Development of a VTE prediction model based on automatically selected features in glioma patients

Abstract

Venous thromboembolism (VTE) is a serious threat to patients undergoing cancer treatment, especially in advanced and metastatic diseases. In neuro-oncology, the incidence of VTE depends on the location and stage of the tumor. Some primary and secondary brain tumors have an increased propensity for thrombotic events. In this study, we applied state-of-the-art machine learning methods, particularly XGBoost, to create models to search for predictors associated with the risk of VTE in glioma patients. By comparing the diagnostic accuracy of our XGBoost models with traditional logistic regression approaches, we aim to advance the understanding of VTE prediction in this patient population. Our results add to the growing body of research on thrombosis risk assessment in cancer patients and may help in the development of personalized prevention and treatment strategies to reduce the risk of VTE in hospitalized glioma patients.