Page 18 - FULL REPORT 30012024

CHAPTER 1

INTRODUCTION

This chapter provides the background of studies which includes the details of the
research significance, problem statements, research objectives and the scope of

research.

1.1 Background of Study

Aside from the fact that stroke has a high mortality rate of about 5.5 million
per year and a high morbidity rate that leaves up to 50% of survivors with

chronic disabilities, stroke is the second leading cause of death in the world

(Donkor, 2018). There are a lot of current methods of assessing stroke risk
such as ultrasound, computed tomography angiography, and magnetic

resonance angiography. However, these methods may be limited in their
accuracy, availability and high cost. By utilizing algorithms that learn from

data, machine learning models can provide more precise and personalized risk

assessments compared to traditional methods.

Recent studies have shown that machine learning models can predict stroke
risk with greater accuracy and incorporate socioeconomic factors to address

health disparities (Amann, 2021). With the increasing use of the internet and

smartphones, tools for health have gained popularity. These applications offer
accessible and user-friendly tools for individuals to monitor their health. By

integrating machine learning algorithms into a tool for stroke risk prediction,
individuals can have personalized and accurate risk assessments, empowering

them to make informed decisions about their health and lifestyle choices.
Therefore, the development of a system utilizing machine learning algorithms

for stroke risk prediction offers numerous benefits, including improved

1