4.4 Review

Artificial Intelligence for Diabetic Retinopathy Screening Using Color Retinal Photographs: From Development to Deployment

Related references

Note: Only part of the references are listed.
Article Biochemistry & Molecular Biology

Reporting guideline for the early-stage clinical evaluation of decision support systems driven by artificial intelligence: DECIDE-AI

Baptiste Vasey et al.

Summary: The article introduces the DECIDE-AI checklist, which includes key items that should be reported in early-stage clinical studies of AI-based decision support systems. It emphasizes the importance of responsible and transparent deployment of AI systems in healthcare. The checklist was developed through a consensus-based process involving multiple stakeholders.

NATURE MEDICINE (2022)

Article Ophthalmology

Training Deep Learning Models to Work on Multiple Devices by Cross-Domain Learning with No Additional Annotations

Yue Wu et al.

Summary: The purpose of this study was to develop an unsupervised cross-domain segmentation algorithm for segmenting intraretinal fluid and retinal layers in OCT images from different manufacturers and camera devices. The study utilized generative adversarial networks (GANs) and limited labeled data for training and validation. The results showed that the GANSeg algorithm performed well in various segmentation tasks and could effectively generalize to other OCT device domains.

OPHTHALMOLOGY (2022)

Article Medicine, General & Internal

A New Deep Learning Algorithm with Activation Mapping for Diabetic Retinopathy: Backtesting after 10 Years of Tele-Ophthalmology

Alicia Pareja-Rios et al.

Summary: This study reports the development of a deep learning algorithm (AI) that can accurately detect signs of diabetic retinopathy (DR) from fundus images. The algorithm shows comparable performance to doctors in determining the presence of DR.

JOURNAL OF CLINICAL MEDICINE (2022)

Article Biology

Influence of Different Types of Retinal Cameras on the Performance of Deep Learning Algorithms in Diabetic Retinopathy Screening

Ramyaa Srinivasan et al.

Summary: This study assessed the performance of regional graders and artificial intelligence algorithms in classifying images as gradable or ungradable using retinal cameras of different specifications. The deep learning algorithm outperformed human graders in sensitivity, specificity, and kappa for all types of non-mydriatic fundus cameras. The deep learning system showed consistent diagnostic performance across images of varying quality and camera types, surpassing human graders.

LIFE-BASEL (2022)

Article Medical Informatics

Real-time diabetic retinopathy screening by deep learning in a multisite national screening programme: a prospective interventional cohort study

Paisan Ruamviboonsuk et al.

Summary: This study evaluated the real-world performance of deploying a deep-learning system for diabetic retinopathy screening in the healthcare system of Thailand. The results showed that the deep-learning system had similar capability as retina specialists in real-time detection of diabetic retinopathy. Socioenvironmental factors and workflows need to be taken into consideration when implementing such a system in large-scale screening programs in low-income and middle-income countries.

LANCET DIGITAL HEALTH (2022)

Article Ophthalmology

Diagnostic accuracy of diabetic retinopathy grading by an artificial intelligence-enabled algorithm compared with a human standard for wide-field true-colour confocal scanning and standard digital retinal images

Abraham Olvera-Barrios et al.

Summary: The study compared the performance of an automated retinal image analysis software (ARIAS) and human grading for diabetic retinopathy screening, finding that the EyeArt identified diabetic retinopathy in EIDON images with similar sensitivity to standard images, serving as an alternative method for large-scale screening programs.

BRITISH JOURNAL OF OPHTHALMOLOGY (2021)

Article Ophthalmology

Comparison of automated and expert human grading of diabetic retinopathy using smartphone-based retinal photography

Tyson N. Kim et al.

Summary: The study aimed to investigate the efficacy of a mobile platform combining smartphone-based retinal imaging with automated grading for determining the presence of referral-warranted diabetic retinopathy. Results showed that automated interpretation achieved a lower sensitivity but higher specificity than trained expert graders, indicating potential for improved efficiency in clinical practice. Feasibility testing with non-ophthalmic personnel in a retina clinic with high disease burden suggested further studies are needed to assess efficacy of screening diabetic patients from general population.
Article Ophthalmology

Prospective evaluation of an artificial intelligence-enabled algorithm for automated diabetic retinopathy screening of 30 000 patients

Peter Heydon et al.

Summary: The study evaluated the performance of an automated artificial intelligence algorithm in triaging retinal images from diabetic retinopathy screening programmes, demonstrating high sensitivity and specificity in distinguishing test-positive/technical failure from test-negative images.

BRITISH JOURNAL OF OPHTHALMOLOGY (2021)

Article Endocrinology & Metabolism

Multicenter, Head-to-Head, Real-World Validation Study of Seven Automated Artificial Intelligence Diabetic Retinopathy Screening Systems

Aaron Y. Lee et al.

Summary: The study compared the performance of seven automated AI-based DR screening algorithms against human graders using real-world retinal imaging data. While most algorithms had similar performance to humans, a few showed higher sensitivity, indicating the need for rigorous testing before clinical implementation.

DIABETES CARE (2021)

Article Public, Environmental & Occupational Health

Causes of blindness and vision impairment in 2020 and trends over 30 years, and prevalence of avoidable blindness in relation to VISION 2020: the Right to Sight: an analysis for the Global Burden of Disease Study

Jaimie D. Steinmetz et al.

Summary: The study assessed the prevalence and relative contribution of avoidable causes of blindness and vision impairment globally from 1990 to 2020, finding that there was no change in the global prevalence of avoidable blindness and vision impairment in adults aged 50 and older. Although the age-standardized prevalence of avoidable blindness decreased, avoidable moderate and severe vision impairment did not change.

LANCET GLOBAL HEALTH (2021)

Article Ophthalmology

In-Person Verification of Deep Learning Algorithm for Diabetic Retinopathy Screening Using Different Techniques Across Fundus Image Devices

Nida Wongchaisuwat et al.

Summary: This study evaluated the clinical performance of an automated diabetic retinopathy screening model at Siriraj Hospital in Bangkok, Thailand. The Eidon images showed better performance in differentiating referable DR compared to Nidek images.

TRANSLATIONAL VISION SCIENCE & TECHNOLOGY (2021)

Article Medicine, General & Internal

Pivotal Evaluation of an Artificial Intelligence System for Autonomous Detection of Referrable and Vision-Threatening Diabetic Retinopathy

Eli Ipp et al.

Summary: This study evaluated the safety and accuracy of an artificial intelligence system in detecting diabetic retinopathy, showing high sensitivity and specificity without dilation. The findings suggest that improved access to accurate diabetic eye examinations may increase patient adherence to annual screenings.

JAMA NETWORK OPEN (2021)

Review Ophthalmology

Global Prevalence of Diabetic Retinopathy and Projection of Burden through 2045 Systematic Review and Meta-analysis

Zhen Ling Teo et al.

Summary: The translated paragraphs provide updated estimates on the global prevalence and number of people with diabetic retinopathy (DR) through 2045. The burden of DR is expected to increase disproportionately in countries in the Middle East and North Africa and the Western Pacific by 2045. Certain ethnic groups such as Hispanics and Middle Easterners with diabetes are more likely to have DR compared to Asians.

OPHTHALMOLOGY (2021)

Article Medicine, General & Internal

Microaneurysm Turnover in Mild Non-Proliferative Diabetic Retinopathy is Associated with Progression and Development of Vision-Threatening Complications: A 5-Year Longitudinal Study

Ana Rita Santos et al.

Summary: The analysis of retinal microaneurysm turnover (MAT) appears to be a predictive biomarker for the progression of diabetic retinopathy (DR) and the development of diabetic macular edema (DME) and proliferative retinopathy (PDR) in individuals with type 2 diabetes. This suggests the potential for early identification and intervention in diabetic eye diseases.

JOURNAL OF CLINICAL MEDICINE (2021)

Review Biology

Deep learning for diabetic retinopathy detection and classification based on fundus images: A review

Nikos Tsiknakis et al.

Summary: This review article discusses the importance of early detection and treatment of diabetic retinopathy, as well as the application of deep learning in detecting and classifying the disease based on fundus images. Various aspects of the detection pipeline, such as datasets, preprocessing techniques, model development, and clinical applications, are thoroughly analyzed and future research directions are suggested.

COMPUTERS IN BIOLOGY AND MEDICINE (2021)

Article Multidisciplinary Sciences

Real-world? artificial intelligence-based opportunistic screening for diabetic retinopathy in endocrinology and indigenous healthcare settings in Australia

Jane Scheetz et al.

Summary: The study demonstrated that the AI-assisted diabetic retinopathy screening model is accurate and well-accepted by patients and clinicians in Australian healthcare settings. Future deployment of AI-assisted screening models would need to consider downstream referral pathways.

SCIENTIFIC REPORTS (2021)

Article Chemistry, Multidisciplinary

Efficient and Accurate Hemorrhages Detection in Retinal Fundus Images Using Smart Window Features

Tamoor Aziz et al.

Summary: The article proposes an algorithm for segmenting and detecting hemorrhages in retinal fundus images, achieving high accuracy through preprocessing, adaptive thresholding, feature extraction, and classification steps. The method proves to be more effective compared to traditional algorithms and can serve as a segmentation tool.

APPLIED SCIENCES-BASEL (2021)

Article Medicine, General & Internal

Analysis and Comparison of Two Artificial Intelligence Diabetic Retinopathy Screening Algorithms in a Pilot Study: IDx-DR and Retinalyze

Andrzej Grzybowski et al.

Summary: Two autonomous DR screening systems, IDx-DR and Retinalyze, showed a high level of agreement in image analysis and were easy to set up and use. However, there were limitations in the current pilot study in terms of sample choice and reference grading that need to be addressed before conducting a more rigorous future study.

JOURNAL OF CLINICAL MEDICINE (2021)

Article Medical Informatics

Predicting the risk of developing diabetic retinopathy using deep learning

Ashish Bora et al.

Summary: The study aimed to predict the risk of diabetic retinopathy development in patients with diabetes using a deep-learning system. The system accurately predicted the risk of disease progression in patients by analyzing color fundus photographs.

LANCET DIGITAL HEALTH (2021)

Article Endocrinology & Metabolism

Artificial intelligence-enabled screening for diabetic retinopathy: a real-world, multicenter and prospective study

Yifei Zhang et al.

BMJ OPEN DIABETES RESEARCH & CARE (2020)

Article Public, Environmental & Occupational Health

Epidemiology of Type 2 Diabetes - Global Burden of Disease and Forecasted Trends

Moien Abdul Basith Khan et al.

JOURNAL OF EPIDEMIOLOGY AND GLOBAL HEALTH (2020)

Article Computer Science, Information Systems

A governance model for the application of AI in health care

Sandeep Reddy et al.

JOURNAL OF THE AMERICAN MEDICAL INFORMATICS ASSOCIATION (2020)

Article Ophthalmology

Microaneurysm turnover is a predictor of diabetic retinopathy progression

Rajeev K. R. Pappuru et al.

BRITISH JOURNAL OF OPHTHALMOLOGY (2019)

Article Computer Science, Information Systems

Diagnostic assessment of deep learning algorithms for diabetic retinopathy screening

Tao Li et al.

INFORMATION SCIENCES (2019)

Article Health Care Sciences & Services

Deep learning versus human graders for classifying diabetic retinopathy severity in a nationwide screening program

Paisan Raumviboonsuk et al.

NPJ DIGITAL MEDICINE (2019)

Article Health Care Sciences & Services

Pivotal trial of an autonomous AI-based diagnostic system for detection of diabetic retinopathy in primary care offices

Michael D. Abramoff et al.

NPJ DIGITAL MEDICINE (2018)

Article Medicine, General & Internal

Development and Validation of a Deep Learning Algorithm for Detection of Diabetic Retinopathy in Retinal Fundus Photographs

Varun Gulshan et al.

JAMA-JOURNAL OF THE AMERICAN MEDICAL ASSOCIATION (2016)

Article Ophthalmology

A Multi-Center Diabetes Eye Screening Study in Community Settings: Study Design and Methodology

Ann P. Murchison et al.

OPHTHALMIC EPIDEMIOLOGY (2016)

Article Ophthalmology

Improved Automated Detection of Diabetic Retinopathy on a Publicly Available Dataset Through Integration of Deep Learning

Michael David Abramoff et al.

INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE (2016)

Article Materials Science, Multidisciplinary

FEEDBACK ON A PUBLICLY DISTRIBUTED IMAGE DATABASE: THE MESSIDOR DATABASE

Etienne Decenciere et al.

IMAGE ANALYSIS & STEREOLOGY (2014)

Article Ophthalmology

Automated Analysis of Retinal Images for Detection of Referable Diabetic Retinopathy

Michael D. Abramoff et al.

JAMA OPHTHALMOLOGY (2013)

Article Ophthalmology

Improved Automated Screening of Diabetic Retinopathy

Carlos M. Oliveira et al.

OPHTHALMOLOGICA (2011)

Article Ophthalmology

Automated detection of fundus photographic red lesions in diabetic retinopathy

M Larsen et al.

INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE (2003)

Article Ophthalmology

Automated detection of diabetic retinopathy in a fundus photographic screening population

N Larsen et al.

INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE (2003)