Machine Learning-Based Non-Communicable Disease Prediction Evaluating the Impact of Hypertension, Diabetes, and Lifestyle Factors on Stroke Risk

Abstract

Chronic diseases such as diabetes, stroke, and heart disease are major challenges in the global health system. Data-driven risk prediction for this disease is important for supporting more precise and effective medical decisions. This study aims to evaluate the main factors contributing to the incidence of diabetes, stroke, and heart disease using logistic regression analysis. The data used are from health sources and includes demographic variables, lifestyle factors, and health indicators. Logistic regression was used to identify variables significantly associated with each health condition studied. The model was evaluated using p-value, regression coefficient, and confidence interval to assess the significance of risk factors. The results of the analysis showed that age, high blood pressure, cholesterol levels, and body mass index (BMI) contributed significantly to the risk of diabetes, stroke, and heart disease. Physical activity and alcohol consumption negatively affected the risk, while smoking factors did not show strong significance in the model. These findings confirm that certain lifestyle factors and health conditions significantly affect the risk of chronic disease. The implications of this research can inform data-driven prevention and early intervention strategies in the health sector.

Ähnliche Arbeiten