mailto:uumlib@uum.edu.my 24x7 Service; AnyTime; AnyWhere

Utilization of an artificial neural network in the prediction of heart disease

Awang, Mohd Khalid and Siraj, Fadzilah (2013) Utilization of an artificial neural network in the prediction of heart disease. International Journal of Bio-Science and Bio-Technology, 5 (4). pp. 159-166. ISSN 2233-7849

[thumbnail of IJBSBT 5 4 2013 159 166.pdf] PDF
Restricted to Registered users only

Download (300kB) | Request a copy

Abstract

This paper intents to assess the application of artificial neural network in predicting the presence of heart disease, mainly the angina in patients.The prediction and detection of angina are significant in determining the most appropriate form of treatment for these patients.The development of the application involves three main phases.The first phase is the development of Heart Disease Management Information System (HDMIS) for data collection and patient management.Then followed by the second phase, which is the development of Neural Network Simulator (NNS) using back propagation neural network for training and testing.The final phase is the development of Prediction System (PS) for prediction on new patient’s data.The best network model produced prediction accuracy of 88.89 percent.Apart from proving the ability of neural network technology in medical diagnosis, this study also shown how the neural network could be incorporated into the hospital information system as a prediction tool.As the pilot project, the application developed could be used as the starting point in building a medical decision support system, particularly in diagnosing the heart disease.

Item Type: Article
Uncontrolled Keywords: Prediction, Neural Network, Heart Disease
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
R Medicine > R Medicine (General)
Divisions: School of Computing
Depositing User: Prof Madya Fadzilah Siraj
Date Deposited: 17 Jan 2017 07:18
Last Modified: 17 Jan 2017 07:18
URI: https://repo.uum.edu.my/id/eprint/20622

Actions (login required)

View Item View Item