Hybrid OCT-OCTA Vessel Visualization for Projection-Free Display of the Intermediate and Deep Retinal Plexuses

Abstract

Purpose : In optical coherence tomography angiography (OCTA), projection artifacts (also shadowing artifacts or decorrelation tails) cause superficial retinal vasculature to appear in the segmented OCTA images of the intermediate and deep retinal plexuses. The projections of these larger superficial vessels obfuscate the unique vascular patterning of the deeper layers. Several algorithms have been proposed to remove these shadows. However, by removing the projected vessels these approaches also decrease the contrast of, or fully eliminate, the underlying vasculature, thereby introducing discontinuities. Thus, in effect, one artifact is replaced with another. The purpose of this study is to develop a projection artifact removal scheme that overcomes this limitation and fully preserves the intermediate/deep retinal vasculature.

Methods : Amplitude decorrelation based OCTA data was collected using a 1050nm swept source OCT system. A hybrid OCT-OCTA vessel visualization scheme (described in Figure 1) was developed. The key advantage of our proposed scheme is that the OCT signal is used to adaptively remove the OCTA signal only in the intercapillary regions.

Results : The hybrid OCT-OCTA approach reduces projection artifacts to a negligible level in the intermediate and deep plexuses without introducing vessel discontinuity artifacts (Figure 2).

Conclusions : Hybrid OCT-OCTA vessel visualization is a promising approach to visualize the unique patterning of the intermediate and deep plexuses, and is likely to be particularly important in diseases such as diabetic retinopathy where it is desirable to separately analyze the retinal plexuses.

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.

 

Flow chart of the hybrid OCT-OCTA vessel visualization algorithm alongside representative en face projections of the deep retinal plexus using (A) OCT, (B) OCTA, and (C) hybrid OCT-OCTA volumes. (D) Volumetric rendering showing the superficial/intermediate/deep plexus in red/yellow/purple color.

 

1st column: OCT; 2nd column: OCTA; 3rd column: hybrid OCT-OCTA. (A) En face projections over the intermediate retinal plexus. (B) En face projections over the deep retinal plexus. (C) B-scans extracted at the dashed lines in B. Note the decorrelation tails in C2 (green arrows). In C3, decorrelation tails are absent, and vessels perpendicular to the B-scan appear as circles. Note also that the different vascular plexuses can easily be distinguished in B-scan C3 (right arrows).