—The elbow method was used to visually determine the optimal number of clusters present in the experimental data. In this method, a number of clusters is selected to which adding an extra one does not explain much more of the variance. Based on this figure, the use of 3 clusters was considered to be optimal. This also made most sense when visually assessing the data as it resulted in gait patterns with clearly distinct gait patterns, with a quite even distribution of legs between cluster (9–14 legs per cluster).

Citation: Journal of Applied Biomechanics 39, 5; 10.1123/jab.2023-0022