Table 3 Multivariable ordinal logistic model (nomogram) features for T-stage prediction, with regression coefficients and odds ratios (OR)
From: Interpretable deep learning for multicenter gastric cancer T staging from CT images
Predictor | Coefficient (β) | OR (95% CI) | p-Value |
|---|---|---|---|
Radscore (continuous) | 0.87 | OR per SD: 2.39 (1.8–3.2) | <0.001** |
Tumor size ≥5 cm | 0.69 | 2.00 (1.45–2.75) | <0.001** |
Poor differentiation | 1.15 | 3.16 (2.10–4.75) | <0.001** |
Diffuse Lauren type | 1.06 | 2.90 (1.95–4.32) | <0.001** |
Moderate differentiation | 0.22 | 1.25 (0.85–1.82) | 0.26 |
Mixed Lauren type | –0.08 | 0.92 (0.65–1.30) | 0.64 |
(Intercept for T1 vs ≥T2) | –3.20 | – | – |
(Intercept for ≤T2 vs ≥T3) | –1.17 | – | – |
(Intercept for ≤T3 vs T4) | 1.24 | – | – |
