• This web page is devoted to the project

    Advancing methods of signal processing for magnetoencephalography

    Funded by the National Science Centre of Poland (POLONEZ 2, grant agreement No 2016/21/P/ST7/03929), co-funded by the European Union's Horizon 2020 Research and Innovation Programme under the Marie Skłodowska-Curie Grant 665778.

  • Background work

    Simultaneous spatio-temporal matching pursuit decomposition of evoked brain responses in MEG

    Biological Cybernetics, 111(1), 69–89; DOI: 10.1007/s00422-016-0707-5

  • Description for the general public

    Magnetoencephalography (MEG) ...

    ... is a technique similar to the well-known electroencephalography (EEG). The two methods are mutually complementary in the sense that EEG measures the electric field, whereas MEG measures the magnetic field, both stemming from the electric currents behind the neuronal activity in the brain. One of the advantages of MEG over EEG is that it does not require electrodes to be in contact with the skin of the examined person. Hence, the time required for preparing a person for the examination is substantially reduced.

    The amount of data acquired from a MEG device is enormous. This is cumbersome for the clinician to arrive at adequate interpretation of the results because it is often difficult to ascertain what constitutes a signal and what can be regarded as noise. Hence, extracting the clinically relevant information from MEG signals is not a trivial task. One approach is to use a stimulus to obtain the so-called evoked responses of the brain (for example, auditory or visual) that after data averaging are easier to interpret. However, interpreting the so-called induced activity in the human brain requires far more sophisticated methods, which are often computationally very intensive.

    During the project we developed new statistical signal processing algorithms that enable reducing the computational burden of data processing and, more importantly, reducing the length of a typical MEG/EEG examination, which directly translates to increased mental and physical comfort of the patient.

    Importantly, the procedures that we have developed can be applied also to other branches of biomedical signal processing, well beyond MEG and EEG; for example to electrocardiography (ECG) and ophthalmic diagnostics.

    The project was performed in collaboration with the best European research centres and universities, including the Leibniz Institute for Neurobiology, Magdeburg, the ICM Brain and Spine Institute and Pierre and Marie Curie University (Sorbonne), Paris, as well as the Lyon Centre for Research in Neuroscience.