Search

AimMethodsResultsConclusionRegular screening of large number of people with diabetes for diabetic retinopathy (DR) with the support of available human resources alone is a global challenge. Digital health innovation is a boon in screening for DR. Multiple artificial intelligence (AI)-based deep learning (DL) algorithms have shown promise for accurate diagnosis of referable DR (RDR). The aim of this review is to evaluate the use of AI for DR screening and the various currently available automated DR detection algorithms.We reviewed articles published up to May 15th 2024, on the use of AI for DR by searching PubMed, Medline, Embase, Scopus, and Google Scholar using keywords like diabetic retinopathy, retinal imaging, teleophthalmology, automated detection, artificial intelligence, deep learning and fundus photography.This narrative review, traces the advent of AI and its use in digital health, the key concepts in AI and DL algorithm development for diagnosis of DR, some crucial AI algorithms that have been validated for detection of DR and the benefits and challenges of use of AI in detection and management of DR. While there are many approved AI algorithms that are in use globally for DR detection, IDx-DR, EyeArt, and AEYE Diagnostic Screening (AEYE-DS) are the algorithms that have been approved so far by USFDA for automated DR screening.AI has revolutionized screening of DR by enabling early automated detection. Continuous advances in AI technology, combined with high-quality retinal imaging, can lead to early diagnosis of sight-threatening DR, appropriate referrals, and better outcomes. [ABSTRACT FROM AUTHOR]

Purpose: Risk factors (RFs), like 'body mass index (BMI),' 'age,' and 'gender' correlate with Diabetic Retinopathy (DR) diagnosis and have been widely studied. This study examines how these three secondary RFs independently affect the predictive capacity of primary RFs. Methods: The dataset consisted of four population-based studies on the prevalence of DR and associated RFs in India between 2001 and 2010. An Autoencoder was employed to categorize RFs as primary or secondary. This study evaluated six primary RFs coupled independently with each secondary RF on five machine-learning models. Results: The secondary RF 'gender' gave a maximum increase in Area under the curve (AUC) score to predict DR when combined separately with 'insulin treatment,' 'fasting plasma glucose,' 'hypertension history,' and 'glycosylated hemoglobin' with a maximum increase in AUC for the Naive Bayes model from 0.573 to 0.646, for the Support Vector Machines (SVM) model from 0.644 to 0.691, for the SVM model from 0.487 to 0.607, and for the Decision Tree model from 0.8 to 0.848, respectively. The secondary RFs 'age' and 'BMI' gave a maximum increase in AUC score to predict DR when combined separately with 'diabetes mellitus duration' and 'systolic blood pressure,' with a maximum increase in AUC for the SVM model from 0.389 to 0.621, and for the Decision Tree model from 0.617 to 0.713, respectively. Conclusion: The risk factor 'gender' was the best secondary RF in predicting DR compared to 'age' and 'BMI,' increasing the predictive power of four primary RFs. [ABSTRACT FROM AUTHOR]

Objective: To evaluate the appropriateness of responses generated by an online chat-based artificial intelligence (AI) model for diabetic retinopathy (DR) related questions. Design: Cross-sectional study. Methods: A set of 20 questions framed from the patient's perspective addressing DR-related queries, such as the definition of disease, symptoms, prevention methods, treatment options, diagnostic methods, visual impact, and complications, were formulated for input into ChatGPT-4. Peer-reviewed, literature-based answers were collected from popular search engines for the selected questions and three retinal experts reviewed the responses. An inter-human agreement was analyzed for consensus expert responses and also between experts. The answers generated by the AI model were compared with those provided by the experts. The experts rated the response generated by ChatGPT-4 on a scale of 0-5 for appropriateness and completeness. Results: The answers provided by ChatGPT-4 were appropriate and complete for most of the DR-related questions. The response to questions on the adverse effects of laser photocoagulation therapy and compliance to treatment was not perfectly complete. The average rating given by the three retina expert evaluators was 4.84 for appropriateness and 4.38 for completeness of answers provided by the AI model. This corresponds to an overall 96.8% agreement among the experts for appropriateness and 87.6% for completeness regarding AI-generated answers. Conclusion: ChatGPT-4 exhibits a high level of accuracy in generating appropriate responses for a range of questions in DR. However, there is a need to improvise the model to generate complete answers for certain DR-related topics. [ABSTRACT FROM AUTHOR]

Purpose: The aim of this study was to report visual and anatomical outcomes following treatment for diabetic macular edema (DME) in clinical practice in India. Methods: Retrospective chart review of patients with DME who were initiated on treatment and followed up for at least 1 year at 9 tertiary eye care centers during 2016–2017 was performed. Data on demographics, systemic illnesses, visual acuity and anatomical characteristics of DME, treatment history were collated and analyzed for change in visual acuity level and central macular thickness at 1 year. Results: A total 1853 patients were diagnosed with treatable DME during study period, 1315 patients were treated and 556 patients (1019 eyes) followed up at one year. Although patients achieved significantly better anatomical outcome (central macular thickness of

Purpose: To evaluate the ability of an artificial intelligence (AI) model, ChatGPT, in predicting the diabetic retinopathy (DR) risk.Methods: This retrospective observational study utilized an anonymized dataset of 111 patients with diabetes who underwent a comprehensive eye examination along with clinical and biochemical assessments. Clinical and biochemical data along with and without central subfield thickness (CST) values of the macula from OCT were uploaded to ChatGPT-4, and the response from the ChatGPT was compared to the clinical DR diagnosis made by an ophthalmologist.Results: The study assessed the consistency of responses provided by ChatGPT, yielding an Intraclass Correlation Coefficient (ICC) value of 0.936 (95% CI, 0.913– 0.954, p < 0.001) (with CST) and 0.915 (95% CI, 0.706– 0.846, p < 0.001) (without CST), both situations indicated excellent reliability. The sensitivity and specificity of ChatGPT in predicting the DR cases were evaluated. The results revealed a sensitivity of 67% with CST and 73% without CST. The specificity was 68% with CST and 54% without CST. However, Cohen's kappa revealed only a fair agreement between ChatGPT predictions and clinical DR status in both situations, with CST (kappa = 0.263, p = 0.005) and without CST (kappa = 0.351, p < 0.001).Conclusion: This study suggests that ChatGPT has the potential of a preliminary DR screening tool with further optimization needed for clinical use. [ABSTRACT FROM AUTHOR]

Aims: To estimate the prevalence of undiagnosed diabetes and suboptimally controlled diabetes and the associated risk factors by community screening in India. Methods: In this multi‐centre, cross‐sectional study, house‐to‐house screening was conducted in people aged ≥40 years in urban and rural areas across 10 states and one union territory in India between November 2018 and March 2020. Participants underwent anthropometry, clinical and biochemical assessments. Capillary random blood glucose and point‐of‐care glycated haemoglobin (HbA1c) were used to diagnose diabetes. The prevalence of undiagnosed diabetes and suboptimal control (HbA1c ≥53 mmol/mol [≥7%]) among those with known diabetes was assessed. Results: Among the 42,146 participants screened (22,150 urban, 19,996 rural), 5689 had known diabetes. The age‐standardised prevalence of known diabetes was 13.1% (95% CI 12.8–13.4); 17.2% in urban areas and 9.4% in rural areas. The age‐standardised prevalence of undiagnosed diabetes was 6.0% (95% CI 5.7–6.2); similar in both urban and rural areas with the highest proportions seen in the East (8.0%) and South (7.8%) regions. When we consider all people with diabetes in the population, 22.8% of individuals in urban areas and 36.7% in rural areas had undiagnosed diabetes. Almost 75% of the individuals with known diabetes had suboptimal glycaemic control. Conclusions: High prevalence of undiagnosed diabetes and suboptimally controlled diabetes emphasises the urgent need to identify and optimally treat people with diabetes to reduce the burden of diabetes. [ABSTRACT FROM AUTHOR]