Table 4 Comparison of direct/model-aided diagnostic performance of the gynecologists in the five-category classification task
From: Deep learning model for classifying endometrial lesions
Category | Sensitivity (%) | Specificity (%) | Precision (%) | F1-score (%) | AUC | Accuracy (%) |
|---|---|---|---|---|---|---|
Gynecologist 4 | ||||||
EH | 59.1 | 94.2 | 68.4 | 63.4 | 0.766 | 64.0 |
AH | 58.6 | 93.8 | 73.9 | 65.4 | 0.762 | |
EC | 56.5 | 89.2 | 54.2 | 55.3 | 0.729 | |
EP | 78.3 | 85.3 | 54.5 | 64.3 | 0.818 | |
SM | 67.9 | 92.8 | 73.1 | 70.4 | 0.803 | |
Model-aided gynecologist 4 | ||||||
EH | 82.1 | 93.8 | 79.3 | 80.7 | 0.880 | 78.4 |
AH | 47.6 | 98.1 | 83.3 | 60.6 | 0.728 | |
EC | 77.8 | 94.9 | 80.8 | 79.2 | 0.863 | |
EP | 92.6 | 90.8 | 73.5 | 82.0 | 0.917 | |
SM | 86.4 | 95.1 | 79.2 | 82.6 | 0.908 | |
Gynecologist 5 | ||||||
EH | 68.0 | 82.5 | 45.5 | 54.5 | 0.754 | 62.4 |
AH | 55.2 | 93.8 | 72.7 | 62.7 | 0.745 | |
EC | 52.2 | 90.2 | 54.5 | 53.3 | 0.712 | |
EP | 73.9 | 88.2 | 58.6 | 65.4 | 0.811 | |
SM | 64.3 | 99.0 | 94.7% | 76.6 | 0.816 | |
Model-aided gynecologist 5 | ||||||
EH | 82.1 | 84.5 | 60.5 | 69.7 | 0.833 | 72.8 |
AH | 42.9 | 92.3 | 52.9 | 47.4 | 0.676 | |
EC | 74.1 | 96.9 | 87.0 | 80.0 | 0.855 | |
EP | 81.5 | 94.9 | 81.5 | 81.5 | 0.882 | |
SM | 77.3 | 97.1 | 85.0 | 81.0 | 0.872 | |
Gynecologist 6 | ||||||
EH | 59.1 | 96.1 | 76.5 | 66.7 | 0.776 | 69.2 |
AH | 58.6 | 85.4 | 54.8 | 56.7 | 0.720 | |
EC | 52.2 | 92.2 | 60.0 | 55.8 | 0.722 | |
EP | 91.3 | 93.1 | 75.0 | 82.4 | 0.922 | |
SM | 78.6 | 92.8 | 75.9 | 77.2 | 0.857 | |
Model-aided gynecologist 6 | ||||||
EH | 75.0 | 96.9 | 87.5 | 80.8 | 0.860 | 77.6 |
AH | 57.1 | 94.2 | 66.7 | 61.5 | 0.757 | |
EC | 77.8 | 93.9 | 77.8 | 77.8 | 0.858 | |
EP | 88.9 | 89.8 | 70.6 | 78.7 | 0.893 | |
SM | 86.4 | 97.1 | 86.4 | 86.4 | 0.917 | |