University of Twente Student Theses
Machine Learning-based Estimation of 3D Kinetics during Normal Walking and Single-leg Hop Tests in Patients with Anterior Cruciate Ligament Injury
Bondt, M.R. de (2023) Machine Learning-based Estimation of 3D Kinetics during Normal Walking and Single-leg Hop Tests in Patients with Anterior Cruciate Ligament Injury.
|
PDF
3MB |
| Abstract: | The analysis of quantitative data, particularly kinetics, can offer valuable insights for the monitoring of the rehabilitation process in cases of anterior cruciate ligament (ACL) injuries. However, the conventional methods of collecting kinetic data, using a force plate and an optical motion capture system, are time-consuming and require an expensive laboratory setup. This study aims to validate existing machine learning (ML) algorithms based on inertial measurement units (IMUs) from literature and proposes three novel ML algorithms for estimating three-dimensional ground reaction forces (GRFs) and net knee joint moments during walking and single-leg hop tests. Nine healthy participants and eight ACL patients were assessed during walking and single-hop tests while wearing thirty reflective markers and eight IMUs. Two algorithms from literature were validated against the collected dataset and three novel movement-specific artificial neural networks were designed. The most important finding of the present study is that medial-lateral GRFs and all net knee moments are still difficult to accurately estimate during a single leg hop test. However, it does show that ACL patients are a suitable population to develop and evaluate ML algorithms on. The current study serves as a first step in providing quantitative data for monitoring the ACL rehabilitation process. |
| Item Type: | Essay (Master) |
| Faculty: | TNW: Science and Technology |
| Subject: | 44 medicine |
| Programme: | Biomedical Engineering MSc (66226) |
| Link to this item: | https://purl.utwente.nl/essays/96820 |
| Export this item as: | BibTeX EndNote HTML Citation Reference Manager |
Show download statistics for this publication
Show download statistics for this publicationRepository Staff Only: item control page