A | Ipsilateral | Contralateral | |||||||
Sensitivity (%) | Specificity (%) | J | ROC-AUC | Sensitivity (%) | Specificity (%) | J | ROC-AUC | ||
Planimetry | Increased (0.2–0.25 cm2) | 79.0 | 63.8 | 0.43 | 0.718* | 84.1 | 63.8 | 0.48 | 0.821* |
Hyperechogenic (>0.25 cm2) | 51.7 | 75.0 | 0.27 | 67.0 | 85.0 | 0.52 | |||
Echogenicity | Grade II | 85.1 | 43.8 | 0.29 | 0.726* | 89.7 | 48.8 | 0.39 | 0.792* |
Grade III | 64.9 | 76.3 | 0.41 | 73.7 | 80.0 | 0.54 |
B | Ipsilateral | Contralateral | |||||||
Sensitivity (%) | Specificity (%) | J | ROC-AUC | Sensitivity (%) | Specificity (%) | J | ROC-AUC | ||
Planimetry | Increased (0.2–0.25 cm2) | 87.8 | 60.3 | 0.48 | 0.740* | 88.1 | 61.8 | 0.50 | 0.833* |
Hyperechogenic (>0.25 cm2) | 61.2 | 71.4 | 0.33 | 73.3 | 83.6 | 0.57 | |||
Echogenicity | Grade II | 90.8 | 47.6 | 0.38 | 0.766* | 90.0 | 47.3 | 0.37 | 0.795* |
Grade III | 77.6 | 71.4 | 0.49 | 78.2 | 78.2 | 0.56 |
Sensitivity, specificity and area under the curve (AUC) of the receiver-operating characteristic (ROC) analysis are presented. Data are given (A) for all measurements (n=256) and (B) for measurements through good bone windows (n=142).
↵* p<0.0001 (Fisher exact test).