Early Detection of Arthritis using Hierarchical Fuzzy Expert System
Anurag Sharma · International Journal for Research in Applied Science and Engineering Technology · 2020
Purpose: This research aims to discover diagnostic tools using fuzzy logic and artificial intelligence for early detection of Arthritis that can be applied in practice.The complexity of medical practice makes traditional quantitative approaches of analysis inadequate.In medicine, the lack of information (patient information, medical history, physical examination and laboratory results), imprecision and contradictory nature are common facts.This makes diagnosis and treatment difficult for the medical practitioner.These procedures are rigorous, the linguistic approach and human reasoning nature of fuzzy logic makes it easier for medical experts.In addition to this, there is a huge time lag between a patient experiencing the symptoms and actually seeking a doctor's help and this can worsen the situation and can lead to long term malfunctioning.Fuzzy logic and Graphical user Interface focus on targeting and solving these issues.The core objective of the research is designing a system that is cheap and easily accessible by the masses.Design and Methodology: Since the research aims at finding feasible diagnostic tool for assisting physicians and orthopaedics, secondary data from books, articles and engineering research journals was used.All the basics of fuzzy logic and the tool MATLAB were deliberated thoroughly.The chapter and verse for the disease Arthritis was analysed carefully so that the input and output parameters for the fuzzy inference system can be chosen precisely.Membership functions were defined for the chosen parameters.A rule base was designed for implementation.The parameters were combined in a hierarchical workspace so as to reduce the complexity of the system and design a compact model.The fuzzy inference system was then linked to Graphical User Interface for a novice friendly operation.The results were calculated and compiled.The results were then compared with those indicated by an orthopaedic to determine the accuracy of the system.Findings: Applying Fuzzy logic and Artificial Intelligence in Medicine can prove of great potential as it is robust and has the ability to deal with imprecise, distorted and erroneous data. Fuzzy expert system helps in making decisions very accurately which in turn helps medical experts in making decisions timely to avoid long term and permanent damage or complications from Arthritis or any other disease for that matter. Overall, the proposed Artificial Intelligence System produced favorable response based on the expected outcome and experimentations. The entire research was carried out under the guidance and supervision of a renowned orthopedic surgeon, who found the system to be useful as it was able to produce quintessential results. Practical Implications: The proposed system can assist the diagnosis of Arthritis at an early stage and can even be modified to detect other diseases as well accordingly. It can be further enhanced by combining with image detection and segmentation to analyse the affected bones and joints. The system can also be used to characterize the subtypes and coexisting causes of Arthritis. Furthermore, Neuro-Fuzzy based portable thermo-graphic system can be created that combine the potential of both thermal imaging and fuzzy logic. Research Limitations/Implications: Sample size is a little under-powered because of lack of patient data but can be increased by seeking data from other orthopaedic practitioners. Lack of available data was an obstacle in finding a trend and a meaningful relationship between the possible causes and the outcomes of the disease.