(21) A 3D FORCE MEASURING INSOLE FOR ATHLETE TRAINING AND READINESS MONITORING

Saturday, July 19, 2025

10:00 AM - 11:30 AM CT

Location: Hall D

Benjamin O. Abell (he/him/his)

Graduate Researcher
University of Kansas
Lawrence, Kansas, United States
EK

Ember Krech

Lead Bioengineer
Axioforce Inc.
Lawrence, Kansas, United States
SS

Seth Siemens

Graduate Research Assistant
University of Kansas
Lawrence, Kansas, United States
JS

Jacob Sosnoff

Associate Dean for Research for the School of Health Professions
University of Kansas Medical Center
Kansas City, Kansas, United States
SH

Stephen Houston

Lead Research and Development Scientist
Axioforce Inc.
Lawrence, Kansas, United States
Andrew C. Fry, PhD, CSCS*D, FNSCA*E

Professor & Director
University of Kansas
Lawrence, Kansas, United States
LF

Lance Frazer

Senior Research Engineer
Southwest Research Institute
San Antonio, Texas, United States
TT

Ty Templin

Senior Research Engineer
Southwest Research Institute
San Antonio, Texas, United States
TE

Travis Eliason

Senior Research Engineer
Southwest Research Institute
San Antonio, Texas, United States
Jonathan D. Miller, PhD

Assistant Research Professor
University of Kansas
Lawrence, Kansas, United States

Purpose: Although the demand for wearable technology for sports performance research and individual athlete monitoring is on the rise¹ the current options for achieving this are limited. GPS and accelerometry based wearables lack important 3D ground reaction force (GRF) based metrics, and force plates are limited to snapshots of data, rather than continuous monitoring during practice or game sessions. The purpose of this project is to validate a novel 3D force and motion sensing shoe insole, based on a smart material technology^2,3, for 3D ground reaction force measurement and gait feature recognition, to prove its potential for providing external load monitoring and athlete readiness monitoring in a single wearable package. Methods: For gait metric validation, 2 pairs of women’s size 6 and 1 pair of men’s size 12 were provided. A GAITRite gait map was used for validating the insole data. The insoles were inserted into standard athletic shoes and worn by the researchers for several trials that were varying in length and included some trials of irregular gait. In total, 27 trials of viable data were collected. For the 3D force validation, 4 women’s size 6 insoles and 4 men’s size 12 insoles were provided. A BERTEC force plate was used to validate the data. The insoles were worn by five researchers over 13 trials of squats, bouncing, and adduction/abduction. The right and left foot data was synced and calibrated through an optimization algorithm.Results: We applied a time-interpolation method to address inconsistencies in the insoles’ data frequency, reducing errors in stance time and cadence. By interpolating single-foot metrics (stance time, gait cycle time) and using raw data for dual-foot metrics (step time, single support time), the insoles achieved 90% accuracy in gait metric predictions. The mean percent error for each movement is as listed: Cadence: 0.3%, Step Time: 9.8%, Single Support Time: 1.0%,Gait Cycle Time: 6.7%, Stance Time: 2.2%. In the 3D GRF validation, for the movements involved, an overall average NRMSE was calculated as 4.34, 3.92, and 6.78 for F_x, F_y, and F_z, respectively. This indicates that the insoles were very accurate in the shear force data, and slightly less accurate for vertical forces. Conclusions: The insoles were successful in showing their ability to accurately predict 3D GRF and gait metrics. This data indicates a strong potential for future external load and athlete readiness monitoring from a single, convenient to use wearable shoe insole. The insoles were able to capture force data with an NRMSE of under 7, being especially accurate in shear force detection. They were also able to correctly time heel strikes and toe offs of walking trials to result in a 90% gait metric accuracy. Practical Applications: This combination of accurate force data and time data makes for a convenient, multi-use, wearable shoe insole to monitor the progress of rehabilitating athletes, track progression of force production, and collect various other types of important information. Acknowledgements: This work was supported by NIH grant R42EB034605-01, as well as the University of Kansas Medical Center, and the Midwest Biomedical Accelerator Consortium.