From: Radiomics assessment of carotid intraplaque hemorrhage: detecting the vulnerable patients
Set | Model | AUC (95%CI) | Sensitivity (%) | Specificity (%) | Accuracy (%) |
---|---|---|---|---|---|
Training set | Clinical model | 0.631 (0.524–0.738) | 61.8 (28/46) | 49.0 (29/60) | 56.7 (60/106) |
Radiomics signature | 0.717 (0.620–0.814) | 68.9 (32/46) | 60.0 (36/60) | 65.1 (69/106) | |
Radiomics nomogram | 0.743 (0.650–0.835) | 69.1 (32/46) | 56.9 (34/60) | 63.2 (67/106) | |
External test set | Clinical model | 0.761 (0.596–0.927) | 80.0 (14/18) | 77.8 (16/20) | 78.9 (30/38) |
Radiomics signature | 0.725 (0.562–0.888) | 62.5 (11/18) | 69.2 (14/20) | 63.2 (24/38) | |
Radiomics nomogram | 0.811 (0.661–0.961) | 88.9 (16/18) | 80.0 (16/20) | 84.2 (32/38) |