This project contains links to experimental data and a manuscript on the comparison of methods for estimating the time profile of metabolic cost within the gait cycle.

Respiratory oxygen consumption measurements allow recording of the average metabolic cost of walking, but the slow rate of these measurements prevents assessing which part of the gait cycle has the highest metabolic cost. Simulation methods allow estimation of the time profile of metabolic cost within the gait cycle. We compared estimations of the metabolic cost of walking using a method that is based on kinematic and electromyography recordings from participants in conjunction with muscle metabolic rate equations and a method based on joint kinematics and kinetics. While both methods are able to accurately estimate large changes in the average metabolic cost from walking on different inclinations, the time profiles estimated by the two methods are different, indicating that estimations of the time profile of metabolic cost are dependent on the estimation method. Both estimation methods matched well with the results from experimental perturbation studies that suggest that the metabolic cost of the swing phase is approximately 10% to 20% of the metabolic cost of the entire gait cycle.

As a follow-up to this project we aim to develop slightly improved estimations of metabolic time profiles by adjusting the estimations to fit rich experimental datasets. We are using a robotic tether that can assist specific parts of the gait cycle with high repeatability and generate very large changes in metabolic cost. Advances in estimations of the time profile of metabolic cost could lead to practical applications such as designing rehabilitation robots that assist specifically during the phase with the highest metabolic cost.

Project manuscript:
Mohammadzadeh Gonabadi, A., Antonellis, P., & Malcolm, P. (2020). Differences between joint-space and musculoskeletal estimations of metabolic rate time profiles. PLOS Computational Biology, 16(10), e1008280.

Supporting MATLAB data:

The 3D musculoskeletal model that was used (Rajagopal2015):

The manuscript describing experimental dataset:
Antonellis, P., Frederick, C. M., Gonabadi, A. M., & Malcolm, P. (2020). Modular footwear that partially offsets downhill or uphill grades minimizes the metabolic cost of human walking. Royal Society open science, 7(2), 191527.

The manuscript describing the robotic system for generating rich perturbation data:
Gonabadi, A. M., Antonellis, P., & Malcolm, P. (2020). A system for simple robotic walking assistance with linear impulses at the center of mass. IEEE Transactions on Neural Systems and Rehabilitation Engineering.