MFC histogram and Poincare plot images for automated gait detection

In this paper, we apply support vector machines (SVMs) for the automatic recognition of young-old gait from their respective gait-patterns. Minimum foot clearance (MFC) data of 30 young and 28 elderly participants were analysed using a PEAK-2D motion analysis system. Gait features extracted from individual MFC histogram-plot and Poincare-plots were used to develop gait classification models using SVMs. Test results indicate that the generalization performance of the SVMs, was on average 83.3% (/spl plusmn/2.9) to differentiate young and elderly gait patterns. Forward feature selection algorithm demonstrated that only 3-5 gait features could differentiate the gait patterns with 90% accuracy. Performance of the gait classifier was evaluated using areas under the receiver operating characteristic plots. Improved performance of the classifier was evident when trained with reduced number of selected good features. These results suggest that SVMs are an efficient gait classifier for recognition of movement pattern changes due to ageing, and has potential for wider applications in gait diagnostics.