41 99.07 75.58 83.93 63.65
AMSMW 14.55 97.26 84.86 92.60 97.33 84.86 79.72 74.42
fold2 RDAU 23.53 96.80 76.49 73.62 99.20 76.47 87.83 66.21
AMSMW 13.92 97.34 86.08 93.33 97.40 86.08 81.14 76.23
fold3 RDAU 30.57 96.66 69.43 71.52 98.80 69.43 78.85 59.61
AMSMW 15.15 97.37 84.85 93.21 97.43 84.85 79.66 74.55
fold4 RDAU 23.50 97.14 76.50 78.32 98.77 77.27 80.92 66.23
AMSMW 15.46 97.35 84.54 92.95 97.55 84.54 79.24 74.20
Average RDAU 26.35 96.84 73.66 74.46 \(\mathbf{98.87} \) 74.55 \(\mathbf{81.90} \) 63.37
AMSMW \(\mathbf{14.73} \) \(\mathbf{97.33} \) \(\mathbf{85.16} \) 93.15 \(\mathbf{97.42} \) \(\mathbf{85.16} \) 79.93 \(\mathbf{74.95} \)
  1. Bold indicates the best results in the current column