Abstract
Background: Multiple sclerosis (MS) is an autoimmune inflammatory disease of the central nervous system (CNS) that affects about 2.9 million people worldwide. Current disease-modifying therapies are focused on delaying the disease progression, treating sensitive attacks, and improving symptoms. However, some patients partially respond or do not respond to MS treatments. So, it is important to determine the degree of response of patients to treatments.
Methods: Expert systems are computer programs that attempt to emulate the reasoning process of some skills of a human expert. A fuzzy expert system incorporates fuzzy logic into its reasoning process to manage uncertain and imprecise information that a binary system could not. So, a based-rule fuzzy expert system based on the Takagi–Sugeno–Kang Fuzzy System (TSKFS) model is proposed to estimate the degree of response to different types of drugs such as Gelenia, Tysabri, Avonex, Betaferon, and Rebif in 60 MS patients, using clinical patient information derived from neurological examinations as input variables.
Results: The results of the proposed fuzzy expert system to estimate the response to MS treatments in MS patients show a high efficiency (100%) compared with conventional classification methods such as the K-Means clustering model (62%).
Conclusion: Expert systems are efficient tools for classifying the response to MS treatments and can support the decision of specialists to prescribe the most appropriate therapy for the individual patient.