DEVELOPMENT OF A CONNECTIONIST EXPERT-SYSTEM TO IDENTIFY FOOT PROBLEMS BASED ON UNDER-FOOT PRESSURE PATTERNS

The first phase of developing a connectionist expert system is described in th is study. Dynamic pressure patterns below the foot obtained by the EMED F system of 18 subjects were classified into one of the three categories: normals, pes cavus and hallux valgus. As opposed to the traditional approach for the analysis of the pressure patterns by quantifying isolated segments, a neural network based expert system was trained to perform objective classification of the patterns while keeping their integrity. The results of the backpropagation paradigm were unexpectedly good considering the limiting factors which hindered the performance of the neural networks. The successes indicate the potential of artificial intelligence in automated analysis of foot pathologies based on the EMED F system. Relevance-This paper describes the first steps towards a novel automated screening system, which classifies the patient with foot abnormalities into the pathological category he or she belongs to, based on the pressure pattern below the foot. The decision making is objective, and shows a practical example of the use of artificial intelligence.