Background:
The National Eye Institute reports that KC affects 1 in every 2000 Americans, and some studies suggest as high as 1 in every 500. Additionally, FFKC patients with mild–to–moderate KC can be misdiagnosed as optical defocus and astigmatism. Recent availability of intervention treatments, such as Intacs, for mild–to–moderate KC offers the potential of improved vision, increased corneal strength, and possibly retardation of disease progression. Therefore, early–stage KC diagnosis becomes important. The development and use of a screening instrument for indications of FFKC and KC would be advantageous for early detection and estimation of the incidence rate in larger and other populations.
Purpose:
To develop an effective screening device for forme fruste keratoconus (FFKC) and keratoconus (KC).
Methods:
Experienced clinicians can detect FFKC from an irregular reflex during a retinoscopy or a direct ophthalmoscopy exam. We adopt a photorefraction (PR) technique that involves similar optical principles. PR has been used extensively for pediatric vision screening. It commonly uses a camera with a flash–source to photograph retinal reflex patterns through the pupil. Computer simulation of PR images under various instrumentation and ocular conditions was used to optimize the design of the KC instrument. We used 20 customized eye models that were based on the corneal topographies of 10 KC and 10 pre–LASIK patients.
Results:
PR simulations results show the spatial irradiance distributions for normal, KC, and refractice–error eyes. The figure shows an example result of a KC and an astigmatic eye on the left and right, respectively. A comparison of the coaxial images on top and eccentric images below shows that off–axis measurements offer greater sensitivity for KC. The eccentricity–angle sensitivity for KC detection and differentiation from refractive errors is clearly observed.
Conclusions:
The multiple–eccentricity and orientation PR design provides a promising future in KC screening. Further investigation in FFKC is required.
Keywords: keratoconus • imaging/image analysis: non-clinical • computational modeling