Table 4 The performance of US, CT and PET/CT in predicting DMI, LNM and CSI
From: Radiomics-based fertility-sparing treatment in endometrial carcinoma: a review
Factors and reference | Model and dataset type | Sensitivity | Specificity | Accuracy | AUC |
|---|---|---|---|---|---|
US | Â | Â | Â | Â | Â |
DMI | Â | Â | Â | Â | Â |
Alcazar et al | Â | Â | Â | Â | Â |
|  | Van Holsbeke’s subjective model |  |  |  |  |
| Â | Only one set | 0.80 (33/41) | 0.90 (103/114) | 0.88 (136/155) | NA |
CT | Â | Â | Â | Â | Â |
DMI | Â | Â | Â | Â | Â |
Ytre-Hauge et al | Â | Â | Â | Â | Â |
| Â | ModelR | Â | Â | Â | Â |
| Â | Only one set | NA | NA | NA | 0.71 |
LNM | Â | Â | Â | Â | Â |
Ytre-Hauge et al | Â | Â | Â | Â | Â |
| Â | ModelR | Â | Â | Â | Â |
| Â | Only one set | NA | NA | NA | 0.69 |
CSI | Â | Â | Â | Â | Â |
Ytre-Hauge et al | Â | Â | Â | Â | Â |
| Â | ModelR | Â | Â | Â | Â |
| Â | Only one set | NA | NA | NA | 0.67 |
DECT | Â | Â | Â | Â | Â |
DMI | Â | Â | Â | Â | Â |
Rizzo et al | Â | Â | Â | Â | Â |
| Â | NA | Â | Â | Â | Â |
| Â | Only one set | 1.00 (13/13) | 0.91 (20/22) | 0.94 (33/35) | NA |
PET/CT | Â | Â | Â | Â | Â |
LNM | Â | Â | Â | Â | Â |
Crivellaro et al | Â | Â | Â | Â | Â |
| Â | NA | Â | Â | Â | Â |
| Â | Training set | NA | NA | NA | 0.77 |
| Â | Test set | 0.43 (6/14) | 0.93 (13/14) | 0.68 (19/28) | NA |
Elisabetta et al | Â | Â | Â | Â | Â |
| Â | ModelR | Â | Â | Â | Â |
| Â | Training set | 0.75 (NA) | 0.81 (NA) | NA | NA |
| Â | Test set | 0.89 (NA) | 0.80 (NA) | NA | NA |