Intelligent neuro fuzzy expert system for autism recognition

Abstract

Most children are not diagnosed with autism until they are around preschool age; the first signs of autism generally appear between 12 and 18 months of age. Autism is a brain disorder that is associated with a wide range of developmental problems, especially in communication, social
interaction and unusual repetitive behavior. However, it is believed that at least some cases involve an inherited or acquired genetic defect. Researchers have proposed that the immune-system, metabolic and environmental factors may play important part as well. A number of other possible causes have been suspected, but not proven. They involve, diet, digestive tract changes, mercury poisoning, the body's inability to properly use vitamins and minerals as well as vaccine sensitivity. The symptoms of autism includes avoiding eye contact, playing alone, not smiling, not responding to names, echolia (only parroting), unusual language, not talking, repetitive movement, self mutilation and reduced sensitivity to pain. Neuro-Fuzzy Logic explores approximation techniques from neural networks to find the parameter of a fuzzy system. In this paper, the traditional procedure of the medical diagnosis of autism employed by physician is analyzed using neuro-fuzzy inference procedure. The proposed system which is self-learning and adaptive is able to handle the
uncertainties often associated with the diagnosis and analysis of autism.

Keywords: Neural network, fuzzy logic, Neuro fuzzy system, autism