Affiliation:
1. The Second Hospital of Hebei Medical University
Abstract
Abstract
PURPOSE
To investigate the reliability, repeatability and consistency of choroidal vascularity index (CVI) measurements provided by an artificial intelligence-based software in normal subjects.
METHODS
One hundred and twenty-five eyes of 125 healthy volunteers were included in this prospective study. The repeatability of the measurements was analyzed based on the in-built ETDRS segmentation with 3D-CVI was obtained automatically on the segmented diagram (CVI1). Ten eyes from this population were selected to study agreement between CVIs obtained from SS-OCT and that from SD-OCT (CVI2) based on the 6*6mm square overlayed over the fundus image analyzed by intraclass correlation coefficient. The influencing factors including age, gender, partition, axial length and spherical equivalent on CVI was further investigated.
RESULTS
Intraclass correlation coefficient between different measurements was 0.934 (95% CI: 0.812–0.956). Intraclass correlation coefficient between centered and CVI1 and CVI2 was 0.887 (95% CI: 0.796–0.938, P value < 0.001). The mean difference between CVI1 and CVI2 was 0.063 ± 0.017. CVI measured with SS-OCTA showed stronger correlations with axial length and age but not correlated with gender.
CONCLUSION
There is good agreement between CVIs obtained from the built-in software that requires less timing in manual quantification. Studies investigating choroidal vascularity can be standardized by the AI-based CVI analyze software.
Publisher
Research Square Platform LLC