Prediction of Stroke using Data Mining Classification Techniques

Ohoud Almadani,Riyad Alshammari

doi:10.14569/ijacsa.2018.090163

Abstract

Stroke is a neurological disease that occurs when a brain cells die as a result of oxygen and nutrient deficiency. Stroke detection within the first few hours improves the chances to prevent complications and improve health care and management of patients. In addition, significant effect of medications that were used as treatment for stroke would appear only if they were given within the first three hours since the beginning of stroke. A framework has been designed based on data mining techniques on Stroke data set that is obtained from Ministry of National Guards Health Affairs hospitals, Kingdom of Saudi Arabia. A data mining model was built with 95% accuracy. Furthermore, this study showed that patient with the following medical conditions, such as heart diseases (hypertension mainly), immunity diseases, diabetes militias, kidney diseases, hyperlipidemia, epilepsy, or blood (platelets) disorders has a higher probability to develop stroke.

Highlights

Knowledge Discovery from Data (KDD) is a growing field of computer science that deals with information gain and decision support through large data analysis and automated extraction of patterns.Information gain from health data may lead to innovative solution or better treatment plan for patients
In order to gain knowledge intelligently from stroke data, a data mining technique is utilized to semi-automatically process data and generate data mining model that can be used by health care professionals [1]
The comparison of the data mining algorithms used with 10-fold cross validation method, were data set first performed on training data set before any attribute reduction methods

Summary

Introduction

Knowledge Discovery from Data (KDD) is a growing field of computer science that deals with information gain and decision support through large data analysis and automated extraction of patterns. Information gain from health data may lead to innovative solution or better treatment plan for patients. In order to gain knowledge intelligently from stroke data, a data mining technique is utilized to semi-automatically process data and generate data mining model that can be used by health care professionals [1]. A stroke is a neurological disease that occurs when a brain cells die because of oxygen and nutrient deficiency. Occlusion of brain blood vessel by a clot or blood vessel rupturing are the major causes of oxygen and nutrient supply deficiency [2]. Cerebro-Vascular Accident (CVA) is the previous name of stroke, which divided nowadays into three types known as Hemorrhagic stroke, Acute Ischemic stroke, or Transient Ischemic Attack [2], [3]

Objectives

Methods

Results

Conclusion