Elektrofiziologičeskaja aktivnost mozga pri upravlenii interfejsom mozg kompjuter, osnovannym na voobraženii dviženij

0476818 - ÚI 2018 RIV RU rus J - Článek v odborném periodiku
Frolov, A. A. - Aziatskaya, G.A. - Bobrov, P.D. - Luykmanov, R. Kh. - Húsek, Dušan - Snášel, V.
Elektrofiziologičeskaja aktivnost mozga pri upravlenii interfejsom mozg kompjuter, osnovannym na voobraženii dviženij.
[Electrophysiological Activity of the Brain During Control of a Motor Imagery Based Brain-Computer Interface.]
Fiziologiya cheloveka. Roč. 43, č. 5 (2017), s. 17-28. ISSN 0131-1646
Institucionální podpora: RVO:67985807
Klíčová slova: brain-computer interface * electroencephalogram * motor imagery * EEG rhythm synchronization and desynchronization * Independent Component Analysis * neurorehabilitaiton
Obor OECD: Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

V dannojj rabote rassmatrivajutsa osobennosti pati patternov elektroencefalogrammy, naibolee stabilno vydeljajuščichsaa metodom nezavisimych komponent pri voobraženii dvžhenijj kistejj ruk vo vremja upravlenija interfejsom mozg-kompjuter. Putem rešenija obratnoj zadači EHEHG s ispolzovaniem individualnoj geometričeskojj modeli golovy pokazyvaetsa,č to istočniki vydelennych patternov nachodjatsa v glubine centralnych borozd oboich polušarij, premotornoj kore levogo polušarija, dopolnitelnoj motornoj oblasti i v oblasti predklinja. Funkcionalnoe značenie ukazannych istočnikov obsuždaetsa v sravnenii s rezultatami meta analiza literaturnych dannych, polučennych s pomoščju funkcionalnojj magnitorezonansnoj tomogrammy. Rezultaty lokalizacii istočnikov sovpadajut dla 7 zdorovych ispytuemych i 4 postinsultnych bolnych s podkorkovym raspoloženiem očaga. Odnako, nesmotrja na sovpadenie lokalizacij, imejutsa značitelnye različija v spektralnych charakteristikach vydelennych patternov u ukazannych grupp, v častnosti, v storonu otsutstvija u pacientov četko vyražennoj aktivnosti v verchnem alfa diapazone i gorazdo menšej stepeni podavlenija ritmov alfa diapazona v pervičnych somatosensornych oblastach vo vremja voobraženija dviženij.

Present study concerns features of five electroencephalogram patterns which are the most frequently extracted by means of Independent Component Analysis from the data recorded while controlling a brain-computer interface with hand movement imagery by 7 healthy subjects and 4 patients with subcortical foci of stroke. Sources of the revealed patterns were localized by solving the inverse EEG problem using individual head models. It is shown that the sources localize at the bottom of both left and right central sulci, in left premotor cortex, in supplementary motor area, and precuneus. Possible functional meaning of the patterns is discussed by comparing localization results with foci of brain activity known from fMRI study. The source locations are found to be the same in the healthy subjects and the patients, but despite that, the source activities in these groups have significantly different spectral properties. In particular, alpha band activity in somatosensory areas is much less suppressed during motor imagery in patients and this activity has less exposed peak in upper alpha band compared to the healthy subjects.
Trvalý link: http://hdl.handle.net/11104/0273234