To evaluate the role of ensemble learning techniques with deep learning in classifying diabetic retinopathy (DR) in optical coherence tomography angiography (OCTA) images and their corresponding co-registered structural images. Methods: A total of 463 volumes from 380 eyes were acquired using the 3 × 3-mm OCTA protocol on the Zeiss Plex Elite system. Enface images of the superficial and deep capillary plexus were exported from both the optical coherence tomography and OCTA data. Component neural networks were constructed using single data-types and fine-tuned using VGG19, ResNet50, and DenseNet architectures pretrained on ImageNet weights. These networks were then ensembled using majority soft voting and stacking techniques. Results were compared with a classifier using manually engineered features. Class activation maps (CAMs) were created using the original CAM algorithm and Grad-CAM. Results: The networks trained with the VGG19 architecture outperformed the networks trained on deeper architectures. Ensemble networks constructed using the four finetuned VGG19 architectures achieved accuracies of 0.92 and 0.90 for the majority soft voting and stacking methods respectively. Both ensemble methods outperformed the highest single data-type network and the network trained on hand-crafted features. Grad-CAM was shown to more accurately highlight areas of disease. Conclusions: Ensemble learning increases the predictive accuracy of CNNs for classifying referable DR on OCTA datasets. Translational Relevance: Because the diagnostic accuracy of OCTA images is shown to be greater than the manually extracted features currently used in the literature, the proposed methods may be beneficial toward developing clinically valuable solutions for DR diagnoses.

This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Optical coherence tomography angiography (OCT-A) permits visualization of the changes to the retinal circulation due to diabetic retinopathy (DR), a microvascular complication of diabetes. We demonstrate accurate segmentation of the vascular morphology for the superficial capillary plexus (SCP) and deep vascular complex (DVC) using a convolutional neural network (CNN) for quantitative analysis. Methods: The main CNN training dataset consisted of retinal OCT-A with a 6 × 6-mm field of view (FOV), acquired using a Zeiss PlexElite. Multiple-volume acquisition and averaging enhanced the vasculature contrast used for constructing the ground truth for neural network training. We used transfer learning from a CNN trained on smaller FOVs of the SCP acquired using different OCT instruments. Quantitative analysis of perfusion was performed on the resulting automated vasculature segmentations in representative patients with DR. Results: The automated segmentations of the OCT-A images maintained the distinct morphologies of the SCP and DVC. The network segmented the SCP with an accuracy and Dice index of 0.8599 and 0.8618, respectively, and 0.7986 and 0.8139, respectively, for the DVC. The inter-rater comparisons for the SCP had an accuracy and Dice index of 0.8300 and 0.6700, respectively, and 0.6874 and 0.7416, respectively, for the DVC. Conclusions: Transfer learning reduces the amount of manually annotated images required while producing high-quality automatic segmentations of the SCP and DVC that exceed inter-rater comparisons. The resulting intercapillary area quantification provides a tool for in-depth clinical analysis of retinal perfusion. Translational Relevance: Accurate retinal microvasculature segmentation with the CNN results in improved perfusion analysis in diabetic retinopathy.

The purpose of this study was to compare perfusion parameters of the parafovea with scans outside the parafovea to find an area most susceptible to changes secondary to diabetic retinopathy (DR). METHODS. Patients with different DR severity levels as well as controls were included in this cross-sectional clinical trial. Seven standardized 3 × 3 mm areas were recorded with Swept Source Optical Coherence Tomography Angiography: one centered on the fovea, three were temporal to the fovea, and three nasally to the optic disc. The capillary perfusion density (PD) of the superficial capillary complex (SCC) and deep capillary complex (DCC) as well as the fractal dimension (FD) were generated. Statistical analyses were done with R software. RESULTS. One hundred ninety-two eyes (33 controls, 51 no-DR, 41 mild DR, 37 moderate/severe DR, and 30 proliferative DR), of which 105 patients with diabetes and 25 healthy controls were included (59 ± 15 years; 62 women). Mean PD of the DCC was significantly less in patients without DR (parafovea = 0.48 ± 0.03; temporal = 0.48 ± 0.02; and nasal = 0.48 ± 0.03) compared to controls (parafovea = 0.49 ± 0.02; temporal = 0.50 ± 0.02; and nasal = 0.50 ± 0.03). With increasing DR severity, PD and FD of the SCC and DCC further decreased. CONCLUSIONS. Capillary perfusion of the retina is affected early by diabetes. PD of the DCC was significantly reduced in patients with diabetes who did not have any clinical signs of DR. The capillary network outside the parafovea was more susceptible to capillary perfusion deficits compared to the capillaries close to the fovea. TRIAL REGISTRATION. clinicaltrial.gov, NCT03765112, https://clinicaltrials.gov/ct2/show/ NCT03765112?term=NCT03765112&rank=1

No abstract