In-silico techniques to inform and improve the personalized prescription of shoe insoles.

Abstract

Background: Corrective shoe insoles are prescribed for a range of foot deformities and are typically designed based on a subjective assessment limiting personalization and potentially leading to sub optimal treatment outcomes. The incorporation of in silico techniques in the design and customization of insoles may improve personalized correction and hence insole efficiency. Methods: We developed an in silico workflow for insole design and customization using a combination of measured motion capture, inverse musculoskeletal modelling as well as forward simulation approaches to predict the kinematic response to specific insole designs. The developed workflow was tested on twenty-seven participants containing a combination of healthy participants (7) and patients with flatfoot deformity (20). Results: Average error between measured and simulated kinematics were 4.7 ± 3.1, 4.5 ± 3.1, 2.3 ± 2.3, and 2.3 ± 2.7° for the chopart obliquity, chopart anterior-posterior axis, tarsometatarsal first ray, and tarsometatarsal fifth ray joints respectively. Discussion: The developed workflow offers distinct advantages to previous modeling workflows such as speed of use, use of more accessible data, use of only open-source software, and is highly automated. It provides a solid basis for future work on improving predictive accuracy by adapting the currently implemented insole model and incorporating additional data such as plantar pressure.