Search

Funding Information: This work is supported by National Medical Research Council Individual Research Grant (NMRC/0975/2005), National Medical Research Council Center Grant (NMRC/CG/C010A/2017_SERI) and Nurturing Clinician Researcher Scheme Program Grant Award (05/FY2021/P2/11-A92). Publisher Copyright: © 2023, The Author(s). Our study aims to identify children at risk of developing high myopia for timely assessment and intervention, preventing myopia progression and complications in adulthood through the development of a deep learning system (DLS). Using a school-based cohort in Singapore comprising of 998 children (aged 6–12 years old), we train and perform primary validation of the DLS using 7456 baseline fundus images of 1878 eyes; with external validation using an independent test dataset of 821 baseline fundus images of 189 eyes together with clinical data (age, gender, race, parental myopia, and baseline spherical equivalent (SE)). We derive three distinct algorithms – image, clinical and mix (image + clinical) models to predict high myopia development (SE ≤ −6.00 diopter) during teenage years (5 years later, age 11–17). Model performance is evaluated using area under the receiver operating curve (AUC). Our image models (Primary dataset AUC 0.93–0.95; Test dataset 0.91–0.93), clinical models (Primary dataset AUC 0.90–0.97; Test dataset 0.93–0.94) and mixed (image + clinical) models (Primary dataset AUC 0.97; Test dataset 0.97–0.98) achieve clinically acceptable performance. The addition of 1 year SE progression variable has minimal impact on the DLS performance (clinical model AUC 0.98 versus 0.97 in primary dataset, 0.97 versus 0.94 in test dataset; mixed model AUC 0.99 versus 0.97 in primary dataset, 0.95 versus 0.98 in test dataset). Thus, our DLS allows prediction of the development of high myopia by teenage years amongst school-going children. This has potential utility as a clinical-decision support tool to identify “at-risk” children for early intervention. publishersversion published

Pathologic myopia is a severe form of myopia that can lead to permanent visual impairment. The recent global increase in the prevalence of myopia has been projected to lead to a higher incidence of pathologic myopia in the future. Thus, imaging myopic eyes to detect early pathological changes, or predict myopia progression to allow for early intervention, has become a key priority. Recent advances in optical coherence tomography (OCT) have contributed to the new grading system for myopic maculopathy and myopic traction maculopathy, which may improve phenotyping and thus, clinical management. Widefield fundus and OCT imaging has improved the detection of posterior staphyloma. Non-invasive OCT angiography has enabled depth-resolved imaging for myopic choroidal neovascularisation. Artificial intelligence (AI) has shown great performance in detecting pathologic myopia and the identification of myopia-associated complications. These advances in imaging with adjunctive AI analysis may lead to improvements in monitoring disease progression or guiding treatments. In this review, we provide an update on the classification of pathologic myopia, how imaging has improved clinical evaluation and management of myopia-associated complications, and the recent development of AI algorithms to aid the detection and classification of pathologic myopia.

The application of artificial intelligence (AI) in medicine and ophthalmology has experienced exponential breakthroughs in recent years in diagnosis, prognosis, and aiding clinical decision-making. The use of digital data has also heralded the need for privacy-preserving technology to protect patient confidentiality and to guard against threats such as adversarial attacks. Hence, this review aims to outline novel AI-based systems for ophthalmology use, privacy-preserving measures, potential challenges, and future directions of each.Several key AI algorithms used to improve disease detection and outcomes include: Data-driven, imagedriven, natural language processing (NLP)-driven, genomics-driven, and multimodality algorithms. However, deep learning systems are susceptible to adversarial attacks, and use of data for training models is associated with privacy concerns. Several data protection methods address these concerns in the form of blockchain technology, federated learning, and generative adversarial networks.AI-applications have vast potential to meet many eyecare needs, consequently reducing burden on scarce healthcare resources. A pertinent challenge would be to maintain data privacy and confidentiality while supporting AI endeavors, where data protection methods would need to rapidly evolve with AI technology needs. Ultimately, for AI to succeed in medicine and ophthalmology, a balance would need to be found between innovation and privacy.

Purpose To develop and evaluate the performance of a 3-dimensional (3D) deep-learning–based automated digital gonioscopy system (DGS) in detecting 2 major characteristics in eyes with suspected primary angle-closure glaucoma (PACG): (1) narrow iridocorneal angles (static gonioscopy, Task I) and (2) peripheral anterior synechiae (PAS) (dynamic gonioscopy, Task II) on OCT scans. Design International, cross-sectional, multicenter study. Participants A total of 1.112 million images of 8694 volume scans (2294 patients) from 3 centers were included in this study (Task I, training/internal validation/external testing: 4515, 1101, and 2222 volume scans, respectively; Task II, training/internal validation/external testing: 378, 376, and 102 volume scans, respectively). Methods For Task I, a narrow angle was defined as an eye in which the posterior pigmented trabecular meshwork was not visible in more than 180° without indentation in the primary position captured in the dark room from the scans. For Task II, PAS was defined as the adhesion of the iris to the trabecular meshwork. The diagnostic performance of the 3D DGS was evaluated in both tasks with gonioscopic records as reference. Main Outcome Measures The area under the curve (AUC), sensitivity, and specificity of the 3D DGS were calculated. Results In Task I, 29.4% of patients had a narrow angle. The AUC, sensitivity, and specificity of 3D DGS on the external testing datasets were 0.943 (0.933–0.953), 0.867 (0.838–0.895), and 0.878 (0.859–0.896), respectively. For Task II, 13.8% of patients had PAS. The AUC, sensitivity, and specificity of 3D DGS were 0.902 (0.818–0.985), 0.900 (0.714–1.000), and 0.890 (0.841–0.938), respectively, on the external testing set at quadrant level following normal clinical practice; and 0.885 (0.836–0.933), 0.912 (0.816–1.000), and 0.700 (0.660–0.741), respectively, on the external testing set at clock-hour level. Conclusions The 3D DGS is effective in detecting eyes with suspected PACG. It has the potential to be used widely in the primary eye care community for screening of subjects at high risk of developing PACG.

The accelerated growth in electronic health records (EHR), Internet-of-Things, mHealth, telemedicine, and artificial intelligence (AI) in the recent years have significantly fuelled the interest and development in big data research. Big data refer to complex datasets that are characterized by the attributes of “5 Vs”—variety, volume, velocity, veracity, and value. Big data analytics research has so far benefitted many fields of medicine, including ophthalmology. The availability of these big data not only allow for comprehensive and timely examinations of the epidemiology, trends, characteristics, outcomes, and prognostic factors of many diseases, but also enable the development of highly accurate AI algorithms in diagnosing a wide range of medical diseases as well as discovering new patterns or associations of diseases that are previously unknown to clinicians and researchers. Within the field of ophthalmology, there is a rapidly expanding pool of large clinical registries, epidemiological studies, omics studies, and biobanks through which big data can be accessed. National corneal transplant registries, genome-wide association studies, national cataract databases, and large ophthalmology-related EHR-based registries (e.g., AAO IRIS Registry) are some of the key resources. In this review, we aim to provide a succinct overview of the availability and clinical applicability of big data in ophthalmology, particularly from the perspective of corneal diseases and cataract, the synergistic potential of big data, AI technologies, internet of things, mHealth, and wearable smart devices, and the potential barriers for realizing the clinical and research potential of big data in this field.

Purpose To determine whether eyes with pathologic myopia can be identified and whether each type of myopic maculopathy lesion on fundus photographs can be diagnosed by deep learning (DL) algorithms. Design A DL algorithm was developed to recognize myopic maculopathy features and to categorize the myopic maculopathy automatically. Participants We examined 7020 fundus images from 4432 highly myopic eyes obtained from the Advanced Clinical Center for Myopia. Methods Deep learning (DL) algorithms were developed to recognize the key features of myopic maculopathy with 5176 fundus images. These algorithms were also used to develop a Meta-analysis for Pathologic Myopia (META-PM) study categorizing system (CS) by adding a specific processing layer. Models and the system were evaluated by 1844 fundus image. The area under the receiver operating characteristic curve (AUC), sensitivity, and specificity were used to determine the performance of each DL algorithm. The rate of correct predictions was used to determine the performance of the META-PM study CS. Main Outcome Measures Four trained DL models were able to recognize the lesions of myopic maculopathy accurately with high sensitivity and specificity. The META-PM study CS also showed a high accuracy and was qualified to be used in a semiautomated way during screening for myopic maculopathy in highly myopic eyes. Results The sensitivity of the DL models was 84.44% for diffuse atrophy, 87.22% for patchy atrophy, 85.10% for macular atrophy, and 37.07% for choroidal neovascularization, and the AUC values were 0.970, 0.978, 0.982, and 0.881, respectively. The rate of total correct predictions from the META-PM study CS was 87.53%, with rates of 90.18%, 95.28%, 97.50%, and 91.14%, respectively, for each type of lesion. The META-PM study CS showed an overall rate of 92.08% in detecting pathologic myopia correctly, which was defined as having myopic maculopathy equal to or more serious than diffuse atrophy. Conclusions The novel DL models and system can achieve high sensitivity and specificity in identifying the different types of lesions of myopic maculopathy. These results will assist in the screening for pathologic myopia and subsequent protection of patients against low vision and blindness caused by myopic maculopathy.

Corneal diseases, uncorrected refractive errors, and cataract represent the major causes of blindness globally. The number of refractive surgeries, either cornea- or lens-based, is also on the rise as the demand for perfect vision continues to increase. With the recent advancement and potential promises of artificial intelligence (AI) technologies demonstrated in the realm of ophthalmology, particularly retinal diseases and glaucoma, AI researchers and clinicians are now channeling their focus toward the less explored ophthalmic areas related to the anterior segment of the eye. Conditions that rely on anterior segment imaging modalities, including slit-lamp photography, anterior segment optical coherence tomography, corneal tomography, in vivo confocal microscopy and/or optical biometers, are the most commonly explored areas. These include infectious keratitis, keratoconus, corneal grafts, ocular surface pathologies, preoperative screening before refractive surgery, intraocular lens calculation, and automated refraction, among others. In this review, we aimed to provide a comprehensive update on the utilization of AI in anterior segment diseases, with particular emphasis on the recent advancement in the past few years. In addition, we demystify some of the basic principles and terminologies related to AI, particularly machine learning and deep learning, to help improve the understanding, research and clinical implementation of these AI technologies among the ophthalmologists and vision scientists. As we march toward the era of digital health, guidelines such as CONSORT-AI, SPIRIT-AI, and STARD-AI will play crucial roles in guiding and standardizing the conduct and reporting of AI-related trials, ultimately promoting their potential for clinical translation.

Retinal blood vessels provide information on the risk of cardiovascular disease (CVD). Here, we report the development and validation of deep-learning models for the automated measurement of retinal-vessel calibre in retinal photographs, using diverse multiethnic multicountry datasets that comprise more than 70,000 images. Retinal-vessel calibre measured by the models and by expert human graders showed high agreement, with overall intraclass correlation coefficients of between 0.82 and 0.95. The models performed comparably to or better than expert graders in associations between measurements of retinal-vessel calibre and CVD risk factors, including blood pressure, body-mass index, total cholesterol and glycated-haemoglobin levels. In retrospectively measured prospective datasets from a population-based study, baseline measurements performed by the deep-learning system were associated with incident CVD. Our findings motivate the development of clinically applicable explainable end-to-end deep-learning systems for the prediction of CVD on the basis of the features of retinal vessels in retinal photographs. Deep-learning models for the automated measurement of retinal-vessel calibre in retinal photographs perform comparably to or better than expert graders in associations of measurements of retinal-vessel calibre with cardiovascular risk factors.

Purpose: This review details the rationale behind recommendations recently published by the American Society of Retina Specialists (ASRS) and outlines best practices for safety of vitreoretinal surgeons and staff while performing vitreoretinal surgery during the coronavirus disease 2019 (COVID-19) pandemic. Methods: The Committee for ASRS Best Practices for Retinal Surgery During the COVID-19 Pandemic reviewed existing evidence and information on severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) transmission and risk factors during vitreoretinal surgery. Recommendations were based on best available published data, cumulative clinical experiences, and recommendations and policies from other organizations. The Grading of Recommendations Assessment, Development and Evaluation approach was used to assess the strength of recommendations and confidence in the evidence. These serve as interim recommendations that will be routinely updated given the current gaps of knowledge and lack of high-quality data on this evolving subject. Results: Relevant existing literature related to methods of transmission and ocular manifestations of SARS-CoV-2 are summarized. The data and clinical experiences driving recommendations for preoperative, intraoperative, and postoperative surgical considerations and anesthesia choice as well as considerations for intravitreal injections are provided. Conclusions: Recommendations are provided with the goal of protecting vitreoretinal surgeons and associated personnel from exposure to SARS-CoV-2 during interventional vitreoretinal procedures. This is a rapidly evolving topic with numerous remaining gaps in our current knowledge. As such, recommendations will evolve and the present article is intended to serve as a foundation for continued dialogue on best practices.

The World Health Organization declared the Coronavirus Disease 2019 (COVID-19) caused by Severe Acute Respiratory Syndrome Coronavirus 2 a “Pandemic” on March 11, 2020. As of June 1, 2020, Severe Acute Respiratory Syndrome Coronavirus 2 has infected >6.2 million people and caused >372,000 deaths, including many health care personnel. It is highly infectious and ophthalmologists are at a higher risk of the infection due to a number of reasons including the proximity between doctors and patients during ocular examinations, microaerosols generated by the noncontact tonometer, tears as a potential source of infection, and some COVID-19 cases present with conjunctivitis. This article describes the ocular manifestations of COVID-19 and the APAO guidelines in mitigating the risks of contracting and/or spreading COVID-19 in ophthalmic practices.

Background Deep learning (DL) has demonstrated human expert levels of performance for medical image classification in a wide array of medical fields, including ophthalmology. In this article, we present the results of our DL system designed to determine optic disc laterality, right eye vs left eye, in the presence of both normal and abnormal optic discs. Methods Using transfer learning, we modified the ResNet-152 deep convolutional neural network (DCNN), pretrained on ImageNet, to determine the optic disc laterality. After a 5-fold cross-validation, we generated receiver operating characteristic curves and corresponding area under the curve (AUC) values to evaluate performance. The data set consisted of 576 color fundus photographs (51% right and 49% left). Both 30° photographs centered on the optic disc (63%) and photographs with varying degree of optic disc centration and/or wider field of view (37%) were included. Both normal (27%) and abnormal (73%) optic discs were included. Various neuro-ophthalmological diseases were represented, such as, but not limited to, atrophy, anterior ischemic optic neuropathy, hypoplasia, and papilledema. Results Using 5-fold cross-validation (70% training; 10% validation; 20% testing), our DCNN for classifying right vs left optic disc achieved an average AUC of 0.999 (±0.002) with optimal threshold values, yielding an average accuracy of 98.78% (±1.52%), sensitivity of 98.60% (±1.72%), and specificity of 98.97% (±1.38%). When tested against a separate data set for external validation, our 5-fold cross-validation model achieved the following average performance: AUC 0.996 (±0.005), accuracy 97.2% (±2.0%), sensitivity 96.4% (±4.3%), and specificity 98.0% (±2.2%). Conclusions Small data sets can be used to develop high-performing DL systems for semantic labeling of neuro-ophthalmology images, specifically in distinguishing between right and left optic discs, even in the presence of neuro-ophthalmological pathologies. Although this may seem like an elementary task, this study demonstrates the power of transfer learning and provides an example of a DCNN that can help curate large medical image databases for machine-learning purposes and facilitate ophthalmologist workflow by automatically labeling images according to laterality.

Background: The introduction of artificial intelligence (AI) in medicine has raised significant ethical, economic, and scientific controversies. Introduction: Because an explicit goal of AI is to perform processes previously reserved for human clinicians and other health care personnel, there is justified concern about the impact on patient safety, efficacy, equity, and liability. Discussion: Systems for computer-assisted and fully automated detection, triage, and diagnosis of diabetic retinopathy (DR) from retinal images show great variation in design, level of autonomy, and intended use. Moreover, the degree to which these systems have been evaluated and validated is heterogeneous. We use the term DR AI system as a general term for any system that interprets retinal images with at least some degree of autonomy from a human grader. We put forth these standardized descriptors to form a means to categorize systems for computer-assisted and fully automated detection, triage, and diagnosis of DR. The components of the categorization system include level of device autonomy, intended use, level of evidence for diagnostic accuracy, and system design. Conclusion: There is currently minimal empirical basis to assert that certain combinations of autonomy, accuracy, or intended use are better or more appropriate than any other. Therefore, at the current stage of development of this document, we have been descriptive rather than prescriptive, and we treat the different categorizations as independent and organized along multiple axes.

BackgroundMany artificial intelligence (AI) studies have focused on development of AI models, novel techniques, and reporting guidelines. However, little is understood about clinicians' perspectives of AI applications in medical fields including ophthalmology, particularly in light of recent regulatory guidelines. The aim for this study was to evaluate the perspectives of ophthalmologists regarding AI in 4 major eye conditions: diabetic retinopathy (DR), glaucoma, age-related macular degeneration (AMD) and cataract.MethodsThis was a multi-national survey of ophthalmologists between March 1st, 2020 to February 29th, 2021 disseminated via the major global ophthalmology societies. The survey was designed based on microsystem, mesosystem and macrosystem questions, and the software as a medical device (SaMD) regulatory framework chaired by the Food and Drug Administration (FDA). Factors associated with AI adoption for ophthalmology analyzed with multivariable logistic regression random forest machine learning.ResultsOne thousand one hundred seventy-six ophthalmologists from 70 countries participated with a response rate ranging from 78.8 to 85.8% per question. Ophthalmologists were more willing to use AI as clinical assistive tools (88.1%, n = 890/1,010) especially those with over 20 years' experience (OR 3.70, 95% CI: 1.10–12.5, p = 0.035), as compared to clinical decision support tools (78.8%, n = 796/1,010) or diagnostic tools (64.5%, n = 651). A majority of Ophthalmologists felt that AI is most relevant to DR (78.2%), followed by glaucoma (70.7%), AMD (66.8%), and cataract (51.4%) detection. Many participants were confident their roles will not be replaced (68.2%, n = 632/927), and felt COVID-19 catalyzed willingness to adopt AI (80.9%, n = 750/927). Common barriers to implementation include medical liability from errors (72.5%, n = 672/927) whereas enablers include improving access (94.5%, n = 876/927). Machine learning modeling predicted acceptance from participant demographics with moderate to high accuracy, and area under the receiver operating curves of 0.63–0.83.ConclusionOphthalmologists are receptive to adopting AI as assistive tools for DR, glaucoma, and AMD. Furthermore, ML is a useful method that can be applied to evaluate predictive factors on clinical qualitative questionnaires. This study outlines actionable insights for future research and facilitation interventions to drive adoption and operationalization of AI tools for Ophthalmology.

Importance: Deep learning (DL) networks require large data sets for training, which can be challenging to collect clinically. Generative models could be used to generate large numbers of synthetic optical coherence tomography (OCT) images to train such DL networks for glaucoma detection. Objective: To assess whether generative models can synthesize circumpapillary optic nerve head OCT images of normal and glaucomatous eyes and determine the usability of synthetic images for training DL models for glaucoma detection. Design, Setting, and Participants: Progressively growing generative adversarial network models were trained to generate circumpapillary OCT scans. Image gradeability and authenticity were evaluated on a clinical set of 100 real and 100 synthetic images by 2 clinical experts. DL networks for glaucoma detection were trained with real or synthetic images and evaluated on independent internal and external test data sets of 140 and 300 real images, respectively. Main Outcomes and Measures: Evaluations of the clinical set between the experts were compared. Glaucoma detection performance of the DL networks was assessed using area under the curve (AUC) analysis. Class activation maps provided visualizations of the regions contributing to the respective classifications. Results: A total of 990 normal and 862 glaucomatous eyes were analyzed. Evaluations of the clinical set were similar for gradeability (expert 1: 92.0%; expert 2: 93.0%) and authenticity (expert 1: 51.8%; expert 2: 51.3%). The best-performing DL network trained on synthetic images had AUC scores of 0.97 (95% CI, 0.95-0.99) on the internal test data set and 0.90 (95% CI, 0.87-0.93) on the external test data set, compared with AUCs of 0.96 (95% CI, 0.94-0.99) on the internal test data set and 0.84 (95% CI, 0.80-0.87) on the external test data set for the network trained with real images. An increase in the AUC for the synthetic DL network was observed with the use of larger synthetic data set sizes. Class activation maps showed that the regions of the synthetic images contributing to glaucoma detection were generally similar to that of real images. Conclusions and Relevance: DL networks trained with synthetic OCT images for glaucoma detection were comparable with networks trained with real images. These results suggest potential use of generative models in the training of DL networks and as a means of data sharing across institutions without patient information confidentiality issues. Agency for Science, Technology and Research (A*STAR) Nanyang Technological University National Medical Research Council (NMRC) National Research Foundation (NRF) Published version This work was funded by grants from the National Medical Research Council (grants CG/C010A/2017_SERI, OFLCG/004c/2018-00, MOH-000249-00, MOH-000647-00, MOH-001001-00, MOH-001015-00, MOH-000500-00, and MOH-000707-00), National Research Foundation Singapore (grants NRF2019-THE002-0006 and NRF-CRP24-2020-0001), A*STAR (grant A20H4b0141), the Singapore Eye Research Institute & Nanyang Technological University (SERI-NTU Advanced Ocular Engineering [STANCE] Program), and the SERI-Lee Foundation (grant LF1019-1) in Singapore.

Introduction Artificial intelligence (AI) and machine learning (ML) are rapidly evolving fields in various sectors, including healthcare. This article reviews AI’s present applications in healthcare, including its benefits, limitations and future scope. Sources of data A review of the English literature was conducted with search terms ‘AI’ or ‘ML’ or ‘deep learning’ and ‘healthcare’ or ‘medicine’ using PubMED and Google Scholar from 2000–2021. Areas of agreement AI could transform physician workflow and patient care through its applications, from assisting physicians and replacing administrative tasks to augmenting medical knowledge. Areas of controversy From challenges training ML systems to unclear accountability, AI’s implementation is difficult and incremental at best. Physicians also lack understanding of what AI implementation could represent. Growing points AI can ultimately prove beneficial in healthcare, but requires meticulous governance similar to the governance of physician conduct. Areas timely for developing research Regulatory guidelines are needed on how to safely implement and assess AI technology, alongside further research into the specific capabilities and limitations of its medical use.

To develop and evaluate the performance of a 3-dimensional (3D) deep-learning-based automated digital gonioscopy system (DGS) in detecting 2 major characteristics in eyes with suspected primary angle-closure glaucoma (PACG): (1) narrow iridocorneal angles (static gonioscopy, Task I) and (2) peripheral anterior synechiae (PAS) (dynamic gonioscopy, Task II) on OCT scans.International, cross-sectional, multicenter study.A total of 1.112 million images of 8694 volume scans (2294 patients) from 3 centers were included in this study (Task I, training/internal validation/external testing: 4515, 1101, and 2222 volume scans, respectively; Task II, training/internal validation/external testing: 378, 376, and 102 volume scans, respectively).For Task I, a narrow angle was defined as an eye in which the posterior pigmented trabecular meshwork was not visible in more than 180° without indentation in the primary position captured in the dark room from the scans. For Task II, PAS was defined as the adhesion of the iris to the trabecular meshwork. The diagnostic performance of the 3D DGS was evaluated in both tasks with gonioscopic records as reference.The area under the curve (AUC), sensitivity, and specificity of the 3D DGS were calculated.In Task I, 29.4% of patients had a narrow angle. The AUC, sensitivity, and specificity of 3D DGS on the external testing datasets were 0.943 (0.933-0.953), 0.867 (0.838-0.895), and 0.878 (0.859-0.896), respectively. For Task II, 13.8% of patients had PAS. The AUC, sensitivity, and specificity of 3D DGS were 0.902 (0.818-0.985), 0.900 (0.714-1.000), and 0.890 (0.841-0.938), respectively, on the external testing set at quadrant level following normal clinical practice; and 0.885 (0.836-0.933), 0.912 (0.816-1.000), and 0.700 (0.660-0.741), respectively, on the external testing set at clock-hour level.The 3D DGS is effective in detecting eyes with suspected PACG. It has the potential to be used widely in the primary eye care community for screening of subjects at high risk of developing PACG.

Purpose The purpose of this study was to assess whether the tractional elements of pathologic myopia (PM; e.g. myopic traction maculopathy [MTM], posterior staphyloma [PS], and aberrant posterior vitreous detachment [PVD]) are associated with myopic macular degeneration (MMD) independent of age and axial length, among highly myopic (HM) eyes. Methods One hundred twenty-nine individuals with 239 HM eyes from the Myopic and Pathologic Eyes in Singapore (MyoPES) cohort underwent ocular biometry, fundus photography, swept-source optical coherence tomography, and ocular B-scan ultrasound. Images were analyzed for PVD grade, and presence of MTM, PS, and MMD. The χ² test was done to determine the difference in prevalence of MMD between eyes with and without PVD, PS, and MTM. Multivariate probit regression analyses were performed to ascertain the relationship between the potential predictors (PVD, PS, and MTM) and outcome variable (MMD), after accounting for possible confounders (e.g. age and axial length). Marginal effects were reported. Results Controlling for potential confounders, eyes with MTM have a 29.92 percentage point higher likelihood of having MMD (P = 0.003), and eyes with PS have a 25.72 percentage point higher likelihood of having MMD (P = 0.002). The likelihood of MMD increases by 10.61 percentage points per 1 mm increase in axial length (P < 0.001). Subanalysis revealed that eyes with incomplete PVD have a 22.54 percentage point higher likelihood of having MMD than eyes with early PVD (P = 0.04). Conclusions Our study demonstrated an association between tractional (MTM, PS, and persistently incomplete PVD) and degenerative elements of PM independent of age and axial length. These data provide further insights into the pathogenesis of MMD.

Inherited retinal disease (IRD) affects approximately 1 in 3000 individuals in North America and Europe, and is a significant cause of visual impairment and blindness among children and working-age adults, with major personal and societal impact.1 2 Accurate clinical phenotypic and genotypic diagnosis of IRD is challenging, but increasingly important and relevant. Traditionally, genotypic diagnosis has been considered ‘nice to have’, but not ‘essential’, with implications usually related to patient prognostication and genetic counselling. However, an accurate genetic diagnosis is now of paramount importance because of rapid advances in potential gene replacement and other therapies for these previously untreatable conditions. In 2017, the first gene therapy for IRD was approved by the US Food and Drug Administration for the treatment of RPE65 -mediated retinal dystrophy, and shortly after by the European Medicines Agency as well.3 Multiple clinical trials are currently underway for other IRDs, including choroideraemia, Stargardt disease and retinitis pigmentosa (RP).4 5 Besides gene replacement therapy, progress in other areas such as antisense oligonucleotide therapy and gene editing with clustered regularly interspaced short palindromic repeats (CRISPR) and CRISPR-associated proteins also rely on accurate genetic diagnosis.5 6 Successful genotypic diagnosis remains elusive for many patients globally, due in part to remaining gaps in knowledge, but also due to limited access to testing, which remains relatively expensive, along with scarcity and an uneven distribution of institutions with expertise in IRD. In certain tertiary centres in the western world, patients have a high chance of an accurate genetic diagnosis. Recent studies have demonstrated the successful characterisation of large cohorts of patients with IRD using systematic clinical phenotyping and genetic testing protocols.7–10 Typically, historical, clinical, electrophysiological and multi-modal imaging data are used to assign each patient a clinical phenotypic category and to facilitate the selection of a genetic …

BackgroundTo develop computer-aided detection (CADe) of ORL abnormalities in the retinal pigmented epithelium, interdigitation zone and ellipsoid zone via optical coherence tomography (OCT).MethodsIn this retrospective study, healthy participants with normal ORL, and patients with abnormality of ORL including choroidal neovascularisation (CNV) or retinitis pigmentosa (RP) were included. First, an automatic segmentation deep learning (DL) algorithm, CADe, was developed for the three outer retinal layers using 120 handcraft masks of ORL. This automatic segmentation algorithm generated 4000 segmentations, which included 2000 images with normal ORL and 2000 (1000 CNV and 1000 RP) images with focal or wide defects in ORL. Second, based on the automatically generated segmentation images, a binary classifier (normal vs abnormal) was developed. Results were evaluated by area under the receiver operating characteristic curve (AUC).ResultsThe DL algorithm achieved an AUC of 0.984 (95% CI 0.976 to 0.993) for individual image evaluation in the internal test set of 797 images. In addition, performance analysis of a publicly available external test set (n=968) had an AUC of 0.957 (95% CI 0.944 to 0.970) and a second clinical external test set (n=1124) had an AUC of 0.978 (95% CI 0.970 to 0.986). Moreover, the CADe highlighted well normal parts of ORL and omitted highlights in abnormal ORLs of CNV and RP.ConclusionThe CADe can use OCT images to segment ORL and differentiate between normal ORL and abnormal ORL. The CADe classifier also performs visualisation and may aid future physician diagnosis and clinical applications.

Artificial intelligence (AI) has been billed as a key component of the Fourth Industrial Revolution. Currently, we are witnessing the growing shift of AI from theoretical ideations to practical applications in healthcare.1 2 Ophthalmology has emerged as one of the focal points of AI research.3–5 Current AI platforms are highly successful in screening for diabetic retinopathy, age-related macular degeneration and glaucoma.6–11 Other fields including cataract screening are similarly producing promising results.12 13 The WHO has identified that least 1 billion suffer from vision impairment that is preventable or treatable—of which myopia is a significant factor. With its growing prevalence in East Asia and many parts of the world, the ‘myopia pandemic’ is estimated to affect 50% (4.7 billion) of the world’s population by 2050, with 10% (1 billion) having high myopia (≤−5.00 D).14–16 This could lead to a staggering number of myopic individuals at risk of developing blinding conditions including myopic macular degeneration (MMD) and macular neovascularisation (MNV).17 However, AI research efforts in the field of refractive errors,18 particularly myopia19 are still relatively under-developed (table 1). View this table: Table 1 Summary of current Artificial Intelligence research in myopia The global attention towards myopia has led to a renewed focus on prediction, prevention, prognostication, early control as well as diagnostic accuracy.20 Early identification of high-risk individuals and unhindered access to appropriate healthcare will be critical in stemming the myopic tide. This has led to greater emphasis to develop dedicated AI models to address these unmet needs, especially for different phenotypes of myopia—childhood and adult myopia (high and pathological myopia). Relevant considerations include age, population size of each segment and measurable dataset, resource allocation, potential social burden, …

Background/aimsTo explore if retinal findings are associated with COVID-19 infection.MethodsIn this prospective cross-sectional study, we recruited participants positive for COVID-19 by nasopharyngeal swab, with no medical history. Subjects underwent retinal imaging with an automated imaging device (3D OCT-1 Maestro, Topcon, Tokyo, Japan) to obtain colour fundus photographs (CFP) and optical coherence tomographic (OCT) scans of the macula. Data on personal biodata, medical history and vital signs were collected from electronic medical records.Results108 patients were recruited. Mean age was 36.0±5.4 years. 41 (38.0%) had symptoms of acute respiratory infection (ARI) at presentation. Of 216 eyes, 25 (11.6%) had retinal signs—eight (3.7%) with microhaemorrhages, six (2.8%) with retinal vascular tortuosity and two (0.93%) with cotton wool spots (CWS). 11 eyes (5.1%) had hyper-reflective plaques in the ganglion cell-inner plexiform layer layer on OCT, of which two also had retinal signs visible on CFP (CWS and microhaemorrhage, respectively). There was no significant difference in the prevalence of retinal signs in symptomatic versus asymptomatic patients (12 (15.0%) vs 13 (9.6%), p=0.227). Patients with retinal signs were significantly more likely to have transiently elevated blood pressure than those without (p=0.03).ConclusionOne in nine had retinal microvascular signs on ocular imaging. These signs were observed even in asymptomatic patients with normal vital signs. These retinal microvascular signs may be related to underlying cardiovascular and thrombotic alternations associated with COVID-19 infection.

Deep learning (DL) has the potential to transform medical diagnostics. However, the diagnostic accuracy of DL is uncertain. Our aim was to evaluate the diagnostic accuracy of DL algorithms to identify pathology in medical imaging. Searches were conducted in Medline and EMBASE up to January 2020. We identified 11,921 studies, of which 503 were included in the systematic review. Eighty-two studies in ophthalmology, 82 in breast disease and 115 in respiratory disease were included for meta-analysis. Two hundred twenty-four studies in other specialities were included for qualitative review. Peer-reviewed studies that reported on the diagnostic accuracy of DL algorithms to identify pathology using medical imaging were included. Primary outcomes were measures of diagnostic accuracy, study design and reporting standards in the literature. Estimates were pooled using random-effects meta-analysis. In ophthalmology, AUC’s ranged between 0.933 and 1 for diagnosing diabetic retinopathy, age-related macular degeneration and glaucoma on retinal fundus photographs and optical coherence tomography. In respiratory imaging, AUC’s ranged between 0.864 and 0.937 for diagnosing lung nodules or lung cancer on chest X-ray or CT scan. For breast imaging, AUC’s ranged between 0.868 and 0.909 for diagnosing breast cancer on mammogram, ultrasound, MRI and digital breast tomosynthesis. Heterogeneity was high between studies and extensive variation in methodology, terminology and outcome measures was noted. This can lead to an overestimation of the diagnostic accuracy of DL algorithms on medical imaging. There is an immediate need for the development of artificial intelligence-specific EQUATOR guidelines, particularly STARD, in order to provide guidance around key issues in this field.

The COVID-19 pandemic has resulted in massive disruptions within health care, both directly as a result of the infectious disease outbreak, and indirectly because of public health measures to mitigate against transmission. This disruption has caused rapid dynamic fluctuations in demand, capacity, and even contextual aspects of health care. Therefore, the traditional face-to-face patient-physician care model has had to be re-examined in many countries, with digital technology and new models of care being rapidly deployed to meet the various challenges of the pandemic. This Viewpoint highlights new models in ophthalmology that have adapted to incorporate digital health solutions such as telehealth, artificial intelligence decision support for triaging and clinical care, and home monitoring. These models can be operationalised for different clinical applications based on the technology, clinical need, demand from patients, and manpower availability, ranging from out-of-hospital models including the hub-and-spoke pre-hospital model, to front-line models such as the inflow funnel model and monitoring models such as the so-called lighthouse model for provider-led monitoring. Lessons learnt from operationalising these models for ophthalmology in the context of COVID-19 are discussed, along with their relevance for other specialty domains.

Summary Background Coronary artery calcium (CAC) score is a clinically validated marker of cardiovascular disease risk. We developed and validated a novel cardiovascular risk stratification system based on deep-learning-predicted CAC from retinal photographs. Methods We used 216 152 retinal photographs from five datasets from South Korea, Singapore, and the UK to train and validate the algorithms. First, using one dataset from a South Korean health-screening centre, we trained a deep-learning algorithm to predict the probability of the presence of CAC (ie, deep-learning retinal CAC score, RetiCAC). We stratified RetiCAC scores into tertiles and used Cox proportional hazards models to evaluate the ability of RetiCAC to predict cardiovascular events based on external test sets from South Korea, Singapore, and the UK Biobank. We evaluated the incremental values of RetiCAC when added to the Pooled Cohort Equation (PCE) for participants in the UK Biobank. Findings RetiCAC outperformed all single clinical parameter models in predicting the presence of CAC (area under the receiver operating characteristic curve of 0·742, 95% CI 0·732–0·753). Among the 527 participants in the South Korean clinical cohort, 33 (6·3%) had cardiovascular events during the 5-year follow-up. When compared with the current CAC risk stratification (0, >0–100, and >100), the three-strata RetiCAC showed comparable prognostic performance with a concordance index of 0·71. In the Singapore population-based cohort (n=8551), 310 (3·6%) participants had fatal cardiovascular events over 10 years, and the three-strata RetiCAC was significantly associated with increased risk of fatal cardiovascular events (hazard ratio [HR] trend 1·33, 95% CI 1·04–1·71). In the UK Biobank (n=47 679), 337 (0·7%) participants had fatal cardiovascular events over 10 years. When added to the PCE, the three-strata RetiCAC improved cardiovascular risk stratification in the intermediate-risk group (HR trend 1·28, 95% CI 1·07–1·54) and borderline-risk group (1·62, 1·04–2·54), and the continuous net reclassification index was 0·261 (95% CI 0·124–0·364). Interpretation A deep learning and retinal photograph-derived CAC score is comparable to CT scan-measured CAC in predicting cardiovascular events, and improves on current risk stratification approaches for cardiovascular disease events. These data suggest retinal photograph-based deep learning has the potential to be used as an alternative measure of CAC, especially in low-resource settings. Funding Yonsei University College of Medicine; Ministry of Health and Welfare, Korea Institute for Advancement of Technology, South Korea; Agency for Science, Technology, and Research; and National Medical Research Council, Singapore.

Purpose To survey pediatric ophthalmologists on their perspectives of artificial intelligence (AI) in ophthalmology. Methods This is a subgroup analysis of a study previously reported. In March 2019, members of the American Association for Pediatric Ophthalmology and Strabismus (AAPOS) were recruited via the online AAPOS discussion board to voluntarily complete a Web-based survey consisting of 15 items. Survey items assessed the extent participants “agreed” or “disagreed” with statements on the perceived benefits and concerns of AI in ophthalmology. Responses were analyzed using descriptive statistics. Results A total of 80 pediatric ophthalmologists who are members of AAPOS completed the survey. The mean number of years since graduating residency was 21 years (range, 0-46). Overall, 91% (73/80) reported understanding the concept of AI, 70% (56/80) believed AI will improve the practice of ophthalmology, 68% (54/80) reported willingness to incorporate AI into their clinical practice, 65% (52/80) did not believe AI will replace physicians, and 71% (57/80) believed AI should be incorporated into medical school and residency curricula. However, 15% (12/80) were concerned that AI will replace physicians, 26% (21/80) believed AI will harm the patient-physician relationship, and 46% (37/80) reported concern over the diagnostic accuracy of AI. Conclusions Most pediatric ophthalmologists in this survey viewed the role of AI in ophthalmology positively.

BackgroundThe ability of deep learning (DL) algorithms to identify eyes with neovascular age-related macular degeneration (nAMD) from optical coherence tomography (OCT) scans has been previously established. We herewith evaluate the ability of a DL model, showing excellent performance on a Korean data set, to generalse onto an American data set despite ethnic differences. In addition, expert graders were surveyed to verify if the DL model was appropriately identifying lesions indicative of nAMD on the OCT scans.MethodsModel development data set—12 247 OCT scans from South Korea; external validation data set—91 509 OCT scans from Washington, USA. In both data sets, normal eyes or eyes with nAMD were included. After internal testing, the algorithm was sent to the University of Washington, USA, for external validation. Area under the receiver operating characteristic curve (AUC) and precision–recall curve (AUPRC) were calculated. For model explanation, saliency maps were generated using Guided GradCAM.ResultsOn external validation, AUC and AUPRC remained high at 0.952 (95% CI 0.942 to 0.962) and 0.891 (95% CI 0.875 to 0.908) at the individual level. Saliency maps showed that in normal OCT scans, the fovea was the main area of interest; in nAMD OCT scans, the appropriate pathological features were areas of model interest. Survey of 10 retina specialists confirmed this.ConclusionOur DL algorithm exhibited high performance for nAMD identification in a Korean population, and generalised well to an ethnically distinct, American population. The model correctly focused on the differences within the macular area to extract features associated with nAMD.

COVID-19 has led to massive disruptions in societal, economic and healthcare systems globally. While COVID-19 has sparked a surge and expansion of new digital business models in different industries, healthcare has been slower to adapt to digital solutions. The majority of ophthalmology clinical practices are still operating through a traditional model of ‘brick-and-mortar’ facilities and ‘face-to-face’ patient–physician interaction. In the current climate of COVID-19, there is a need to fuel implementation of digital health models for ophthalmology. In this article, we highlight the current limitations in traditional clinical models as we confront COVID-19, review the current lack of digital initiatives in ophthalmology sphere despite the presence of COVID-19, propose new digital models of care for ophthalmology and discuss potential barriers that need to be considered for sustainable transformation to take place.

Purpose of review The use of artificial intelligence (AI) in ophthalmology has increased dramatically. However, interpretation of these studies can be a daunting prospect for the ophthalmologist without a background in computer or data science. This review aims to share some practical considerations for interpretation of AI studies in ophthalmology. Recent findings It can be easy to get lost in the technical details of studies involving AI. Nevertheless, it is important for clinicians to remember that the fundamental questions in interpreting these studies remain unchanged - What does this study show, and how does this affect my patients? Being guided by familiar principles like study purpose, impact, validity, and generalizability, these studies become more accessible to the ophthalmologist. Although it may not be necessary for nondomain experts to understand the exact AI technical details, we explain some broad concepts in relation to AI technical architecture and dataset management. Summary The expansion of AI into healthcare and ophthalmology is here to stay. AI systems have made the transition from bench to bedside, and are already being applied to patient care. In this context, 'AI education' is crucial for ophthalmologists to be confident in interpretation and translation of new developments in this field to their own clinical practice.

Purpose of review Diabetic retinopathy is the most common specific complication of diabetes mellitus. Traditional care for patients with diabetes and diabetic retinopathy is fragmented, uncoordinated and delivered in a piecemeal nature, often in the most expensive and high-resource tertiary settings. Transformative new models incorporating digital technology are needed to address these gaps in clinical care. Recent findings Artificial intelligence and telehealth may improve access, financial sustainability and coverage of diabetic retinopathy screening programs. They enable risk stratifying patients based on individual risk of vision-threatening diabetic retinopathy including diabetic macular edema (DME), and predicting which patients with DME best respond to antivascular endothelial growth factor therapy. Summary Progress in artificial intelligence and tele-ophthalmology for diabetic retinopathy screening, including artificial intelligence applications in 'real-world settings' and cost-effectiveness studies are summarized. Furthermore, the initial research on the use of artificial intelligence models for diabetic retinopathy risk stratification and management of DME are outlined along with potential future directions. Finally, the need for artificial intelligence adoption within ophthalmology in response to coronavirus disease 2019 is discussed. Digital health solutions such as artificial intelligence and telehealth can facilitate the integration of community, primary and specialist eye care services, optimize the flow of patients within healthcare networks, and improve the efficiency of diabetic retinopathy management.

International audience; Introduction: Routine medical and ophthalmic care is being drastically curtailed in the context of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pandemic. Uveitis patients require particular attention because of their theoretical risk of viral infection, in the context of therapeutic immunosuppression.Areas covered: This collaborative work proposes practical management and follow-up criteria for uveitis patients in the context of the ongoing SARS-CoV-2 pandemic.Expert opinion: Management should proceed as usual when access to health care possible in patients who do not belong to a group at high risk of severe SARS-CoV-2 infection, and in uncontrolled uveitis cases. In case of reduced access to eye clinics or high risk of SARS-CoV-2 infection, patients' management should be stratified based on their clinical presentation. In non-severe uveitis cases, the use of systemic steroids should be avoided, and local steroids preferred whenever possible. In uncontrolled situations where there is real risk of permanent visual loss, high-dose intravenous steroids and/or systemic immunosuppressants and/or biotherapies can be administered depending on the severity of eye disease. Immunosuppressive therapy should not be withheld, unless the patient develops SARS-CoV2 infection.

Coronavirus disease 19 (COVID-19) was first reported in Wuhan, China, in December 2019, and has since become a global pandemic. Singapore was one of the first countries outside of China to be affected and reported its first case in January 2020. Strategies that were deployed successfully during the 2003 outbreak of severe acute respiratory syndrome have had to evolve to contain this novel coronavirus. Like the rest of the health care services in Singapore, the practice of ophthalmology has also had to adapt to this rapidly changing crisis. This article discusses the measures put in place by the 3 largest ophthalmology centers in Singapore's public sector in response to COVID-19, and the challenges of providing eye care in the face of stringent infection control directives, staff redeployments and “social distancing.” The recently imposed “circuit breaker,” effectively a partial lockdown of the country, has further limited our work to only the most essential of services. Our staff are also increasingly part of frontline efforts in the screening and care of patients with COVID-19. However, this crisis has also been an opportunity to push ahead with innovative practices and given momentum to the use of teleophthalmology and other digital technologies. Amidst this uncertainty, our centers are already planning for how ophthalmology in Singapore will be practiced in this next stage of the COVID-19 pandemic, and beyond.