Motor-parietal cortical neuroprosthesis with somatosensory feedback for restoring hand and arm functions in tetraplegic patients.
Medical Physics, Biomedical Technology
Final Report Abstract
In this project I worked on the development of a brain-computer interface (BCI) platform and on establishing a BCI laboratory in Germany. A brain-computer interface is used to help severely paralyzed patients to regain autonomy by directly using brain signals to control a robotic device. Since the spinal cord of these patients is damaged information from the brain cannot reach the patient's own limbs. The BCI is used to pick-up the user’s movement intentions from the brain and relay them to a device, for example a robotic limb or an exoskeleton. The project was divided into two phases. In the first phase I worked with patients who were implanted temporarily with electrode grids that are placed on top of the brain prior to epilepsy surgery and record brain signals from them. These patients performed specific tasks in a virtual reality environment while neural signals were recorded with dedicated recording hardware. This neural data could be used to train a decoding algorithm to control an extracorporeal device. In phase two it was planned to implant a paralyzed patient with a chronic electrode implant, but due to regulatory delays this phase was not finished. With the data from the first phase of the project and data that was acquired beforehand, a new spike sorting algorithm based on deep learning methods was developed. Spike sorting is an important step in the BCI pipeline to identify relevant neural information that could further improve future brain-computer interfaces. The algorithm called “Spikedeeptector” was published in a peer reviewed journal. In another experiment we used virtual reality with healthy volunteers to simulate certain aspects of the control problems associated with BCIs. We discovered by using a virtual reality task that a particular type of control error in brain-computer interfaces could be attributed to the positioning of the controlled robotic arm. These novel results were also published in a peer reviewed journal. The discoveries during the project phase led to several new projects for which third party funding was acquired. In one of the projects, we develop a exoskeleton that is supposed to be controlled by patients. In another we use virtual reality for neurorehabilitation and in a very recent project we use the results from our spike sorter to build a hardware system that implements the algorithm.
Publications
-
Invasive Brain-Computer Interfaces and Neural Recordings From Humans. Handbook of Behavioral Neuroscience, 527-539. Elsevier.
Klaes, Christian
-
Was steckt hinter den Versprechen des Transhumanismus? In: Göcke BP, Meier-Hamidi F editors. Designobjekt Mensch: Die Agenda des Transhumanismus auf dem Prüfstand. 1st ed. Freiburg im Breisgau: Verlag Herder GmbH; 2018. p. 379-408
Klaes C.
-
SpikeDeeptector: a deep-learning based method for detection of neural spiking activity. Journal of Neural Engineering, 16(5), 056003.
Saif-ur-Rehman, Muhammad; Lienkämper, Robin; Parpaley, Yaroslav; Wellmer, Jörg; Liu, Charles; Lee, Brian; Kellis, Spencer; Andersen, Richard; Iossifidis, Ioannis; Glasmachers, Tobias & Klaes, Christian
-
SpikeDeep-Classifier: A deep-learning based fully automatic offline spike sorting algorithm. Journal of Neural Engineering.
Saif-ur-Rehman, Muhammad; Ali, Omair; Dyck, Susanne; Lienkämper, Robin; Metzler, Marita; Parpaley, Yaroslav; Wellmer, Jörg; Liu, Charles; Lee, Brian; Kellis, Spencer; Andersen, Richard A.; Iossifidis, Ioannis; Glasmachers, Tobias & Klaes, Christian
-
Quantifying the alignment error and the effect of incomplete somatosensory feedback on motor performance in a virtual brain–computer-interface setup. Scientific Reports, 11(1).
Lienkämper, Robin; Dyck, Susanne; Saif-ur-Rehman, Muhammad; Metzler, Marita; Ali, Omair & Klaes, Christian
-
Enhancing the decoding accuracy of EEG signals by the introduction of anchored-STFT and adversarial data augmentation method. Scientific Reports, 12(1).
Ali, Omair; Saif-ur-Rehman, Muhammad; Dyck, Susanne; Glasmachers, Tobias; Iossifidis, Ioannis & Klaes, Christian
