Fuzzy expert system for the intelligent recognition of cerebral palsy

Abstract

This study describes a fuzzy system for intelligent recognition and estimation of possibility of suffering from cerebral palsy (CP) in children between the ages of 3 months and 2 years. The hallmark symptoms of CP are disturbances of movement and/or posture which are manifested as failure to meet appropriate motor developmental milestones such as head control, rolling, sitting, crawling, standing, walking and running. Fuzzy expert systems comprising fuzzification interface, fuzzy assessment methodology and defuzzification interface for CP were used. Motor developmental milestones in children were the input variables with time of acquisition (early, average, late) as fuzzy set while the possibility of developing CP was the output variable. Triangular function and the diagnostic framework for CP were implemented using MATLAB 7.6.0 in order to arrive at a definite conclusion from ambiguous fuzzy sets and numbers. Developmental milestones diagnostic data for children with CP were evaluated and the computed results were in the range of the predefine limits of probability of suffering from cerebral palsy by the domain expert. The result indicates that fuzzy system is very effective in improving the accuracy for diagnosis of cerebral palsy through estimation of occurrence of CP as very low, low or high.

Keywords: fuzzy system, cerebral palsy, motor milestones, medical diagnosis