Purchase this article with an account.
B.A. Mazinani, T.D. Waberski, P. Walter; Improving the Signal–To–Noise–Ratio in MfVEP by Source Dipole Analysis . Invest. Ophthalmol. Vis. Sci. 2006;47(13):1677.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
It is recommended to use multiple recording channels in mfVEP (e. g. three channels recorded and three derived by subtraction) to cover the highest possible amount of source vector directions on the cortex surface and obtain a "best–of"–mfVEP that has a better SNR than the single recordings1. It is possible to predict VEP signals on any point of the head surface on the base of recordings with three electrodes by source dipole calculation with good precision2. Purpose of the study was to test, if the SNR of "best–of"–mfVEPs can be improved by implementing this "mini–brainmapping"–method compared to recordings with three recordings and three derivations.
From 11 healthy subjects (24–29, mean 26 years) mfVEPs were recorded on three channels with the inion as reference with derivation of three additional channels by subtraction and with electrodes at the positions POz’, O1’ and O2’ (reference Fz). A source dipole vector was calculated for each time point of the latter recordings and VEP responses were back projected for 96 electrode positions. The total SNR of the "best–of"–mfVEPs of both methods were compared.
The conventional six channel mfVEP had a mean SNR of 2,9, the new recording method based on "mini–brainmapping" had a mean SNR of 3,4 (p<0,005).
The "Mini–Brainmapping" model permits to improve the signal–to–noise ratio of mfVEPs significantly compared to common recording procedures. 1Hood DC, Zhang X, Hong JE, Chen CS. Quantifying the benefits of additional channels of multifocal VEP recording. Doc Ophthalmol. 2002 May;104(3):303–20 2Mazinani BAE, Waberski T, Walter P. Extraction of topographical information from visual evoked Potentials with three active electrodes. IOVS 46: 5686 Suppl. S 2005 (ARVO abstract)
This PDF is available to Subscribers Only