Dictionary-based Method for Vascular Segmentation for OCTA Images

Abstract

Optical coherence tomography angiography (OCTA) is an imaging technique that allows for non-invasive investigation of the microvasculature in the retina. OCTA uses laser light reflectance to measure moving blood cells. Hereby, it visualizes the blood flow in the retina and can be used for determining regions with more or less blood flow. OCTA images contain the capillary network together with larger blood vessels, and in this paper we propose a method that segments larger vessels, capillaries and background. The segmentation is obtained using a dictionary-based machine learning method that requires training data to learn the parameters of the segmentation model. Here, we give a detailed description of how the method is applied to OCTA images, and we demonstrate how it robustly labels capillaries and blood vessels and hereby provides the basis for quantifying retinal blood flow.