International Journal of Computer Applications (0975 – 8887) Volume 180 – No.38, May 2018 45 Diagnosis of Hepatitis using Adaptive Neuro-Fuzzy Inference System (ANFIS) Rahmon Ibrahim Babcock University Dept. of Computer Science Ilisan Remo, Ogun State, Nigeria Omotosho Olawale Babcock University Dept. of Computer Science Ilisan Remo, Ogun State, Nigeria Kasali Funmilayo Babcock University Dept. of Computer Science Ilisan Remo, Ogun State, Nigeria ABSTRACT Hepatitis B is one of the liver diseases that is difficult to discover at an early stage of its attack and prominent public health problem. As at 2017, medical statistic recorded that over 23 million of Nigerians were living with Hepatitis B. Several decision support systems used in diagnosing liver diseases derived their efficiencies from artificial intelligence techniques in tackling the challenges facing physician in respect to complexity of the numerous variables involved in liver diseases diagnosis. In this paper, Adaptive Neuro-Fuzzy Inference System (ANFIS) was employed to invoke neural network that provided structures for fuzzy inference engine (FIE) in order to learn information about the normalized dataset on hepatitis B. The neural network (NN) triggers backpropagation and least square methods for tuning the membership functions at the fuzzification stage while the center of area (COA) was used as defuzzification method to compute the weighted average of the fuzzy set and intensity level of the disease for each record. The system was implemented with technical computing language, MATHLAB, on a dataset that consists of 155 instances and 20 attributes of which only the most five liver function tests (LFTs) attributes were selected as input parameters and the corresponding linguistic values and intensity levels were generated as output in order to identify the severity level of the infection. After the system was evaluated, the performance metric gave accuracy of 90.2%. General Terms Neural Networks, Fuzzy Logic, Linguistic Values, Liver Function Test, Adaptive Neuro-Fuzzy Inference System Keywords Hepatitis B, Intensity Level, Decision Support System 1. INTRODUCTION The relevance of intelligence systems in medical diagnosis is tremendously increasing. Information taken from patients and decisions of experts are the most important factors in diagnosis of diseases (Neshat and Zadeh, 2013). The increase in scientific knowledge and computer technology is of great benefits to medical practitioners, in diagnosing life- threatening diseases and administering treatment for patients. The most commonly known liver disease, hepatitis B is a condition when a liver is inflamed as a result of viral infection. Among the possible causes are wrong intake of medications and drugs, toxins, and excessive alcohol. Hepatitis B could also be transmitted through infectious body fluids and sexual intercourse with infected person (Friedman, 2004). Artificial intelligence in medicine is essentially concerned with the development of artificial-based techniques programs that relate disease entities with patients‟ symptoms in form of a model in order to provide a basis for series of diagnosis and precise therapy recommendations for medical management. Numerous researches have shown that, the most successful applications in artificial intelligence (AI) are clinical decision support system used to diagnose patients with liver and kidney related diseases (Neshat and Yaghobi, 2009). Neuro-fuzzy is an integral aspect of artificial intelligence, which combines the strength of neural network and fuzzy logic together by utilizing the approximation method of neural-network to compute the parameters of a fuzzy system (Obi and Imianvan, 2011). Obviously, medical field has explored a quite number of techniques and methods provided by artificial intelligence for emergency care units and surgical operations Yardimci (2001). Several researches have been conducted on fuzzy-based expert system for diagnosis of diseases while neural network models have also been explored by experts for prediction and classification of liver diseases diseases. The task of disease diagnosis and management is complex because of the numerous variables involved. The traditional method of diagnosing liver diseases is characterized with a lot of subjective decision-making, and logical thinking of medical practitioners, which lead to inappropriate use of inefficient tool (Smita et al. 2012). One of the key problems encountered in the medical field during the course of delivering proper diagnosis of liver disease is the inability of the physicians to derive comprehensive information as a result of available imprecise medical data set. Therefore, in this paper, an Adaptive Neuro-Fuzzy Inference System (ANFIS) based model is being used to develop architecture model as problem solving for classifying and determining the intensity levels of hepatitis liver disease. The main objectives are to implement and evaluate the proposed architecture model with performance metrics. The general architecture of ANFIS as a classifier model is presented in Section 2. In Section 3, few related works regarding the use of Neural Network, Fuzzy Logic and Neuro- Fuzzy model based architecture in diagnosing liver diseases are reviewed. while the methodology adopted for the proposed architecture model is systematically presented in Section 4. In section 5, the metrics used to evaluate the performance of the proposed model are presented while the implementation and discussion of results are drawn in section 6. 2. OVERVIEW OF ANFIS ARCHITECTURE MODEL ANFIS adopt a neural network technique that can adjust the membership functions parameters and linguistic rules directly from data in order to enhance the system performance. The ANFIS architecture contains a five-layer feedforward neural