AN EXPERT SYSTEM FOR DIAGNOSING GAIT IN CEREBRAL PALSY PATIENTS

Many first generation expert systems in medicine assumed that a single fault was the cause of the patient''s problems. However, this is not always so and in the domain of gait analysis this is usually not the case. This work looks at an expert system for performing gait analysis on cerebral palsy patients. The system is able to handle cases where there are many interacting faults causing the patient''s gait deviations. A qualitative-torque model of gait is used as a framework for analyzing the patient''s disorders. This model allows the effects of individual faults to be examined and also explicitly captures the interaction between various faults. The incorporation of this model into an expert system leads to a more complete and robust expert system for gait analysis than any existing previously. The qualitative-torque model of gait models gait by using torque equations represented as incremental qualitative (IQ) algebra equations. Because everything is not known about the torques, the system has to make assumptions to arrive at some values. These assumptions are kept consistent by using an assumption-based truth maintenance system. The only part of the model specific to cerebral palsy is the set of heuristics used to score competing assumptions. Thus, the gait model developed for this system could easily be adapted for use in an expert system diagnosing gait disorders of any given etiology.