Adaptive Neuro Fuzzy Expert System for Diagnosing HIV

Abstract

Human Immunodeficiency Virus (HIV) is a retrovirus that causes Acquired Immune Deficiency Syndrome (AIDS) by infecting helper T cells or Lymphocyte of the immune system. HIV is transmitted primarily by exposure to contaminated body fluids, especially blood and semen. Other means of transmission of HIV include sharing contaminated sharp objects and blood transfusion. HIV symptoms can include a headache, chronic cough, diarrhea, swollen glands, lack of energy, and loss of appetite, weight loss, frequent fevers, frequent yeast infections, skin rashes, pelvic/abdominal cramps, sores on certain parts of your body and short-term memory loss. The focal point of this research is to describe and illustrate the application of fuzzy logic system to the diagnosis of HIV. It involves a sequence of methodological and analytical decision steps that enhance the quality and meaning of the logic produced. The system eliminates the uncertainties often associated with the analysis of HIV test data. To actualize the objective of this study, data collected from University Teaching Hospital, Ibadan, Nigeria were used as a set of parameters for diagnosis, and the software used for the development of necessary Graphical User interfaces (GUI) for fuzzy modeling using the fuzzy inference system editor, membership function editor, rule editor, rule editor, rule viewer and surface viewer was MATLAB, while MAMDANI editor GUI; a type of an Adaptive Neural Fuzzy Inference Systems was used for building and analyzing Mamdani