Paper Contents
Abstract
Stroke is one of the leading causes of death and long-term disability worldwide, making early detection essential for improving patient outcome. In this study, we present a machine learning framework designed to predict stroke risk using patients health records. The model considers important factors, such as age, sex, hypertension, heart disease, body mass index (BMI), smoking status, and glucose levels, to identify individuals who may be at risk.After data cleaning, feature selection, and model optimization, several algorithms were tested, with the Random Forest algorithm showing the best results. The model achieved 95% accuracy and an AUC-ROC score of 0.99, proving highly reliable in capturing the complex relationships between health attributes and the incidence of stroke. To make the predictions more practical, patients were categorized into three risk levels: low, mid, and high. Low-risk individuals are unlikely to develop stroke, mid-risk individuals may require lifestyle changes, and high-risk individuals require immediate medical attention.The analysis also showed that age, glucose level, and BMI were the most important predictors. By integrating this framework into healthcare systems, doctors can take preventive action earlier, provide personalized care, and optimize resources, ultimately helping to reduce the burden of stroke on both patients and healthcare providers.
Copyright
Copyright © 2025 DILLI GANESH A. This is an open access article distributed under the Creative Commons Attribution License.
