Abstract
Visible-light optical coherence tomography (vis-OCT) enables retinal oximetry by measuring the oxygen saturation of hemoglobin (sO2) from within individual retinal blood vessels. The sO2 calculation requires reliable estimation of the true spectrum of backscattered light from the posterior vessel wall. Unfortunately, subject motion and image noise make averaging from multiple A-lines at the same depth position challenging, and lead to inaccurate sO2 estimation. In this study, we developed an algorithm to reliably extract the backscattered light's spectrum. We used circumpapillary scanning to sample the vessels repeatedly at the same location. A combination of cross-correlation and graph-search based segmentation extracted the posterior wall locations. Using measurements from 100 B-scans as a gold standard, we demonstrated that our method achieved highly accurate measures of sO2 with minimal bias. In addition, we also investigated how the number of repeated measurements affects the accuracy of sO2 measurement. Our method sets the stage for large-scale studies of retinal oxygenation in animals and humans.
Highlights
Visible-light optical coherence tomography adds functional information to OCT, such as providing oxygenation measurements in the living retina [1, 2]
We show that graph-search segmentation can be used to simultaneously help with two key tasks: (1) locating the vessel wall to extract the backscattered spectrum and (2) increasing the accuracy of the OCT signal when there is axial motion
3.1 Comparison of methods for OCT amplitude spectra extraction We compared four averaging methods to extract the spectrum from the posterior wall: (1) blind averaging with manually placed straight slab, (2) cross-correlation with manually placed manual straight slab (X-Corr), (3) cross-correlation with automatic graph-search (X-Corr + gold standard (GS)), and (4) cross-correlation with manual segmentation (X-Corr + Manual)
Summary
Visible-light optical coherence tomography (vis-OCT) adds functional information to OCT, such as providing oxygenation measurements in the living retina [1, 2]. In human studies and patient imaging [6, 7], vis-OCT may eventually identify oxygenation related biomarkers that correlate with progression of disease or a response to treatment. In spectroscopic OCT, the spectral interferogram is divided into sub-bands, which are individually processed to produce a set of reduced axial-resolution images. Conventional OCT uses the entire band of the spectral interferogram to form a single, high axial-resolution image. The axial-resolution is sacrificed in spectroscopic OCT, each image corresponds to a different wavelength band, giving the spectrum of the backscattered light that conventional OCT does not provide. The spectrum of the backscattered light can be used to calculate sO2 by fitting it against a model of light attenuation [10]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
