Table 3.
The Performance of Models in Training and Testing cohorts.
| Model | AUC (95%CI) | SEN | SPE | ACC | TP | FN | FP | TN |
|---|---|---|---|---|---|---|---|---|
| Morphological features model | ||||||||
| Training cohort | — | 0.667 | 0.883 | 0.796 | 42 | 21 | 11 | 83 |
| Testing cohort | — | 0.690 | 0.9 | 0.823 | 20 | 9 | 5 | 45 |
| Radiomic model | ||||||||
| Training cohort | 0.954 (0.926-0.982) | 0.857 | 0.883 | 0.873 | 54 | 9 | 11 | 83 |
| Testing cohort | 0.940 (0.888-0.992) | 0.793 | 0.92 | 0.873 | 23 | 6 | 4 | 46 |
| Combined model | ||||||||
| Training cohort | 0.957 (0.928-0.987) | 0.889 | 0.915 | 0.904 | 56 | 7 | 8 | 86 |
| Testing cohort | 0.962 (0.927-0.997) | 0.828 | 0.94 | 0.899 | 24 | 5 | 3 | 47 |
| Senior radiologist | ||||||||
| Training cohort | — | 0.571 | 0.947 | 0.796 | 36 | 27 | 5 | 89 |
| Testing cohort | — | 0.517 | 0.980 | 0.810 | 15 | 14 | 1 | 49 |
| Junior radiologist | ||||||||
| Training cohort | — | 0.492 | 0.904 | 0.739 | 31 | 32 | 9 | 85 |
| Testing cohort | — | 0.448 | 0.960 | 0.772 | 13 | 16 | 2 | 48 |
AUC, area under the curve; CI, confidence interval; SEN, Sensitivity; SPE, Specificity; ACC, Accuracy; TP, True Positive; FN, False Negative; FP, False Positive; TN, True Negative.