STROKE RISK PREDICTION

Abstract

Stroke is one of the leading causes of mortality and long-term disability worldwide, making early risk detection crucial for effective prevention and treatment. This project focuses on developing a machine learning-based model to predict the risk of stroke using demographic and symptom-related features. A clinically inspired and balanced dataset, consisting of 50% high-risk and 50% low-risk cases, is utilized to ensure unbiased model training and evaluation. The methodology involves data preprocessing, exploratory data analysis, and the application of various machine learning algorithms to identify the most effective predictive model. Among the tested approaches, the Random Forest classifier demonstrated superior performance, achieving an accuracy of 95%, precision of 95%, and recall of 97% on the test dataset. These results indicate the model’s strong ability to correctly identify individuals at risk of stroke while minimizing false predictions. In addition, statistical analysis such as correlation studies and hypothesis testing (t-test) is conducted to understand the influence of different features, with age emerging as a significant factor in stroke risk. The developed model highlights the potential of machine learning in healthcare by providing accurate and reliable predictions.