Different effect of media opacity on automated and manual measurement of foveal avascular zone of optical coherence tomography angiographies. Issue 6 (23rd July 2020)

• Record Type: Journal Article
• Authors: Zhang, Jinyu; Tang, Fang Yao; Cheung, Carol; Chen, Xiaolin; Chen, Haoyu
• Is Part Of: British journal of ophthalmology. Volume 105:Issue 6(2021)
• Page Start: 812
• Page End: 818
• Publication Date: 2020-07-23
• Subjects: Imaging -- Retina -- Glaucoma -- Diagnostic tests/Investigation -- Angiogenesis -- Macula -- Trauma; Ophthalmology -- Periodicals; 617.7
• DOI: 10.1136/bjophthalmol-2019-315780
• Languages: English
• ISSNs: 0007-1161

Abstract

Abstract : Background: Optical coherence tomography angiography (OCTA) provides not only visualisation but also quantitative measurement of foveal avascular zone (FAZ). Media opacity is common in elderly subjects with cataracts. This study aimed to investigate the impact of media opacity on automated and manual FAZ measurement. Methods: Cirrus 5000 OCTA and Triton OCTA were used to image FAZ using a 3×3 mm scanning protocol from 30 eyes of 30 healthy normal subjects. Media opacity was simulated with neutral-density filters (optical density (OD): 0.10–0.48 in Cirrus 5000 and 0.15–0.51 in Triton). Signal strength (SS) and signal strength intensity (SSI) were provided by the built-in software in Cirrus 5000 and Triton, respectively. FAZ area, perimeter and circularity were measured automatically using the built-in software as well as a customised MATLAB program. FAZ metrics were also measured manually. The correlations between the OD, SS/SSI and FAZ metrics were analysed using Spearman correlation. Results: Increased OD significantly correlated with decreased SS/SSI ( rs =−0.602 and −0.925, respectively, both p<0.001), decreased automated FAZ area ( rs =−0.344 and −0.766, respectively, both p<0.001), but increased manual FAZ area in both Cirrus 5000 and Triton ( rs =0.423 and 0.543, respectively, both p<0.001). Similar results were found for FAZ perimeter and circularity. There was a positive correlation between SS/SSI with the automated FAZ area but negative correlation with the manual FAZ area. Conclusions: The effect of media opacity on quantitative measurement of FAZ is different between automated and manual measurements. Cautions must be taken when interpreting FAZ measurement in eyes with media opacity.