Published May 12, 2018 | Version v1
Presentation Open

Using a serious game to find distinct electromyogram patterns

  • 1. University of Groningen, University Medical Center Groningen, Department of Rehabilitation Medicine, Groningen, the Netherlands
  • 2. University of Groningen, University Medical Center Groningen, Center for Human Movement Sciences, Groningen, the Netherlands


Using a serious game to find distinct electromyogram patterns
M. B. Kristoffersen, A. Franzke, A. Murgia, R. Bongers, C. van der Sluis;
University Medical Center Groningen, University of Groningen, Groningen, Netherlands.

Background: Pattern Recognition (PR) control for upper limb prosthetics is based on electromyogram (EMG) patterns that are commonly generated by (phantom) movements matching the desired movements of the prosthesis. Effective PR control requires EMG patterns for different hand movements to be distinct, which is an important goal of training. Conventional training of PR control is long and tedious for the patient and requires supervision. Serious games might make this training more efficient, fun and independent. We compared a serious game with conventional training and ask if serious game training can lead to better outcomes and more distinct EMG patterns than conventional training.

Materials and methods: Able-bodied participants (N=50) were fitted with 8 electrodes around their forearm and performed 5 training sessions of 30 minutes on 5 consecutive days. The game group (n=13, mean age 21.3, 6 female) trained with the game using a direct mapping from electrode orientation to avatar movement in 2-dimensional space. Players explored different movements in order to learn which movements cause the avatar to move in the desired direction. Mastering all directions indicates the player has found 8 (phantom) movements with distinct EMG patterns. After each game session participants performed a system training followed by the Motion Test using 7 movements that they learned in the game (Linear Discriminant Analysis classifier, Hudgins feature set). The conventional group (n=37, mean age 21.6, 18 females) trained by performing a system training followed by the Motion Test using 7 predefined movements. Outcome variables were the outcome of the Motion Test and EMG pattern separability measured as the average Mahalanobis distance between each pattern and all of its neighbours in the feature space.

Results: The game group achieved comparable results to the conventional group in the Motion Test. EMG patterns were significantly more distinct for the game group than the conventional group. However, some participants in the game group had trouble transferring the movements learned in the game to the test without feedback from the game.

Conclusions: A serious game can train participants to find the set of (phantom) movements that provide the most distinct EMG patterns. As such, the game seems to be a promising training tool. However, it remains to be seen if such movements are also suitable for people with an upper-limb defect and if transfer to actual prosthesis use is possible.


Using a Serious Game to Find Distinct Electromyogram Patterns - Abstract.txt

Additional details


INPUT – Intuitive Natural Prosthesis UTilization 687795
European Commission