Table 4.
A list of independent predictors that demonstrated significant clinical association with the (a) 3-month and (b) 6-month response rate assessment. Statistical significance is indicated by false discovery rate (FDR)-adjusted p-value, after applying the Benjamini–Hochberg (BH) procedure for multiple test corrections. AUC, sensitivity, specificity, and p-value obtained from logistic regression for each of these independent significant predictors are also reported. The superscript a denotes FDR-adjusted p-values obtained after applying the BH procedure.
| (a) | ||||
|---|---|---|---|---|
| Features | FDR-Adjusted p-Value a | AUC | Sensitivity | Specificity |
| DeltaP radiomic features | ||||
| Second-order feature | ||||
| GLCM_Joint Entropy | 0.063 | 0.690 (0.527–0.843) |
0.625 | 0.667 |
| GLRLM_Run Entropy | 0.044 | 0.734 (0.573–0.869) |
0.625 | 0.667 |
| GLSZM_Gray-Level Non-Uniformity Normalized | 0.038 | 0.734 (0.566–0.881) |
0.625 | 0.667 |
| GLSZM_Small Area Emphasis | 0.038 | 0.766 (0.600–0.912) |
0.625 | 0.667 |
| (b) | ||||
| Features | FDR-Adjusted p-Value a | AUC | Sensitivity | Specificity |
| Shape and size features | ||||
| Major Axis Length | 0.074 | 0.724 (0.529–0.891) |
0.714 | 0.607 |
| AP radiomic features | ||||
| First-order feature | ||||
| Kurtosis | 0.028 | 0.750 (0.589–0.895) |
0.786 | 0.643 |
| Maximum | 0.028 | 0.727 (0.564–0.879) |
0.714 | 0.607 |
| Second-order feature | ||||
| GLRLM_Short Run High Gray-Level Emphasis | 0.028 | 0.773 (0.606–0.917) |
0.857 | 0.679 |
| GLDM_Small Dependence High Gray-Level Emphasis | 0.055 | 0.719 (0.533–0.859) |
0.786 | 0.643 |
| DeltaP radiomic features | ||||
| First-order feature | ||||
| Range | 0.047 | 0.699 (0.518–0.865) |
0.786 | 0.643 |
| Second-order feature | ||||
| GLCM_Joint Entropy | 0.070 | 0.707 (0.527–0.877) |
0.643 | 0.571 |
| GLRLM_Run Entropy | 0.047 | 0.747 (0.593–0.883) |
0.714 | 0.607 |
| GLSZM_Gray-Level Non-Uniformity Normalized | 0.047 | 0.788 (0.633–0.909) |
0.786 | 0.643 |