Abstract
Gait Analysis (GA) is often defined as the study of human walking; typically involving computerised and instrumented measurement of the movement patterns that make up walking. GA can reveal the timing and pattern of activation of muscles and joints, of body segment motions, and the forces that act on them. It can facilitate objective comparison of pathological gait with normal gait, and monitoring of progress in rehabilitation. However, although raw results can be printed in minutes, the clinical team may spend hours in interpreting the data. The success of this approach is limited mainly by the ability of clinicians to handle large sets of data, their expertise with respect to the biomechanics of gait, and their individual experience with the characteristics of a particular population. In addition, it is recognised that the interpretation of data varies from clinician to clinician and institution to institution which may have an impact on clinical decision-making. Also, the techniques used in the interpretation of gait data often do not provide information about possible causes for gait abnormalities. Improving the efficiency of patient testing will greatly enhance the productivity of gait laboratories and improve patient care. For this reason, the focus in this project is in developing a technique for the analysis of gait data to aid clinical interpretation. A software package, also called expert system, is developed based on automating the Rancho Observational Gait Analysis (OGA) approach used to denote gait deviations. Causes related to deviations are listed and the result of additional tests that may help prove or refute any cause is also included. A report is then generated that includes all the above. The software is tested with data from a group of Cerebral Palsy patients to check its efficiency. Results showed that the expert system was capable in denoting deviations and overcoming a number of major challenges in gait data interpretation. However, many limitations still present such as the need to test it on other pathologies and consider more parameters, e. g. kinetics and Electromyography (EMG) data.