| Download | - View final version: Effects of electrode position targeting in noninvasive electromyography technologies for finger and hand movement prediction (PDF, 2.1 MiB)
|
|---|
| DOI | Resolve DOI: https://doi.org/10.1007/s40846-023-00823-x |
|---|
| Author | Search for: Wang, Michelle1; Search for: Khundrakpam, Budhachandra1; Search for: Vaughan, Thomas1ORCID identifier: https://orcid.org/0009-0004-3214-3847 |
|---|
| Affiliation | - National Research Council of Canada. Medical Devices
|
|---|
| Funder | Search for: NSERC-CREATE; Search for: Fonds de recherche du Québec – Nature et technologies; Search for: National Research Council Canada |
|---|
| Format | Text, Article |
|---|
| Subject | electromyography; machine learning; armband; stroke rehabilitation |
|---|
| Abstract | Purpose: Stroke patients may need to undergo rehabilitation therapy to improve their mobility. Electromyography (EMG) can be used to improve the efectiveness of at-home therapy programs, as it can assess recovery progress in the absence of a health professional. In particular, EMG armbands have the advantage of being easy to use compared to other EMG technologies, which could allow patients to complete therapy programs without external assistance. However, it is unclear whether there are drawbacks associated with the fxed electrode placement imposed by current armband designs. This study compared the hand gesture prediction capabilities of an of-the-shelf EMG armband with fxed electrode placement and an EMG setup with fexible electrode positioning.
Methods: Ten able-bodied participants performed a series of hand and fnger gestures with their dominant hand, once with an EMG armband (Untargeted condition) and once with electrodes deliberately placed on specifc muscles (Targeted condition). EMG features were extracted from overlapping sliding windows and were used to (1) classify the gestures and (2) predict fnger joint positions as measured by a robotic hand exoskeleton.
Results: For the classifcation task, a logistic regression model performed signifcantly better ( p < 0.001) for the Targeted condition (55.8% ± 10.1%) compared to the Untargeted condition (47.9% ± 11.6%). For the regression task, a k-nearest neighbours model obtained signifcantly lower ( p = 0.007) mean RMSE values for the Targeted condition (0.260 ± 0.037) compared to the Untargeted condition (0.270 ± 0.043).
Conclusion: We observed a trade-of between predictive accuracy and ease-of-use of the EMG devices used in this study. It is important to consider such a trade-of when developing clinical applications such as at-home stroke rehabilitation therapy programs. |
|---|
| Publication date | 2023-09-26 |
|---|
| Publisher | Springer |
|---|
| Licence | |
|---|
| In | |
|---|
| Language | English |
|---|
| Peer reviewed | Yes |
|---|
| Export citation | Export as RIS |
|---|
| Report a correction | Report a correction (opens in a new tab) |
|---|
| Record identifier | 4655854e-b1fc-4d22-9d3a-703750de4cec |
|---|
| Record created | 2024-04-19 |
|---|
| Record modified | 2024-04-19 |
|---|
