Table 2 Performance of various radiomics-guided DL models.
From: Radiomics-guided deep neural networks stratify lung adenocarcinoma prognosis from CT scans
Training | |||
|---|---|---|---|
Hazard ratio (95% CI) | p-value | C-index | |
Radiomics | 4.1384 (1.9140–8.9479) | 0.0006 | 0.7459 |
Yolo | 3.1728 (1.4675–6.8596) | 0.0061 | 0.7994 |
DenseNet | 4.4383 (2.0468–9.6241) | 0.0003 | 0.7410 |
VGG | 4.2454 (1.9658–9.1683) | 0.0005 | 0.7660 |
Test | |||
Hazard ratio (95% CI) | p-value | C-index | |
Radiomics | 5.0566 (1.5379–16.6263) | 0.0180 | 0.6837 |
Yolo | 7.6277 (2.2870–25.4399) | 0.0027 | 0.7593 |
DenseNet | 5.3620 (1.7011–18.5774) | 0.0116 | 0.7214 |
VGG | 4.7060 (1.4333–15.4520) | 0.0244 | 0.8009 |
Validation | |||
Hazard ratio (95% CI) | p-value | C-index | |
Radiomics | – | – | – |
Yolo | 6.5362 (2.1773–19.6213) | 0.0022 | 0.7696 |
DenseNet | 4.2277 (1.4800–12.0768) | 0.0153 | 0.7112 |
VGG | 3.5488 (1.2409–10.1494) | 0.0362 | 0.6757 |
