Incorporating an Integrated Software System for Stroke Prediction using Machine Learning Algorithms and Artificial Neural Network

A stroke is a condition where there is no flow of blood in the brain, depriving it of oxygen and resulting in long-term brain damage and loss of function. Most typically, a clot in an artery delivering blood to the brain is the culprit. Hemorrhage, in which blood leaks into the brain due to a ruptured vessel, is another possible reason. A stroke can result in long-term harm, such as partial paralysis, speech impairment, cognitive impairment, and memory loss. The second leading cause of mortality is a stroke, which is also a significant contributor to disability worldwide. According to the World Stroke Organization (WSO), a stroke will occur in 1 out of every four adults over the age of 25. The number of stroke victims worldwide exceeds 110 million. The objective of the thesis is to incorporate an integrated software system that will predict stroke using artificial neural networks and machine learning algorithms like logistic regression, decision tree classification, random forest classification, K-nearest neighbors, support vector machines, and Nave Bayes classification. The rate of strokes is already alarming. Having an intuitive system to conveniently predict stroke could help reduce the rate of stroke by educating people, which is the focus of the research.

I. Introduction

The American Heart Association claims that when a blood vessel that carries oxygen and nutrients to the brain is either blocked by a clot or breaks, a stroke results. According to the Centers for Disease Control and Prevention (CDC), In 2020, stroke caused 1 in 6 fatalities from cardiovascular disease. Over 795,000 people in the US have a stroke each year [1]. About 185,000 strokes, or close to one in four, affect people who already have one. Ischemic strokes make up about 87% of all strokes; there is no blood flow to the brain. For long-term disability, stroke is the leading cause of worldwide, which is an alarming situation [2].