Evaluating the Risk of Postoperative Venous Thromboembolism in Cervical Cancer Patients

Development and Validation of Machine Learning Models to Evaluate the Postoperative Venous Thromboembolism Risk of Cervical Cancer Patients in China

The aim of this study is to develop a machine learning model to accurately predict the risk of venous thromboembolism in patients with cervical cancer after surgery.

Venous thromboembolism (VTE) is a common and life-threatening complication in patients with cervical cancer following surgery. The objective of this study is to develop a machine learning model with the potential to predict the risk of VTE in these patients postoperatively. We plan to employ partial dependence (PD) curves, breakdown (BD) curves, Ceteris-paribus (CP), and SHapley additive exPlanations (SHAP) values for a comprehensive analysis. The goal is to explore how different machine learning algorithms can be utilized as tools for personalized postoperative VTE risk assessment in cervical cancer patients.

Inclusion Criteria:

• Confirmation of cervical cancer through pathological examination.
• Receipt of surgical treatment for cervical cancer at Chongqing University Cancer Hospital in China.
• Provision of comprehensive case information.

Exclusion Criteria:

• Patients under the age of 18.
• History of VTE caused by other reasons before surgery.
• Secondary cervical cancer or accompanying primary malignant tumor.