4.6 Review

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

Related references

Note: Only part of the references are listed.
Article Ophthalmology

Clinically relevant factors associated with a binary outcome of diabetic macular ischaemia: an OCTA study

Da Wei Yang et al.

Summary: This study investigated the factors associated with diabetic macular ischaemia (DMI) as assessed by optical coherence tomography angiography (OCTA). The results showed that age, visual acuity, ganglion cell-inner plexiform layer thickness, diabetic retinopathy severity, haemoglobin A1c level, estimated glomerular filtration rate, and low-density lipoprotein cholesterol level were associated with SCP-DMI. In addition, the presence of diabetic macular oedema and shorter axial length were associated with DCP-DMI.

BRITISH JOURNAL OF OPHTHALMOLOGY (2023)

Article Ophthalmology

Parapapillary Choroidal Microvasculature Predicts Diabetic Retinopathy Progression and Diabetic Macular Edema Development: A Three-Year Prospective Study

Xiao Guo et al.

Summary: This study shows that the microcirculation of the optic nerve head can predict the risk of diabetic retinopathy progression and diabetic macular edema development using swept-source optical coherence tomography angiography.

AMERICAN JOURNAL OF OPHTHALMOLOGY (2023)

Review Radiology, Nuclear Medicine & Medical Imaging

A review of explainable and interpretable AI with applications in COVID-19 imaging

Jordan D. Fuhrman et al.

Summary: The development of medical imaging AI for evaluating COVID-19 patients shows potential in enhancing clinical decision making, with developers utilizing explainability techniques to increase user trust and clinical translation potential.

MEDICAL PHYSICS (2022)

Review Ophthalmology

Diabetic macular ischaemia- a new therapeutic target?

Chui Ming Gemmy Cheung et al.

Summary: This review discusses the definition and grading issues of diabetic macular ischaemia (DMI), proposing vascular phenotypes of DMI based on relative perfusion deficits. The review highlights mechanistic pathways of DMI, including the role of hypoxia and the complex relation between neurons, glia, and microvasculature. Utilising optical coherence tomography angiography (OCTA) to identify reversible DMI phenotypes and developing new trial endpoints for disease progression are crucial. Natural history studies of DMI are needed to define phenotypes and understand the development of diabetic macular oedema.

PROGRESS IN RETINAL AND EYE RESEARCH (2022)

Article Ophthalmology

Optical coherence tomography angiography metrics predict normal tension glaucoma progression

Yu Meng Wang et al.

Summary: This study aimed to determine the relationship between optical coherence tomography angiography (OCTA) metrics and visual field (VF) progression in eyes with normal tension glaucoma (NTG). The findings showed that lower superotemporal cpVD measured from OCTA in the peripapillary region improves the prediction of VF deterioration in NTG.

ACTA OPHTHALMOLOGICA (2022)

Article Ophthalmology

Deep Learning Image Analysis of Optical Coherence Tomography Angiography Measured Vessel Density Improves Classification of Healthy and Glaucoma Eyes

Christopher Bowd et al.

Summary: Deep learning analysis of en face vessel density images using convolutional neural networks improves classification of healthy and glaucomatous eyes compared to feature-based gradient boosting classifier models.

AMERICAN JOURNAL OF OPHTHALMOLOGY (2022)

Article Medicine, General & Internal

Hypertensive eye disease

Carol Y. Cheung et al.

Summary: Hypertensive eye disease refers to eye diseases associated with high blood pressure. It includes various pathological changes, with hypertensive retinopathy being the most well-known. Research has greatly improved our understanding of the epidemiology and clinical implications of hypertensive eye disease. Advances in technology, such as optical coherence tomography angiography and artificial intelligence, have contributed to better assessment and management of hypertension and cardiovascular risk using ocular imaging.

NATURE REVIEWS DISEASE PRIMERS (2022)

Article Computer Science, Interdisciplinary Applications

Disentangled Representation Learning for OCTA Vessel Segmentation With Limited Training Data

Yihao Liu et al.

Summary: Optical coherence tomography angiography (OCTA) is a useful modality for analyzing retinal vasculature. However, accurate segmentation of capillaries in en face OCTA images using deep learning approaches faces challenges such as acquiring manual delineations for training and dealing with contrast-related artifacts. In this study, a disentanglement approach is proposed to learn the anatomy component and local contrast component from paired OCTA scans, improving the performance of vessel segmentation.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2022)

Article Computer Science, Interdisciplinary Applications

DLTTA: Dynamic Learning Rate for Test-Time Adaptation on Cross-Domain Medical Images

Hongzheng Yang et al.

Summary: Test-time adaptation is important for addressing the distribution shift in medical images from different institutions. Previous methods using fixed learning rates fail to adapt to frequent changes in distribution shift. To solve this problem, we propose a dynamic learning rate adjustment method based on discrepancy estimation.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2022)

Article Multidisciplinary Sciences

Deep learning for quality assessment of optical coherence tomography angiography images

Rahul M. Dhodapkar et al.

Summary: Optical coherence tomography angiography (OCTA) is a promising non-invasive technique for imaging the retinal vasculature. In this study, a deep learning-based system using a pretrained neural network classifier was developed to classify OCTA images based on image quality, demonstrating outstanding performance compared to traditional methods.

SCIENTIFIC REPORTS (2022)

Article Ophthalmology

Correlation of Diabetic Disease Severity to Degree of Quadrant Asymmetry in En Face OCTA Metrics

Jesse J. Jung et al.

Summary: The severity of diabetic retinopathy affects the asymmetry of optical coherence tomography angiography metrics, with deeper layers and perfusion density being more affected.

INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE (2022)

Proceedings Paper Computer Science, Artificial Intelligence

Frequency-Aware Inverse-Consistent Deep Learning for OCT-Angiogram Super-Resolution

Weiwen Zhang et al.

Summary: Optical Coherence Tomography Angiography (OCTA) is a novel imaging modality that captures retinal and choroidal microvasculature non-invasively. This paper proposes an inverse-consistent generative adversarial network (GAN) for achieving super-resolution of 6 mm x 6 mm OCTA images, trained with unpaired 3 mm x 3mm and 6 mm x 6 mm OCTA image datasets. Compared with other methods, our approach outperforms them on various performance metrics.

MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2022, PT II (2022)

Article Ophthalmology

A Deep Learning Network for Classifying Arteries and Veins in Montaged Widefield OCT Angiograms

Min Gao et al.

Summary: The purpose of this study was to differentiate arteries from veins in montaged widefield OCT angiography (OCTA) using a deep-learning-based method. The proposed method, called CAVnet, achieved high accuracy in classifying and identifying arteries and veins, and showed potential clinical utility in analyzing vessel type-specific features in eye diseases such as branch retinal artery occlusion (BRAO) and branch retinal vein occlusion (BRVO).

OPHTHALMOLOGY SCIENCE (2022)

Article Ophthalmology

Standardisation of optical coherence tomography angiography nomenclature in uveitis: first survey results

Francesco Pichi et al.

Summary: There is considerable agreement in the terminology used by all uveitis experts regarding OCT-A findings, but significant differences are noted between users and experts, indicating the need for standardization of nomenclature among all uveitis specialists.

BRITISH JOURNAL OF OPHTHALMOLOGY (2021)

Article Ophthalmology

Early Detection of Microvascular Impairments With Optical Coherence Tomography Angiography in Diabetic Patients Without Clinical Retinopathy: A Meta-analysis

Bilei Zhang et al.

Summary: The study evaluated microvascular impairments in the eyes of diabetic patients with no diabetic retinopathy using optical coherence tomography angiography (OCTA). The results showed that retinal microvascular impairments might have occurred before clinically visible diabetic retinopathy and could be detected early by OCTA, but the manifestations could be inconsistent according to the types of diabetes mellitus.

AMERICAN JOURNAL OF OPHTHALMOLOGY (2021)

Article Ophthalmology

Evaluation of the clinical utility of optical coherence tomography angiography in age-related macular degeneration

Melina Cavichini et al.

Summary: This study retrospectively reviewed 152 eyes of 106 AMD patients and found that OCTA had a specificity and sensitivity of 85.4% and 79.3%, respectively, in identifying the presence or absence of CNV. However, OCTA was only 69.5% accurate in recognizing active CNV.

BRITISH JOURNAL OF OPHTHALMOLOGY (2021)

Review Ophthalmology

Optical coherence tomography angiography in diabetic retinopathy: an updated review

Zihan Sun et al.

Summary: OCTA technology allows for quantification of microvascular changes in diabetic retinopathy, including classic features and early microvascular abnormalities, making it a promising objective tool for early detection, staging, and progression of DR, ultimately leading to improved individualized management and prevention of visual impairment.
Review Engineering, Biomedical

Interpretation and visualization techniques for deep learning models in medical imaging

Daniel T. Huff et al.

Summary: This review provides insights into methods for interpreting image models in medical imaging, categorizing them into understanding model structure and function, and understanding model output. It discusses various interpretation techniques, limitations of current methods, and offers recommendations for DL practitioners looking to incorporate model interpretation into their tasks in medical imaging contexts.

PHYSICS IN MEDICINE AND BIOLOGY (2021)

Article Multidisciplinary Sciences

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

Reza Mirshahi et al.

Summary: This study introduced a new deep learning model for accurate segmentation of the fovea avascular zone in OCTA images of diabetic patients and healthy subjects. The deep learning and manual measurements performed better than the automated measurements of the built-in software.

SCIENTIFIC REPORTS (2021)

Article Computer Science, Interdisciplinary Applications

ROSE: A Retinal OCT-Angiography Vessel Segmentation Dataset and New Model

Yuhui Ma et al.

Summary: In this study, a new dataset ROSE for retinal vessel OCTA images was constructed, and a novel vessel segmentation network OCTA-Net was proposed with superior performance. Experimental results demonstrated potential advantages in studying neurodegenerative diseases through fractal dimension analysis.

IEEE TRANSACTIONS ON MEDICAL IMAGING (2021)

Article Ophthalmology

Optical Coherence Tomography Angiography Quality Across Three Multicenter Clinical Studies of Diabetic Retinopathy

Brandon J. Lujan et al.

Summary: This study investigated the quality of OCTA scans in multicenter clinical studies, showing that the majority of scans were of good quality. However, improvements are needed in signal strength and motion artifacts.

TRANSLATIONAL VISION SCIENCE & TECHNOLOGY (2021)

Review Radiology, Nuclear Medicine & Medical Imaging

Federated learning: a collaborative effort to achieve better medical imaging models for individual sites that have small labelled datasets

Dianwen Ng et al.

Summary: Federated learning is a new learning framework that allows different medical institutions to collaboratively train a global model to improve the application of AI in radiology. By aggregating training results from multiple sites, the model receives better supervision and global detection capabilities, while maintaining patient privacy.

QUANTITATIVE IMAGING IN MEDICINE AND SURGERY (2021)

Article Multidisciplinary Sciences

A deep learning model for identifying diabetic retinopathy using optical coherence tomography angiography

Gahyung Ryu et al.

Summary: With the increasing prevalence of diabetes, the use of artificial intelligence for diabetic retinopathy screening has become more widespread. Studies have shown that convolutional neural network models based on optical coherence tomography angiography images exhibit high accuracy and reliability in diagnosing DR.

SCIENTIFIC REPORTS (2021)

Article Ophthalmology

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

Min Gao et al.

Summary: The study proposed a deep learning-based image reconstruction algorithm to generate high-resolution OCTA of ICP and DCP. The DCARnet-enhanced angiograms showed significantly reduced noise intensity, improved vascular connectivity, and enhanced Weber contrast without generating false flow signal.

TRANSLATIONAL VISION SCIENCE & TECHNOLOGY (2021)

Editorial Material Ophthalmology

Updating the Staging System for Diabetic Retinal Disease

Jennifer K. Sun et al.

OPHTHALMOLOGY (2021)

Article Endocrinology & Metabolism

A Multitask Deep-Learning System to Classify Diabetic Macular Edema for Different Optical Coherence Tomography Devices: A Multicenter Analysis

Fangyao Tang et al.

Summary: A deep learning system was developed for accurate classification of DME using OCT images, achieving high AUROC values. This system showed excellent performance in automated DME classification, demonstrating potential as an effective screening tool for patients with diabetes.

DIABETES CARE (2021)

Article Engineering, Biomedical

DcardNet: Diabetic Retinopathy Classification at Multiple Levels Based on Structural and Angiographic Optical Coherence Tomography

Pengxiao Zang et al.

Summary: A CNN-based DR classification framework using en face OCT and OCTA was proposed in this study, with overall classification accuracies of 95.7%, 85.0%, and 71.0% for the three levels achieved through 10-fold cross-validation using 10% of the data.

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING (2021)

Review Radiology, Nuclear Medicine & Medical Imaging

A review of medical image data augmentation techniques for deep learning applications

Phillip Chlap et al.

Summary: Data augmentation has become a popular method for training deep learning models in the field of radiology and radiotherapy, where limited medical image datasets are often encountered. By generating additional training data, data augmentation can improve model performance and has been widely used in state-of-the-art deep learning models.

JOURNAL OF MEDICAL IMAGING AND RADIATION ONCOLOGY (2021)

Article Computer Science, Artificial Intelligence

Automated cardiac segmentation of cross-modal medical images using unsupervised multi-domain adaptation and spatial neural attention structure

Jinping Liu et al.

Summary: This work proposes a cross-modal cardiac segmentation scheme using a symmetric full convolutional neural network (SFCNN) with unsupervised multi-domain adaptation (UMDA) and spatial neural attention (SNA), achieving superior performance compared to state-of-the-art methods on a benchmark dataset.

MEDICAL IMAGE ANALYSIS (2021)

Article Biochemical Research Methods

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

Yukun Guo et al.

Summary: This study presents a deep-learning-based method for segmenting retinal capillary plexuses directly from volumetric OCTA data, showing feasibility and potential for simplifying OCTA data processing for retinal vasculature segmentation. The proposed technique includes a custom projection module to connect retinal layer segmentation and vasculature segmentation modules, establishing an end-to-end method without human intervention.

BIOMEDICAL OPTICS EXPRESS (2021)

Article Biochemical Research Methods

Glaucoma classification in 3 x 3 mm en face macular scans using deep learning in a different plexus

Julia Schottenhamml et al.

Summary: Glaucoma is a leading cause of irreversible blindness worldwide, and early diagnosis and treatment are crucial. This study analyzes computational features from images to differentiate glaucoma patients from healthy controls effectively. Using convolutional neural networks (CNNs) to extract features shows promising results in distinguishing between the two groups.

BIOMEDICAL OPTICS EXPRESS (2021)

Article Ophthalmology

Standardization of OCT Angiography Nomenclature in Retinal Vascular Diseases: First Survey Results

Marion R. Munk et al.

Summary: The survey aimed to develop a consensus nomenclature for OCT angiography findings in retinal vascular diseases. There were agreements in certain areas but significant differences found in many aspects of OCTA terminology, suggesting the need for standardization among all specialists in the field.

OPHTHALMOLOGY RETINA (2021)

Article Ophthalmology

Quantification of Nonperfusion Area in Montaged Widefield OCT Angiography Using Deep Learning in Diabetic Retinopathy

Yukun Guo et al.

Summary: A deep learning-based algorithm was developed to accurately detect and quantify nonperfusion areas in montaged widefield OCTA, showing potential for improved detection of clinically important diabetic retinopathy and assessment of DR severity.

OPHTHALMOLOGY SCIENCE (2021)

Review Ophthalmology

Optical Coherence Tomography Angiography in Diabetes

Enrico Borrelli et al.

Summary: The introduction of optical coherence tomography angiography (OCTA) has greatly improved our understanding of ocular vascular alterations in diabetes, particularly in the retinal vessels and choriocapillaris. The recent use of widefield technology and 3-dimensional visualization in OCTA imaging has also been discussed.

ASIA-PACIFIC JOURNAL OF OPHTHALMOLOGY (2021)

Article Engineering, Biomedical

MedGAN: Medical image translation using GANs

Karim Armanious et al.

COMPUTERIZED MEDICAL IMAGING AND GRAPHICS (2020)

Article Biochemical Research Methods

Automated diagnosis and segmentation of choroidal neovascularization in OCT angiography using deep learning

Jie Wang et al.

BIOMEDICAL OPTICS EXPRESS (2020)

Article Ophthalmology

Prevalence and Severity of Artifacts in Optical Coherence Tomographic Angiograms

Ian C. Holmen et al.

JAMA OPHTHALMOLOGY (2020)

Article Ophthalmology

Insights into Systemic Disease through Retinal Imaging-Based Oculomics

Siegfried K. Wagner et al.

TRANSLATIONAL VISION SCIENCE & TECHNOLOGY (2020)

Article Biochemical Research Methods

Reconstruction of high-resolution 6 x 6-mm OCT angiograms using deep learning

Min Gao et al.

BIOMEDICAL OPTICS EXPRESS (2020)

Article Biochemistry & Molecular Biology

Predicting conversion to wet age-related macular degeneration using deep learning

Jason Yim et al.

NATURE MEDICINE (2020)

Article Ophthalmology

Transfer Learning for Automated OCTA Detection of Diabetic Retinopathy

David Le et al.

TRANSLATIONAL VISION SCIENCE & TECHNOLOGY (2020)

Article Biochemical Research Methods

AV-Net: deep learning for fully automated artery-vein classification in optical coherence tomography angiography

Minhaj Alam et al.

BIOMEDICAL OPTICS EXPRESS (2020)

Article Multidisciplinary Sciences

Global assessment of arteriolar, venular and capillary changes in normal tension glaucoma

Timothy P. H. Lin et al.

SCIENTIFIC REPORTS (2020)

Article Ophthalmology

NOVEL OPTICAL COHERENCE TOMOGRAPHY ANGIOGRAPHY BIOMARKER IN BRANCH RETINAL VEIN OCCLUSION MACULAR EDEMA

Ling Yeung et al.

RETINA-THE JOURNAL OF RETINAL AND VITREOUS DISEASES (2019)

Article Ophthalmology

Correlation of retinal vascular perfusion density with dark adaptation in diabetic retinopathy

Chia-Chieh Hsiao et al.

GRAEFES ARCHIVE FOR CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY (2019)

Article Ophthalmology

Automated OCT angiography image quality assessment using a deep learning algorithm

J. L. Lauermann et al.

GRAEFES ARCHIVE FOR CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY (2019)

Review Ophthalmology

Deep learning in ophthalmology: The technical and clinical considerations

Daniel S. W. Ting et al.

PROGRESS IN RETINAL AND EYE RESEARCH (2019)

Review Ophthalmology

Artificial intelligence and deep learning in ophthalmology

Daniel Shu Wei Ting et al.

BRITISH JOURNAL OF OPHTHALMOLOGY (2019)

Article Biochemistry & Molecular Biology

A guide to deep learning in healthcare

Andre Esteva et al.

NATURE MEDICINE (2019)

Article Biochemistry & Molecular Biology

The practical implementation of artificial intelligence technologies in medicine

Jianxing He et al.

NATURE MEDICINE (2019)

Article Computer Science, Interdisciplinary Applications

Automatic quantification of superficial foveal avascular zone in optical coherence tomography angiography implemented with deep learning

Menglin Guo et al.

VISUAL COMPUTING FOR INDUSTRY BIOMEDICINE AND ART (2019)

Article Biochemistry & Molecular Biology

Identifying Medical Diagnoses and Treatable Diseases by Image-Based Deep Learning

Daniel S. Kermany et al.

Article Ophthalmology

OCTA vessel density changes in the macular zone in glaucomatous eyes

C. Lommatzsch et al.

GRAEFES ARCHIVE FOR CLINICAL AND EXPERIMENTAL OPHTHALMOLOGY (2018)

Review Ophthalmology

Retinal vasculature development in health and disease

Senthil Selvam et al.

PROGRESS IN RETINAL AND EYE RESEARCH (2018)

Review Ophthalmology

Optical coherence tomography angiography

Richard F. Spaide et al.

PROGRESS IN RETINAL AND EYE RESEARCH (2018)

Article Biochemistry & Molecular Biology

Clinically applicable deep learning for diagnosis and referral in retinal disease

Jeffrey De Fauw et al.

NATURE MEDICINE (2018)

Article Multidisciplinary Sciences

OCT-angiography: A qualitative and quantitative comparison of 4 OCT-A devices

Marion R. Munk et al.

PLOS ONE (2017)

Article Multidisciplinary Sciences

Determinants of Quantitative Optical Coherence Tomography Angiography Metrics in Patients with Diabetes

Fang Yao Tang et al.

SCIENTIFIC REPORTS (2017)

Article Endocrinology & Metabolism

Retinal vascular geometry and 6 year incidence and progression of diabetic retinopathy

Carol Yim-lui Cheung et al.

DIABETOLOGIA (2017)

Review Multidisciplinary Sciences

Deep learning

Yann LeCun et al.

NATURE (2015)

Review Ophthalmology

Current Choroidal Neovascularization Treatment

Liu Jian et al.

OPHTHALMOLOGICA (2013)

Article Ophthalmology

The Effects of Macular Ischemia on Visual Acuity in Diabetic Retinopathy

Dawn A. Sim et al.

INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE (2013)

Review Peripheral Vascular Disease

Retinal Microvasculature as a Model to Study the Manifestations of Hypertension

Carol Yim-lui Cheung et al.

HYPERTENSION (2012)