Purchase this article with an account.
D.L. Guyton, O.H. Y. Zalloum, K.R. Gemp, D.G. Hunter, B.I. Gramatikov; Simplified Polarization Birefringence Technique for Detecting Eye Fixation . Invest. Ophthalmol. Vis. Sci. 2004;45(13):3489.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Purpose: Previously described methods of birefringence–based detection of eye fixation have successfully utilized the form birefringence of the Henle fibers radiating from the fovea, but these methods require an optical scanning system and may be adversely affected by the amplitude and orientation of the corneal birefringence. An eye fixation detection method without these limitations is needed. Methods: Poincaré sphere analyses are performed of the changes in the polarization state of polarized near–infrared light reflected from the ocular fundus in the auto–conjugate arrangement that occurs when an eye focuses in the plane of the light source. Polarization changes are produced by the corneal birefringence, by the form birefringence of the Henle fibers and retinal nerve fibers, and upon reflection from the fundus. Alternative polarization states of the incoming light are also analyzed, as well as alternative methods of detecting the reflected light. Results: Of numerous designs for a no–moving–parts eye fixation detector, the simplest and most promising design uses four fixed spots of linearly polarized light aligned with the two "bright" and two "dark" arms of the polarization cross centered on the fovea. Differential polarization detection is used to measure the S3 component of the Stokes vector representaton of the polarization state of the reflected light from each spot. By subtracting the "dark" arm signals from the "bright" arm signals, and then by dividing by the sum of all four signals, a normalized signal is obtained that is sensitive to foveal fixation and is insensitive to overall fundus reflectance. With proper arrangement of light source(s) and detectors, interference from corneal birefringence, over the reported range of normal values of amplitude and orientation, is minimal. Conclusions: A no–moving–parts eye fixation detector is designed that is only minimally influenced by the amplitude and orientation of corneal birefringence.
This PDF is available to Subscribers Only