Abstract
Purpose :
Optical coherence tomography angiography (OCTA) provides non-invasive imaging of retinal capillary network and may be useful for detection of vascular morphological alterations due to diseases. The purpose of the current study is to report an image analysis method for quantification of retinal vessel tortuosity in healthy and sickle cell retinopathy (SCR) subjects based on OCTA images.
Methods :
OCTA was performed in one eye of 5 healthy control (NC) and 5 SCR subjects. Images of the superficial retinal capillary network were generated in a 3 mm × 3 mm region centered on the fovea. A dedicated customized software algorithm was developed for assessment of tortuosity in retinal vessels. K-means clustering algorithm and morphological image processing were used for vessel segmentation. Vessel centerlines between bifurcation points were extracted using distance transformations. A vessel tortuosity index (VTI) was derived using variation of in-plane curvature of points along the vessel centerline, number of inflection points (twists) and the magnitude of curvature. The method was validated using images of sinusoidal waves with variable amplitude and frequency. VTI measurements were compared in the 2 groups of subjects using unpaired t- test. Significant was accepted at P < 0.05.
Results :
As expected, for the sinusoidal waves at a constant frequency, VTI increased with increasing the magnitude. Similarly, for the sinusoidal waves at a constant magnitude, VTI increased with increasing the frequency. There was no statistically significant age difference between the NC (38±12 years) and SCR subjects (48±12 years) (P=0.2). Compiled data yielded a total of VTI measurements in 181 and 154 vessels in NC and SCR subjects, respectively. VTI in SCR (0.5±0.3) was significantly higher than NC (0.3±0.2) subjects (P < 0.001).
Conclusions :
A novel method for assessment of tortuosity of retinal vessels in OCTA was demonstrated. This method can be useful for detection of retinal vascular tortuosity alterations due to systemic and ocular diseases.
This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.