Table 2 Multivariate logistic regression analysis of the radiomics score and clinical candidate predictors in the training set
Variables and intercept | Univariate model | Radiomic-clinical multivariate model | Clinical multivariate model | ||||||
|---|---|---|---|---|---|---|---|---|---|
β | OR (95% CI) | P | β | OR (95% CI) | P | β | OR (95% CI) | P | |
The radiomics score (per 0.1 increase) | 0.207 | 1.230 (1.152 to 1.336) | < 0.001* | 0.204 | 1.226 (1.146 to 1.332) | < 0.001* | – | – | – |
Sex (male vs. female) | 0.423 | 1.526 (0.314 to 7.982) | 0.283 | – | – | – | – |  |  |
Age, years (continuous) | 0.004 | 1.004 (0.978 to 1.032) | 0.759 | – | – | – | – | – | – |
Symptom number | 0.824 | 2.279 (1.487 to 3.594) | < 0.001* | 0.602 | 1.826 (1.013 to 3.280) | 0.042* | 0.852 | 2.344 (1.511 to 3.731) | < 0.001* |
Hypertension (no vs. yes) | 0.201 | 1.222 (0.585 to 2.574) | 0.593 | – | – | – | – | – | – |
Smoker (no vs. yes) | − 0.597 | 0.550 (0.124 to 1.742) | 0.359 | – | – | – | – | – | – |
Tumor location (left vs. right) | 0.692 | 1.998 (0.951 to 4.317) | 0.071 | – | – | – | – | – | – |
MRI-determined tumor size, cm (continuous) | 0.173 | 1.189 (1.056 to 1.346) | 0.005* | – | – | – | 0.182 | 1.200 (1.060 to 1.365) | 0.004* |
Hyperintense on a T2 weighted MRI (no vs. yes) | 2.014 | 7.495 (2.139 to 47.541) | 0.007* | – | – | – | – | – | – |
Intercept | – | – | – | 0.781 |  |  | − 2.586 | – | – |