A New Choroidal Layer Segmentation Method Based on Graph Theory and Directional Curvelet Transform in OCT Images

Abstract

Background: Automatic segmentation of the choroid on Optical Coherence Tomography (OCT) images helps ophthalmologists in diagnosing eye pathologies. In nature, it is not as exhausting as manual segmentation and does not depend on human errors. In this study, sixty EDI-OCT (Enhanced Depth Imaging Optical Coherence Tomography) images of both normal and abnormal eyes gathered from Isfahan Feiz Medical Center were used. The data were manually segmented by a retinal ophthalmologist to draw comparison with the proposed automatic segmentation technique.Methods: In this study, curvelet transform based KSVD dictionary learning and Lucy-Richardson algorithm was used to remove speckle noise from OCT images. The Outer / Inner Choroidal Boundaries (O/ICB) were determined utilizing graph theory. The area between ICB and OCB was considered as choroidal region.Results: The method was evaluated on the EDI-OCT images and the average Dice Similarity Coefficient (DSC) was calculated to be 92.14% ± 3.30% between automatic and manual segmented regions. Moreover, by applying the latest presented open-source algorithm by Mazzaferri et al on our dataset the mean DSC was calculated to be 55.75% ± 14.54%. Conclusions: A significant similarity was observed between automatic and manual segmentations, in both normal and abnormal eyes. Automatic segmentation of the choroidal layer could be also utilized in large-scale quantitative studies of the choroid.