Purchase this article with an account.
R Suryakumar, WR Bobier; Brightness profiles in Eccentric Photorefraction crescents . Invest. Ophthalmol. Vis. Sci. 2002;43(13):2670.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: When an infrared source is offset from a camera aperture, a brightness profile (crescent) is formed in the margin of the pupil from light reflected from the eye. This study measured the slope of this brightness profile at 5 adjacent pixel offsets (0.28mm width each) on either side of the pupil center. The purpose was to determine the consistency of refractive measures taken from intensity gradients offset from the pupil center. Methods: Eccentric photorefraction measures were taken from three subjects (age=27.6 ± 2.64 yrs, SE = -0.83 ± 1.0D) using an extended source set inferiorly from a camera aperture. Ophthalmic lenses were introduced in 0.50D steps for a range of ±4D. The images were captured and analyzed offline using a Visual C++ algorithm. The slope of the brightness measures across the pupil for an offset of ±5 pixels on either side of the pupil center was determined for each induced refractive error. Conversion factors were determined for each subject that defined refractive error as a function of slope. Complete pixel light distributions for crescents formed at +4 and -4D defocus were studied. Results: In both crescent measures, patches of uneven intensity not predicted by current theory 1 were found. However, the average slopes of the brightness profile across the ten offsets from the pupil center (+5 to -5) differed only by 0.004, 0.01, and -0.031 for plano, +4 and -4 D defocus respectively which corresponds to dioptric differences of only -0.07, -0.14, and 0.20D. The largest difference for an individual was +0.74D. The smallest variations were found for infocus images. For this condition, two repeated measures of the slope of the brightness gradient through the pupil center showed no significant differences (average refractive error at test = -0.56 ± 0.47D and re-test = -0.63 ± 0.54D, p=0.24). Conclusion: The patterns of variable brightness observed within a crescent in this small sample do not upon average seriously affect the refractive measures. However, the discrepancy does vary with the individual. It would appear that brightness measures in eccentric photorefraction need not be limited to pupil center. (1) Kusel R et al. J Opt Soc.Am A Opt Image Sci Vis. 1998;15:1500-1511.
This PDF is available to Subscribers Only