Neuroimaging Methods Group

Group Leader - Alexei Ossadtchi 
 

Key words: EEG, MEG, BCI, neurofeedback, digital signal processing, inverse problem, brain mapping, brain connectivity analysis

The main goal of our group is to develop methods for efficient handling of non-invasively collected magneto- and electroencephalographic data. We work in a close collaboration with other experimental groups of the Center, providing data analysis and experimental design support.

Our work

Methods for source analysis and source space connectivity analysis of MEG, EEG data for cognitive and clinical uses
MEG and EEG-based brain-imaging technology allows us to visualize and study neuronal processes non-invasively, with very high temporal and reasonable spatial resolution. Currently, this method is used not only to localize neuronal activity but also to recover the dynamics of the entire neural network comprised of several distant cortical regions subserving the cognitive phenomenon at hands. The accuracy and reproducibility of such inferences crucially depend on the methods and computational approaches used for analysis of the MEG and EEG sensor data. Development of the novel methods and verification of the existing approaches allow us to ensure that the information present in the data is utilized with maximal efficiency. The use of the subject-specific probabilistic models obtained via data fusion form various modalities including fMRI, DTI, Optical tomography, TMS bear a promise of improving the non-invasively achieved resolution and further open "the non-invasive window" into the brain's function. The clinical portion of this work is done in close collaboration with the Moscow MEG center, Burdenko Institute of Neurosurgery and the Institute for problems of Mechanical Engineering, RAS, and is supported by an RFFI grant.
Selected publications:

1. A. Ossadtchi, D. Altukhov, K. Jerbi (2017). Power and shift invariant detection of dynamically coupled networks (PSIICOS) from non-invasive MEG data. BioRxiv, 129155
2. A. Kuznetsova, E. Krugliakova, A. Ossadtchi (2017). Localizing hidden regularities with known temporal structure in the EEG evoked response data. BioRxiv, 093922
3. Kozunov VV and Ossadtchi A (2015) GALA: group analysis leads to accuracy, a novel approach for solving the inverse problem in exploratory analysis of group MEG recordings. Front. Neuroscience 9 : 107. doi: 10.3389/fnins.2015.00107
4. D. Altukhov and A. Ossadtchi, GO-PSIICOS: Globally Optimized Power and Shift Independent Imaging of Coherent Sources, Brain connectivity workshop (BCW 2015), Sand Diego, CA, June 2015
5. A. Ossadtchi, Interaction Space RAP-MUSIC for estimation of transient networks from MEG data, Proceedings of the 19th international conference on biomagnetism, Halifax, Canada, 2014
6. A. Ossadtchi, R. Greenblatt, Cross-term deprived covariance approach as an extension of DICS for detection of cross-frequency coupling, Proceedings of the 18th international conference on biomagnetism, Paris, France, 2012
7. A. Ossadtchi, P. Pronko, S. Baillet, M. Pflieger, T. Stroganova, Mutual information spectrum for selection of event-related spatial components. Application to eloquent motor cortex mapping, Frontiers in Neuroinformatics, January, 2014, doi: 10.3389/fninf.2013.00053
8. R.E. Greenblatt, M.E. Pflieger, A. Ossadtchi, Connectivity measures applied to human brain electrophysiological data, Journal of Neuroscience Methods 207 (2012) 1– 16

Real-time EEG: neurofeedback and neurointerfaces

Neurofeedback is a promising tool for non-pharmacological therapy for depression, ADHD, epilepsy and other psychiatric diseases and syndromes. This technique is also used for peak performance training and in relaxation practices. Within the neurofeedback paradigm a person is put into a closed loop where his/her brain state is presented as a feedback signal via one of the sensory modalities and is used to consciously (or not so consciously) modify the dynamics of his/her brain rythms. In this project we will consider this closed-loop system from the standpoint of control theory (both continuous and discrete) and will focus on developing novel model-based techniques to improve the efficiency and spatial specificity of neurofeedback therapy. This research will pave the way towards a new generation of more efficient and more natural brain-computer interfaces 
Selected publications:

1. N. Smetanin, A. Ossadtchi (2017). Express estimation of brain rhythm power for low-latency neurofeedback. The First Biannual Neuroadaptive Technology Conference, Berlin
2.  
3. V. Minkov, N. Smetanin, N. Markina, I. Dybushkin, A. Ossadtchi (2017). Neurophysiological correlates of efficient learning in the neurofeedback paradigm. The 1st biannual neuroadaptive technology conference, 153
4. N. Dagaev, K. Volkova, A. Ossadtchi (2017). Latent variable method for automatic adaptation to background states in motor imagery BCI. Journal of Neural Engineering
5. R. van Lutterveld, S. D. Houlihan, P. Pal, M. D. Sacchet, C. McFarlane-Blake, P. R. Patel, J. S. Sullivan, Alex Ossadtchi, S. Druker, C. Bauer, J. A. Brewer (2017). Source-space EEG neurofeedback links subjective experience with brain activity during meditation, Neuroimage 151, 117-127
6. A. Ossadtchi, T. Shamaeva, E. Okorokova, V. Moiseeva, MA Lebedev (2017). Neurofeedback learning modifies the incidence rate of alpha spindles, but not their duration and amplitude. Scientific Reports 7
7. Okorokova E., Lebedev M., Linderman M. and Ossadtchi A. (2015). A dynamical model improves reconstruction of handwriting from multichannel electromyographic recordings Front. Neurosci. 9 : 389 doi: 10.3389/fnins.2015.00389
8. S. L. Shishkin, A. A. Fedorova, Y. O. Nuzhdin, I. P. Ganin, A. E. Ossadtchi, B. B. Velichkovskiy, A. Y. Kaplan, B. M. Velichkovsliy. Towards high-speed eye-brain-computer interfaces: Combining the “single stimulus” paradigm and saccades to stimulus location, Vestnik Moskovskogo Universiteta, Seriya 14. Psikhologiya. 2013. N4
9. I. Ovod, A. Ossadtchi, A. Pupyshev, A. Fradkov, Forming neurofeedback signal based on the adaptive model of the EEG observed human brain activity, Neurocomputers and applications, February, 2012 (in Russian).

Brain-Computer Interface 

Brain computer Interface (BCI, Human-Computer Interface HCI) is a system which allows to control an external device with the signals generated by the brain, without using muscle activity.

Most Brain-Computer Interfaces work the following way:

1. Record brain activity with some neuroimaging technique (EEG, MEG, EcoG, etc)
2. Extract useful information from the signals
3. Classify discrete commands (actions) based on the differentiation of the patterns of the brain activity.

The Neuroimaging Methods Group of the CDM has developed and tested the EEG-based Brain Compuer Interface, which can easily classify up to 6 real and imaginary movements: the arms, the legs, the tongue and the resting state. The Interface is based on a well-established principle of the body parts representations in the servomotor cortex, which can help to infer patterns of electrical activity, differentiating various motor states.

Selected publications:

1. M. A. Lebedev and A. Ossadtchi, Bi-Directional Neural Interfaces, Brain–Computer Interfaces Handbook: Technological and Theoretical Advances, Eds: Chang S. Nam, Anton Nijholt, Fabien Lotte, CRC Press, 2018
2. Волкова К. В., Дагаев Н. И., Киселёв А., Касумов В., Александров М., Осадчий А. Е. Интерфейс мозг-компьютер: опыт построения, использования и возможные пути повышения рабочих характеристик // Журнал высшей нервной деятельности им. И.П. Павлова. 2017. Т. 67. № 4. С. 504-520

Collaboration with other CCDM groups

Zubarev I., Shestakova A., Klucharev V., Ossadtchi A., Moiseeva V. (2017). MEG Signatures of a Perceived Match or Mismatch between Individual and Group Opinions. // Frontiers in Neuroscience. 2017. No. 11. P. 1-9