Advantages of a cross-correlation technique for ERG measurement in children

The measurement of ERGs of infants and children suspected of having retinal dysfunction can present some significant practical problems. Often the degree of cooperation is less than ideal, the stimulus actually reaching the retina may vary, and the level of background electrical activity can mask small responses. Single flash ERGs often are not adequate in such circumstances, and usually it is advantageous to use some form of noise reduction. Conventional averaging techniques can be used, but one disadvantage is the time required to average a significant number of responses. The use of rapid, pseudorandomly timed stimuli and a cross-correlation signal recovery procedure offers some theoretical and practical advantages. Usually a better defined response (with a higher signal-to-noise ratio) can be obtained in a relatively short time, sometimes in a few seconds. In addition, the pseudorandomly timed flash stimuli inherently contain a wide range of frequencies, so that in effect the frequency response of the outer layers of the retina is being measured. Some clinical examples will be described to illustrate the practical application of this procedure for measuring the cone and rod ERGs of young patients.