Analysis and Classification of Occurrence of Brain Stroke Types using Machine Learning Approaches

Strokes are the second leading cause of mortality and also long-term disability. A blockage in the blood flow to the brain or a broken supply line to the brain causes the sudden death of cerebrum cells; this is known as brain-stroke. There are three primary types of brain stroke viz., I) Ischemic stroke, ii) Hemorrhagic stroke and iii) Transient ischemic stroke. This paper proposes the analysis and classification of occurrence of brain stroke types using machine learning approaches. In this paper, dataset is taken from a medical institute containing 13485 records. In the pre-processing stage, duplicate records and inconsistent data are eliminated and missing data are imputed in the dataset. Later records in the dataset are classified into three types of strokes using machine learning methods. The Random Forest classifier gave the best accuracy of 95.46% with the dataset split into 80:20 ratios for training and testing data respectively. This study focuses on developing a system that helps to take precautionary measures by the patient at an earlier stage for a healthy life.