Automated Segmentation of Retinal Fluid Volumes From Structural and Angiographic Optical Coherence Tomography Using Deep Learning

Evaluation of OCT biomarker changes in treatment-naive neovascular AMD using a deep semantic segmentation algorithm

OBJECTIVES

To determine real-life quantitative changes in OCT biomarkers in a large set of treatment naive patients in a real-life setting undergoing anti-VEGF therapy. For this purpose, we devised a novel deep learning based semantic segmentation algorithm providing the first benchmark results for automatic segmentation of 11 OCT features including biomarkers for neovascular age-related macular degeneration (nAMD).

METHODS

Training of a Deep U-net based semantic segmentation ensemble algorithm for state-of-the-art semantic segmentation performance which was used to analyze OCT features prior to, after 3 and 12 months of anti-VEGF therapy.

RESULTS

High F1 scores of almost 1.0 for neurosensory retina and subretinal fluid on a separate hold-out test set with unseen patients. The algorithm performed worse for subretinal hyperreflective material and fibrovascular PED, on par with drusenoid PED, and better in segmenting fibrosis. In the evaluation of treatment naive OCT scans, significant changes occurred for intraretinal fluid (mean: 0.03 µm3 to 0.01 µm3, p < 0.001), subretinal fluid (0.08 µm3 to 0.01 µm3, p < 0.001), subretinal hyperreflective material (0.02 µm3 to 0.01 µm3, p < 0.001), fibrovascular PED (0.12 µm3 to 0.09 µm3, p = 0.02) and central retinal thickness C0 (225.78 µm3 to 169.40 µm3). The amounts of intraretinal fluid, fibrovascular PED, and ERM were predictive of poor outcome.

CONCLUSIONS

The segmentation algorithm allows efficient volumetric analysis of OCT scans. Anti-VEGF provokes most potent changes in the first 3 months while a gradual loss of RPE hints at a progressing decline of visual acuity. Additional research is required to understand how these accurate OCT predictions can be leveraged for a personalized therapy regimen.

Precision Segmentation of Subretinal Fluids in OCT Using Multiscale Attention-Based U-Net Architecture

This paper presents a deep-learning architecture for segmenting retinal fluids in patients with Diabetic Macular Oedema (DME) and Age-related Macular Degeneration (AMD). Accurate segmentation of multiple fluid types is critical for diagnosis and treatment planning, but existing techniques often struggle with precision. We propose an encoder-decoder network inspired by U-Net, processing enhanced OCT images and their edge maps. The encoder incorporates Residual and Inception modules with an autoencoder-based multiscale attention mechanism to extract detailed features. Our method shows superior performance across several datasets. On the RETOUCH dataset, the network achieved F1 Scores of 0.82 for intraretinal fluid (IRF), 0.93 for subretinal fluid (SRF), and 0.94 for pigment epithelial detachment (PED). The model also performed well on the OPTIMA and DUKE datasets, demonstrating high precision, recall, and F1 Scores. This architecture significantly enhances segmentation accuracy and edge precision, offering a valuable tool for diagnosing and managing retinal diseases. Its integration of dual-input processing, multiscale attention, and advanced encoder modules highlights its potential to improve clinical outcomes and advance retinal disease treatment.

Novel Method to Measure Volumes of Retinal Specific Entities

Objectives: The aim of this study is to describe and validate an optical-coherence-tomography (OCT)-based method to easily calculate specific volumes, addressing the limitations of current OCT software in automating volumetric analysis for specific entities in retinal pathologies. Methods: After manually drawing the specific entity on linear OCT scans using the calliper function and automated measurement of its area, the following formula was used for volumetric calculation: Volume [mm3] = ∑area [mm2] × OCT-scan distance [mm]. Retinal volume (RV) was measured by two independent observers in eyes with a normal foveal profile (NFP) and was compared with the automated measurements performed by the OCT software (Engineering GmbH, Heidelberg, Germany); the same process was repeated for the volume of the foveal cavity (FC) or foveoschisis (FS) in eyes with lamellar macular holes (LMHs). Power calculations were conducted to ensure adequate sample size. The measurements were re-acquired after six weeks. Intra- and inter-observer variability as well as comparison to automated RV calculations were analysed. Results: This study included a total of 62 eyes divided into two groups: the NFP (30 eyes) and LMH (32 eyes) groups. The Bland-Altman plots showed a high degree of agreement in both groups for inter-observer and intra-observer agreement. In addition, in the NFP group, a high degree of agreement was demonstrated between human observers and the OCT software (Spectralis). Conclusions: An easy, reliable, and widely applicable method to calculate volumes is described and validated in this paper, showing excellent inter- and intra-observer agreement, which can be applied to any entity requiring a specific study in the context of retinal pathologies.

DISCOVER: 2-D multiview summarization of Optical Coherence Tomography Angiography for automatic diabetic retinopathy diagnosis

Diabetic Retinopathy (DR), an ocular complication of diabetes, is a leading cause of blindness worldwide. Traditionally, DR is monitored using Color Fundus Photography (CFP), a widespread 2-D imaging modality. However, DR classifications based on CFP have poor predictive power, resulting in suboptimal DR management. Optical Coherence Tomography Angiography (OCTA) is a recent 3-D imaging modality offering enhanced structural and functional information (blood flow) with a wider field of view. This paper investigates automatic DR severity assessment using 3-D OCTA. A straightforward solution to this task is a 3-D neural network classifier. However, 3-D architectures have numerous parameters and typically require many training samples. A lighter solution consists in using 2-D neural network classifiers processing 2-D en-face (or frontal) projections and/or 2-D cross-sectional slices. Such an approach mimics the way ophthalmologists analyze OCTA acquisitions: (1) en-face flow maps are often used to detect avascular zones and neovascularization, and (2) cross-sectional slices are commonly analyzed to detect macular edemas, for instance. However, arbitrary data reduction or selection might result in information loss. Two complementary strategies are thus proposed to optimally summarize OCTA volumes with 2-D images: (1) a parametric en-face projection optimized through deep learning and (2) a cross-sectional slice selection process controlled through gradient-based attribution. The full summarization and DR classification pipeline is trained from end to end. The automatic 2-D summary can be displayed in a viewer or printed in a report to support the decision. We show that the proposed 2-D summarization and classification pipeline outperforms direct 3-D classification with the advantage of improved interpretability.

Perfused and Nonperfused Microaneurysms Identified and Characterized by Structural and Angiographic OCT

PURPOSE

Microaneurysms (MAs) have distinct, oval-shaped, hyperreflective walls on structural OCT, and inconsistent flow signal in the lumen with OCT angiography (OCTA). Their relationship to regional macular edema in diabetic retinopathy (DR) has not been quantitatively explored.

DESIGN

Retrospective, cross-sectional study.

PARTICIPANTS

A total of 99 participants, including 23 with mild, nonproliferative DR (NPDR), 25 with moderate NPDR, 34 with severe NPDR, and 17 with proliferative DR.

METHODS

We obtained 3 × 3-mm scans with a commercial device (Solix, Visionix/Optovue) in 99 patients with DR. Trained graders manually identified MAs and their location relative to the anatomic layers from cross-sectional OCT. Microaneurysms were first classified as perfused if flow signal was present in the OCTA channel. Then, perfused MAs were further classified into fully and partially perfused MAs based on the flow characteristics in en face OCTA. The presence of retinal fluid based on OCT near MAs was compared between perfused and nonperfused types. We also compared OCT-based MA detection to fundus photography (FP)- and fluorescein angiography (FA)-based detection.

MAIN OUTCOME MEASURES

OCT-identified MAs can be classified according to colocalized OCTA flow signal into fully perfused, partially perfused, and nonperfused types. Fully perfused MAs may be more likely to be associated with diabetic macular edema (DME) than those without flow.

RESULTS

We identified 308 MAs (166 fully perfused, 88 partially perfused, 54 nonperfused) in 42 eyes using OCT and OCTA. Nearly half of the MAs identified in this study straddle the inner nuclear layer and outer plexiform layer. Compared with partially perfused and nonperfused MAs, fully perfused MAs were more likely to be associated with local retinal fluid. The associated fluid volumes were larger with fully perfused MAs compared with other types. OCT/OCTA detected all MAs found on FP. Although not all MAs seen with FA were identified with OCT, some MAs seen with OCT were not visible with FA or FP.

CONCLUSIONS

OCT-identified MAs with colocalized flow on OCTA are more likely to be associated with DME than those without flow.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Interpretable Diabetic Retinopathy Diagnosis Based on Biomarker Activation Map

OBJECTIVE

Deep learning classifiers provide the most accurate means of automatically diagnosing diabetic retinopathy (DR) based on optical coherence tomography (OCT) and its angiography (OCTA). The power of these models is attributable in part to the inclusion of hidden layers that provide the complexity required to achieve a desired task. However, hidden layers also render algorithm outputs difficult to interpret. Here we introduce a novel biomarker activation map (BAM) framework based on generative adversarial learning that allows clinicians to verify and understand classifiers' decision-making.

METHODS

A data set including 456 macular scans were graded as non-referable or referable DR based on current clinical standards. A DR classifier that was used to evaluate our BAM was first trained based on this data set. The BAM generation framework was designed by combing two U-shaped generators to provide meaningful interpretability to this classifier. The main generator was trained to take referable scans as input and produce an output that would be classified by the classifier as non-referable. The BAM is then constructed as the difference image between the output and input of the main generator. To ensure that the BAM only highlights classifier-utilized biomarkers an assistant generator was trained to do the opposite, producing scans that would be classified as referable by the classifier from non-referable scans.

RESULTS

The generated BAMs highlighted known pathologic features including nonperfusion area and retinal fluid.

CONCLUSION/SIGNIFICANCE

A fully interpretable classifier based on these highlights could help clinicians better utilize and verify automated DR diagnosis.

Alterations of outer retinal reflectivity in diabetic patients without clinically detectable retinopathy

PURPOSE

This study aimed to investigate alterations of outer retinal reflectivity on spectral-domain optical coherence tomography (OCT) in diabetic patients without clinically detectable retinopathy (NDR).

METHODS

In this retrospective study, 64 NDR patients and 71 controls were included. Relative reflectivity (RR) of the ellipsoid zone (EZ), photoreceptor outer segment (OS) and inner segment (IS), and outer nuclear layer (ONL) at the foveola and at 500 μm, 1000 μm, and 2000 μm nasal (N), temporal (T), superior (S), and inferior (I) to the foveola was measured by cross-line OCT and ImageJ. Retinal vessel densities (VD) in fovea, parafovea, and perifovea areas were detected by OCT angiography (OCTA).

RESULTS

EZ RR in most retinal locations was significantly lower in NDR eyes compared to controls (all P < 0.05), except the foveola. Compared with controls, NDR eyes also displayed lower RR at N2000, T2000, S1000, and I1000 of OS, at S500 and I500 of IS, and at I500 of ONL (all P < 0.05). Negative correlations could be observed between retinal RR and diabetes duration, HbA1c, and best-corrected visual acuity (BCVA) (r =  - 0.303 to - 0.452). Compared to controls, EZ, OS, and IS RR of the NDR eyes showed lower correlation coefficients with whole image SCP and DCP VD of parafovea and perifovea regions.

CONCLUSION

Outer retinal reflectivity, along with the coefficients between retinal reflectivity and VD, is reduced in NDR patients and is correlated with diabetes duration, HbA1c, and BCVA. The reduction of outer retinal reflectivity may be a potential biomarker of early retinal alterations in diabetic patients.

Optical coherence tomography angiography in diabetic retinopathy

There remain many unanswered questions on how to assess and treat the pathology and complications that arise from diabetic retinopathy (DR). Optical coherence tomography angiography (OCTA) is a novel and non-invasive three-dimensional imaging method that can visualize capillaries in all retinal layers. Numerous studies have confirmed that OCTA can identify early evidence of microvascular changes and provide quantitative assessment of the extent of diseases such as DR and its complications. A number of informative OCTA metrics could be used to assess DR in clinical trials, including measurements of the foveal avascular zone (FAZ; area, acircularity, 3D para-FAZ vessel density), vessel density, extrafoveal avascular zones, and neovascularization. Assessing patients with DR using a full-retinal slab OCTA image can limit segmentation errors and confounding factors such as those related to center-involved diabetic macular edema. Given emerging data suggesting the importance of the peripheral retinal vasculature in assessing and predicting DR progression, wide-field OCTA imaging should also be used. Finally, the use of automated methods and algorithms for OCTA image analysis, such as those that can distinguish between areas of true and false signals, reconstruct images, and produce quantitative metrics, such as FAZ area, will greatly improve the efficiency and standardization of results between studies. Most importantly, clinical trial protocols should account for the relatively high frequency of poor-quality data related to sub-optimal imaging conditions in DR and should incorporate time for assessing OCTA image quality and re-imaging patients where necessary.

Single-shot OCT and OCT angiography for slab-specific detection of diabetic retinopathy

In this study, we present an optical coherence tomographic angiography (OCTA) prototype using a 500 kHz high-speed swept-source laser. This system can generate a 75-degree field of view with a 10.4 µm lateral resolution with a single acquisition. With this prototype we acquired detailed, wide-field, and plexus-specific images throughout the retina and choroid in eyes with diabetic retinopathy, detecting early retinal neovascularization and locating pathology within specific retinal slabs. Our device could also visualize choroidal flow and identify signs of key biomarkers in diabetic retinopathy.

Perfused and Nonperfused Microaneurysms Identified and Characterized by Structural and Angiographic OCT

Purpose

Microaneurysms (MAs) have distinct, oval-shaped, hyperreflective walls on structural OCT, and inconsistent flow signal in the lumen with OCT angiography (OCTA). Their relationship to regional macular edema in diabetic retinopathy (DR) has not been quantitatively explored.

Design

Retrospective, cross-sectional study.

Participants

A total of 99 participants, including 23 with mild, nonproliferative DR (NPDR), 25 with moderate NPDR, 34 with severe NPDR, and 17 with proliferative DR.

Methods

We obtained 3 × 3-mm scans with a commercial device (Solix, Visionix/Optovue) in 99 patients with DR. Trained graders manually identified MAs and their location relative to the anatomic layers from cross-sectional OCT. Microaneurysms were first classified as perfused if flow signal was present in the OCTA channel. Then, perfused MAs were further classified into fully and partially perfused MAs based on the flow characteristics in en face OCTA. The presence of retinal fluid based on OCT near MAs was compared between perfused and nonperfused types. We also compared OCT-based MA detection to fundus photography (FP)- and fluorescein angiography (FA)-based detection.

Main Outcome Measures

OCT-identified MAs can be classified according to colocalized OCTA flow signal into fully perfused, partially perfused, and nonperfused types. Fully perfused MAs may be more likely to be associated with diabetic macular edema (DME) than those without flow.

Results

We identified 308 MAs (166 fully perfused, 88 partially perfused, 54 nonperfused) in 42 eyes using OCT and OCTA. Nearly half of the MAs identified in this study straddle the inner nuclear layer and outer plexiform layer. Compared with partially perfused and nonperfused MAs, fully perfused MAs were more likely to be associated with local retinal fluid. The associated fluid volumes were larger with fully perfused MAs compared with other types. OCT/OCTA detected all MAs found on FP. Although not all MAs seen with FA were identified with OCT, some MAs seen with OCT were not visible with FA or FP.

Conclusions

OCT-identified MAs with colocalized flow on OCTA are more likely to be associated with DME than those without flow.

Financial Disclosures

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article. Ophthalmology Retina 2023;■:1-8 © 2023 by the American Academy of Ophthalmology. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

OCT angiography and its retinal biomarkers [Invited]

Optical coherence tomography angiography (OCTA) is a high-resolution, depth-resolved imaging modality with important applications in ophthalmic practice. An extension of structural OCT, OCTA enables non-invasive, high-contrast imaging of retinal and choroidal vasculature that are amenable to quantification. As such, OCTA offers the capability to identify and characterize biomarkers important for clinical practice and therapeutic research. Here, we review new methods for analyzing biomarkers and discuss new insights provided by OCTA.

Deep-Learning-Based Visualization and Volumetric Analysis of Fluid Regions in Optical Coherence Tomography Scans

Retinal volume computation is one of the critical steps in grading pathologies and evaluating the response to a treatment. We propose a deep-learning-based visualization tool to calculate the fluid volume in retinal optical coherence tomography (OCT) images. The pathologies under consideration are Intraretinal Fluid (IRF), Subretinal Fluid (SRF), and Pigmented Epithelial Detachment (PED). We develop a binary classification model for each of these pathologies using the Inception-ResNet-v2 and the small Inception-ResNet-v2 models. For visualization, we use several standard Class Activation Mapping (CAM) techniques, namely Grad-CAM, Grad-CAM++, Score-CAM, Ablation-CAM, and Self-Matching CAM, to visualize the pathology-specific regions in the image and develop a novel Ensemble-CAM visualization technique for robust visualization of OCT images. In addition, we demonstrate a Graphical User Interface that takes the visualization heat maps as the input and calculates the fluid volume in the OCT C-scans. The volume is computed using both the region-growing algorithm and selective thresholding technique and compared with the ground-truth volume based on expert annotation. We compare the results obtained using the standard Inception-ResNet-v2 model with a small Inception-ResNet-v2 model, which has half the number of trainable parameters compared with the original model. This study shows the relevance and usefulness of deep-learning-based visualization techniques for reliable volumetric analysis.

The Present and Future of Artificial Intelligence-Based Medical Image in Diabetes Mellitus: Focus on Analytical Methods and Limitations of Clinical Use

Artificial intelligence (AI)-based diagnostic technology using medical images can be used to increase examination accessibility and support clinical decision-making for screening and diagnosis. To determine a machine learning algorithm for diabetes complications, a literature review of studies using medical image-based AI technology was conducted using the National Library of Medicine PubMed, and the Excerpta Medica databases. Lists of studies using diabetes diagnostic images and AI as keywords were combined. In total, 227 appropriate studies were selected. Diabetic retinopathy studies using the AI model were the most frequent (85.0%, 193/227 cases), followed by diabetic foot (7.9%, 18/227 cases) and diabetic neuropathy (2.7%, 6/227 cases). The studies used open datasets (42.3%, 96/227 cases) or directly constructed data from fundoscopy or optical coherence tomography (57.7%, 131/227 cases). Major limitations in AI-based detection of diabetes complications using medical images were the lack of datasets (36.1%, 82/227 cases) and severity misclassification (26.4%, 60/227 cases). Although it remains difficult to use and fully trust AI-based imaging analysis technology clinically, it reduces clinicians' time and labor, and the expectations from its decision-support roles are high. Various data collection and synthesis data technology developments according to the disease severity are required to solve data imbalance.

Optical Coherence Tomography Angiography in Retinal Vascular Disorders

Traditionally, abnormalities of the retinal vasculature and perfusion in retinal vascular disorders, such as diabetic retinopathy and retinal vascular occlusions, have been visualized with dye-based fluorescein angiography (FA). Optical coherence tomography angiography (OCTA) is a newer, alternative modality for imaging the retinal vasculature, which has some advantages over FA, such as its dye-free, non-invasive nature, and depth resolution. The depth resolution of OCTA allows for characterization of the retinal microvasculature in distinct anatomic layers, and commercial OCTA platforms also provide automated quantitative vascular and perfusion metrics. Quantitative and qualitative OCTA analysis in various retinal vascular disorders has facilitated the detection of pre-clinical vascular changes, greater understanding of known clinical signs, and the development of imaging biomarkers to prognosticate and guide treatment. With further technological improvements, such as a greater field of view and better image quality processing algorithms, it is likely that OCTA will play an integral role in the study and management of retinal vascular disorders. Artificial intelligence methods-in particular, deep learning-show promise in refining the insights to be gained from the use of OCTA in retinal vascular disorders. This review aims to summarize the current literature on this imaging modality in relation to common retinal vascular disorders.

Automated Macular Fluid Volume As a Treatment Indicator for Diabetic Macular Edema

Introduction: To assess the diagnostic accuracy of automatically quantified macular fluid volume (MFV) for treatment-required diabetic macular edema (DME). Methods: This retrospective cross-sectional study included eyes with DME. The commercial software on optical coherence tomography (OCT) produced the central subfield thickness (CST), and a custom deep-learning algorithm automatically segmented the fluid cysts and quantified the MFV from the volumetric scans of an OCT angiography system. Retina specialists treated patients per standard of care based on clinical and OCT findings without access to the MFV. The main outcome measures were the area under the receiver operating characteristic curve (AUROC), sensitivity, and specificity of the CST, MFV, and visual acuity (VA) for treatment indication. Results: Of 139 eyes, 39 (28%) were treated for DME during the study period and 101 (72%) were previously treated. The algorithm detected fluid in all eyes; however, only 54 eyes (39%) met the DRCR.net criteria for center-involved ME. The AUROC of MFV predicting a treatment decision of 0.81 was greater than that of CST (0.67) (P = .0048). Untreated eyes that met the optimal threshold for treatment-required DME based on MFV (>0.031 mm3) had better VA than treated eyes (P = .0053). A multivariate logistic regression model showed that MFV (P = .0008) and VA (P = .0061) were significantly associated with a treatment decision, but CST was not. Conclusions: MFV had a higher correlation with the need for treatment for DME than CST and may be especially useful for ongoing management of DME.

RGC-Net: An Automatic Reconstruction and Quantification Algorithm for Retinal Ganglion Cells Based on Deep Learning

Purpose

The purpose of this study was to develop a deep learning-based fully automated reconstruction and quantification algorithm which automatically delineates the neurites and somas of retinal ganglion cells (RGCs).

Methods

We trained a deep learning-based multi-task image segmentation model, RGC-Net, that automatically segments the neurites and somas in RGC images. A total of 166 RGC scans with manual annotations from human experts were used to develop this model, whereas 132 scans were used for training, and the remaining 34 scans were reserved as testing data. Post-processing techniques removed speckles or dead cells in soma segmentation results to further improve the robustness of the model. Quantification analyses were also conducted to compare five different metrics obtained by our automated algorithm and manual annotations.

Results

Quantitatively, our segmentation model achieves average foreground accuracy, background accuracy, overall accuracy, and dice similarity coefficient of 0.692, 0.999, 0.997, and 0.691 for the neurite segmentation task, and 0.865, 0.999, 0.997, and 0.850 for the soma segmentation task, respectively.

Conclusions

The experimental results demonstrate that RGC-Net can accurately and reliably reconstruct neurites and somas in RGC images. We also demonstrate our algorithm is comparable to human manually curated annotations in quantification analyses.

Translational Relevance

Our deep learning model provides a new tool that can trace and analyze the RGC neurites and somas efficiently and faster than manual analysis.

On Machine Learning in Clinical Interpretation of Retinal Diseases Using OCT Images

Optical coherence tomography (OCT) is a noninvasive imaging technique that provides high-resolution cross-sectional retina images, enabling ophthalmologists to gather crucial information for diagnosing various retinal diseases. Despite its benefits, manual analysis of OCT images is time-consuming and heavily dependent on the personal experience of the analyst. This paper focuses on using machine learning to analyse OCT images in the clinical interpretation of retinal diseases. The complexity of understanding the biomarkers present in OCT images has been a challenge for many researchers, particularly those from nonclinical disciplines. This paper aims to provide an overview of the current state-of-the-art OCT image processing techniques, including image denoising and layer segmentation. It also highlights the potential of machine learning algorithms to automate the analysis of OCT images, reducing time consumption and improving diagnostic accuracy. Using machine learning in OCT image analysis can mitigate the limitations of manual analysis methods and provide a more reliable and objective approach to diagnosing retinal diseases. This paper will be of interest to ophthalmologists, researchers, and data scientists working in the field of retinal disease diagnosis and machine learning. By presenting the latest advancements in OCT image analysis using machine learning, this paper will contribute to the ongoing efforts to improve the diagnostic accuracy of retinal diseases.

Deep Learning in Optical Coherence Tomography Angiography: Current Progress, Challenges, and Future Directions

Optical coherence tomography angiography (OCT-A) provides depth-resolved visualization of the retinal microvasculature without intravenous dye injection. It facilitates investigations of various retinal vascular diseases and glaucoma by assessment of qualitative and quantitative microvascular changes in the different retinal layers and radial peripapillary layer non-invasively, individually, and efficiently. Deep learning (DL), a subset of artificial intelligence (AI) based on deep neural networks, has been applied in OCT-A image analysis in recent years and achieved good performance for different tasks, such as image quality control, segmentation, and classification. DL technologies have further facilitated the potential implementation of OCT-A in eye clinics in an automated and efficient manner and enhanced its clinical values for detecting and evaluating various vascular retinopathies. Nevertheless, the deployment of this combination in real-world clinics is still in the "proof-of-concept" stage due to several limitations, such as small training sample size, lack of standardized data preprocessing, insufficient testing in external datasets, and absence of standardized results interpretation. In this review, we introduce the existing applications of DL in OCT-A, summarize the potential challenges of the clinical deployment, and discuss future research directions.

A vision transformer architecture for the automated segmentation of retinal lesions in spectral domain optical coherence tomography images

Neovascular age-related macular degeneration (nAMD) is one of the major causes of irreversible blindness and is characterized by accumulations of different lesions inside the retina. AMD biomarkers enable experts to grade the AMD and could be used for therapy prognosis and individualized treatment decisions. In particular, intra-retinal fluid (IRF), sub-retinal fluid (SRF), and pigment epithelium detachment (PED) are prominent biomarkers for grading neovascular AMD. Spectral-domain optical coherence tomography (SD-OCT) revolutionized nAMD early diagnosis by providing cross-sectional images of the retina. Automatic segmentation and quantification of IRF, SRF, and PED in SD-OCT images can be extremely useful for clinical decision-making. Despite the excellent performance of convolutional neural network (CNN)-based methods, the task still presents some challenges due to relevant variations in the location, size, shape, and texture of the lesions. This work adopts a transformer-based method to automatically segment retinal lesion from SD-OCT images and qualitatively and quantitatively evaluate its performance against CNN-based methods. The method combines the efficient long-range feature extraction and aggregation capabilities of Vision Transformers with data-efficient training of CNNs. The proposed method was tested on a private dataset containing 3842 2-dimensional SD-OCT retina images, manually labeled by experts of the Franziskus Eye-Center, Muenster. While one of the competitors presents a better performance in terms of Dice score, the proposed method is significantly less computationally expensive. Thus, future research will focus on the proposed network's architecture to increase its segmentation performance while maintaining its computational efficiency.

Volume-based, layer-independent, disease-agnostic detection of abnormal retinal reflectivity, nonperfusion, and neovascularization using structural and angiographic OCT

Optical coherence tomography (OCT) is widely used in ophthalmic practice because it can visualize retinal structure and vasculature in vivo and 3-dimensionally (3D). Even though OCT procedures yield data volumes, clinicians typically interpret the 3D images using two-dimensional (2D) data subsets, such as cross-sectional scans or en face projections. Since a single OCT volume can contain hundreds of cross-sections (each of which must be processed with retinal layer segmentation to produce en face images), a thorough manual analysis of the complete OCT volume can be prohibitively time-consuming. Furthermore, 2D reductions of the full OCT volume may obscure relationships between disease progression and the (volumetric) location of pathology within the retina and can be prone to mis-segmentation artifacts. In this work, we propose a novel framework that can detect several retinal pathologies in three dimensions using structural and angiographic OCT. Our framework operates by detecting deviations in reflectance, angiography, and simulated perfusion from a percent depth normalized standard retina created by merging and averaging scans from healthy subjects. We show that these deviations from the standard retina can highlight multiple key features, while the depth normalization obviates the need to segment several retinal layers. We also construct a composite pathology index that measures average deviation from the standard retina in several categories (hypo- and hyper-reflectance, nonperfusion, presence of choroidal neovascularization, and thickness change) and show that this index correlates with DR severity. Requiring minimal retinal layer segmentation and being fully automated, this 3D framework has a strong potential to be integrated into commercial OCT systems and to benefit ophthalmology research and clinical care.

Association Between Visual Acuity and Residual Retinal Fluid Following Intravitreal Anti-Vascular Endothelial Growth Factor Treatment for Neovascular Age-Related Macular Degeneration: A Systematic Review and Meta-analysis

Importance

The association between residual subretinal fluid (SRF) and intraretinal fluid (IRF) and visual acuity following anti-vascular endothelial growth factor (VEGF) treatment is not well understood.

Objective

To examine the association of residual retinal fluid, SRF, and IRF with visual acuity following anti-VEGF treatment in patients with neovascular age-related macular degeneration (nAMD).

Data Sources

A systematic literature search was performed from January 2005 to August 2021 using Ovid MEDLINE, Embase, and the Cochrane Library.

Study Selection

Peer-reviewed articles reporting on visual acuity stratified by the presence or absence of any residual SRF, IRF, or any retinal fluid at last study observation after intravitreal bevacizumab, ranibizumab, aflibercept, or brolucizumab in patients with nAMD were included. Studies that were noncomparative, included fewer than 10 eyes, or reported on other anti-VEGF agents were excluded.

Data Extraction and Synthesis

Two independent reviewers conducted data extraction and synthesis. The Cochrane risk of bias tool 2 and ROBINS-I were used to assess risk of bias and GRADE evaluation was conducted to assess certainty of evidence.

Main Outcomes and Measures

Primary outcomes were BCVA at last study observation, change in BCVA from baseline, and retinal thickness at last study observation.

Results

In this systematic review and meta-analysis, 11 studies (6 randomized clinical trials [RCTs]) comprising 3092 eyes were included in our analysis. Across all included studies, the BCVA of eyes with residual SRF was better than eyes without SRF (weighted mean difference [WMD], 3.1 letter score; 95% CI, 0.05 to 6.18; P = .05; GRADE, low certainty of evidence; 6 studies; 1931 eyes) but similar in RCTs (WMD, 2.7 letter score; 95% CI, -2.40 to 7.84; P = .30; GRADE, low certainty of evidence; 3 studies; 1406 eyes). The BCVA of eyes with residual IRF was worse than that of eyes without IRF (WMD, -8.2 letter score; 95% CI, -11.79 to -4.50; P < .001; GRADE, low; 7 studies; 2114 eyes).

Conclusions and Relevance

The findings suggest that the presence of residual SRF was associated with slightly better BCVA at last study observation; however, baseline differences in BCVA existed and this conclusion was primarily driven by 1 study. The presence of residual IRF was associated with substantially worse BCVA at last study observation and less improvement of BCVA from baseline. The conclusions are limited by the inclusion of data from observational studies, heterogeneity, and a low certainty of evidence.

Association Between Fluid Volume in Inner Nuclear Layer and Visual Acuity in Diabetic Macular Edema

PURPOSE

In diabetic macular edema (DME), the correlation between visual acuity (VA) and central subfield thickness (CST) is weak. We hypothesize that fluid volume (FV) in the inner nuclear layer (INL) may correlate more strongly with VA.

DESIGN

Retrospective, cross-sectional study.

METHODS

One eye each of diabetic patients with DME was included. We measured intraretinal fluid volume that was detected by automated fluid detection algorithm on 3- × 3-mm optical coherence tomography angiogram volume scans. The detected fluid was subdivided into inner FV, bounded by the INL, and outer FV, the fluid between the outer border of INL to the ellipsoid zone.

RESULTS

We enrolled 125 patients with DME (60 women; mean age, 61 years). The mean detected inner FV was 0.013 mm3 in 109 eyes (87%). The mean detected outer FV was 0.042 mm3 in 124 eyes (99%). Univariate analysis demonstrated that the VA significantly correlated with the inner FV (P < .0001), whole macular FV (P = .010), and CST (P = .036). Multivariate analysis demonstrated that the inner FV was the only significant factor (β = -0.41, P = .004). These correlations were consistent when the treatment-naïve group (n = 33) and the eyes without previous laser treatments (n = 93) were analyzed separately. The area under the receiver operating characteristic curve of inner FV for VA of 20/32 or worse was significantly higher than that for CST (0.66 vs 0.54, P = .018).

CONCLUSIONS

The inner FV has a stronger association with VA than other OCT biomarkers in DME and may be more clinically useful.

ADC-Net: An Open-Source Deep Learning Network for Automated Dispersion Compensation in Optical Coherence Tomography

Chromatic dispersion is a common problem to degrade the system resolution in optical coherence tomography (OCT). This study is to develop a deep learning network for automated dispersion compensation (ADC-Net) in OCT. The ADC-Net is based on a modified UNet architecture which employs an encoder-decoder pipeline. The input section encompasses partially compensated OCT B-scans with individual retinal layers optimized. Corresponding output is a fully compensated OCT B-scan with all retinal layers optimized. Two numeric parameters, i.e., peak signal to noise ratio (PSNR) and structural similarity index metric computed at multiple scales (MS-SSIM), were used for objective assessment of the ADC-Net performance and optimal values of 29.95 ± 2.52 dB and 0.97 ± 0.014 were obtained respectively. Comparative analysis of training models, including single, three, five, seven and nine input channels were implemented. The mode with five-input channels was observed to be optimal for ADC-Net training to achieve robust dispersion compensation in OCT.

Directional analysis of intensity changes for determining the existence of cyst in optical coherence tomography images

Diabetic retinopathy (DR) is an important cause of blindness in people with the long history of diabetes. DR is caused due to the damage to blood vessels in the retina. One of the most important manifestations of DR is the formation of fluid-filled regions between retinal layers. The evaluation of stage and transcribed drugs can be possible through the analysis of retinal Optical Coherence Tomography (OCT) images. Therefore, the detection of cysts in OCT images and the is of considerable importance. In this paper, a fast method is proposed to determine the status of OCT images as cystic or non-cystic. The method consists of three phases which are pre-processing, boundary pixel determination and post-processing. After applying a noise reduction method in the pre-processing step, the method finds the pixels which are the boundary pixels of cysts. This process is performed by finding the significant intensity changes in the vertical direction and considering rectangular patches around the candidate pixels. The patches are verified whether or not they contain enough pixels making considerable diagonal intensity changes. Then, a shadow omission method is proposed in the post-processing phase to extract the shadow regions which can be mistakenly considered as cystic areas. Then, the pixels extracted in the previous phase that are near the shadow regions are removed to prevent the production of false positive cases. The performance of the proposed method is evaluated in terms of sensitivity and specificity on real datasets. The experimental results show that the proposed method produces outstanding results from both accuracy and speed points of view.

Open-source deep learning-based automatic segmentation of mouse Schlemm's canal in optical coherence tomography images

The purpose of this study was to develop an automatic deep learning-based approach and corresponding free, open-source software to perform segmentation of the Schlemm's canal (SC) lumen in optical coherence tomography (OCT) scans of living mouse eyes. A novel convolutional neural network (CNN) for semantic segmentation grounded in a U-Net architecture was developed by incorporating a late fusion scheme, multi-scale input image pyramid, dilated residual convolution blocks, and attention-gating. 163 pairs of intensity and speckle variance (SV) OCT B-scans acquired from 32 living mouse eyes were used for training, validation, and testing of this CNN model for segmentation of the SC lumen. The proposed model achieved a mean Dice Similarity Coefficient (DSC) of 0.694 ± 0.256 and median DSC of 0.791, while manual segmentation performed by a second expert grader achieved a mean and median DSC of 0.713 ± 0.209 and 0.763, respectively. This work presents the first automatic method for segmentation of the SC lumen in OCT images of living mouse eyes. The performance of the proposed model is comparable to the performance of a second human grader. Open-source automatic software for segmentation of the SC lumen is expected to accelerate experiments for studying treatment efficacy of new drugs affecting intraocular pressure and related diseases such as glaucoma, which present as changes in the SC area.

An Open-Source Deep Learning Network for Reconstruction of High-Resolution OCT Angiograms of Retinal Intermediate and Deep Capillary Plexuses

Purpose

We propose a deep learning-based image reconstruction algorithm to produce high-resolution optical coherence tomographic angiograms (OCTA) of the intermediate capillary plexus (ICP) and deep capillary plexus (DCP).

Methods

In this study, 6-mm × 6-mm macular scans with a 400 × 400 A-line sampling density and 3-mm × 3-mm scans with a 304 × 304 A-line sampling density were acquired on one or both eyes of 180 participants (including 230 eyes with diabetic retinopathy and 44 healthy controls) using a 70-kHz commercial OCT system (RTVue-XR; Optovue, Inc., Fremont, California, USA). Projection-resolved OCTA algorithm removed projection artifacts in voxel. ICP and DCP angiograms were generated by maximum projection of the OCTA signal within the respective plexus. We proposed a deep learning-based method, which receives inputs from registered 3-mm × 3-mm ICP and DCP angiograms with proper sampling density as the ground truth reference to reconstruct 6-mm × 6-mm high-resolution ICP and DCP en face OCTA. We applied the same network on 3-mm × 3-mm angiograms to enhance these images further. We evaluated the reconstructed 3-mm × 3-mm and 6-mm × 6-mm angiograms based on vascular connectivity, Weber contrast, false flow signal (flow signal erroneously generated from background), and the noise intensity in the foveal avascular zone.

Results

Compared to the originals, the Deep Capillary Angiogram Reconstruction Network (DCARnet)-enhanced 6-mm × 6-mm angiograms had significantly reduced noise intensity (ICP, 7.38 ± 25.22, P < 0.001; DCP, 11.20 ± 22.52, P < 0.001), improved vascular connectivity (ICP, 0.95 ± 0.01, P < 0.001; DCP, 0.96 ± 0.01, P < 0.001), and enhanced Weber contrast (ICP, 4.25 ± 0.10, P < 0.001; DCP, 3.84 ± 0.84, P < 0.001), without generating false flow signal when noise intensity lower than 650. The DCARnet-enhanced 3-mm × 3-mm angiograms also reduced noise, improved connectivity, and enhanced Weber contrast in 3-mm × 3-mm ICP and DCP angiograms from 101 eyes. In addition, DCARnet preserved the appearance of the dilated vessels in the reconstructed angiograms in diabetic eyes.

Conclusions

DCARnet can enhance 3-mm × 3-mm and 6-mm × 6-mm ICP and DCP angiogram image quality without introducing artifacts.

Translational Relevance

The enhanced 6-mm × 6-mm angiograms may be easier for clinicians to interpret qualitatively.

Artificial intelligence for improving sickle cell retinopathy diagnosis and management

Sickle cell retinopathy is often initially asymptomatic even in proliferative stages, but can progress to cause vision loss due to vitreous haemorrhages or tractional retinal detachments. Challenges with access and adherence to screening dilated fundus examinations, particularly in medically underserved areas where the burden of sickle cell disease is highest, highlight the need for novel approaches to screening for patients with vision-threatening sickle cell retinopathy. This article reviews the existing literature on and suggests future research directions for coupling artificial intelligence with multimodal retinal imaging to expand access to automated, accurate, imaging-based screening for sickle cell retinopathy. Given the variability in retinal specialist practice patterns with regards to monitoring and treatment of sickle cell retinopathy, we also discuss recent progress toward development of machine learning models that can quantitatively track disease progression over time. These artificial intelligence-based applications have great potential for informing evidence-based and resource-efficient clinical diagnosis and management of sickle cell retinopathy.

An end-to-end network for segmenting the vasculature of three retinal capillary plexuses from OCT angiographic volumes

The segmentation of en face retinal capillary angiograms from volumetric optical coherence tomographic angiography (OCTA) usually relies on retinal layer segmentation, which is time-consuming and error-prone. In this study, we developed a deep-learning-based method to segment vessels in the superficial vascular plexus (SVP), intermediate capillary plexus (ICP), and deep capillary plexus (DCP) directly from volumetric OCTA data. The method contains a three-dimensional convolutional neural network (CNN) for extracting distinct retinal layers, a custom projection module to generate three vascular plexuses from OCTA data, and three parallel CNNs to segment vasculature. Experimental results on OCTA data from rat eyes demonstrated the feasibility of the proposed method. This end-to-end network has the potential to simplify OCTA data processing on retinal vasculature segmentation. The main contribution of this study is that we propose a custom projection module to connect retinal layer segmentation and vasculature segmentation modules and automatically convert data from three to two dimensions, thus establishing an end-to-end method to segment three retinal capillary plexuses from volumetric OCTA without any human intervention.

Machine learning in optical coherence tomography angiography

Optical coherence tomography angiography (OCTA) offers a noninvasive label-free solution for imaging retinal vasculatures at the capillary level resolution. In principle, improved resolution implies a better chance to reveal subtle microvascular distortions associated with eye diseases that are asymptomatic in early stages. However, massive screening requires experienced clinicians to manually examine retinal images, which may result in human error and hinder objective screening. Recently, quantitative OCTA features have been developed to standardize and document retinal vascular changes. The feasibility of using quantitative OCTA features for machine learning classification of different retinopathies has been demonstrated. Deep learning-based applications have also been explored for automatic OCTA image analysis and disease classification. In this article, we summarize recent developments of quantitative OCTA features, machine learning image analysis, and classification.

Automated Quantification of Pathological Fluids in Neovascular Age-Related Macular Degeneration, and Its Repeatability Using Deep Learning

Purpose

To develop a reliable algorithm for the automated identification, localization, and volume measurement of exudative manifestations in neovascular age-related macular degeneration (nAMD), including intraretinal (IRF), subretinal fluid (SRF), and pigment epithelium detachment (PED), using a deep-learning approach.

Methods

One hundred seven spectral domain optical coherence tomography (OCT) cube volumes were extracted from nAMD eyes. Manual annotation of IRF, SRF, and PED was performed. Ninety-two OCT volumes served as training and validation set, and 15 OCT volumes from different patients as test set. The performance of our fluid segmentation method was quantified by means of pixel-wise metrics and volume correlations and compared to other methods. Repeatability was tested on 42 other eyes with five OCT volume scans acquired on the same day.

Results

The fully automated algorithm achieved good performance for the detection of IRF, SRF, and PED. The area under the curve for detection, sensitivity, and specificity was 0.97, 0.95, and 0.99, respectively. The correlation coefficients for the fluid volumes were 0.99, 0.99, and 0.91, respectively. The Dice score was 0.73, 0.67, and 0.82, respectively. For the largest volume quartiles the Dice scores were >0.90. Including retinal layer segmentation contributed positively to the performance. The repeatability of volume prediction showed a standard deviations of 4.0 nL, 3.5 nL, and 20.0 nL for IRF, SRF, and PED, respectively.

Conclusions

The deep-learning algorithm can simultaneously acquire a high level of performance for the identification and volume measurements of IRF, SRF, and PED in nAMD, providing accurate and repeatable predictions. Including layer segmentation during training and squeeze-excite block in the network architecture were shown to boost the performance.

Translational Relevance

Potential applications include measurements of specific fluid compartments with high reproducibility, assistance in treatment decisions, and the diagnostic or scientific evaluation of relevant subgroups.