Coordinated finger movements predicted from intracranial brain activity
Marc M. Van Hulle* D.Sc. in Engineering, Professor, Laboratory for Neuro- & Psychophysiology, Department of Neurosciences, University of Leuven, Academician of Royal Belgian Academy, Member of European Academy of Nature Sciences, member of Royal Pharmaceutical Academy (Spain), Member of Neurobiological Society (USA), Member of IEEE, Leuven, Belgium firstname.lastname@example.org * The corresponding author
Axel Faes PhD Student, Laboratory for Neuro- & Psychophysiology, Department of Neurosciences, University of Leuven email@example.com
Abstract Several studies have successfully employed Brain Computer Interfaces (BCIs) to replace the function of a lost or impaired limb by circumventing disconnected neural pathways. Electrocorticography (ECoG) offers unique perspectives for long-term brain activity recording while providing high temporal, spatial and spectral resolution. ECoG uses electrodes placed on the exposed cortical surface, thus without entering the cortical tissue which could lead to scarring and other histological processes eventually affecting signal quality.
Despite advances in individual finger decoding based on ECoG, convincing demonstrations of coordinated finger actions are still lacking. In this contribution, we report on our advances in accurately predicting self-paced individual- and coordinated finger movements from ECoG activity recorded in temporarily implanted epileptic patients capable of performing finger movements.
Our long-term ambition is to transfer trained hand-motor BCIs to decode ECoG activity evoked by imagined finger movements as it could serve patients suffering from paralysis due to spinal cord injury, brain stem stroke or a degenerative disorder such as amyotrophic lateral sclerosis, but that are otherwise fully conscious of the intended actions.
Keywords: Brain Computer Interface, Electrocorticography, Finger movement decoding, Tensor regression
Acknowledgements AF is supported by a fundamental research grant awarded by the Research Foundation – Flanders (1157021N)MMVH is supported by research grants received from the European Union's Horizon 2020 research and innovation programme under grant agreement No. 857375, the special research fund of the KU Leuven (C24/18/098), the Belgian Fund for Scientific Research − Flanders (G0A4118N, G0A4321N, G0C1522N), the Interuniversity Attraction Poles Programme − Belgian Science Policy (IUAP P7/11), and the Hercules Foundation (AKUL 043).