An Experimental Analysis of Brain Stroke Prediction Using Machine Learning Algorithms

The brain is a fascinating and complex organ. The skull shields the brain, which consists of the brainstem, cerebellum, and cerebrum. Because stroke is one of the world’s most prevalent causes of death, it must be treated quickly to prevent brain damage. A brain stroke can be prevented with early identification, which in turn reduces the mortality rates. One approach is to use machine learning algorithms to identify risk factors. Having a high-quality data collection and cleaning process can streamline the prediction process and help improve the accuracy of predicting brain stroke. This paper proposes a model to achieve an accurate brain stroke forecast. The "Stroke Prediction Dataset" collected from Kaggle was used to train the models. The dataset contains 5110 observations with 12 attributes. Among the seven models used, the gradient-boosting classifier outperformed the rest achieving the highest accuracy of approximately 97.2%. The model can be implemented on cloud-based systems to perform real-time predictions to assist professionals in detecting brain stroke in human beings.