A New Texture-Based Segmentation Method for Optical Coherence Tomography Images Publisher Pubmed



Monemian M¹ ; Rabbani H¹ Show Affiliations
Source: Proceedings of the Annual International Conference of the IEEE Engineering in Medicine and Biology Society# EMBS Published:2019

Abstract

Optical Coherence Tomography (OCT) is an imaging modality which facilitates capturing pictures from biological organs like retina. Accurate segmentation and verification of OCT images leads to the identification and treatment of harmful retinal diseases such as glaucoma. The main fact used for segmentation in this paper is that a considerable number of boundary pixels have similar features from texture point-of-view. Thus, a novel low-complexity segmentation method for OCT images is proposed paying attention to the texture feature of pixels on the boundaries. The simulation results show that the proposed method provides acceptable values for mean signed and unsigned errors compared to the result of manual segmentation. © 2019 IEEE.