Purchase this article with an account.
Abdolreza Rashno, Keshab K Parhi, Behzad Nazari, Saeed Sadri, Hossein Rabbani,, Paul Drayna, Dara D Koozekanani; Automated intra-retinal, sub-retinal and sub-RPE cyst regions segmentation in age-related macular degeneration (AMD) subjects. Invest. Ophthalmol. Vis. Sci. 2017;58(8):397.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Exudative age related macular degeneration (EAMD) is characterized by the growth of abnormal blood vessels from the choroidal vasculature, and the resultant fluid leakage into the intra-retinal, sub-retinal, and sub-retinal pigment epithelium (RPE) spaces. The standard treatment for this condition is guided by the presence and quantity of this fluid. The fluid quantity cannot be routinely measured in clinical practice, because commercial algorithms do not directly detect fluid. Automated detection and segmentation of this fluid to address this, an automated method for three mentioned cyst regions for EAMD subjects is presented in this work.
The proposed segmentation method is described as follows: (1) the inner limiting membrane (ILM) and the retinal pigment epithelium (RPE) layers are segmented using a proposed method based on graph shortest path in the neutrosophic (NS) domain. A “flattened” RPE boundary is calculated such that all three types of cyst regions are located above it. (2) Seed points for Object (fluid regions) and Background (tissue) are initialized by the proposed automated method. 3) A new cost function is proposed in the kernel space, and then it is minimized with graph cut algorithms, leading to a binary segmentation. Representative results of the proposed segmentation method for three types of fluid (intra-retinal, sub-retinal, and sub-RPE) are shown in Fig. 1.
Segmentation results of the proposed method are compared with manual segmentation results by two ophthalmologists in both a publicly available dataset (Optima) and a local dataset from our University (UMN) clinic. The proposed algorithm achieves sensitivity of 87.85 % and 77.9 %, precision of 85.79% and 90.49%, and dice coefficients of 81.22% and 78.97 % for the Optima, and the UMN datasets, respectively.
Our segmentation method automatically and very accurately segments fluid/cyst regions in OCT images of EAMD subjects. Accurate segmentation of fluid in different tissue planes allows a quantitative measure of treatment effect with anti-VEGF agents, which may help guide treatment regimens. This may also facilitate the investigation of OCT biomarkers in EAMD images.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.
Fig. 1. The results of the proposed cyst segmentation scheme for three classes of fluid location in sample B-scans: (a) intera-retinal, (b) sub-retinal and (c) sub-RPE fluid.
This PDF is available to Subscribers Only