Automated retinal layer segmentation and characterization

Jonathan Luisi; David Briley; Adam Boretsky; Massoud Motamedi

doi:10.1117/12.2057814

19 May 2014 Automated retinal layer segmentation and characterization

Jonathan Luisi, David Briley, Adam Boretsky, Massoud Motamedi

Proceedings Volume 9155, Translational Biophotonics; 91551N (2014) https://doi.org/10.1117/12.2057814
Event: SPIE Translational Biophotonics, 2014, Houston, Texas, United States

Abstract

Spectral Domain Optical Coherence Tomography (SD-OCT) is a valuable diagnostic tool in both clinical and research settings. The depth-resolved intensity profiles generated by light backscattered from discrete layers of the retina provide a non-invasive method of investigating progressive diseases and injury within the eye. This study demonstrates the application of steerable convolution filters capable of automatically separating gradient orientations to identify edges and delineate tissue boundaries. The edge maps were recombined to measure thickness of individual retinal layers. This technique was successfully applied to longitudinally monitor changes in retinal morphology in a mouse model of laser-induced choroidal neovascularization (CNV) and human data from age-related macular degeneration patients. The steerable filters allow for direct segmentation of noisy images, while novel recombination of weaker segmentations allow for denoising post-segmentation. The segmentation before denoising strategy allows the rapid detection of thin retinal layers even under suboptimal imaging conditions.

Citation Download Citation

Jonathan Luisi, David Briley, Adam Boretsky, and Massoud Motamedi "Automated retinal layer segmentation and characterization", Proc. SPIE 9155, Translational Biophotonics, 91551N (19 May 2014); https://doi.org/10.1117/12.2057814

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available