Research on the Prediction Model of Sudden Death Risk in Coronary Heart Disease Based on XGBoost and Random Forest

This study explored the feasibility of applying XGBoost and random forest algorithms to predict the risk of sudden death from coronary heart disease. From the perspective of symmetry, the human body′s physiological and pathological states can be considered to have a certain dynamic balance, akin to a form of biological symmetry. Sudden death from coronary heart disease disrupts this inherent balance, representing extreme asymmetry in the body′s state. Our study aims to restore a degree of symmetry in the decision-making process for medical professionals by providing accurate prediction models. By adding the fuzzy comprehensive evaluation method for data preprocessing, the prediction models for sudden death from coronary heart disease based on XGBoost and random forests were optimized and constructed. The results indicated that XGBoost and random forest algorithms could be effectively applied to predict the risk of sudden death from coronary heart disease. The promotion and application of these models could serve as an auxiliary tool to provide additional insights that may assist physicians in their decision-making, especially for those with relatively less clinical experience in grassroots units, enable early intervention for high-risk patients, and thereby reduce the occurrence and mortality risk of sudden death from coronary heart disease.