The Development Of Obesity Diagnosis Expert System Using Bayesian Network Model

Low, Han Wai (2011) The Development Of Obesity Diagnosis Expert System Using Bayesian Network Model. Project Report. UTeM, Melaka, Malaysia. (Submitted)

	PDF (24 pages) The_Development_Of_Obesity_Diagnosis_Expert_System__Using_Bayersian_Ntework_Model_-_24_pages.pdf - Submitted Version Download (523kB)
	PDF (Full Text) The_Development_Of_Obesity_Diagnosis_Expert_System__Using_Bayersian_Ntework_Model_-_Full_Text.pdf - Submitted Version Restricted to Registered users only Download (8MB)

Abstract

Health Minister Datuk Seri Liow Tiong Lai said obesity is becoming the major health issue with the number of obese people almost tripling in the past 15 years from four per cent in 1996 to 14 per cent in 2010. Besides that, current web application expert system exists now just to determine the obesity based on the BMI calculation. BMI can be misleading in terms of a person's body fat as it solely depends on the net weight and height of a person. The purpose of this project is to develop a web-based medical expert system that incorporates Bayesian Network Model and works in the fields of obesity disease. Bayesian Network formalism offers a natural way to represent the uncertainties involved in medicine when dealing with diagnosis, treatment selection, planning, and prediction. The system is called Obesity Diagnosis Expert System (ODES). This system is to let user know and do the prediction based on their current lifestyle that will causes them obesity or not in the future. The key parameter involve in prediction are age range, eating habit, exercise rate, family genetic problem and sleeping lifestyle. The system will do the prediction whether they will get obesity or not in the future based on their answer given to system. ODES are constructing based on System Development Life Cycle (SDLC). ODES is undergoes four stages of testing which is Unit Testing, Integration Testing, System Testing and User Acceptable Testing which are under white-box and black box testing strategy. The strength of the system is not only consists obesity prediction function that uses the AI technique which is Bayesian Network Model in evaluating the obesity risk but also provides the suggestion based on their answer of the question given to the system. Besides that, the system also provides a lot of obesity information to user for their references. With this system, the system users can easily identify about their obesity risk and gain the information related to health and obesity. The weakness of ODES are only take into consider five common causes of obesity as key parameter prediction but other causes are not included as node in the system. Besides that, ODES only take into consider only one expert knowledge. Probability gets that use in prediction in the system may give an unexpected result in some condition. ODES stills can be expandable. A more accuracy prediction system can be developed where combination two or more expert knowledge. Although Bayesian Network Model is a good method in medical prediction field, there are still a lot of methods that available which can do medical prediction and will get more accurate prediction result. © Unlveraltl

Actions (login required)

View Item