MEG-BCI: Adaptive online-decoding of movement direction and application in rehabilitation
-
1
University of Tübingen, MEG Center, Germany
-
2
Freiburg University, Institute of Biology I, Germany
-
3
Freiburg University, Bernstein Centre for Computational Neuroscience, Germany
-
4
University of Trento, CIMEC, Italy
Decoding brain states which can be reliably used for Brain-Computer-Interfaces (BCI’s) is one of the most challenging topics in neuroscientific research. A recent study showed that hand movement direction in a center-out task can be inferred by non-invasive recordings using linear discriminant analysis (LDA) offline [2]. In this study it was also shown that magnetoencephalography (MEG) compared to electroencephalography led to a higher decoding accuracy for movement direction. These findings raise hope to overcome frequent limitations of BCI’s which apply sophisticated predefined classification schemes. In our approach we seek to construct an online-adaptive MEG-BCI which decodes hand movement direction in a center-out task. Low-pass filtered MEG signal of 52 channels over sensorimotor areas were used to predict the movement direction online based on LDA. The classification parameters were adapted on a single-trial basis in order to compensate for non-stationarities during recordings. Furthermore, we provided subjects with continuous visual feedback of the current decoding result to include subject-specific learning effects. The preliminary results showed that we were able to reach a mean decoding accuracy of 72%. Moreover, the combination of machine- and feedback-learning increased decoding performance compared to earlier results based on a non-adaptive approach. These promising findings indicate that a non-invasive online-BCI for hand movements can be achieved without extensive subject training or demanding computational processes. Future applications therefore might include pre-surgery testing of ECoG-patients or rehabilitation after stroke. This is supported by recent results of our group where a MEG-based µ-rhythm BCI has been applied to train motor functions in chronic stroke patients suffering from subcortical stroke in the pyramidal tract [1]. We were able to show that all patients could increase motor-related brain activity in motor and pre-motor cortex of the lesioned hemisphere. Furthermore, discriminability of brain activities related to opening and closing the hand was increased over training sessions and led to clinically relevant increase of motor function in one subject. Combining our new ‘direct’ BCI setup with these patient studies is therefore likely to further increase the clinical benefit.
References
1. Buch E, et al. (2008). Think to Move: a Neuromagnetic Brain-Computer Interface (BCI) System for Chronic Stroke. Stroke, 39(3), 910-917.
2. Waldert S, et al. (2008). Hand Movement Direction Decoded from MEG and EEG. J Neurosci, 28(4), 1000-1008
Conference:
Bernstein Symposium 2008, Munich, Germany, 8 Oct - 10 Oct, 2008.
Presentation Type:
Poster Presentation
Topic:
All Abstracts
Citation:
Witte
M,
Heiden
L,
Caria
A,
Birbaumer
N,
Mehring
C and
Braun
C
(2008). MEG-BCI: Adaptive online-decoding of movement direction and application in rehabilitation.
Front. Comput. Neurosci.
Conference Abstract:
Bernstein Symposium 2008.
doi: 10.3389/conf.neuro.10.2008.01.081
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
17 Nov 2008;
Published Online:
17 Nov 2008.
*
Correspondence:
Matthias Witte, University of Tübingen, MEG Center, Tübingen, Germany, matthias.witte@med.uni-tuebingen.de