Individualized hand motor training under suitable brain states to improve performance and learning after stroke
Image and Language Processing, Computer Graphics and Visualisation, Human Computer Interaction, Ubiquitous and Wearable Computing
Biomedical Systems Technology
Human Cognitive and Systems Neuroscience
Final Report Abstract
The SuitAble project investigated how brain activity measured in real-time can possibly be used to improve the efficiency of rehabilitation training for patients, who suffer from a hand motor impairment that had been induced by a stroke. We developed novel data analysis algorithms to cope with specific challenges of this patient scenario and improved existing algorithms for detecting the relevant brain activity, to make a brain state dependent closed-loop patient training technically feasible. Furthermore, we developed tools for clinicians and therapists which make it easy to select individual patterns of brain activities per patient, which are relevant for the training. We found that we could successfully predict individual hand motor performance in a single trial not only for 20 healthy aged subjects, but also for five out of seven patients with chronic stroke. Waiting a few seconds for an informative brain activity pattern, we thus could determine the starting time point for the repeated hand motor training tasks. Waiting for specifically suitable or unsuitable brain states allowed us to influence a patient’s following hand motor performance. While all necessary algorithmic tools and closed-loop software for a randomized, controlled patient rehabilitation study were developed, the study could not reach the originally foreseen number of patient participants due to COVID-19. Thus a future project with a larger number of stroke participants will need to determine whether the proposed brain state dependent rehabilitation training actually results in improved clinical outcomes for patients. The data analysis algorithms developed within the project, however, are in use already in other applications that analyze brain activity in real-time. The “SuitAble” project sparked publication of three peer-reviewed journal articles, two conference articles and provided the scientific challenges for two PhD projects and three MSc projects.
Publications
-
Extremely Reduced Data Sets Indicate Optimal Stimulation Parameters for Classification in Brain-Computer Interfaces. 2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), 2256-2260. IEEE.
Sosulski, Jan & Tangermann, Michael
-
Mining Within-Trial Oscillatory Brain Dynamics to Address the Variability of Optimized Spatial Filters. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 27(3), 378-388.
Meinel, Andreas; Kolkhorst, Henrich & Tangermann, Michael
-
Spatial Filters for Auditory Evoked Potentials Transfer Between Different Experimental Conditions. In Proceedings of the 8th Graz Brain-Computer Interface Conference 2019. Verlag der Technischen Universität Graz. (2019)
Sosulski, J. & Tangermann, M.
-
Improving Covariance Matrices Derived from Tiny Training Datasets for the Classification of Event-Related Potentials with Linear Discriminant Analysis. Neuroinformatics, 19(3), 461-476.
Sosulski, Jan; Kemmer, Jan-Philipp & Tangermann, Michael
-
Manipulating Single-Trial Motor Performance in Chronic Stroke Patients by Closed-Loop Brain State Interaction. IEEE Transactions on Neural Systems and Rehabilitation Engineering, 29(2021), 1806-1816.
Meinel, Andreas; Sosulski, Jan; Schraivogel, Stephan; Reis, Janine & Tangermann, Michael
-
Optimization of Stimulation Speed in an Auditory Brain-Computer Interface under Time Constraints. (2021).
Sosulski, J., Hübner, D., Klein, A. & Tangermann, M.
