Purchase this article with an account.
Justin V Migacz, Benjamin Musheyev, Eric Zhong, Oscar Otero, Iwona Gorczynska, Brennan Marsh-Armstrong, Robert J Zawadzki, Ravi Sankar Jonnal, John S Werner, Toco Yuen Ping Chui, Richard B Rosen; Quantitative Choriocapillaris Vessel Analysis of Age-Related Macular Degeneration Patients with Ultrahigh-Speed OCTA. Invest. Ophthalmol. Vis. Sci. 2021;62(8):2493.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
To quantitively assess, in-vivo, the choriocapillaris (CC) vessels of patients with age-related macular degeneration (AMD) in regions under and near pathological features such as drusen.
Imaging was performed with a custom swept-source OCTA prototype, operating at 1.65MHz A-scan rate. Between 2 and 7 volumes were acquired, centered at the fovea or a nearby region of interest, with a field of view of 2x2mm across. Each volume required a 1 second capture time. En face projections of the choriocapillaris (CC) and retinal vessels were generated for each volume, and the images were averaged to create a high-contrast angiogram. Images were then registered to angiograms from a commercial OCT system () to account for motion distortions. The lateral scale was adjusted using a separate eye length measurement. CC images were binarized and skeletonized using a previously published algorithm (PMID: 31646049). Image portions affected by projection artifacts from retinal vessels were removed from analysis and the vessel density (VD), the number of vessel branches per area, was calculated. These metrics were computed for areas outside and directly below manually segmented pathological features such as drusen and CNV.
Five dry AMD subjects, one suspected wet AMD subject, and 4 age-matched normal subjects were imaged and the CC vessel images skeletonized for quantitative analysis. Fig. 1 shows an example of an AMD subject with the commercial images (a-b), the equivalent images shown with our custom system (c-d), the skeleton segmentation of the CC vessels (e), and the CC vessel map with an overlayed vessel density color map (f). Drusen areas are outlined in magenta. Another example of an AMD subject with drusen and pseudodrusen is shown in Fig. 2, in which more obscuration of the underlying vessels is possible due to eye dryness or blurring caused by the drusen. The scalebars are 1mm.
The ultrahigh-speed OCTA system demonstrated clear visibility of most CC vessels in the CC layer, allowing for effective vessel quantification. Images with this system may provide more insight into the manifestation and progression of AMD pathology.
This is a 2021 ARVO Annual Meeting abstract.
This PDF is available to Subscribers Only