Machine Learning-Based Heart Disease Prediction Model: Application of Data Science in Health Economics

This article introduces a study combining machine learning and health economics, which uses the UCI Heart Disease Dataset to build a prediction model and explores the mechanisms by which physiological, demographic, and lifestyle factors affect heart disease risk. The study uses logistic regression and random forest algorithms, analyzes the value of disease prediction from the perspectives of economics and public health, and demonstrates the significance of interdisciplinary research.

Heart disease is a major global health threat, causing millions of deaths each year and imposing a heavy economic burden on healthcare systems. Traditional risk assessment relies on clinical experience and simple statistics, making it difficult to capture complex interactions. Machine learning can handle high-dimensional data, capture non-linear relationships, automate feature learning, and has strong scalability, providing new tools for disease prediction.

Data Sources: Uses the Heart Disease Dataset from the UCI Machine Learning Repository, containing medical records of hundreds of patients. Research Questions: 1. Identification of the strongest predictors; 2. Impact of lifestyle variables; 3. Demographic differences; 4. Contribution of variables to prediction accuracy. Model Selection: Logistic regression (high interpretability) and random forest (captures non-linear interactions, high prediction accuracy).

Key Risk Factors: Physiological indicators (blood pressure, cholesterol, etc.), symptom manifestations (chest pain type, etc.), demographics (age, gender), lifestyle (exercise capacity). Age and Gender Differences: Risk increases with age, and there are differences in risk patterns between men and women. Lifestyle Correlations: Individuals with impaired exercise capacity have higher risk; chest pain characteristics are related to the degree of lesions. Model Performance: Logistic regression has excellent interpretability, while random forest has higher prediction accuracy; combining both is a best practice.

Cost-Effectiveness: Improves screening efficiency, reduces medical costs, optimizes resource allocation, supports insurance pricing and policy formulation. Health Equity: Need to implement differentiated screening and interventions for high-risk groups, focus on modifiable risk factors, and identify subpopulations requiring attention through models.

Limitations: Limited sample representativeness, data timeliness issues, causal inference constraints, clinical practicality to be verified. Future Directions: Model optimization (advanced algorithms, feature engineering), external validation, dynamic prediction, causal analysis, intervention effect evaluation.

This study demonstrates the application potential of data science in the medical field. By combining machine learning and economic perspectives, it provides new tools for heart disease risk assessment and helps understand the socioeconomic impact of the disease. In the context of the integration of AI and healthcare, such research will promote the realization of precision medicine and a fair health society.