Table 4.
AUC | SE | 95% CI | P value | |
---|---|---|---|---|
1st session | ||||
Reader 1 | ||||
Method 1 | 0.568 | 0.075 | 0.423–0.705 | |
Method 2 | 0.679 | 0.067 | 0.535–0.802 | |
Difference | 0.111 | 0.076 | −0.038-0.261 | 0.145 |
Reader 2 | ||||
Method 1 | 0.636 | 0.071 | 0.491–0.765 | |
Method 2 | 0.816 | 0.054 | 0.684–0.909 | |
Difference | 0.180 | 0.075 | 0.033–0.327 | 0.016 |
2nd session | ||||
Reader 1 | ||||
Method 1 | 0.633 | 0.072 | 0.488–0.763 | |
Method 2 | 0.700 | 0.068 | 0.557–0.819 | |
Difference | 0.067 | 0.064 | −0.059-0.193 | 0.299 |
Reader 2 | ||||
Method 1 | 0.647 | 0.07 | 0.503–0.779 | |
Method 2 | 0.753 | 0.063 | 0.614–0.862 | |
Difference | 0.106 | 0.064 | −0.020-0.232 | 0.099 |
Note. Data represent AUC values, standard errors, 95% confidence intervals for the diagnostic performance of each method, classified as intact or tear based on MRI grade using ROC curves. P-value represents the statistical significance of the differences between the two AUCs using methods 1 and 2, calculated using the DeLong test