Digital analysis of high resolution fundus images

Abstract

Fundus photography is a common procedure in ophthalmology providing high resolution images of the inside back portion of the eye to diagnose diseases of the retina and the optic nerve, and to record their progress over time. In many instances, objective, quantitative, reproducible and reliable interpretation of fundus images requires their computerized analysis. A comprehensive system for digital analysis of high resolution fundus images has to address virtuallly all engineering aspects of medical image processing: restoration, segmentation, pattern recognition, and registration. Based on the specific application of investigating the tapetal-like reflex, a retinal reflection uniquely present in carriers of X-linked retinitis pigmentosa (XLRP), novel approaches to the various stages of image processing are presented, and applications in other areas of medical diagnostics are outlined.