EMG-informed neuromusculoskeletal models accurately predict knee loading measured using instrumented implants

Simple item page
dc.contributor.authorBennett, K.J.
dc.contributor.authorPizzolato, C.
dc.contributor.authorMartelli, S.
dc.contributor.authorBahl, J.S.
dc.contributor.authorSivakumar, A.
dc.contributor.authorAtkins, G.J.
dc.contributor.authorSolomon, L.B.
dc.contributor.authorThewlis, D.
dc.date.issued2022
dc.description.abstractObjective: Using a musculoskeletal modelling framework, we aimed to (1) estimate knee joint loading using static optimization (SO); (2) explore different calibration functions in electromyogram (EMG)-informed models used in estimating knee load; and (3) determine, when using an EMG-informed stochastic method, if the measured joint loadings are solutions to the muscle redundancy problem when investigating only the uncertainty in muscle forces. Methods: Musculoskeletal models for three individuals with instrumented knee replacements were generated. Muscle forces were calculated using SO, EMG-informed, and EMGinformed stochastic methods. Measured knee joint loads from the prostheses were compared to the SO and EMGinformed solutions. Root mean square error (RMSE) in joint load estimation was calculated, and the muscle force ranges were compared. Results: The RMSE ranged between 192-674 N, 152-487 N, and 7-108 N for the SO, the calibrated EMG-informed solution, and the best fit stochastic result, respectively. The stochastic method produced solution spaces encompassing the measured joint loading up to 98% of stance. Conclusion: Uncertainty in muscle forces can account for total knee loading and it is recommended that, where possible, EMG measurements should be included to estimate knee joint loading. Significance: This work shows that the inclusion of EMG-informed modelling allows for better estimation of knee joint loading when compared to SO.
dc.description.statementofresponsibilityKieran J. Bennett, Claudio Pizzolato, Saulo Martelli, Jasvir S. Bahl, Arjun Sivakumar, Gerald J. Atkins, Lucian Bogdan Solomon, and Dominic Thewlis
dc.identifier.citationIEEE Transactions on Biomedical Engineering, 2022; 69(7):2268-2275
dc.identifier.doi10.1109/TBME.2022.3141067
dc.identifier.issn0018-9294
dc.identifier.issn1558-2531
dc.identifier.orcidBennett, K.J. [0000-0001-5411-0289]
dc.identifier.orcidBahl, J.S. [0000-0002-3267-0098]
dc.identifier.orcidSivakumar, A. [0000-0002-7446-3679]
dc.identifier.orcidAtkins, G.J. [0000-0002-3123-9861]
dc.identifier.orcidSolomon, L.B. [0000-0001-6254-2372]
dc.identifier.orcidThewlis, D. [0000-0001-6614-8663]
dc.identifier.urihttps://hdl.handle.net/2440/135800
dc.language.isoen
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)
dc.relation.granthttp://purl.org/au-research/grants/arc/DP180103146
dc.relation.granthttp://purl.org/au-research/grants/arc/FT180100338
dc.relation.granthttp://purl.org/au-research/grants/arc/IC190100020
dc.relation.granthttp://purl.org/au-research/grants/nhmrc/1126229
dc.rights© 2022 IEEE. Personal use is permitted, but republication/redistribution requires IEEE permission. See https://www.ieee.org/publications/rights/index.html for more information.
dc.source.urihttps://doi.org/10.1109/tbme.2022.3141067
dc.subjectBiomechanics; Biomechanical simulation; neuromusculoskeletal model; electromyography
dc.subject.meshBiomechanical Phenomena
dc.subject.meshElectromyography
dc.subject.meshGait
dc.subject.meshHumans
dc.subject.meshKnee Joint
dc.subject.meshModels, Biological
dc.subject.meshMuscle, Skeletal
dc.subject.meshProstheses and Implants
dc.subject.meshWalking
dc.titleEMG-informed neuromusculoskeletal models accurately predict knee loading measured using instrumented implants
dc.typeJournal article
pubs.publication-statusPublished

Files

Collections

Medicine publications