Quantifying the benefits of additional channels of multifocal VEP recording

Abstract

For some individuals and for some locations, multifocal visual evoked potentials (mfVEP) may be too small or appear 'too noisy' to be reliably measured. By adding electrodes, especially electrodes placed lateral to the midline, and by recording with multiple channels, the amplitude of the signal can be increased in some field locations. However, the addition of electrodes involves certain costs; the set-up time is longer and the data analysis more time consuming and complex. The objective of this study was to assess the benefits of adding electrodes by quantifying these benefits using a signal-to-noise measure. In addition to the typical midline placement of electrodes, two electrodes were placed 1 cm above and 4 cm lateral to the inion on each side. This allowed for 3 channels of recording and 3 additional, derived channels. The mfVEPs were recorded with a 60 sector, pattern-reversing display presented to one eye. Two 7 min records were obtained from 14 individuals with no known visual problems. The two records were averaged and a signal-to-noise (SNR) measure was obtained for every response from all 6 channels. For each sector of the display and each subject, the benefits of additional electrodes were quantified by comparing the SNR from the traditional midline channel to the best SNR from amongst the 6 channels. The number of responses exceeding any given criterion SNR value was increased with the additional channels. For example, 79% of the responses for the typical midline channel exceeded a SNR of 0.6 (a false positive rate of about 2.5%) and this increased to 93% when the best SNR value was used. As expected, summing the mfVEP responses from contiguous sectors also increased the SNR values. Additional electrodes and multiple channels of recording substantially improve the quality of the mfVEP records and the SNR measure provides a useful metric for assessing these benefits.