Development of Decision Support System for the Diagnosis of Arthritis Pain for Rheumatic Fever Patients: Based on the Fuzzy Approach. Issue 3 (September 2015)

• Record Type: Journal Article
• Authors: Pandey, Sanjib Raj; Ma, Jixin; Lai, Choi-Hong
• Is Part Of: Journal of algorithms & computational technology. Volume 9:Issue 3(2015)
• Page Start: 265
• Page End: 290
• Publication Date: 2015-09
• Subjects: Rheumatic Fever -- Arthritis Pain -- Fuzzy Logic -- Fuzzy Rules -- Knowledge-Based -- Fuzzy Inference -- Defuziffication; Computer algorithms -- Periodicals; Numerical calculations -- Periodicals; Computer algorithms; Numerical calculations; Periodicals; 518.1
• DOI: 10.1260/1748-3018.9.3.265
• Languages: English
• ISSNs: 1748-3018

Abstract

Developing a Decision Support System (DSS) for Rheumatic Fever (RF) is complex due to the levels of vagueness, complexity and uncertainty management involved, especially when the same arthritis symptoms can indicate multiple diseases. It is this inability to describe observed symptoms precisely that necessitates our approach to developing a Decision Support System (DSS) for diagnosing arthritis pain for RF patients using fuzzy logic. In this paper we describe how fuzzy logic could be applied to the development of a DSS application that could be used for diagnosing arthritis pain (arthritis pain for rheumatic fever patients only) in four different stages, namely: Fairly Mild, Mild, Moderate and Severe. Our approach employs a knowledge-base that was built using WHO guidelines for diagnosing RF, specialist guidelines from Nepal and a Matlab fuzzy tool box as components to the system development. Mixed membership functions (Triangular and Trapezoidal) are applied for fuzzification and Mamdani-type is used for the fuzzy reasoning process. Input and output parameters are defined based on the fuzzy set rules.