Development, investigation, and software implementation of a new mathematical method for automating the analysis of corneal endothelium images

Abstract

A new mathematical method is presented for processing and analyzing microscopic images of the epithelium posterius (endothelium) in the human eye cornea. As initial data, we use images of endothelial cells that are obtained noninvasively using a confocal microscope. Each image is a black-and-white photograph of the endothelial layer in the eye cornea with a bright-lighted center and dark periphery. The endothelial cells depicted in the images are light-colored, mostly hexagonal figures with a dark periphery. The proposed method is intended for detecting endothelial cells and determining some of their morphometric characteristics. The method yields an image with marked cells of hexagonal, pentagonal, and other (tetragonal and heptagonal) shapes, as well as a set of their characteristics. The proposed approach combines methods of mathematical morphology and image segmentation. The software implementation of the developed method, as well as the results of its experimental testing, is presented.