Heart Stroke Prediction using Machine Learning

The system proposed in this paper specifies. An overlook that monitors stroke prediction. Prediction is done based on the condition of the patient, the ascribe, the diseases he has, and the influences of those diseases that lead to a stroke, early prediction of heart stroke risk can help in timely Intercede to minimize the risk of stroke, by making use of Machine learning algorithms, for stroke prediction. This study aims to develop a machine-learning model that can accurately predict a stroke. The data collected for the study include the clinical and demographic characteristics of the patients. Either had a stroke or are in danger of having one. On the preprocessed dataset, various machine learning models are trained, including logistic regression, decision trees, random forests, and the KNN model, Naive Bayes. Accuracy, precision, recall, and F1-score are a few examples of standard assessment metrics that are used to assess each model's performance. The models are also tested using a confusion matrix to determine how well they can distinguish between individuals at high and low risk of having a stroke. Stroke is a leading cause of death worldwide, and early identification of individuals at risk can significantly improve outcomes, and help people be cautious and take preventative measures. Machine learning algorithms have been well suited and their flexibility in predicting stroke risk by analyzing large datasets of patient information. This review provides an outlook on recent research on stroke prediction using machine learning, including the types of data used, the algorithms employed, and the performance metrics reported. Machine learning algorithms have been applied to these data sources to identify patterns and develop predictive models. F1-Score of 96% was achieved in this study using the XGBoost classifier. The applied machine learning algorithms show potential for improving stroke risk prediction, by signifying a pattern, studying the case study, and outlying factors.