Analyzing the Performance of Brain Stroke Prediction Using Various Machine Learning Classification Algorithms

Abstract

In today’s day and age, strokes are officially regarded as the leading cause of death and disability globally, per the World Health Organization (WHO). A stroke, also known as a cerebrovascular accident, is a medical emergency that occurs when blood flow to a part of the brain is interrupted or reduced, leading to damage or death of brain cells. In many cases, the medical diagnosis of a stroke isn’t attained until after its onset, which, more often than not, leads to fatal consequences. Prompt medical attention is critical when treating a stroke to minimize brain damage and prevent long-term disability or death. Lately, machine learning has been viewed as a significant advancement towards preemptive stroke diagnosis. Machine learning (ML) algorithms analyze vast amounts of medical data, including electronic health records, medical imaging, genetic information, and real-time patient monitoring data, to uncover patterns and insights that were previously unattainable. This research paper investigates the application of machine learning models at their fundamental level for stroke prediction. The paper employs a supervised machine learning model, applying regression algorithms to a collected patient dataset comprising demographic, clinical, and lifestyle factors of patients. Various classifiers, including logistic regression, decision trees, support vector machines (SVM), k-nearest neighbors (KNN), and random forest, were employed to develop predictive models. The study aimed to assess the performance of these classifiers and identify the most accurate model for stroke prediction. Results indicated that the random forest classifier achieved the highest accuracy among all models evaluated, with 99.81% accuracy. This finding underscores the efficacy of ensemble learning techniques in capturing complex interactions and non-linear relationships within the data. The research highlights the potential of ML-based approaches for identifying high-risk individuals for stroke and guiding targeted preventive interventions in clinical practice.