eduzhai > Applied Sciences > Engineering >

Spatial sampling of MEG and EEG revisited From spatial-frequency spectra to model-informed sampling

  • Save

... pages left unread,continue reading

Document pages: 18 pages

Abstract: In this paper, we analyze spatial sampling of electro- (EEG)magnetoencephalography (MEG), where the electric or magnetic field is typicallysampled on a curved surface such as the scalp. Using simulated measurements, westudy the spatial-frequency content in EEG as well as in on- and off-scalp MEG.The analysis suggests that on-scalp MEG would generally benefit from threetimes more samples than EEG or off-scalp MEG. Based on the theory of Gaussianprocesses and experimental design, we suggest an approach to obtain samplinglocations on surfaces that are optimal with respect to prior assumptions.Additionally, the approach allows to control, e.g., the uniformity of thesampling locations in the grid. By simulating the performance of gridsconstructed with different priors, we show that for a low number of spatialsamples, model-informed non-uniform sampling can be beneficial. For a largenumber of samples, uniform sampling grids yield nearly the same totalinformation as the model-informed grids.

Please select stars to rate!

         

0 comments Sign in to leave a comment.

    Data loading, please wait...
×