Artificial Intelligence-Based Medical Diagnostic Expert System For Malaria And The Related Ailments

This work explored the potentials of information technology to develop a Medical Diagnostic Expert System (AUTOMEDES), which can be used, by qualified medical personnel or even unqualified medical personnel in the absence of qualified ones, without endangering human life. The researchers adopted the Structured System Analysis and Design Method (SSADM) and the Expert System Methodology to develop the system. The programming language used in developing the system is TURBO PROLG. The system developed was found to be able to diagnose malaria parasites in patients and recommend treatment procedures for such patients. The level of efficiency and precision of the system developed in diagnosing and treating malaria parasites compared favourably with the diagnosis and prescriptions of human experts in the field of study. Furthermore, in the class of diagnostic expert systems, literature has shown that no expert system on the disease malaria has been developed. Therefore, when the system is launched in the World Wide Web, it would provide facilities for linkages to various other Internet based medical expert systems in related areas. The linkages are necessary as they offer one who is shopping for the appropriate medical expert system, the opportunity of getting the desired help. DEVELOPMENT OF AN AUTOMATED MEDICAL DIAGNOSTIC EXPERT SYSTEM FOR MALARIA AND THE RELATED AILMENTS 1.0 INTRODUCTION It is an indisputable fact that health care delivery services in Nigeria are far from being adequate when compared with the minimum standards set by the World Health Organization (WHO). A number of factors are responsible for the inadequacies. Some of them include; insufficient qualified medical personnel, hospitals, particularly in less urbanized localities without social amenities, equipments and the like. Indeed, some of these factors listed have contributed to avoidable deaths, particularly, the inadequacy of qualified medical personnel. This work explored the potentials of information technology to develop a medical Diagnostic Expert system, which can be used by qualified medical personnel or even unqualified medical person4 Artificial Intelligence-Based Medical Diagnostic Expert System for Malaria and the Related Ailments Sylvanus O. Anigbogu & Hyacinth C. Inyiama nel in the absence of qualified one, for the diagnosis and treatment of malaria parasites without endangering human lives. An Expert system is a subsystem of cognitive science, which is a branch of Artificial intelligence. Meanwhile, Artificial intelligence (AI) is science and technology whose goal is to develop computers that can think, see, perceive, hear, talk and feel etc (1). In other words, artificial intelligence involves developing in a machine (computer system), the functions that are normally associated with human intelligence, which include; reasoning, inference, hearing and problem solving etc (2). Indeed, human intelligence attributes include; ability to plan, create model, search, reason with certain/uncertain data, solve problems when no algorithms exist and learn from environment to fulfill the intended goals (3). 2.0 LITERATURE REVIEW The Expert System technology has evolved over the years. Expert system first emerged from the research laboratories of a few leading United States universities during the 1960’s and 1970’s. They were developed as specialized problem solvers, which emphasized the use of knowledge rather than algorithms and general search methods (4). The first Expert system to be developed was DENDRAL, at the Stanford University. And the first medical expert system to emerge was the MYCIN, which diagnoses infectious blood diseases and determines a recommended list of therapies for the patient. Other medical diagnostic expert systems that have emerged include PUFF for diagnosis of lung disease, BLUE Box for depression etc (5). Malaria is a tropical disease. Indeed, the disease has assumed an endemic dimension within the tropics. Today, malaria and its related diseases are responsible for the avoidable death of a good percentage of Nigerians in particular and other people from the region of the tropics in general (6) In fact, the Nigerian government and other governments in the infested regions of the world had set out programs in order to fight the scourge, which include; the Roll Back Malaria (RMB); Multilateral Initiative on Malaria (MIM), Drive Against Malaria (DAM) etc (7). Other recent work in Artificial intelligent-based related systems include: ASUDA and EGRESS. ASUDA is a program, which has the potentials to examine software to see if it is capable of handling the tasks one intended to perform. Similarly, EGROSS is also a program or an expert system, which can study human reactions to accidents (8). Literature also reported a work on the Soa architecture, which is an expert system, intended to provide both the basis for building integrated intelligent systems and a unified theory of human cognition. Furthermore, literature reported of several Diagnostic medical expert systems for different diseases, without one for malaria and the related ailments. Indeed, it was this gap that the researchers intended to close by developing a medical expert system called Automated medical Diagnostic Expert system for Diagnosis and treatment of malaria (AUTOMEDES) 3.0 MATERIALS AND METHODS THE research method adopted is the Structured System Analysis and Design (SSADM), which is an accepted software Engineering principle for designing software. Also adopted is the Expert system methodol5 Artificial Intelligence-Based Medical Diagnostic Expert System for Malaria and the Related Ailments Sylvanus O. Anigbogu & Hyacinth C. Inyiama ogy, which involved knowledge engineering process of Inference and Knowledge base. Twelve Domain experts were randomly selected and two medical laboratory scientists based on consent from six medical clinics/hospitals within Anambra state of Nigeria. The choice of these medical personnel for the work was not necessarily by any known statistical sampling method. Standardized statistical methods could not be used because of the unwillingness of some of the domain experts to freely take part in the work due to high level of ignorance on the part of our medical personnel on the potentials of computer technology in enhancing medical practice. Therefore, the domain experts used in the study were those who gave their consent to participate in the research work. The knowledge Engineering process involved building a knowledge base and an inference engine for the expert system. The relevant facts or information were elicited from the domain experts through interviews, observations, study of relevant documents and structured questionnaires. The instrument entitled: Questionnaires for Eliciting of Information on Diagnosis and Treatment of Malaria (QEIDTM) was developed and subjected to Known standard reliability and validity tests before being used for the work. . 4.0 WORKING PRINCIPLE /DESIGN The Components of the system include: • Knowledge base • Database • Inference engine • User interface The knowledge base: This contains the domain knowledge of malaria acquired from the domain experts. It is designed based on rules, which combine antecedents (symptoms) in order to arrive at conclusion (disease) as follows: If patient has body/joint pains And has weakness And has fever And has evidence of head ache Then patient has malaria Then treatment is choroquine or quinine etc The Database This contains all relevant data on the current situation for diagnosis. It is a dynamic database, which holds data about the patient whose medical problem is being diagnosed. Again, each patient complaint is designated as Sympt for symptom, and they are numbered serially. The content of the Database is designated as follows: DOMAINS Passwords = string File = patfile DATABASE Sympt.1(symbol) Sympt 2(symbol) etc The inference engine This is the reasoning components of the system developed. It is the software, which processes the facts and rules stored in the knowledge base of the expert system in order to derive a conclusion. For instance, in combing the antecedents (arguments) stored in the knowledge base, the inference engine is able to arrive at conclusion (goals) for problem solving. The Backward-chaining rule based technique was used to design the inference engine as presented hereunder