Kristoffersen, Morten Bak
Franzke, Andreas
Murgia, Alessio
van der Sluis, Corry
Bongers, Raoul
2017-08-15
<p>Please note there was a problem with the upload. The file is a pdf, but the file ending is missing. Manually opening the file in a pdf reader should work. Otherwise feel free to contact me. </p>
<p>Abstract</p>
<p>Individuals with upper-limb deficiency who are fitted<br>
with a prosthesis are normally trained in the use of such<br>
device. This is even true for individuals who are fitted with a<br>
myoelectric prosthesis that uses control algorithms based on<br>
pattern-recognition, despite the intent of pattern-recognition<br>
control of exploiting “intuitive” phantom movements.<br>
Conventionally, training individuals for pattern-recognition<br>
control usually involves an expert who guides the user to<br>
produce electromyogram (EMG) signals that optimize<br>
pattern recognition. In the training the individual is<br>
stimulated to adapt their EMG signals as to make them more<br>
distinct in terms of the resulting patterns. To achieve this, for<br>
instance, small movements can be added to the basic pattern,<br>
such as flexing the little finger during open hand. Although<br>
training improves online accuracy it still involves<br>
considerable trial and error. Moreover, expert guidance is<br>
currently done based on visual perusal of EMG patterns or<br>
features thereof and not based on specific metrics<br>
characterizing those EMG signal patterns. Rather than using<br>
intuitive phantom movements for control, we instead propose<br>
to use those phantom movements which are most distinct in<br>
terms of EMG. To find the set of phantom movements that<br>
provides the most distinct EMG activation patterns, we<br>
propose to use a serious game. Using a game, we can train<br>
individuals to make EMG patterns distinct while performing<br>
them in a robust manner. This game is controlled using the<br>
EMG captured from 8 electrodes positioned around the<br>
forearm. Inspired by the work of Radhakrishnan et. al and<br>
Pistohl et. al, the EMG from each electrode is mapped to a<br>
direction of the game avatar in the 2D environment. We<br>
hypothesize that this training will make individuals utilize<br>
their EMG activation space to a greater extent and become<br>
better at generating only EMG activity at specific electrode<br>
sites so that patterns are more distinct.<br>
We are currently conducting an experiment in which 4<br>
experimental groups receive different kinds of training.<br>
Group 1 receives conventional training without coaching.<br>
Group 2 receives conventional training with feedback. Group<br>
3 receives training with the proposed serious game and group<br>
4 receives training without any feedback (control). The<br>
learning effects between groups are analysed using the<br>
metrics proposed by Bunderson et al. and the motion test.</p>
https://doi.org/10.5281/zenodo.1217832
oai:zenodo.org:1217832
Zenodo
https://zenodo.org/communities/eu
https://doi.org/10.5281/zenodo.1217828
info:eu-repo/semantics/openAccess
Creative Commons Attribution 4.0 International
https://creativecommons.org/licenses/by/4.0/legalcode
MEC, Myoelectric Controls Symposium, Fredericton, New Brunswick, Canada, 15-08-2017 - 18-08-2017
Serious games
User training
Motor learning
Prosthesis
Myocontrol
Pattern-recognition
Feedback matters – or doesn't it? User training for pattern-recognition controlled prostheses (Abstract title: User training for pattern-recognition based myoelectric prostheses using a serious game)
info:eu-repo/semantics/conferencePoster