Abstract
Purpose: :
To evaluate three different approaches to the quantification of human OPs and their relationship to stimulus intensity. In previous work we demonstrated that OP amplitudes in an animal model were intensity dependent, and together with OP timing, provided a good indicator of retinal pathology. However this type of analysis is labor–intensive.
Methods: :
ERGs were recorded at different intensities covering a log unit above and below the ISCEV standard for OPs in five human subjects who were free of known ocular pathologies. OPs were isolated by band–pass filtering between 100–300Hz. A novel analysis of the area and the power of the OPs was performed.
Results: :
Area of the OPs was found to be less noisy than measurement of OP amplitude. Similarly, analysis of the spectral power had a good signal to noise ratio. The power analysis showed a peak frequency at about 150 Hz, which increased in amplitude up to about 3 cd.s/m2. For stimuli above this intensity, a second peak at about 100 Hz arose. Amplitude, area, and power analyses were all found to have roughly the same shape when plotted against log stimulus intensity.
Conclusions: :
The results of both the area and power analyses closely resemble amplitude versus intensity function but required much less operator judgment and may have a better signal to noise ratio. Thus, either of these approaches may be better suited for clinical applications, wherein automated analysis is preferable.
Keywords: electroretinography: non-clinical • electroretinography: clinical • retina