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

Artificial intelligence for diabetic retinopathy screening in Africa

Artificial intelligence using deep learning to screen for referable and vision-threatening diabetic retinopathy in Africa: a clinical validation study

The objective of this study is to evaluate the sensitivity, specificity, and accuracy of artificial intelligence (AI) for diabetic retinopathy (DR) screening in community in Binh Dinh Province in Vietnam. This retrospective, descriptive, cross-sectional study assessed the DR screening efficacy of EyeArt system v2.0 by analyzing 2332 nonmydriatic digital fundus pictures of 583 diabetic patients from hospitals and health centers in Binh Dinh province. First, we selected thirty patients with 120 digital fundus pictures to perform the Kappa index by two eye doctors who would be responsible for the DR clinical feature evaluation and DR severity scale classification. Second, all digital fundus pictures were coded and then sent to the two above-mentioned eye doctors for the evaluation and classifications according to the International Committee of Ophthalmology's guidelines. Finally, DR severity scales with EyeArt were compared with those by eye doctors as a reference standard for EyeArt's effectiveness. All the data were analyzed using the SPSS software version 20.0. Values (with confidence interval 95%) of sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were calculated according to DR state, referable or not and vision-threatening DR state or not. P < 0.05 was considered statistically significant. The sensitivity and specificity of EyeArt for DR screening were 94.1% and 87.2%. The sensitivity and specificity for referable DR and vision-threatening DR were 96.6%, 90.1%, and 100.0%, 92.2%. Accuracy for DR screening, referable DR, and vision-threatening DR were 88.9%, 91.4%, and 93.0%, respectively. EyeArt AI was effective for DR screening in community.

Robust integration of diabetic retinopathy (DR) screening within health systems is essential to prevent DR-related blindness. This, however, remains a challenge in the developing world. The aim of this study was to evaluate two models of DR screening programs within rural general health-care services. This was a retrospective observational study from two rural health centers. Demographic and clinical data of patients completing DR screening were analyzed. Patients were screened in regular ophthalmology clinics (ROC) or integrated diabetic clinics (IDC). Referral and treatment completion data were retrieved from the clinical charts at the base hospital. A total of 2535 DR screenings were conducted for 2296 patients. The total population prevalence for any DR was 14.2% (95% confidence interval [CI]: 12.8%-15.6%) and vision-threatening DR (VTDR) was 4.7% (95% CI: 3.8%-5.6%). In the ROC and IDC groups, respectively, the prevalence of any DR was 20.4% and 8.2%, VTDR, 7.8% and 1.7%, and blindness, 1.4% and 0.4% (all P < 0.001). Referral completion rates were higher in the ROC group (44.8% vs. 25.2%, P < 0.001), while treatment completion in both was similar (69.6% vs. 70.6%). Referral and treatment completion rates for referable DR were 61.2% and 48.2%, and for VTDR, 62% and 38.8%, respectively. Only 11.45% of patients completed the repeat screening follow-up. Patients attending IDCs had a significantly lower prevalence of any DR, VTDR, and blindness demonstrating the advantages of integrated diabetic care in a rural setting. However, referral uptake and DR treatment completion need strengthening.

Retrospective studies on artificial intelligence (AI) in screening for diabetic retinopathy (DR) have shown promising results in addressing the mismatch between the capacity to implement DR screening and increasing DR incidence. This review sought to evaluate the diagnostic test accuracy (DTA) of AI in screening for referable diabetic retinopathy (RDR) in real-world settings. We searched CENTRAL, PubMed, CINAHL, Scopus, and Web of Science on 9 February 2023. We included prospective DTA studies assessing AI against trained human graders (HGs) in screening for RDR in patients with diabetes. Two reviewers independently extracted data and assessed methodological quality against QUADAS-2 criteria. We used the hierarchical summary receiver operating characteristics (HSROC) model to pool estimates of sensitivity and specificity and, forest plots and SROC plots to visually examine heterogeneity in accuracy estimates. From our initial search results of 3899 studies, we included 15 studies comprising 17 datasets. Meta-analyses revealed a sensitivity of 95.33% (95%CI: 90.60-100%) and specificity of 92.01% (95%CI: 87.61-96.42%) for patient-level analysis (10 datasets, N = 45,785) while, for the eye-level analysis, sensitivity was 91.24% (95%CI: 79.15-100%) and specificity, 93.90% (95%CI: 90.63-97.16%) (7 datasets, N = 15,390). Subgroup analyses did not provide variations in the diagnostic accuracy of country classification and DR classification criteria. However, a moderate increase was observed in diagnostic accuracy in the primary-level healthcare settings: sensitivity of 99.35% (95%CI: 96.85-100%), specificity of 93.72% (95%CI: 88.83-98.61%) and, a minimal decrease in the tertiary-level healthcare settings: sensitivity of 94.71% (95%CI: 89.00-100%), specificity of 90.88% (95%CI: 83.22-98.53%). Sensitivity analyses did not show any variations in studies that included diabetic macular edema in the RDR definition, nor studies with ≥3 HGs. This review provides evidence, for the first time from prospective studies, for the effectiveness of AI in screening for RDR in real-world settings. The results may serve to strengthen existing guidelines to improve current practices.

Highlights from the Diabetes UK Professional Conference 2017

Prevalence of Diabetic Retinopathy and Blindness in Indonesian Adults With Type 2 Diabetes

Objectives: This study investigated the visual search behaviour of optometrists and fellowship-trained ophthalmologists when screening for diabetic retinopathy in retinal photographs. Methods: Participants assessed and graded retinal photographs on a computer screen while a Gazepoint GP3 HD eye tracker recorded their eye movements. Areas of interest were derived from the raw data using Hidden Markov modelling. Fixation strings were extracted by matching raw fixation data to areas of interest and resolving ambiguities with graph search algorithms. Fixation strings were clustered using Affinity Propagation to determine search behaviours characteristic of the correct and incorrect response groups. Results: A total of 23 participants (15 optometrists and 8 ophthalmologists) completed the grading task, with each assessing 20 images. Visual search behaviour differed between correct and incorrect responses, with data suggesting correct responses followed a visual search strategy incorporating the optic disc, macula, superior arcade, and inferior arcade as areas of interest. Data from incorrect responses suggest search behaviour driven by saliency or a search pattern unrelated to anatomical landmarks. Referable diabetic retinopathy was correctly identified in 86% of cases. Grader accuracy was 64.8% with good inter-grader agreement (α = 0.818). Conclusions: Our study suggests that a structured visual search strategy is correlated with higher accuracy when assessing retinal photographs for diabetic retinopathy. Referable diabetic retinopathy is detected at high rates; however, there is disagreement between clinicians when determining a precise severity grade.

Application of deep learning image assessment software VeriSee™ for diabetic retinopathy screening

OBJECTIVETo compare the efficacy of the diabetic retinopathy (DR) screening with digital camera by endocrinologists with that by specialist and resident ophthalmologists in terms of sensitivity, specificity, and level of “loss of chance.”RESEARCH DESIGN AND METHODSIn a cross-sectional study, 500 adult diabetic patients (1,000 eyes) underwent three-field retinal photography with a digital fundus camera following pupillary dilatation. Five endocrinologists and two ophthalmology residents underwent 40 h of training on screening and grading of DR and detection of associated retinal findings. A κ test compared the accuracy of endocrinologist and ophthalmology resident screening with that performed by experienced ophthalmologists. Screening efficiency of endocrinologists was evaluated in terms of “loss of chance,” i.e., missed diagnoses that required ophthalmologist referrals.RESULTSThe mean weighted κ of DR screening performed by endocronologists was similar to that of ophthalmology residents (0.65 vs. 0.73). Out of 456 DR eyes, both endocrinologists and ophthalmology residents misdiagnosed only stage 1 DR (36 and 14, respectively), which did not require ophthalmologist referral. There were no significant differences between endocrinologists and ophthalmology residents in terms of diabetic maculopathy and incidental findings except for papillary cupping and choroidal lesions, which were not the main purpose of the study or of the training.CONCLUSIONSThe endocrinologist with specific training for DR detection using a three-field digital fundus camera with pupillary dilatation can perform a reliable DR screening without any loss of chance for the patients when compared with identical evaluation performed by experienced ophthalmologists.

Purpose : Diabetic retinopathy (DR) is a leading cause of blindness in low- and middle-income countries. This study aims to assess the agreement and diagnostic test accuracy of DR grading using fundus photographs by the allied medical personnel (AMP) as compared to ophthalmologist at DR screening program. Methods : Two 45 degree disc-centred and macula centred fundus photographs captured using digital fundus camera from 1344 eyes were graded by six AMP twice at three months and six months of training. DR grading was done using Early Treatment Diabetic Retinopathy Study criteria. Retina specialist grading was taken as the reference standard. Percentage agreement using kappa coefficient (k) and diagnostic test accuracy were assessed. Results : Intra-rater agreement of most AMP was almost perfect for detection of any DR and vision-threatening diabetic retinopathy (VTDR) at three and six months. Inter-rater agreement at 3 months at second grading was substantial for any DR (k = 0.62; 95% confidence interval (CI): 0.67–0.68) and VTDR (k = 0.63; 95% CI: 0.47–0.75). Inter-rater agreement in 6 months at second grading was also substantial for any DR (k = 0.66; 95% CI: 0.61–0.71) and VTDR (k = 0.64; 95% CI: 0.52–0.75). The sensitivity and specificity for any DR at second grading in 3 months was 94.99% and 63.97% and at 6 months was 95.55% and 66.91%, respectively. Conclusion : The inter-rater agreement of AMP as compared to ophthalmologists for any DR or VTDR was substantial at both 3 and 6 months. AMP can support as first-level DR graders for timely referral and treatment of VTDR.

BackgroundDiabetic Retinopathy (DR) is a leading cause of blindness worldwide, affecting people with diabetes. The timely diagnosis and treatment of DR are essential in preventing vision loss. Non-mydriatic fundus cameras and artificial intelligence (AI) software have been shown to improve DR screening efficiency. However, few studies have compared the diagnostic performance of different non-mydriatic cameras and AI software.MethodsThis clinical study was conducted at the endocrinology clinic of Akdeniz University with 900 volunteer patients that were previously diagnosed with diabetes but not with diabetic retinopathy. Fundus images of each patient were taken using three non-mydriatic fundus cameras and EyeCheckup AI software was used to diagnose more than mild diabetic retinopathy, vision-threatening diabetic retinopathy, and clinically significant diabetic macular oedema using images from all three cameras. Then patients underwent dilation and 4 wide-field fundus photography. Three retina specialists graded the 4 wide-field fundus images according to the diabetic retinopathy treatment preferred practice patterns of the American Academy of Ophthalmology. The study was pre-registered on clinicaltrials.gov with the ClinicalTrials.gov Identifier: NCT04805541.ResultsThe Canon CR2 AF AF camera had a sensitivity and specificity of 95.65% / 95.92% for diagnosing more than mild DR, the Topcon TRC-NW400 had 95.19% / 96.46%, and the Optomed Aurora had 90.48% / 97.21%. For vision threatening diabetic retinopathy, the Canon CR2 AF had a sensitivity and specificity of 96.00% / 96.34%, the Topcon TRC-NW400 had 98.52% / 95.93%, and the Optomed Aurora had 95.12% / 98.82%. For clinically significant diabetic macular oedema, the Canon CR2 AF had a sensitivity and specificity of 95.83% / 96.83%, the Topcon TRC-NW400 had 98.50% / 96.52%, and the Optomed Aurora had 94.93% / 98.95%.ConclusionThe study demonstrates the potential of using non-mydriatic fundus cameras combined with artificial intelligence software in detecting diabetic retinopathy. Several cameras were tested and, notably, each camera exhibited varying but adequate levels of sensitivity and specificity. The Canon CR2 AF emerged with the highest accuracy in identifying both more than mild diabetic retinopathy and vision-threatening cases, while the Topcon TRC-NW400 excelled in detecting clinically significant diabetic macular oedema. The findings from this study emphasize the importance of considering a non mydriatic camera and artificial intelligence software for diabetic retinopathy screening. However, further research is imperative to explore additional factors influencing the efficiency of diabetic retinopathy screening using AI and non mydriatic cameras such as costs involved and effects of screening using and on an ethnically diverse population.

This review assesses the current status of the different methods used in screening for diabetic retinopathy. This update is particularly timely because the incidence of diabetes is rising rapidly and the number of patients with vision-threatening diabetic retinopathy is increasing. We evaluate the different methods used and their results in improving the delivery of eye care to patients with diabetic retinopathy. In populations with poor access to ophthalmic care, screening techniques such as the nonmydriatic camera used in offices of primary care physicians may be useful in identifying lesions of diabetic retinopathy requiring treatment. One of the limitations is the lack of dilation and cataract formation, which may result in ungradable photographs. Patients with treatable lesions as well as those with ungradable photographs should be referred for comprehensive ocular examination. Screening techniques do not replace the eye examination. Ophthalmologists can play an important role in diabetic care apart from treating eye disease. Counseling can be provided to patients regarding the importance of blood glucose and blood pressure control and may motivate patients to achieve strict glucose and blood pressure control.

Background: Current manual diabetic retinopathy (DR) screening using eye care experts cannot scale to screen the growing population of diabetes patients who are at risk for vision loss. EyeArt system is an automated, cloud-based artificial intelligence (AI) eye screening technology designed to easily detect referral-warranted DR immediately through automated analysis of patient's retinal images.Methods: This retrospective study assessed the diagnostic efficacy of the EyeArt system v2.0 analyzing 850,908 fundus images from 101,710 consecutive patient visits, collected from 404 primary care clinics. Presence or absence of referral-warranted DR (more than mild nonproliferative DR [NPDR]) was automatically detected by the EyeArt system for each patient encounter, and its performance was compared against a clinical reference standard of quality-assured grading by rigorously trained certified ophthalmologists and optometrists.Results: Of the 101,710 visits, 75.7% were nonreferable, 19.3% were referable to an eye care specialist, and in 5.0%, the DR level was unknown as per the clinical reference standard. EyeArt screening had 91.3% (95% confidence interval [CI]: 90.9–91.7) sensitivity and 91.1% (95% CI: 90.9–91.3) specificity. For 5446 encounters with potentially treatable DR (more than moderate NPDR and/or diabetic macular edema), the system provided a positive “refer” output to 5363 encounters achieving sensitivity of 98.5%.Conclusions: This study captures variations in real-world clinical practice and shows that an AI DR screening system can be safe and effective in the real world. This study demonstrates the value of this easy-to-use, automated tool for endocrinologists, diabetologists, and general practitioners to address the growing need for DR screening and monitoring.

Introduction and Objective: Between March and September 2020, routine healthcare in the UK, including diabetic retinopathy (DR) screening, was paused due to the COVID-19 pandemic. This created a natural experiment to explore the impact of extended screening intervals. Methods: Data from Diabetic Eye Screening Wales (DESW) and primary care were linked using the Secure Anonymised Information Linkage databank (SAIL). Individuals screened within 12 months pre-lockdown and rescreened between March 2020 and August 2022 were included. Results: Of 12,424 individuals with no DR pre-lockdown, 84.3% remained free of retinopathy post-lockdown, 15.8% progressed to background retinopathy, and 0.3% to referable retinopathy. Background Retinopathy (Pre-Lockdown): Of 11,694 individuals with background DR pre-lockdown, 30.8% regressed to no retinopathy, 64.9% remained with background DR, and 4.4% progressed to referable diabetic retinopathy (RDR). Of 731 individuals with RDR pre-lockdown, 3.3% had no DR post-lockdown, 36.0% regressed to background DR, and 60.7% remained at referable level. Of 24,866 individuals with no maculopathy pre-lockdown, 94% remained maculopathy-free, while 6.0% developed diabetic maculopathy. Diabetic Of 257 individuals with diabetic maculopathy pre-lockdown, 50.6% regressed to no maculopathy, while 49.4% remained unchanged. Conclusion: The risk of progression from no DR to RDR or maculopathy was low (0.3% and 6.0%, respectively). Only 10% of those referred with suggested maculopathy required treatment. This study provides further evidence that screening biennially is considered safe for those without DR Disclosure R.L. Thomas: None. W. Cheung: None. J. Rafferty: None. L. Pratt: None. J. Cherry: None. R. Reynolds: None. G. Williams: None. S.C. Bain: Research Support; Novo Nordisk A/S, Lilly Diabetes. Speaker's Bureau; Menarini, AstraZeneca. Research Support; Boehringer-Ingelheim, Abbott Diagnostics, Amgen Inc. D.R. Owens: None. Funding Health and Care Research Wales (HRG-20-1776(P))