We use cookies.
By using the site, you agree to our Privacy Policy.

Invited researcher Mikhail Albertovich Lebedev
Contract number
14.641.31.0003
Time span of the project
2018-2020

As of 30.01.2020

18
Number of staff members
8
scientific publications
1
Objects of intellectual property
General information

Name of the project: Bidirectional electrocorticographic brain-computer interfaces for control, stimulation and communications

Strategy for Scientific and Technological Development Priority Level: а, в


Goals and objectives

Research directions: Bidirectional brain-computer communications with somatosensor feedback using electrostimmulation or sensory substitution, creation of neurointerfaces and biological prosthetics.

Project objective: Development of an information technology of bidirectional communications with the human brain electrocorticographic interface in combination with modern methods of processing of multi-dimensional data ans somatosensory feedback by stimulation or sensory replacement.


The practical value of the study

  • We have developed and implemented algorithms of pre-cleaning of electrocorticography images of artifacts related to movement of eyes and interparoxysmal activity.
  • Our researchers have implemented methods of reduction of dimensionality of feature space.
  • We have implemented algorithms for decoding parameters of movement based on classical methods of evaluation of parameters of signals.
  • Our researchers have developed and implemented deep learning architectures used for decoding parameters of movement.
  • Methods have been developed for interpretation of parameters of decoding algorithms.
  • We have created and tested an experimental device for synchronous registration of continuous movements (using a three-axis accelerometer) and multi-channel electrocorticography.
  • Our researchers have obtained synchronous records of electrocorticography signals, parameters of movement and videotapes of the experiment.
  • We have implemented decoding of parameters of movement from electrocorticography signals using deep learning methods and classical methods of signal processing.
  • Our researchers have compared precision of decoding in case of decoding using existing and developed algorithms. We have obtained results of comparison of existing algorithms and the ones developed by the Laboratory within the project. It has been demonstrated significant advantage of methods based on convolutional networks for solving the problem of decoding kinematic parameters from electrocorticography records. We have designed software for supporting decoding in real time.
  • An experimental device has been developed and prepared for conducting research of decoding of movement parameters in real time. We have obtained synchronous records of electrocorticography signals, parameters of movement and a videotape of the experiment, and completed successful decoding of movement in real time. A unique methodology has been developed for patient training. Our researchers have obtained results of decoding of data recorded during experiments.
  • We have implemented online control of an avatar of a limb based on electrocorticography signal decoding with a deep neural network architecture.
  • Our researchers have created additional software for supporting decoding in real time and supporting the procedure training patients using the interface.
  • We have developed a module for NFBLab that implements the center-out paradigm, obtained a synchronized recording of electrocorticography signals, cursor and a videotape of the experiment, characterized invoked responses related to stimuli of the paradigm. We have determined channels and features that allow for pairwise comparison of movements correlated with the direction of movement in the interval of displaying of target and after the moment of the start of movement.
  • We have developed criteria and algorithms of selection of patients for practical testing of neurointerfaces.

Implemented results of research:

  • Developed algorithms of analysis of brain activity are being used by doctors of the Moscow State University of Medicine and Dentistry Clinic for pre-surgery mapping of neurosurgery patients for localization of epileptogenic zone and functionally irretrievable cortical areas.
  • A patent application has been filed for the invention called «A method of evaluation of power of oscillatory components of EEG signals in psychophysiological states based on quantile analysis».
  • Employees of the Center regularly participate in «Neurothlon» international competitions of handicapped persons who use assistive technologies. The goal of such competitions is popularization of high technologies, expansion of capabilities of human body, and supplementing lost body parts.
  • Education and career development: The following courses have been compiled and are being read: the «Data analysis and artificial intelligence=» course at the Faculty of Computer Science of the Higher School of Economics, the «Digital Signal Processing (DSP)» course in English for first-year students of the «Cognitive sciences and technologies» master program of the Department of Psychology of the Higher School of Economics.

Organizational and structural changes: A modern laboratory of invasive interface has been created at the Higher School of Economics and the Moscow State University of Medicine and Dentistry for which we have purchased and launched state-of-the-art modern equipment and a scientific device has been created.

Other results:

  • An integrated system has been created that is based on a neurointerface and designed for controlling exoskeletons of the lower limbs «Exoatlet» for usage in neurorehabilitation.
  • In November 2018 we submitted an application for patent for the invention called «A method of evaluation of power of oscillation components of EEG signals in psychophysiological states based on quantile analysis».

Collaborations:

  • A. I. Evdokimov Medicine and Dentistry (Russian): joint laboratory of invasive interfaces
  • A. P. Polenov Neurosurgical Institute (branch of the V. A. Almazov National Medical Research Center of the Ministry of Health of Russia), University of Tübingen (Germany),    TU Berlin (Germany), Exoatlet (Russia): joint research and experimental work.

Hide Show full
. Koshkin R., Shtyrov Y., Myachykov A., Ossadtchi A.
Testing the Efforts Model of Simultaneous Interpreting: An ERP Study. PLoS ONE 13(10): e0206129 (2018).
Smetanin N., Volkova K., Zabodaev S., Lebedev M., Ossadtchi A.
NFBLab – A Versatile Software for Neurofeedback and Brain-Computer Interface Research. Frontiers in Neuroinformatics 12: 100 (2018).
Volk D., Dubinin I., Myasnikova A., Gutkin B., Nikulin V.
Generalized Cross-Frequency Decomposition: A Method for the Extraction of Neuronal Components Coupled at Different Frequencies. Frontiers in Neuroinformatics 12: 72 (2018).
Lebedev M., Pimashkin A., Ossadtchi A.
Navigation Patterns and Scent Marking: Underappreciated Contributors to Hippocampal and Entorhinal Spatial Representations? Frontiers in Behavioral Neuroscience 12: 98 (2018).
Other laboratories and scientists
Hosting organization
Field of studies
City
Invited researcher
Time span of the project
Computer Synthesis of Chemical Compounds

  G. A. Krestov Institute of Solution Chemistry of the Russian Academy of Sciences

Computer science and information technologies

Ivanovo

Tetko Igor

2021-2023

Multi-scale Neurodynamics for Smart Systems

Skolkovo Institute of Science and Tehnology

Computer science and information technologies

Moscow

Wang Jun

2021-2023

Laboratory of Non-linear and Microwave Photonics

Ulyanovsk State University

Computer science and information technologies

Ulyanovsk

Taylor James Roy

United Kingdom

2021-2023