Hypertension Predictive Model with a Neural Network Approach: A Case Study of Kumasi Metropolis

Loading...
Thumbnail Image
Date
2012-12-10
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Most countries face high and increasing rates of Cardiovascular Disease. Hypertension is a major risk factor for many ardiovascular diseases the number one killer disease in Ghana today and a leading cause of deaths in adults. The objective of this study is to identify differences in the measurements recorded between the group that develop hypertension and the group that does not develop it. Secondly, develop a predictive mathematical model to estimate whether a person will develop hypertension using Artificial Neural Networks (ANN) from the Data and Tracies Data Exploration Studio (Software). ANN Feed Forwarding with Back-propagation methodology was adopted to extract significant patterns from a dataset containing 1027 observations. Using, Age, Body Mass Index, Systolic and Diastolic Blood Pressures as input descriptors. The data used for this study was collected by Healthy life Education, in the Kumasi metropolis, Ashanti Region of Ghana, West Africa. The findings of this study revealed that, the model built from ANN during a designed experiment, using different iterations gave the overall best model accuracy of 97.5% with a specificity of 98.2% and a sensitivity of 92.9%, showing that 16 people will be hypertensive. In conclusion, the study showed that the age group now for hypertensive state was below 50 years being overweight and pre-hypertensive. It was also predicted that 1.85% people will be hypertensive; hence increasing the state of hypertension. This means Artificial Neural networks techniques can be used efficiently to predict hypertension cases, therefore the outcome of this study can be used as an assistant tool by cardiologists to help them to make more consistent diagnosis for hypertension.
Description
A thesis submitted to the Department of Mathematics, Kwame Nkrumah University of Science and Technology in partial fulfilment of the requirements for the degree of MASTER OF PHILOSOPHY Faculty of physical Science College of Science. December 2012
Keywords
Citation