Table 1 Summary of DSC and HD
From: Evaluation of deep learning-based autosegmentation in breast cancer radiotherapy
Contour | P-value* | ||||||
|---|---|---|---|---|---|---|---|
(1) Manual (n = 100) | (2) Corrected auto (n = 100) | (3) Auto (n = 10) | (1) vs. (2) | (1) vs. (3) | (2) vs. (3) | ||
DSC (mean ± SE) | Thyroid | 0.80 ± 0.01 | 0.80 ± 0.01 | 0.79 ± 0.02 | .953 | .042 | .014 |
Lung_right | 0.98 ± 0.00 | 0.97 ± 0.00 | 0.97 ± 0.00 | < .001 | < .001 | .262 | |
Lung_left | 0.97 ± 0.00 | 0.96 ± 0.00 | 0.96 ± 0.01 | < .001 | < .001 | .801 | |
Breast_right | 0.81 ± 0.00 | 0.90 ± 0.00 | 0.91 ± 0.00 | < .001 | < .001 | .596 | |
Breast_left | 0.83 ± 0.00 | 0.90 ± 0.00 | 0.90 ± 0.01 | < .001 | < .001 | .280 | |
Spinal cord | 0.82 ± 0.01 | 0.85 ± 0.00 | 0.85 ± 0.01 | .001 | .001 | .136 | |
Esophagus | 0.84 ± 0.00 | 0.83 ± 0.00 | 0.82 ± 0.01 | .001 | < .001 | < .001 | |
Heart | 0.92 ± 0.00 | 0.94 ± 0.00 | 0.95 ± 0.00 | < .001 | < .001 | .029 | |
Liver | 0.94 ± 0.00 | 0.94 ± 0.00 | 0.93 ± 0.01 | < .001 | < .001 | < .001 | |
HD (mean ± SE) | Thyroid | 3.82 ± 0.20 | 4.12 ± 0.26 | 4.28 ± 0.83 | 0.33 | 0.056 | 0.021 |
Lung_right | 2.37 ± 0.15 | 2.46 ± 0.09 | 2.42 ± 0.29 | > .999 | > .999 | 0.472 | |
Lung_left | 3.61 ± 0.35 | 2.93 ± 0.18 | 2.99 ± 0.67 | 0.077 | 0.115 | 0.538 | |
Breast_right | 12.44 ± 0.80 | 8.06 ± 0.35 | 7.54 ± 0.65 | < .001 | < .001 | 0.284 | |
Breast_left | 10.85 ± 0.58 | 8.14 ± 0.33 | 8.15 ± 1.05 | < .001 | < .001 | > .999 | |
Spinal cord | 3.95 ± 0.54 | 2.20 ± 0.04 | 2.21 ± 0.11 | 0.005 | 0.005 | 0.475 | |
Esophagus | 3.95 ± 0.36 | 3.16 ± 0.06 | 3.46 ± 0.21 | 0.1 | 0.574 | < .001 | |
Heart | 8.26 ± 0.55 | 5.42 ± 0.29 | 4.73 ± 0.33 | < .001 | < .001 | 0.034 | |
Liver | 6.92 ± 0.64 | 5.95 ± 0.32 | 9.74 ± 3.16 | 0.335 | 0.085 | 0.002 | |