Dynamic foot morphology explained through 4D scanning and shape modeling.

A detailed understanding of foot morphology can enable the design of more comfortable and better fitting footwear. However, foot morphology varies widely within the population, and changes dynamically as the foot is loaded during stance. This study presents a parametric statistical shape model from 4D foot scans to capture both the inter- and intra-individual variability in foot morphology. Thirty subjects walked on a treadmill while 4D scans of their right foot were taken at 90 frames-per second during stance phase. Each subject's height, weight, foot length, foot width, arch length, and sex were also recorded. The 4D scans were all registered to a common high-quality foot scan, and a principal component analysis was done on all processed 4D scans. Elastic-net linear regression models were built to predict the principal component scores, which were then inverse transformed into 4D scans. The best performing model was selected with leave-one-out cross-validation. The chosen model predicts foot morphology across stance phase with a root-mean-square error of 5.2 ± 2.0 mm and a mean Hausdorff distance of 25.5 ± 13.4 mm. This study shows that statistical shape modeling can be used to predict dynamic changes in foot morphology across the population. The model can be used to investigate and improve foot-footwear interaction, allowing for better fitting and more comfortable footwear.