Stroke Prediction Using Ensemble Learning

Abstract

Stroke is one of the leading causes of death and long-term disability worldwide, making early prediction and diagnosis essential for improving patient outcomes. Recent advancements in Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL) have enabled the development of intelligent systems capable of predicting stroke risk using patient healthcare data. This review analyzes various research works conducted between 2020 and 2025 on brain stroke prediction using machine learning and neural network-based approaches. Several studies employed algorithms such as Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), and Artificial Neural Networks (ANN) to identify stroke risk factors and predict stroke occurrence. The reviewed works achieved prediction accuracies ranging from 88% to 92%, demonstrating the effectiveness of data-driven predictive models in healthcare applications. Furthermore, deep learning and predictive analytics approaches showed improved performance compared to traditional machine learning techniques. Despite these advancements, challenges such as limited dataset size, lack of real-time clinical validation, inadequate feature optimization, and poor generalization across diverse patient populations remain significant concerns. The findings indicate that integrating advanced deep learning models, feature engineering techniques, and real-time healthcare monitoring systems can further enhance stroke prediction accuracy and support early diagnosis. This review highlights the importance of intelligent predictive systems in reducing stroke related mortality and assisting healthcare professionals in clinical decision-making.