Predicting Long-Term Survival after Endovascular Aneurysm Repair Using Machine Learning-Based Decision Tree Analysis.

Area of Science:

• Vascular Surgery
• Medical Informatics
• Machine Learning in Healthcare

Background:

• Endovascular aneurysm repair (EVAR) is a preferred minimally invasive treatment for abdominal aortic aneurysms (AAA).
• Identifying predictors of long-term patient survival post-EVAR is essential for improving prognosis.
• Machine learning (ML) offers potential for uncovering complex patterns in patient data to predict outcomes.

Purpose of the Study:

• To investigate the efficacy of ML-based decision tree analysis (DTA) in predicting long-term survival (over 5 years) after EVAR for AAA.
• To identify key patient factors influencing long-term survival following EVAR.

Main Methods:

• Retrospective analysis of 142 patients undergoing elective EVAR for AAA.
• Utilized a dataset of 24 variables including demographics, comorbidities, and surgical details.
• Developed and validated a decision tree classifier using Python 3.7 and scikit-learn.

Main Results:

• Poor nutritional status emerged as the most significant predictor of long-term survival.
• Other key predictors included compromised immunity, active cancer, octogenarian status, chronic kidney disease, and COPD.
• The DTA model achieved 76.1% accuracy, 81.8% sensitivity, and an AUC of 0.84.

Conclusions:

• ML-based DTA is effective in predicting long-term survival after EVAR.
• Comprehensive preoperative assessments are crucial for identifying high-risk patients.
• Personalized management strategies informed by DTA can improve patient outcomes.