Attenuation correction assisted automatic segmentation for assessing choroidal thickness and vasculature with swept-source OCT

Abstract

Swept source optical coherence tomography (SS-OCT) is being used more widely in clinical studies to investigate the choroid due to its deeper penetration under the retinal pigment epithelium and improved image quality compared with spectral domain OCT. However, automatic methods to reliably assess choroidal thickness and vasculature are still limited. This paper reports an approach that applies attenuation correction on SS-OCT structural scans to facilitate accurate automatic segmentation of the choroid and provides visualization of the choroidal vasculature without the necessity of OCT angiography. After attenuation correction, enhanced interlayer contrast at the choroidal-scleral interface was observed (from 0.13 ± 0.05 to 0.29 ± 0.10; P < 0.001). An algorithm that segmented the choroid from attenuation compensated B-scans achieved significantly higher accuracy when compared with an automated segmentation performed on regular OCT scans (91.8 ± 3.7% vs.74.5 ± 8.0%; P < 0.01). After attenuation correction, en face images of choroidal vessels were achieved with fewer artifacts from retinal vessels. Measurements of mean choroidal thickness and vessel density showed high repeatability. The attenuation correction assisted segmentation of the choroid and visualization of the choroidal vasculature will be helpful in studying the quantitative changes that occur in a myriad of diseases involving the choroid such as age-related macular degeneration, polypoidal choroidal vasculopathy, pathologic myopia, central serous chorioretinopathy, and inflammatory eye conditions.

The Effect of 0.01% Atropine Eye Drops Combined with Auricular Acupoint Stimulation on Myopia Progression and Choroidal Thickness: The Preliminary Results from a Randomized Controlled Trial

“…The analysis for OCT imaging in terms of choroidal volume was processed in three steps: attenuation compensation, choroid segmentation, and en-face mapping. The attenuation compensation algorithm was proposed [28] to enhance the visibility of the sclera-choroid interface and to minimise the projection shadows of retinal vessels in previous publications [29,30]. We then employed the U-shape convolutional network (U-Net) [31,32] to automatically segment the choroid in OCT.…”

Review of methods and applications of attenuation coefficient measurements with optical coherence tomography

“…However, due to the physiological location of choroid, light is mostly attenuated by the RPE before reaching choroidal structures, causing low contrast in these regions on OCT images. To address this issue, Zhou et al 30 proposed an AC-based correction method to improve the contrast, which allowed successful segmentation of the choroid layer and imaging of choroidal vasculatures. In their approach, the DR method was used to obtain a pixelwise depth-reflectivity profile.…”

“…The paper reports excellent intravisit repeatability for both choroidal thickness maps and vasculature by calculating the coefficient of correlation (CV) for mean thickness (CV ¼ 1.7 AE 0.7%) and vessel density (CV ¼ 0.41 AE 0.18%). 30 However, a limitation of this method is that it can only correctly represent choroidal vessel patterns in healthy eyes with normal RPE, because the minimum intensity projection approach used to visualize choroidal vessel is based on the light scattering in RPE layer. In disease states, such as glaucoma, there is a notable reduce in some regions of the RPE layer, which may affect the accuracy in choroidal vasculature detection with the proposed method.…”

Robust deep learning method for choroidal vessel segmentation on swept source optical coherence tomography images

“…For SS-OCT segmentation, Li Zhang et al [5] proposed to use shape model of the Bruch's membrane with soft-constraint graph-search for segmenting choroidal boundaries. Zhou et al [6] proposed to use attenuation correction approach to denoise the input SS-OCT images which thereby improved the segmentation of the choroidal boundaries. Although these automated segmentation methods have demonstrated accurate results.…”