Table 4.
Predictive performance of the radiomics signature and the two prediction models for the discrimination of metastasis in ovarian cancer.
| Models | Training cohort | Validation cohort | ||||
|---|---|---|---|---|---|---|
| SEN | SPE | AUC (95% CI) | SEN | SPE | AUC (95% CI) | |
| Radiomics model | 0.76 | 0.80 | 0.84 (0.74-0.93) | 0.90 | 0.70 | 0.82 (0.66-0.98) |
| Clinical model | 0.84 | 0.65 | 0.70 (0.58-0.83) | 0.71 | 0.80 | 0.83 (0.67-0.95) |
| Combined model | 0.86 | 0.85 | 0.88 (0.80-0.96) | 0.81 | 0.80 | 0.86 (0.72-0.99) |
SEN, sensitivity; SPE, specificity; AUC, are under curve; 95% CI, 95% confidence interval.