Development and validation of a multiparametric MRI-based radiomics nomogram for the tripartite discrimination of primary benign, primary malignant, and metastatic lumbar spinal tumors

Objective To develop and validate a multiparametric magnetic resonance imaging-based radiomics nomogram for the non-invasive preoperative differentiation of primary benign, primary malignant, and metastatic lumbar spinal tumors. Methodology This retrospective study enrolled 100 patients with pathologically confirmed lumbar tumors. Radiomics features were extracted from T1-weighted, T2-weighted, and fat-suppressed T2-weighted sequences. A radiomics signature was constructed using a two-step feature selection method comprising minimum redundancy maximum relevance and least absolute shrinkage and selection operator regression. Clinical predictors were selected via univariate and multivariate analysis. An integrated nomogram was developed by combining the radiomics signature and independent clinical predictors within a multinomial logistic regression model. The model’s performance was evaluated regarding discrimination, calibration, and clinical utility. Results The radiomics signature comprised 11 stable features. Five independent clinical predictors were identified. The integrated nomogram demonstrated robust discrimination, with a macro-average area under the curve of 0.887 (95% CI: 0.832–0.931) in the independent test set. The nomogram showed good calibration and provided a superior net benefit across a wide range of threshold probabilities in decision curve analysis. Conclusion The proposed nomogram, integrating radiomic and clinical data, serves as a robust and non-invasive tool for preoperatively differentiating the three types of lumbar spinal tumors, holding significant potential to support clinical decision-making.