Abstract
Abstract: :
Purpose: Eye tracker for refractive corneal laser surgery are aiming for the center of the pupil. The constriction of the pupil is not really symmetrical. There is a shift of the center of the pupil towards nasal and up during constriction. This was to determine the speed of the shift and its position during constriction. Methods: The pupils of 6 persons were captured by infrared videography colinear to a fixation diode. The adaption for at least one minute was done at 16 lux, the retina was illuminated up to 5000 lux to constrict the pupil. Digital image processing using eye tracking algorithms was done to determine the size, the elliptical shape and the position of the pupil and the limbus. Using the position of the limbus as a reference leads to the amount of pupil center shift. Linear regression analysis was done to determine the speed of shift. Results: Average pupil size adapted to dark was 5.3±0.8 mm (SD), the eccentricity was 0.21. Under light conditions the average was 3.07±0.6 mm, the eccentricity was 0.02. The mean shift of the pupil was 0.14 mm. The shift was not correlated with pupil size, the ability of shifting was maximum under dilated conditions and decreased to 0.047mm for pupil size under 3.5 mm. The speed of pupil center shift could be estimated to 0.097 mm/s during constriction and 0.081 mm/s during dilation. Conclusions: The center of the pupil is not constant. It varies with pupil size, but is not related to it. Till now, it seems there is no reliable value yet to predict the pupil center. Although eye tracker used in refractive surgery are fast enough to determine the pupil center, we don't know the relation to the center of the cornea. To provide precise excimer treatments, eye tracker systems should also detect other ocular structures, such as the limbus.
Keywords: refractive surgery: complications • pupil • imaging methods (CT, FA, ICG, MRI, OCT, RTA, S