Abstract
Purpose :
To describe retinal capillary microvasculature in young healthy participants using quantitative macular microvasculature density (MMD) and retinal microvasculature fractal dimension (MFD) in Swept Source-Optical Coherence Tomography Angiography (SS-OCTA). Secondary objective was to investigate the correlation between MMD and retinal MFD metrics at the level of the superficial capillary plexus (SCP).
Methods :
Young healthy volunteers were investigated using Plex Elite 9000 SS-OCTA (Carl Zeiss Meditec AG). Retinal examinations consisted of macular SS-OCTA 6X6 mm and 12X12mm acquisitions. Vessel density (VD) and vessel perfusion (VP) were calculated using automated quantitative MMD algorithm (Carl Zeiss Meditec AG) on the superficial capillary plexus (SCP) and the deep capillary plexus (DCP) of 6 x 6 mm SS-OCTA images. The retinal MFD were measured on the SCP of 6 x 6 mm and 12 x 12 mm SS-OCTA images using 4 different algorithms of the retinal image vasculature assessment software (RIVAS software): Fourier Fractal Dimension (FFD) and 3D Box-count Fractal dimension (FD) using gray scale OCTA and binary box-count FD on segmented image (FDbin) as well as on its skeletonized (FDskel) version.
Results :
Overall, 20 eyes of 20 healthy participants (65% female) were enrolled. The mean age was 26.44 ± 3.17 and mean axial length was 24.10 ± 0.85mm. At the SCP level, mean VD and VP were respectively 14.47 ± 0.43 (mm-1) and 0.45 ± 0.01 (unitless). Concerning retinal MFD of the SCP, mean FFD, 3D Boxcount FD, FDbin and FDskel were correspondingly 1.33 ± 0.04, 2.74 ± 0.01, 1.87 ± 0.01 and 1.68 ± 0.01. There was only weak correlation (Spearman) between VD and FDskel (r = 0.55) and VD with 3D Boxcount FD (r = 0.49).
Conclusions :
We found that retinal MFD and MMD measurements using SS-OCTA are weakly correlated. The significance of this interesting finding remains uncertain but will become clearer with larger numbers.
The retinal MFD in SS-OCTA could be an interesting quantitative metrics compared to other validated features. A normative database of fractal dimension could be of interest for future studies.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.