Purchase this article with an account.
Boris I Gramatikov, Kristina Irsch, Yi-Kai Wu, David L Guyton; New Pediatric Vision Screener – Data Analysis and Validation. Invest. Ophthalmol. Vis. Sci. 2016;57(12):3094.
Download citation file:
© 2017 Association for Research in Vision and Ophthalmology.
We have developed and described an improved pediatric vision screener (PVS) that can reliably detect central fixation (CF), eye alignment and focus. It uses the birefringence of the human fovea and identifies risk factors for amblyopia, namely eye misalignment and defocus. An early version of the PVS from our lab was tested at Children’s Hospital, Boston. It is being commercialized as an instrument that detects eye alignment but not defocus (REBIScan, Boston, MA). Meanwhile, development of retinal birefringence scanning (RBS) continued in our lab, resulting ultimately in an improved PVS that combines polarization-modulated RBS (PMRBS) for detecting strabismus, with technology for assessing proper focus of both eyes simultaneously. PMRBS is an optimized upgrade of RBS, based upon our theoretical and experimental research and computer modeling, to yield high signals across the entire population. In addition, using phase-shift-subtraction, the new PVS eliminated the need for initial background measurement.
We studied 18 patients with known vision abnormalities, and 19 controls with proven lack of vision issues.Calibration for CF. RBS in the foveal region results in the generation of five distinct frequencies in the signal obtained from the returned light. Calibration for CF was performed in a preliminary study, based on data from five normal volunteers. Two algorithms were developed for classification into one of the two classes – CF vs para-CF, using the available signal power measurements: a crawling threshold and discriminant analysis.Calibration for focus detection (FD) was based on a normalized focus signal and determining monocular focus curves: the normalized focus signal is plotted while ophthalmic trial lenses are placed in front of the eye, simulating refractive errors. This enables us to find a threshold signal level under which the subject fails the focus screen (within ±1D), thus identifying those subjects where the amount of optical blur is abnormal.
Both CF and FD detection criteria worked robustly and allowed reliable separation between normal test subjects and symptomatic subjects. The sensitivity of the instrument was 100% for both CF and FD. The specificity was 100% for CF and 89.5% for FD. The overall sensitivity was 100% and the overall specificity was 94.7%.
We believe that the PVS design and the analysis methods employed will prove valuable for mass screening of children.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.
This PDF is available to Subscribers Only