Table 6.
Discrimination for predicting pathological response using different criteria.
| SUVmax | SUVpeak | SULmax | SULpeak | End-GLDM-LDHGLE | TLG | LLR | 5PS | |
|---|---|---|---|---|---|---|---|---|
| CPR | ||||||||
| AUC | 0.851 (0.670-0.955) | 0.851 (0.670-0.955) | 0.817 (0.630-0.935) | 0.827 (0.641-0.941) | 0.861 (0.682-0.960) | 0.880 (0.705-0.970) | 0.803 (0.614-0.926) | 0.832 (0.647-0.944) |
| Sensitivity | 76.9 (46.2-95.0) | 76.9 (46.2-95.0) | 76.9 (46.2-95.0) | 69.2 (38.6-90.9) | 92.3 (64.0-99.8) | 100.0 (75.3-100.0) | 69.2 (38.6-90.9) | 92.3 (64.0-99.8) |
| Specificity | 81.3 (54.4-96.0) | 87.5 (61.7-98.4) | 81.3 (54.4-96.0) | 93.8 (69.8-99.8) | 75.0 (47.6-92.7) | 68.8 (41.3-89.0) | 87.5 (61.7-98.4) | 50.0 (24.7-75.3) |
| Accuracy | 79.3 (60.3-92.0) | 82.8 (64.2-94.2) | 79.3 (60.3-92.0) | 82.8 (64.2-94.2) | 82.8 (64.2-94.2) | 82.8 (64.2-94.2) | 79.3 (60.3-92.0) | 69.0 (49.2-84.7) |
| Threshold | 4.20 | 3.65 | 3.32 | 3.00 | 349.5 | 3.92 | 1.50 | 4 |
| MPR | ||||||||
| AUC | 0.928 (0.768-0.990) | 0.956 (0.807-0.998) | 0.894 (0.724-0.977) | 0.933 (0.776-0.992) | 0.822 (0.636-0.938) | 0.917 (0.753-0.986) | 0.889 (0.717-0.975) | 0.894 (0.724-0.977) |
| Sensitivity | 100.0 (66.4-100.0) | 100.0 (66.4-100.0) | 100.0 (66.4-100.0) | 100.0 (66.4-100.0) | 77.8 (40.0-97.2) | 100.0 (66.4-100.0) | 88.9 (51.8-99.7) | 100.0 (66.4-100.0) |
| Specificity | 80.0 (56.3-94.3) | 85.0 (62.1-96.8) | 80.0 (56.3-94.3) | 80.0 (56.3-94.3) | 85.0 (62.1-96.8) | 75.0 (50.9-91.3) | 85.0 (62.1-96.8) | 45.0 (23.1-68.5) |
| Accuracy | 86.2 (68.3-96.1) | 89.7 (72.6-97.8) | 86.2 (68.3-96.1) | 86.2 (68.3-96.1) | 82.8 (64.2-94.2) | 82.8 (64.2-94.2) | 86.2 (68.3-96.1) | 62.1 (42.3-79.3) |
| Threshold | 4.20 | 3.65 | 3.32 | 2.64 | 674.3 | 7.77 | 1.50 | 4 |