Abstract
Purpose :
To quantify red blood cell velocity and velocity pulsatility in retinal capillaries. Cells can be visualized in the retinal capillaries but typical imaging techniques to study blood flow velocity require high frame rates or can only calculate its average velocity. We developed a semi-automated technique to assess red blood cell velocity and study its variation in the retinal capillaries.
Methods :
We used a dual channel Adaptive Optics Scanning Laser Ophthalmoscope (AOSLO) to image the retina with two wavelengths (785 and 810 nm). A spatial shift of the two beams on the retina produces a temporal separation between the two images of a given location that is shorter than the frame rate of the videos. With a spatial separation of 71 lines we introduced a temporal separation of 4.4 ms between the two channels and recorded videos of 100 frames at 30 Hz in three young healthy subjects. An algorithm was developed to detect the image of the cells travelling through a blood vessel in one of the channels and extract the surrounding area from the corresponding location in the other channel. The average image of the cells detected in a given vessel was computed and the change in position between the channels was used to calculate their speed.
Results :
The average cell velocity was calculated using the average image of all the cells detected in the whole video (of about 3 seconds). Measured average speeds in foveal and parafoveal capillaries ranged from 0.4 to 4.1 mm/s. In some capillaries, the cell velocity could be calculated in groups of frames or even on a frame by frame basis showing a change in velocity consistent with the cardiac cycle as can be seen in the figure.
Conclusions :
A dual channel AOSLO can be used to measure the red blood cell velocity in capillaries in the retina. The temporal resolution is sufficient to quantify changes in velocity across the cardiac cycle and allow measurement of pulsatility of velocity within the capillaries.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.