Stacking Model-Based Preoperative Prediction of Lymphovascular Invasion in Lung Adenocarcinoma by Integrating Intratumoral Heterogeneity Score and Clinicoradiological Features: A Multicenter Study.
Authors
Affiliations (6)
Affiliations (6)
- Department of Radiology, The Fifth People's Hospital of Xiangtan City, Xiangtan, Hunan, China.
- Department of Radiology, Xiangtan Central Hospital, Xiangtan, Hunan, China.
- Department of Radiology, The Third Xiangya Hospital of Central South University, Changsha, Hunan, China.
- Department of Radiology, The Affiliated Hospital of Southwest Medical University, Luzhou, Sichuan, China.
- School of Mathematics and Computational Science, Xiangtan University, Xiangtan, Hunan, China.
- College of Information and Intelligence, Hunan Agricultural University, Changsha, Hunan, China.
Abstract
IntroductionLymphovascular invasion (LVI) is associated with poor outcomes in lung adenocarcinoma, yet accurate preoperative prediction in clinical stage IA disease remains difficult. We aimed to develop and validate a multicenter ensemble machine-learning model that integrates intratumoral heterogeneity (ITH) score and clinicoradiological features for preoperative LVI prediction.MethodsIn this retrospective multicenter prediction-model development and external validation study, 1,527 patients with surgically confirmed stage IA lung adenocarcinoma from three tertiary centers were included. Patients from two centers formed the training cohort (n = 1,068), and patients from the third center formed the external validation cohort (n = 459). The intratumoral heterogeneity (ITH) score was quantified from preoperative computed tomography (CT) by integrating global and local pixel-distribution patterns. Ten base machine-learning algorithms and 31 stacking configurations were evaluated. Model performance was assessed using area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, specificity, positive predictive value, negative predictive value, and F1 score. SHapley Additive exPlanations (SHAP) were used for model interpretation and feature selection.ResultsThe XGBoost plus random forest stacking model ranked first overall according to the comprehensive score and achieved the best integrated diagnostic performance. In comparative analyses, the stacking model showed a more balanced diagnostic profile than the radiomics-only, ITH-only, and clinicoradiological-only models, with the highest overall mean performance, although the radiomics-only model showed a slightly higher AUC and sensitivity. SHAP analysis identified the ITH score as the dominant contributor to model output.ConclusionA stacking model integrating ITH score and clinicoradiological features enables reliable preoperative prediction of LVI in stage IA lung adenocarcinoma and may support individualized risk stratification and surgical planning.