Table 3.
Analysis, validation, and results for csPCa prediction in the selected studies based on handcrafted radiomics as the feature extraction method.
| Reference, Year | Analysis | Validation | Sequence for the Best Model | Best Radiomic Model [CI, 95%] a | PI-RADS Cut-Off | PI-RADS Model [CI, 95%] a | ||||
|---|---|---|---|---|---|---|---|---|---|---|
| AUC | Sensitivity | Specificity | AUC | Sensitivity | Specificity | |||||
| Dominguez et al. 2023 [24] | Index | CV//Hold-out set | ADC | 0.81 [0.56–0.94]//0.71 | NR | NR | NR | 0.66 [0.57–0.74]//NR | NR | NR |
| Prata et al. 2023 [25] | Index | CV | ADC | 0.77 | NR | NR | NR | 0.68 | NR | NR |
| Jin et al. 2023 [26] | Index | Hold-out set//External (1 set) | T2 + ADC + DWI (b2000) | 0.80//0.80 | 0.80//0.73 | 0.65//0.92 | NA | NA | NA | NA |
| Jing et al. 2022 [29] | Index | Hold-out set//External (2 sets) | T2 (prostate) + DWI b1500 (lesion) | 0.96 [0.90, 1.00]//0.95 [0.87, 1.00]//0.94 [0.90, 0.99] b | 0.95//0.98//0.86 b | 0.94//0.86//0.91 b | NR | 0.84 [0.74, 0.95]//0.82 [0.72, 0.93]//0.80 [0.71, 0.88] | 0.98//0.98//0.50 | 0.56//0.52//0.94 |
| Lim et al. 2021 [33] | All | CV | ADC | 0.68 [0.65–0.72] | NR | NR | NA | NA | NR | NR |
| Hectors et al. 2021 [34] | Index | Hold-out set | T2 | 0.76 [0.60–0.92] | 0.75 | 0.8 | NA | NA | NA | NA |
| Castillo et al. 2021 [35] | Index | CV//External | T2 + ADC + DWI (highest-b value) | 0.72 [0.64, 0.79]//0.75 | 0.76 [0.66, 0.89]//0.88 | 0.55 [0.44, 0.66]//0.63 | ≥3 | 0.50//0.44 (2 radiologists, External Validation) | 0.76//0.88 | 0.25//0 |
| Li et al. 2020 [36] | Index | Hold-out set | T2 + ADC | 0.98 [0.97–1.00] | 0.95 | 0.87 | NA | NA | NA | NA |
All = all lesions, AUC = area under the curve, CI = confidence interval, csPCa = clinically significant prostate cancer, CV = cross-validation, Index = index lesion, NA = not applicable, NR = not reported, PI-RADS = Prostate Imaging Reporting and Data System. a Data are expressed in the corresponding metric and the CI, 95% for each validation method separated by//. If the CI is not included, it means that it was not reported in the study. b The combined model (radiomic model + PI-RADS) is included since there are no data for the radiomic model.