Abstract
Purpose :
Clinical study demonstrates non-vascular OCTA signal from pathological retinas, correlated with hyperreflective fluid seen on structural OCT that is distinguished from hyporeflective fluid. This study is to test the hypothesis that the non-vascular OCTA signal is generated by the motion of particulate matter suspended in intraretinal fluids [ie, suspended scattering particles in motion (SPPiM)], using a lab SD-OCT with similar specs to commercial systems and custom designed fluid scattering phantoms.
Methods :
Scattering phantoms were fabricated to simulate hyperreflective fluids within retinal tissue background. Solidified 25% gelatin blended with 0.04% highly scattering TiO2 particles was used to simulate the tissue background. In this background, a fluid reservoir was made to hold the viscous scattering fluids with varying strengths of stiffness. The scattering fluid was made by mixing water and gelatin with 1% scattering intralipid particles that simulate the lipid and protein macro-molecules leaked from vulnerable vessels. To simulate the visco-elastic fluid, 0%, 1%, 2%, 4%, 6%, 7%, 8%, 9%, 10%, 12% and 15% gelatin were each mixed with 1% intralipid. In addition, two separate fluid phantoms with 0% intralipid but 2% and 10% gelatin were made, simulating hyporeflective fluid. A lab SD-OCT with 7µm axial and 15µm lateral resolution was used to image the phantoms. OMAG and SSADA algorithms were used to generate OCTA images, and maximum projection images were used to quantify signal to noise ratio (SNR). During imaging, the phantoms were still, meaning flow was not possible.
Results :
With 0% intralipid phantom, no OCTA signals were observed. OCTA signals were generated by all the fluid phantoms containing 1% intralipid. With the increase of the gelatin concentration, the strength of OCTA signals was gradually decreased, and eventually diminished with 25% gelatin concentration. For the same phantom, OCTA signal generated by OMAG is stronger and with higher SNR than by SSADA. See Figure.
Conclusions :
The SPPiM in the hyperreflective fluid generates Brownian-like signal detectable by OCTA system. The strength of the non-flow OCTA signal is dependent on visco-elastic property of the hyperreflective fluid, i.e. higher visco-elastic fluid gives weaker OCTA signal. OMAG algorithm is more sensitive to the SPPiM than SSADA.
This is an abstract that was submitted for the 2018 ARVO Annual Meeting, held in Honolulu, Hawaii, April 29 - May 3, 2018.