Detection of Brain Stroke Using Machine Learning Algorithm

The organ known as the brain, which is securely protected within the skull and consists of three main parts, namely the cerebrum, cerebellum, and brainstem, is an incredibly complex and intriguing component of the human body. Stroke, a condition that ranks as the second leading cause of death worldwide, necessitates immediate treatment in order to prevent any potential damage to the brain. The early identification of a brain stroke can prove instrumental in mitigating its severity or even preventing it altogether, thereby reducing the mortality rates associated with this ailment. One promising approach to achieve this is the application of machine learning algorithms that have the ability to detect and analyze risk variables. The present research endeavors to provide a comprehensive model that encompasses a systematic method for accurately predicting the occurrence of brain strokes. In order to ensure the efficiency and reliability of our proposed model, a meticulous and thorough process of data collection, data preprocessing, and data transformation techniques have been meticulously employed. To construct the model, a dataset specifically tailored to brain strokes was utilized, and a standardized technique was applied to normalize the data. During the training and testing phases, various classifiers such as KNN, Naïve Bias, Random Forest were implemented to discern their individual performance levels. Based on the results obtained when utilizing the RF classifier, it can be confidently concluded that our proposed model exhibits the highest degree of accuracy.