Exploratory analysis of the contributing factors of stroke

Abstract

Stroke is a medical condition that occurs when the blood supply to the brain is interrupted, which causes cell death. Previous studies have shown that the leading risk factors of stroke include elevated systolic blood pressure, poor diet, high body mass index, high fasting plasma glucose, ambient particulate matter pollution, smoking, high low-density lipoprotein, kidney dysfunction, alcohol use, and low physical activity. However, how these factors differentially contribute to the occurrence of stroke remains elusive. In this study, 5110 cases and 11 risk factors were investigated and machine learning models were built to predict the occurrence of stroke and the importance of each of the factors. It was found that the random forest model showed the best performance in predicting stroke and age, glucose level, and body mass index were the top three most important risk factors underlying stroke. These findings shed light on future research in the prevention and diagnosis of stroke.