BRAIN STROKE PREDICTION USING MACHINE LEARNING

Abstract

The present document is a project carried out in the field of engineering to create a system for the prediction of brain stroke using machine learning. A major cause of death and disability worldwide, including in places with limited resources like Bangladesh, is brain stroke. Improving patient outcomes requires early prediction and action. Using a variety of datasets, this research creates a machine learning-based method for brain stroke prediction. The system analyzes patient data using a variety of machine learning algorithms, such as Support Vector Machines (SVM), Random Forest (RF), and Artificial Neural Networks (ANN). incorporating both clinical features (e.g., blood pressure, cholesterol levels) and demographic information (e.g., age, gender). Data was collected from Kaggle and supplemented with data gathered from local hospitals and NGOs in Bangladesh to enhance dataset diversity and model generalizability. Metrics including F1-score, recall, accuracy, and precision are used to assess the models' performance. This research aims to create a practical, accessible tool for healthcare providers and individuals to facilitate early detection, personalized healthcare, and timely interventions, ultimately contributing to a reduction in stroke-related mortality and morbidity. The potential applications include early The present document is a project carried out in the field of engineering to create a system for the prediction of brain stroke using machine learning. A major cause of death and disability worldwide, including in places with limited resources like Bangladesh, is brain stroke. Improving patient outcomes requires early prediction and action. Using a variety of datasets, this research creates a machine learning-based method for brain stroke prediction. The system analyzes patient data using a variety of machine learning algorithms, such as Support Vector Machines (SVM), Random Forest (RF), and Artificial Neural Networks (ANN). incorporating both clinical features (e.g., blood pressure, cholesterol levels) and demographic information (e.g., age, gender). Data was collected from Kaggle and supplemented with data gathered from local hospitals and NGOs in Bangladesh to enhance dataset diversity and model generalizability. Metrics including F1-score, recall, accuracy, and precision are used to assess the models' performance. This research aims to create a practical, accessible tool for healthcare providers and individuals to facilitate early detection, personalized healthcare, and timely interventions, ultimately contributing to a reduction in stroke-related mortality and morbidity. The potential applications include early diagnosis, personalized treatment plans, automated risk assessment, public health insights, and telemedicine integration, particularly benefiting under-resourced communities diagnosis, personalized treatment plans, automated risk assessment, public health insights, and telemedicine integration, particularly benefiting under resourced communities