Mathematical analysis of texture indicators for the segmentation of optical coherence tomography images

Abstract

Optical Coherence Tomography (OCT) is a non-invasive technology which facilitates the process of capturing images from light-scattering organs like retina. Retina is a layered structure each layer of which has its own morphological properties. The segmentation of retinal layers helps to identify retinal diseases. In this paper, a novel mathematical model is proposed which can extract boundary pixels located on the borders between layers. The new model uses texture properties of pixels to extract distinguishing characteristics for boundary pixels. It is explored that boundary pixels provide certain values for texture indicators leading to the existence of special relation between neighbor pixels' intensities. Using the new model which is based on Laplace distribution, it is possible to compute the probability of being a boundary pixel for each pixel. The numerical results show that the proposed model is capable of identifying retinal layers' boundaries in normal cases with acceptable accuracy.