Purpose
Visual acuity (VA) is an important parameter to test visual function. It is commonly assessed with eye (e. g. Snellen-) charts, which use optotypes of different sizes. This technique relies on the subjects' cooperation, and hence, its results are always subjective. VA can also be estimated using Visual Evoked Potentials (VEP): amplitude size of VEPs is correlated to the spatial frequency of a stimulus pattern. This provides an objective method for assessing VA, especially in subjects with low compliance. Two different paradigms are commonly used: the Visus VEP and the Sweep VEP. Here we present a best-of-breed approach, which combines advantages of both techniques. It can be used with current electrophysiological recording systems.
Methods
A custom software was developed to present repetitive sequences of 11 checkerboards with increasing spatial frequencies (.6/.9/1.4/2.1/3.3/4.9/7.3/10.4/18.2/24.3/36.5cpd) using a 21” CRT monitor (92% contrast, on/off: 40/300 ms, isoluminant). Sweep VEPs were recorded using an Espion e2 system (Diagnosys LLC). Averaged results (n=50), were evaluated for trough-to-peak amplitudes of the spindle-shaped signal. The limiting spatial frequency and the resulting VA, were estimated by fitting modified Gaussian functions to the data. Goodness of fit was evaluated using root-mean-square error (RMSE). VA estimated by Gaussian fit, 2nd order polynomial fit and subjectively measured VA were compared.
Results
Nine healthy, non-myopic volunteers (4f/5m, 30±9.3y) were tested. Best corrected visual acuity (BCVA) and simulated deterioration of VA using lenses (1, 2, 3dpt) was measured using an eye chart. Sweep VEP was recorded and objective VA was estimated. Goodness of fit showed better results for modified Gaussian compared to 2nd order polynomial (RMSE 4.14, SD3.61 vs. 5.74, SD2.93). Comparison between subjective and objective VA was performed by Bland-Altman plot (r=0.73).
Conclusions
The Sweep VEP allows for an objective assessment of VA almost independent of subject compliance. Using custom developed software we were able to extend the method and to implement it in our electrophysiologyl recording system. By combining advantages of two well-established techniques we improved the objective estimation of the VA. Next steps include automated marker placement, improvement of curve fitting and definition of a mapping between spacial frequency and VA in a larger parameter space.
Keywords: 507 electrophysiology: clinical •
754 visual acuity •
629 optic nerve