Quality Control for Doppler Optical Coherence Tomography of Retinal Blood Flow

Abstract

 

Purpose:

 

Doppler optical coherence tomography (OCT) with the circumpapillary double circular scan pattern is a novel method of measuring retinal blood flow. The major source of measurement error in this method is the variation in the Doppler angle due to eye motion during the scan. The purpose of this study was to develop quality criteria to improve the precision of blood flow measurement.

 

Methods:

 

Five normal subjects were recruited. Both eyes of each subject was scanned with a Fourier-domain OCT (RTVue by Optovue, Fremont, CA) using the experimental blood flow double ring (BFDR) scan pattern. The pattern consists of two concentric rings with diameters of 3.4 and 3.75mm centered on the optic disc, with each blood flow measurement based on 6 consecutive double circular scans recorded over 2 seconds. Ten measurements were made on each subject. The reflectance and Doppler data on the OCT images were used to identify retinal veins and their Doppler angles. The Doppler angle is the angle between the OCT beam and the vessel orientation minus 90º (normal incidence is defined as Doppler angle = 0º). Blood flow was then computed for each vein from the Doppler shift profile and Doppler angle. Motion error can be estimated by the coefficient of variation of the Doppler angle (CVDA) between the 6 sets of double circular scans. Retinal blood flow and CVDA was averaged or pooled over whole scan area.

 

Results:

 

Doppler scans with higher CVDA generated more widely varying blood flow measurements. The population standard deviation in scans with CVDA >= 0.3 was much higher than those with CVDA < 0.3 (18.3 vs 8.9 µl/min, p=0.099 by F test) though the population average is similar (49.6 vs 46.1 µl/min, p=0.64 by t-test)

 

Conclusions:

 

The variability of Doppler OCT retinal blood flow measurement is highly dependent on the variability of the Doppler angle. Scans with a CVDA >= 0.3 are probably unreliable and should generally not be selected. A study with larger population is needed to verify the results.  

 

Keywords: imaging methods (CT, FA, ICG, MRI, OCT, RTA, SLO, ultrasound) • imaging/image analysis: clinical • blood supply