A quantitative comparison of five optical coherence tomography angiography systems in clinical performance

Abstract

Aim: To compare the clinical performance of 4 spectral-domain (SD) optical coherence tomography angiography (OCTA) systems: AngioVue™, AngioPlex™, Spectralis^® OCTA, AngioScan, and 1 swept-source (SS) OCTA SS OCT Angio™.

Methods: Twenty-seven undilated right eyes of 27 participants underwent OCTA examination using five different systems respectively for both 3×3 and 6×6 mm² scan pattern (Spectralis OCTA for 3×3 mm² scan only). Image quality, including vessel valid visibility and the number of motion artifacts, and acquisition time were evaluated. Repeated measures analysis of variance (ANOVA) with Bonferroni's post-test and Friedman test with Dunn's post-test were used to compare measurements.

Results: The age of the subjects was 28.19±5.55y (range, 23-49y). The spherical equivalent refraction was -2.55±1.84 D (range, 0.00 to -5.25 D). Significant difference was observed in the evaluation of vessel valid visibility (AngioVue the highest: 0.111±0.031 for 3×3 mm² scan and 0.128±0.020 for 6×6 mm² scan), number of motion artifacts (AngioVue the fewest: 0.778±1.086 for 3×3 mm² scan and 0.333±0.620 for 6×6 mm² scan) and acquisition time (AngioPlex the shortest: 8.537±1.921s for 3×3 mm² scan and 8.298±1.741s for 6×6 mm² scan; all P<0.001).

Conclusion: There is poor agreement of measurements among systems. AngioVue provides images with the highest vessel valid visibility and the fewest motion artifacts. AngioPlex achieves the shortest acquisition.

Keywords: acquisition time; devices; motion artifacts; optical coherence tomography angiography; vessel valid visibility.

Figure 1. Illustration of default layer segmentation of AngioVue, AngioPlex, Spectralis OCTA, AngioScan, and SS OCT Angio

Spectral-domain optical coherence tomography (SD-OCT) image shows the structure of a healthy human retina (aspect ratio 10:1). All five devices share similar layer definitions of superficial retinal layer (SRL; red), deep retinal layer (DRL; green) and outer retinal layer (ORL; blue) in existence of small deviations. Meanwhile, slab definitions differ greatly at vitreous and choroid level: AngioVue, AngioPlex, and SS OCT Angio segment choriocapillaris (CC; yellow), AngioPlex and AngioScan segment choroid (orange) and AngioPlex segments vitreoretinal interface (VRI; purple).

Figure 2. Illustration of evaluating vessel valid visibility

A: Original angiographic image of SRL (scan size, 3×3 mm²); B: The vesselness measure of A using Frangi filter; C: Binary vessel mask by thresholding B to distinguish vessels from noise; D: Skeleton of C; E: Effective vessel with length longer than threshold; F: Calculated vessels in a ring shape ROI around the fovea.

Figure 3. Illustration of detecting motion artifacts

A: Original angiographic image of SRL (scan size, 3×3 mm²); B: Plot of correlation for each horizontal line with respect to its next neighbor line, 0.4 set as a threshold. In this sample image, 4 outliers less than 0.4 were detected and marked with arrows; C: Angiogram with motion artifacts are marked as red. The location of artifacts correspond to B.

Figure 4. Results of vessel valid visibility measure of 3×3 mm² scans of a same subject

A-E: Original SRL angiograms overlaid with a ring showing ROI, with inner diameter of 0.36 mm (red) and outer diameter of 1.00 mm (green) of 5 systems; F-J: Images after vesselness filter and binary vessel mask of A-E; K-O: Calculated vessels in ROI after skeletonization and thresholding of F-J.

Figure 5. Results of vessel valid visibility measure of 6×6 mm² scans of a same subject

A-D: Original SRL angiograms overlaid with ring showing ROI, with inner diameter of 0.72 mm (red) and outer diameter of 2.00 mm (green) of 4 systems; E-H: Images after vesselness filter and binary vessel mask of (A-D);. I-L: Calculated vessels in ROI after skeletonization and thresholding of E-H.

Figure 6. Results of motion artifacts detection of 3×3 mm² scans of a same subject

A-E: Original SRL angiographic images obtained from different systems; F-J: Angiogram overlaid with red line marking detected motion artifacts.

Figure 7. Results of motion artifacts detection of 6×6 mm² scans of a same subject

A-D: Original SRL angiographic images obtained from different systems; E-H: Angiogram overlaid with red line marking detected motion artifacts.