Automated Quantification of Retinal Microvasculature from OCT Angiography Using Dictionary-Based Vessel Segmentation

Abstract

Investigations in how the retinal microvasculature correlates with ophthalmological conditions necessitate a method for measuring the microvasculature. Optical coherence tomography angiography (OCTA) depicts the superficial and the deep layer of the retina, but quantification of the microvascular network is still needed. Here, we propose an automatic quantitative analysis of the retinal microvasculature. We use a dictionary-based segmentation to detect larger vessels and capillaries in the retina and we extract features such as densities and vessel radius. The method is validated on repeated OCTA scans from healthy subjects, and we observe high intraclass correlation coe cients and high agreement in a Bland-Altman analysis. The quantification method is also applied to pre-and postoperative scans of cataract patients. Here, we observe a higher variation between the measurements, which can be explained by the greater variation in scan quality. Statistical tests of both the healthy subjects and cataract patients show that our method is able to di↵erentiate subjects based on the extracted microvascular features.

Automatic Segmentation and Classification Methods Using Optical Coherence Tomography Angiography (OCTA): A Review and Handbook

“…In the case of OCTA image segmentation, the majority of the analyzed studies used pixel intensity as a way to group together objects, using common methods such as k-means clustering [63][64][65], or other clustering algorithms such as fuzzy c-means clustering [66] and a modified CLIQUE clustering technique [67]. An interesting study that used local features for clustering and not pixel intensity is a method by Engberg et al [68] which was based on building a dictionary using pre-annotated data and then processing the unseen images by computing the similarity/dissimilarity.…”

“…Clustering methods were employed in two clinical applications: general eye vasculature segmentation and choroidal neovascularization (CNV)/Choriocapillaris segmentation. The study by Engberg et al [68] was a rare study that provided a quantitative validation of general eye vessel segmentation, even though only one image was used for validation. On this image, the DSC was equal to 0.82 for larger vessels and 0.71 for capillaries.…”

Automated Quantification of Macular Vasculature Changes from OCTA Images of Hematologic Patients

“…To detect larger vessels (arterioles and venules), background and capillaries we are using a dictionary-based segmentation approach proposed in [15]. Here, it is assumed that the appearance of the vascular network can be represented as a combination of characteristic image features.…”

“…The approach is described in detail in [16]. We use the same parameters and dictionary as in [15], although we here apply an initial adaptive histogram equalization to improve the contrast of the images. The DRL consists only of capillaries, so we combine the detected capillaries and larger vessels in this layer into one class.…”

Dictionary-based Method for Vascular Segmentation for OCTA Images

“…Instead, we propose to use the dictionary-based segmentation method from [2,3,4], where the segmentation model is learned from annotated training data. We have used this method for segmenting retinal microvasculature from OCTA images in [9,10,11]. In the segmentation part, we assign the dictionary to an input image (S1) and then compute pixel-wise probabilities of the labels using the dictionary labels (S2).…”