Machine Learning for Personalisation of Biomechanical Movement Simulations (C01)

Acronym: SFB 1483 EmpkinS C01
Project leader: Anne Koelewijn
Project members: Eva Dorschky, Markus Gambietz, Marlies Nitschke
Start date: 1. July 2021
End date: 30. June 2025
Funding source: DFG / Sonderforschungsbereich (SFB)

Abstract

The extent to which a neural network can be used to effectively personalise gait simulations using motion data is explored. We first investigate the influence of body parameters on gait simulation. An initial version of the personalisation is trained with simulated motion data, since ground truth data is known for this purpose. We then explore gradient-free methods to fit the network for experimental motion data. The resulting network is validated with magnetic resonance imaging, electromyography and intra-body variables.

Publications

Nitschke M., Marzilger R., Leyendecker S., Eskofier B., Koelewijn A.:
Change the direction: 3D optimal control simulation by directly tracking marker and ground reaction force data
In: PeerJ (2023)
ISSN: 2167-8359
DOI: 10.7717/peerj.14852
URL: https://peerj.com/articles/14852/
BibTeX: Download

Nitschke M., Marzilger R., Koelewijn A.:
3D full-body optimal control simulations with change of direction directly driven by motion capture data
17th International Symposium of 3-D Analysis of Human Movement (3D-AHM) (Tokyo, Japan, 16. July 2022 - 19. July 2022)
URL: https://www.youtube.com/watch?v=3ZFwDhZqZPU
BibTeX: Download

Gambietz M., Nitschke M., Miehling J., Koelewijn A.:
What should a metabolic energy model look like? Sensitivity of metabolic energy model parameters during gait
9th World Congress of Biomechanics 2022 Taipei (Taipei, 10. July 2022 - 14. July 2022)
BibTeX: Download