Foveal avascular zone segmentation in optical coherence tomography angiography images using a deep learning approach

Abstract

The purpose of this study was to introduce a new deep learning (DL) model for segmentation of the fovea avascular zone (FAZ) in en face optical coherence tomography angiography (OCTA) and compare the results with those of the device’s built-in software and manual measurements in healthy subjects and diabetic patients. In this retrospective study, FAZ borders were delineated in the inner retinal slab of 3 × 3 enface OCTA images of 131 eyes of 88 diabetic patients and 32 eyes of 18 healthy subjects. To train a deep convolutional neural network (CNN) model, 126 enface OCTA images (104 eyes with diabetic retinopathy and 22 normal eyes) were used as training/validation dataset. Then, the accuracy of the model was evaluated using a dataset consisting of OCTA images of 10 normal eyes and 27 eyes with diabetic retinopathy. The CNN model was based on Detectron2, an open-source modular object detection library. In addition, automated FAZ measurements were conducted using the device’s built-in commercial software, and manual FAZ delineation was performed using ImageJ software. Bland–Altman analysis was used to show 95% limit of agreement (95% LoA) between different methods. The mean dice similarity coefficient of the DL model was 0.94 ± 0.04 in the testing dataset. There was excellent agreement between automated, DL model and manual measurements of FAZ in healthy subjects (95% LoA of − 0.005 to 0.026 mm2 between automated and manual measurement and 0.000 to 0.009 mm2 between DL and manual FAZ area). In diabetic eyes, the agreement between DL and manual measurements was excellent (95% LoA of − 0.063 to 0.095), however, there was a poor agreement between the automated and manual method (95% LoA of − 0.186 to 0.331). The presence of diabetic macular edema and intraretinal cysts at the fovea were associated with erroneous FAZ measurements by the device’s built-in software. In conclusion, the DL model showed an excellent accuracy in detection of FAZ border in enfaces OCTA images of both diabetic patients and healthy subjects. The DL and manual measurements outperformed the automated measurements of the built-in software.

Retinal Microvasculature and Conjunctival Vessel Alterations in Patients With Systemic Lupus Erythematosus—An Optical Coherence Tomography Angiography Study

“…The scan speed was set to 70,000 A scans per second, the center wavelength was 840 nm, the bandwidth was 45 nm, the axial resolution was 5 mm, and the horizontal resolution rate was 22 μm. A B-scan (along the x-axis) in a 3 × 3-mm scan pattern with five repetitions of angiography was used to image at 216 raster positions (along the y-axis), focusing on the fovea, and the acquisition time was 3.9 s. We captured a 1080b scan (216y position × 5 position) at 270 frames per second ( 14 ). We obtained a 3 × 3 mm OCTA image through a series of four volume scans using two horizontal and two vertical rasters (933120a scans in total).…”

“…A similar length-based metric was used as a measure of blood vessel density. By taking the average of the skeletonized slab in the region of interest and considering the pixel distance (512 pixels per 3 mm), scaling results were used to calculate the density of blood vessels from the center of the macula to the edge of detection of the 3 * 3 mm image of the brightness gradient ( 14 , 15 ). Then a series of customized segmentation algorithms were used to process the image, including inversion, balance, and removal of background noise and non-vascular structures, to generate a binary image.…”

Differentiating features of OCT angiography in diabetic macular edema

“…Segmentation error correction of different retinal layers was manually performed as described elsewhere 16 , 17 . Foveal avascular zone was delineated in full retinal slab, SCP and DCP using a previously reported deep learning approach 18 .…”

Machine learning in optical coherence tomography angiography

“…. [45][46][47][48] Recent development and the implementation of quantitative OCTA features for machine learning classification indicate that OCTA images contain the necessary information to identify different retinopathies and perform disease staging. In principle, the CNN can automatically perform the feature extraction and classification, thereby reducing the burden for manual feature engineering.…”