Diabetic retinopathy screening in the emerging era of artificial intelligence

The efficacy of artificial intelligence in diabetic retinopathy screening: a systematic review and meta-analysis

BACKGROUND

To evaluate the efficacy of artificial intelligence (AI) in screening for diabetic retinopathy (DR) using fundus images and optical coherence tomography (OCT) in comparison to traditional screening methods.

METHODS

This systematic review was registered with PROSPERO (ID: CRD42024560750). Systematic searches were conducted in PubMed Medline, Cochrane Central, ScienceDirect, and Web of Science using keywords such as "diabetic retinopathy," "screening," and "artificial intelligence." Only studies published in English from 2019 to July 22, 2024, were considered. We also manually reviewed the reference lists of relevant reviews. Two independent reviewers assessed the risk of bias using the QUADAS-2 tool, resolving disagreements through discussion with the principal investigator. Meta-analysis was performed using MetaDiSc software (version 1.4). To calculate combined sensitivity, specificity, summary receiver operating characteristic (SROC) plots, forest plots, and subgroup analyses were performed according to clinician type (ophthalmologists vs. retina specialists) and imaging modality (fundus images vs. fundus images + OCT).

RESULTS

18 studies were included. Meta-analysis showed that AI systems demonstrated superior diagnostic performance compared to doctors, with the pooled sensitivity, specificity, diagnostic odds ratio, and Cochrane Q index of the AI being 0.877, 0.906, 0.94, and 153.79 accordingly. The Fagan nomogram analysis further confirmed the strong diagnostic value of AI. Subgroup analyses revealed that factors like imaging modality, and doctor expertise can influence diagnostic performance.

CONCLUSION

AI systems have demonstrated strong diagnostic performance in detecting diabetic retinopathy, with sensitivity and specificity comparable to or exceeding traditional clinicians.

Artificial Intelligence in the Training of Radiology Residents: a Multicenter Randomized Controlled Trial

The aim of the present study was to compare the effectiveness of AI-assisted training and conventional human training in clinical practice. This was a multicenter, randomized, controlled clinical trial conducted in five national-level residency training hospitals. Residents from five hospitals participated, divided into three groups: conventional training (Group A), conventional plus specialty training (Group B), and conventional plus AI-assisted training (Group C). The content of the training was ultrasound diagnosis of thyroid nodules. The training lasted for 18 months, and the three groups of participants were phase-tested every 3 months to compare the effect of the training. The diagnostic accuracy of all three groups gradually increased with increasing training time. Among the three groups, groups B and C had higher accuracy than group A (P < .001), and there was no significant difference between groups B and C (P = .64). Over the training period, diagnostic confidence increased in all groups. Negative activating emotions decreased significantly over time in all groups (95% CI, - 0.81 to - 0.37; P < .001), while positive activating emotions increased significantly (95% CI, 0.18 to 0.53; P < .001). Current research shows that all three approaches are viable for training radiology residents. Furthermore, the AI-assisted approach had no negative emotional impact on the trainees, suggesting that integrating AI into radiology training programs could provide a reliable and effective means of achieving the educational goals of medical education.

Evolution of ophthalmological care in adult with diabetes in France between 2010 and 2022: a nationwide study

PURPOSE

The aim of this study is to describe ophthalmological care of adults with diabetes in France and its evolution between 2010 and 2022.

METHODS

In this study, we used the ESND, a representative permanent random sample of 2/100th of the entire French population. Ophthalmological care was defined by the combination of ophthalmological procedures (fundus examination, color fundus photography, Optical Coherence Tomography..) and/or ophthalmological treatment (intravitreal injection or laser treatment) during the year. Changes in annual rates during the study period were assessed using linear regression models excluding 2020.

RESULTS

From 2010 to 2022, the number of adults treated for diabetes in the ENSD increased from 48 329 patients (mean age 65.3 ± 13.0, 46.3% women) to 68 397 patients (mean age 67.0 ± 13.2, 44.8% women). Among them, the annual rate of ophthalmological care was stable (46.5% in 2010 and 48.5% in 2022) and the difference was not significant (β = 0.10% per year, p = 0.11). The yearly ophthalmological treatment rate increased significantly (3.3% in 2010 and 5.3% in 2022, β = 0.2% per year, p < 0.0001). Rates were lower during the COVID-19 outbreak in 2020.Women, individuals aged between 66-80 years, those living in the least deprived areas and those treated with combined insulin and non-insulin treatment had higher yearly ophthalmological care rate.

CONCLUSION

In this large nationwide representative study with recent and updated data, although ophthalmological treatment rate has increased over the decade mainly due to intravitreal injections, less than half of the diabetic patients receive yearly ophthalmological care.

Performance and limitation of machine learning algorithms for diabetic retinopathy screening and its application in health management: a meta-analysis

BACKGROUND

In recent years, artificial intelligence and machine learning algorithms have been used more extensively to diagnose diabetic retinopathy and other diseases. Still, the effectiveness of these methods has not been thoroughly investigated. This study aimed to evaluate the performance and limitations of machine learning and deep learning algorithms in detecting diabetic retinopathy.

METHODS

This study was conducted based on the PRISMA checklist. We searched online databases, including PubMed, Scopus, and Google Scholar, for relevant articles up to September 30, 2023. After the title, abstract, and full-text screening, data extraction and quality assessment were done for the included studies. Finally, a meta-analysis was performed.

RESULTS

We included 76 studies with a total of 1,371,517 retinal images, of which 51 were used for meta-analysis. Our meta-analysis showed a significant sensitivity and specificity with a percentage of 90.54 (95%CI [90.42, 90.66], P < 0.001) and 78.33% (95%CI [78.21, 78.45], P < 0.001). However, the AUC (area under curvature) did not statistically differ across studies, but had a significant figure of 0.94 (95% CI [- 46.71, 48.60], P = 1).

CONCLUSIONS

Although machine learning and deep learning algorithms can properly diagnose diabetic retinopathy, their discriminating capacity is limited. However, they could simplify the diagnosing process. Further studies are required to improve algorithms.

Valuable insights into general practice staff's experiences and perspectives on AI-assisted diabetic retinopathy screening-An interview study

Aim

This study explores the hands-on experiences and perspectives of general practice staff regarding the feasibility of conducting artificial intelligence-assisted (AI-assisted) diabetic retinopathy screenings (DRS) in general practice settings.

Method

The screenings were tested in 12 general practices in the North Denmark Region and were conducted as part of daily care routines over ~4 weeks. Subsequently, 21 staff members involved in the DRS were interviewed.

Results

Thematic analysis generated four main themes: (1) Experiences with DRS in daily practice, (2) Effective DRS implementation in general practice in the future, (3) Trust and approval of AI-assisted DRS in general practice, and (4) Implications of DRS in general practice. The findings suggest that general practice staff recognise the potential for AI-assisted DRS to be integrated into their clinical workflows. However, they also emphasise the importance of addressing both practical and systemic factors to ensure successful implementation of DRS within the general practice setting.

Conclusion

Focusing on the practical experiences and perspectives of general practice staff, this study lays the groundwork for future research aimed at optimising the implementation of AI-assisted DRS in general practice settings, while recognising that the insights gained may also inform broader primary care contexts.

Grading of diabetic retinopathy using a pre-segmenting deep learning classification model: Validation of an automated algorithm

PURPOSE

To validate the performance of autonomous diabetic retinopathy (DR) grading by comparing a human grader and a self-developed deep-learning (DL) algorithm with gold-standard evaluation.

METHODS

We included 500, 6-field retinal images graded by an expert ophthalmologist (gold standard) according to the International Clinical Diabetic Retinopathy Disease Severity Scale as represented with DR levels 0-4 (97, 100, 100, 103, 100, respectively). Weighted kappa was calculated to measure the DR classification agreement for (1) a certified human grader without, and (2) with assistance from a DL algorithm and (3) the DL operating autonomously. Using any DR (level 0 vs. 1-4) as a cutoff, we calculated sensitivity, specificity, as well as positive and negative predictive values (PPV and NPV). Finally, we assessed lesion discrepancies between Model 3 and the gold standard.

RESULTS

As compared to the gold standard, weighted kappa for Models 1-3 was 0.88, 0.89 and 0.72, sensitivities were 95%, 94% and 78% and specificities were 82%, 84% and 81%. Extrapolating to a real-world DR prevalence of 23.8%, the PPV were 63%, 64% and 57% and the NPV were 98%, 98% and 92%. Discrepancies between the gold standard and Model 3 were mainly incorrect detection of artefacts (n = 49), missed microaneurysms (n = 26) and inconsistencies between the segmentation and classification (n = 51).

CONCLUSION

While the autonomous DL algorithm for DR classification only performed on par with a human grader for some measures in a high-risk population, extrapolations to a real-world population demonstrated an excellent 92% NPV, which could make it clinically feasible to use autonomously to identify non-DR patients.

Exploring general practice staff perspectives on a teaching concept based on instruction videos for diabetic retinopathy screening - an interview study

OBJECTIVE

The aim of this study is to explore general practice staff perspectives regarding a teaching concept based on instructional videos for conducting DR screenings. Furthermore, this study aims to investigate the competencies acquired by the staff through this teaching concept.

DESIGN AND SETTING

Qualitative cross-sectional study conducted in general practice clinics in the North Denmark Region.

METHOD

A teaching concept was developed based on instruction videos to teach general practice staff to conduct diabetic retinopathy screenings with automated grading through artificial intelligence. Semi-structured interviews were performed with 16 staff members to investigate their perspectives on the concept and acquired competencies.

RESULTS

This study found no substantial resistance to the teaching concept from staff; however, participants' satisfaction with the methods employed in the instruction session, the progression of learning curves, screening competencies, and their acceptance of a known knowledge gap during screenings varied slightly among the participants.

CONCLUSION

This study showed that the teaching concept can be used to teach general practice staff to conduct diabetic retinopathy screenings. Staffs' perspectives on the teaching concept and acquired competencies varied, and this study suggest few adjustments to the concept to accommodate staff's preferences and establish more consistent competencies.

Machine learning-based identification and validation of immune-related biomarkers for early diagnosis and targeted therapy in diabetic retinopathy

The early diagnosis of diabetic retinopathy (DR) is challenging, highlighting the urgent need to identify new biomarkers. Immune responses play a crucial role in DR, yet there are currently no reports of machine learning (ML) algorithms being utilized for the development of immune-related molecular markers in DR. Based on the datasets GSE102485 and GSE160306, differentially expressed genes (DEGs) were screened using Weighted Gene Co-expression Network Analysis (WGCNA). Five ML algorithms including Bayesian, Learning Vector Quantization (LVQ), Wrapper (Boruta), Random Forest (RF), and Logistic Regression were employed to select immune-related genes associated with DR (DR.Sig). Seven ML algorithms including Naive Bayes (NB), RF, Support Vector Machine (SVM), AdaBoost Classification Trees (AdaBoost), Boosted Logistic Regressions (LogitBoost), K-Nearest Neighbors (KNN), and Cancerclass were utilized to construct a predictive model for DR. The relationship between DR.Sig genes and immune cells was analyzed using single-sample Gene Set Enrichment Analysis (ssGSEA). Additionally, drug sensitivity prediction of DR.Sig genes and molecular docking were performed. Through the utilization of 5 ML algorithms, 6 immune-related biomarkers closely related to the occurrence of DR were identified, including FCGR2B, CSRP1, EDNRA, SDC2, TEK, and CIITA. The DR predictive model constructed based on these 6 DR.Sig genes using the Cancerclass algorithm demonstrated superior predictive performance compared to 4 previously published DR-related biomarkers. In vivo and in vitro experiments also provided strong validation of the expression of the 6 genes in DR. Positive correlations were observed between these genes and 22 types of immune cells. Molecular docking results revealed that CSRP1, EDNRA, and TEK exhibited the highest affinities with the small molecule compounds etoposide, FR-139317, and camptothecin, respectively. The models constructed based on various ML algorithms can effectively predict the occurrence of DR events and hold potential for targeted drug therapies, providing a basis for the early diagnosis and targeted treatment of DR.

Therapeutic Interventions for Diabetes Mellitus-associated Complications

BACKGROUND

Diabetes Mellitus (DM) is an alarming health concern, affecting approximately 537 million people worldwide. As a leading cause of morbidity and mortality, DM demands a comprehensive understanding of its diverse pathophysiological mechanisms and disease progression.

METHODS

This traditional review has consolidated literature on the pathogenesis of hyperglycemia, its progression into complications, and advances in optimal treatment strategies. The literature in the last two decades has been reviewed using several keywords, including "diabetes," "diabetes-associated complications", "novel therapeutic interventions for diabetes-associated diseases", "phyto-extracts as antidiabetic drugs", etc. in prominent databases, such as PubMed, Scopus, Google Scholar, Web of Science, and ClinicalTrials.gov.

RESULTS

We have discussed macrovascular and microvascular complications, such as atherosclerosis, cardiovascular disease, Peripheral Arterial Disease (PAD), stroke, diabetic nephropathy, retinopathy, and neuropathy, as well as various pharmacological and non-pharmacological interventions that are currently available for the management of DM. We have also focused on the potential of natural products in targeting molecular mechanisms involved in carbohydrate metabolism, insulin production, repair of pancreatic cells, and reduction of oxidative stress, thereby contributing to their antidiabetic activity. Additionally, novel therapeutic approaches, like genetic, stem cell, and immunomodulatory therapies, have been explored. We have also discussed the benefits and limitations of each intervention, emerging research and technologies, and precision medicine interventions.

CONCLUSION

This review has emphasized the need for an improved understanding of these advancements, which is essential to enhance clinicians' ability to identify the most effective therapeutic interventions.

Autonomous artificial intelligence versus teleophthalmology for diabetic retinopathy

Purpose: To assess the role of artificial intelligence (AI) based automated software for detection of Diabetic Retinopathy (DR) compared with the evaluation of digital retinography by two double masked retina specialists. Methods: Two-hundred one patients (mean age 65 ± 13 years) with type 1 diabetes mellitus or type 2 diabetes mellitus were included. All patients were undergoing a retinography and spectral domain optical coherence tomography (SD-OCT, DRI 3D OCT-2000, Topcon) of the macula. The retinal photographs were graded using two validated AI DR screening software (Eye Art TM and IDx-DR) designed to identify more than mild DR. Results: Retinal images of 201 patients were graded. DR (more than mild DR) was detected by the ophthalmologists in 38 (18.9%) patients and by the AI-algorithms in 36 patients (with 30 eyes diagnosed by both algorithms). Ungradable patients by the AI software were 13 (6.5%) and 16 (8%) for the Eye Art and IDx-DR, respectively. Both AI software strategies showed a high sensitivity and specificity for detecting any more than mild DR without showing any statistically significant difference between them. Conclusions: The comparison between the diagnosis provided by artificial intelligence based automated software and the reference clinical diagnosis showed that they can work at a level of sensitivity that is similar to that achieved by experts.

Two Handheld Retinograph Devices Evaluated by Ophthalmologists and an Artificial Intelligence Algorithm

Background/Objectives: Telemedicine in diabetic retinopathy (RD) screening is effective but does not reach the entire diabetes population. The use of portable cameras and artificial intelligence (AI) can help in screening diabetes. Methods: We evaluated the ability of two handheld cameras, one based on a smartphone and the other on a smartscope, to obtain images for comparison with OCT. Evaluation was carried out in two stages: the first by two retina specialists and the second using an artificial intelligence algorithm that we developed. Results: The retina specialists reported that the smartphone images required mydriasis in all cases, compared to 73.05% of the smartscope images and 71.11% of the OCT images. Images were ungradable in 27.98% of the retinographs with the smartphone and in 7.98% with the smartscope. The detection of any DR using the AI algorithm showed that the smartphone obtained lower recall values (0.89) and F1 scores (0.89) than the smartscope, with 0.99. Low results were also obtained using the smartphone to detect mild DR (146 retinographs), compared to using the smartscope (218 retinographs). Conclusions: we consider that the use of handheld devices together with AI algorithms for reading retinographs can be useful for DR screening, although the ease of image acquisition through small pupils with these devices needs to be improved.

Synchronous Diagnosis of Diabetic Retinopathy by a Handheld Retinal Camera, Artificial Intelligence, and Simultaneous Specialist Confirmation

PURPOSE

Diabetic retinopathy (DR) is a leading cause of preventable blindness, particularly in underserved regions where access to ophthalmic care is limited. This study presents a proof of concept for utilizing a portable handheld retinal camera with an embedded artificial intelligence (AI) platform, complemented by a synchronous remote confirmation by retina specialists, for DR screening in an underserved rural area.

DESIGN

Retrospective cohort study.

SUBJECTS

A total of 1115 individuals with diabetes.

METHODS

A retrospective analysis of a screening initiative conducted in 4 municipalities in Northeastern Brazil, targeting the diabetic population. A portable handheld retinal camera captured macula-centered and disc-centered images, which were analyzed by the AI system. Immediate push notifications were sent out to retina specialists upon the detection of significant abnormalities, enabling synchronous verification and confirmation, with on-site patient feedback within minutes. Referral criteria were established, and all referred patients underwent a complete ophthalmic work-up and subsequent treatment.

MAIN OUTCOME MEASURES

Proof-of-concept implementation success.

RESULTS

Out of 2052 invited individuals, 1115 participated, with a mean age of 60.93 years and diabetes duration of 7.52 years; 66.03% were women. The screening covered 2222 eyes, revealing various retinal conditions. Referable eyes for DR were 11.84%, with an additional 13% for other conditions (diagnoses included various stages of DR, media opacity, nevus, drusen, enlarged cup-to-disc ratio, pigmentary changes, and other). Artificial intelligence performance for overall detection of referable cases (both DR and other conditions) was as follows: sensitivity 84.23% (95% confidence interval (CI), 82.63-85.84), specificity 80.79% (95% CI, 79.05-82.53). When we assessed whether AI matched any clinical diagnosis, be it referable or not, sensitivity was 85.67% (95% CI, 84.12-87.22), specificity was 98.86 (95% CI, 98.39-99.33), and area under the curve was 0.92 (95% CI, 0.91-0.94).

CONCLUSIONS

The integration of a portable device, AI analysis, and synchronous medical validation has the potential to play a crucial role in preventing blindness from DR, especially in socially unequal scenarios.

FINANCIAL DISCLOSURE(S)

Proprietary or commercial disclosure may be found in the Footnotes and Disclosures at the end of this article.

Integrated image-based deep learning and language models for primary diabetes care

Primary diabetes care and diabetic retinopathy (DR) screening persist as major public health challenges due to a shortage of trained primary care physicians (PCPs), particularly in low-resource settings. Here, to bridge the gaps, we developed an integrated image-language system (DeepDR-LLM), combining a large language model (LLM module) and image-based deep learning (DeepDR-Transformer), to provide individualized diabetes management recommendations to PCPs. In a retrospective evaluation, the LLM module demonstrated comparable performance to PCPs and endocrinology residents when tested in English and outperformed PCPs and had comparable performance to endocrinology residents in Chinese. For identifying referable DR, the average PCP's accuracy was 81.0% unassisted and 92.3% assisted by DeepDR-Transformer. Furthermore, we performed a single-center real-world prospective study, deploying DeepDR-LLM. We compared diabetes management adherence of patients under the unassisted PCP arm (n = 397) with those under the PCP+DeepDR-LLM arm (n = 372). Patients with newly diagnosed diabetes in the PCP+DeepDR-LLM arm showed better self-management behaviors throughout follow-up (P < 0.05). For patients with referral DR, those in the PCP+DeepDR-LLM arm were more likely to adhere to DR referrals (P < 0.01). Additionally, DeepDR-LLM deployment improved the quality and empathy level of management recommendations. Given its multifaceted performance, DeepDR-LLM holds promise as a digital solution for enhancing primary diabetes care and DR screening.

Danger, Danger, Gaston Labat! Does zero-shot artificial intelligence correlate with anticoagulation guidelines recommendations for neuraxial anesthesia?

INTRODUCTION

Artificial intelligence and large language models (LLMs) have emerged as potentially disruptive technologies in healthcare. In this study GPT-3.5, an accessible LLM, was assessed for its accuracy and reliability in performing guideline-based evaluation of neuraxial bleeding risk in hypothetical patients on anticoagulation medication. The study also explored the impact of structured prompt guidance on the LLM's performance.

METHODS

A dataset of 10 hypothetical patient stems and 26 anticoagulation profiles (260 unique combinations) was developed based on American Society of Regional Anesthesia and Pain Medicine guidelines. Five prompts were created for the LLM, ranging from minimal guidance to explicit instructions. The model's responses were compared with a "truth table" based on the guidelines. Performance metrics, including accuracy and area under the receiver operating curve (AUC), were used.

RESULTS

Baseline performance of GPT-3.5 was slightly above chance. With detailed prompts and explicit guidelines, performance improved significantly (AUC 0.70, 95% CI (0.64 to 0.77)). Performance varied among medication classes.

DISCUSSION

LLMs show potential for assisting in clinical decision making but rely on accurate and relevant prompts. Integration of LLMs should consider safety and privacy concerns. Further research is needed to optimize LLM performance and address complex scenarios. The tested LLM demonstrates potential in assessing neuraxial bleeding risk but relies on precise prompts. LLM integration should be approached cautiously, considering limitations. Future research should focus on optimization and understanding LLM capabilities and limitations in healthcare.

Diabetic retinopathy screening through artificial intelligence algorithms: A systematic review

Diabetic retinopathy (DR) poses a significant challenge in diabetes management, with its progression often asymptomatic until advanced stages. This underscores the urgent need for cost-effective and reliable screening methods. Consequently, the integration of artificial intelligence (AI) tools presents a promising avenue to address this need effectively. We provide an overview of the current state of the art results and techniques in DR screening using AI, while also identifying gaps in research for future exploration. By synthesizing existing database and pinpointing areas requiring further investigation, this paper seeks to guide the direction of future research in the field of automatic diabetic retinopathy screening. There has been a continuous rise in the number of articles detailing deep learning (DL) methods designed for the automatic screening of diabetic retinopathy especially by the year 2021. Researchers utilized various databases, with a primary focus on the IDRiD dataset. This dataset consists of color fundus images captured at an ophthalmological clinic situated in India. It comprises 516 images that depict various stages of DR and diabetic macular edema. Each of the chosen papers concentrates on various DR signs. Nevertheless, a significant portion primarily focused on detecting exudates, which remains insufficient to assess the overall presence of this disease. Various AI methods have been employed to identify DR signs. Among the chosen papers, 4.7 % utilized detection methods, 46.5 % employed classification techniques, 41.9 % relied on segmentation, and 7 % opted for a combination of classification and segmentation. Metrics calculated from 80 % of the articles employing preprocessing techniques demonstrated the significant benefits of this approach in enhancing results quality. In addition, multiple DL techniques, starting by classification, detection then segmentation. Researchers used mostly YOLO for detection, ViT for classification, and U-Net for segmentation. Another perspective on the evolving landscape of AI models for diabetic retinopathy screening lies in the increasing adoption of Convolutional Neural Networks for classification tasks and U-Net architectures for segmentation purposes; however, there is a growing realization within the research community that these techniques, while powerful individually, can be even more effective when integrated. This integration holds promise for not only diagnosing DR, but also accurately classifying its different stages, thereby enabling more tailored treatment strategies. Despite this potential, the development of AI models for DR screening is fraught with challenges. Chief among these is the difficulty in obtaining the high-quality, labeled data necessary for training models to perform effectively. This scarcity of data poses significant barriers to achieving robust performance and can hinder progress in developing accurate screening systems. Moreover, managing the complexity of these models, particularly deep neural networks, presents its own set of challenges. Additionally, interpreting the outputs of these models and ensuring their reliability in real-world clinical settings remain ongoing concerns. Furthermore, the iterative process of training and adapting these models to specific datasets can be time-consuming and resource-intensive. These challenges underscore the multifaceted nature of developing effective AI models for DR screening. Addressing these obstacles requires concerted efforts from researchers, clinicians, and technologists to develop new approaches and overcome existing limitations. By doing so, a full potential of AI may transform DR screening and improve patient outcomes.

Current Status, Hotspots, and Prospects of Artificial Intelligence in Ophthalmology: A Bibliometric Analysis (2003-2023)

PURPOSE

Artificial intelligence (AI) has gained significant attention in ophthalmology. This paper reviews, classifies, and summarizes the research literature in this field and aims to provide readers with a detailed understanding of the current status and future directions, laying a solid foundation for further research and decision-making.

METHODS

Literature was retrieved from the Web of Science database. Bibliometric analysis was performed using VOSviewer, CiteSpace, and the R package Bibliometrix.

RESULTS

The study included 3,377 publications from 4,035 institutions in 98 countries. China and the United States had the most publications. Sun Yat-sen University is a leading institution. Translational Vision Science & Technology"published the most articles, while "Ophthalmology" had the most co-citations. Among 13,145 researchers, Ting DSW had the most publications and citations. Keywords included "Deep learning," "Diabetic retinopathy," "Machine learning," and others.

CONCLUSION

The study highlights the promising prospects of AI in ophthalmology. Automated eye disease screening, particularly its core technology of retinal image segmentation and recognition, has become a research hotspot. AI is also expanding to complex areas like surgical assistance, predictive models. Multimodal AI, Generative Adversarial Networks, and ChatGPT have driven further technological innovation. However, implementing AI in ophthalmology also faces many challenges, including technical, regulatory, and ethical issues, and others. As these challenges are overcome, we anticipate more innovative applications, paving the way for more effective and safer eye disease treatments.

Automated Identification of Different Severity Levels of Diabetic Retinopathy Using a Handheld Fundus Camera and Single-Image Protocol

Purpose

To evaluate the performance of artificial intelligence (AI) systems embedded in a mobile, handheld retinal camera, with a single retinal image protocol, in detecting both diabetic retinopathy (DR) and more-than-mild diabetic retinopathy (mtmDR).

Design

Multicenter cross-sectional diagnostic study, conducted at 3 diabetes care and eye care facilities.

Participants

A total of 327 individuals with diabetes mellitus (type 1 or type 2) underwent a retinal imaging protocol enabling expert reading and automated analysis.

Methods

Participants underwent fundus photographs using a portable retinal camera (Phelcom Eyer). The captured images were automatically analyzed by deep learning algorithms retinal alteration score (RAS) and diabetic retinopathy alteration score (DRAS), consisting of convolutional neural networks trained on EyePACS data sets and fine-tuned using data sets of portable device fundus images. The ground truth was the classification of DR corresponding to adjudicated expert reading, performed by 3 certified ophthalmologists.

Main Outcome Measures

Primary outcome measures included the sensitivity and specificity of the AI system in detecting DR and/or mtmDR using a single-field, macula-centered fundus photograph for each eye, compared with a rigorous clinical reference standard comprising the reading center grading of 2-field imaging protocol using the International Classification of Diabetic Retinopathy severity scale.

Results

Of 327 analyzed patients (mean age, 57.0 ± 16.8 years; mean diabetes duration, 16.3 ± 9.7 years), 307 completed the study protocol. Sensitivity and specificity of the AI system were high in detecting any DR with DRAS (sensitivity, 90.48% [95% confidence interval (CI), 84.99%-94.46%]; specificity, 90.65% [95% CI, 84.54%-94.93%]) and mtmDR with the combination of RAS and DRAS (sensitivity, 90.23% [95% CI, 83.87%-94.69%]; specificity, 85.06% [95% CI, 78.88%-90.00%]). The area under the receiver operating characteristic curve was 0.95 for any DR and 0.89 for mtmDR.

Conclusions

This study showed a high accuracy for the detection of DR in different levels of severity with a single retinal photo per eye in an all-in-one solution, composed of a portable retinal camera powered by AI. Such a strategy holds great potential for increasing coverage rates of screening programs, contributing to prevention of avoidable blindness.

Financial Disclosures

F.K.M. is a medical consultant for Phelcom Technologies. J.A.S. is Chief Executive Officer and proprietary of Phelcom Technologies. D.L. is Chief Technology Officer and proprietary of Phelcom Technologies. P.V.P. is an employee at Phelcom Technologies.

Application of artificial intelligence system for screening multiple fundus diseases in Chinese primary healthcare settings: a real-world, multicentre and cross-sectional study of 4795 cases

BACKGROUND/AIMS

This study evaluates the performance of the Airdoc retinal artificial intelligence system (ARAS) for detecting multiple fundus diseases in real-world scenarios in primary healthcare settings and investigates the fundus disease spectrum based on ARAS.

METHODS

This real-world, multicentre, cross-sectional study was conducted in Shanghai and Xinjiang, China. Six primary healthcare settings were included in this study. Colour fundus photographs were taken and graded by ARAS and retinal specialists. The performance of ARAS is described by its accuracy, sensitivity, specificity and positive and negative predictive values. The spectrum of fundus diseases in primary healthcare settings has also been investigated.

RESULTS

A total of 4795 participants were included. The median age was 57.0 (IQR 39.0-66.0) years, and 3175 (66.2%) participants were female. The accuracy, specificity and negative predictive value of ARAS for detecting normal fundus and 14 retinal abnormalities were high, whereas the sensitivity and positive predictive value varied in detecting different abnormalities. The proportion of retinal drusen, pathological myopia and glaucomatous optic neuropathy was significantly higher in Shanghai than in Xinjiang. Moreover, the percentages of referable diabetic retinopathy, retinal vein occlusion and macular oedema in middle-aged and elderly people in Xinjiang were significantly higher than in Shanghai.

CONCLUSION

This study demonstrated the dependability of ARAS for detecting multiple retinal diseases in primary healthcare settings. Implementing the AI-assisted fundus disease screening system in primary healthcare settings might be beneficial in reducing regional disparities in medical resources. However, the ARAS algorithm must be improved to achieve better performance.

TRIAL REGISTRATION NUMBER

NCT04592068.

Exploration of AI-powered DenseNet121 for effective diabetic retinopathy detection

OBJECTIVE

Diabetic Retinopathy (DR) is a severe complication of diabetes that damages the retina and affects approximately 80% of patients with diabetes for 10 years or more. This condition primarily impacts young and productive individuals, resulting in significant long-term medical complications for patients and society. The early stages of diabetic retinopathy often advance without noticeable symptoms, resulting in delayed identification and intervention. Therefore, develop approaches employing transfer learning methodologies to enhance early detection capabilities, facilitating timely diagnosis and intervention to mitigate the progression of diabetic retinopathy.

METHODS

This study introduces a transfer learning approach for detecting four stages of DR: No DR, Mild, Moderate, and Severe. The methods AlexNet, VGG16, ResNet50, Inception v3, and DenseNet121 are utilized and trained using the Kaggle DR dataset.

RESULTS

To assess the efficiency of the suggested improved network, the Kaggle dataset is employed to analyze four performance metrics: Sensitivity, Precision, Accuracy, and F1 score. DenseNet121 demonstrated superior accuracy among the two models, outperforming other models, making it a suitable option for automatic DR sign detection.

CONCLUSION

The integration of the DenseNet121 model shows great promise in transforming the timely identification and treatment of DR, resulting in enhanced patient results in the long run and alleviating the burden on society.

Clinical utility of handheld fundus and smartphone-based camera for monitoring diabetic retinal diseases: a review study

Diabetic retinopathy (DR) is the leading global cause of vision loss, accounting for 4.8% of global blindness cases as estimated by the World Health Organization (WHO). Fundus photography is crucial in ophthalmology as a diagnostic tool for capturing retinal images. However, resource and infrastructure constraints limit access to traditional tabletop fundus cameras in developing countries. Additionally, these conventional cameras are expensive, bulky, and not easily transportable. In contrast, the newer generation of handheld and smartphone-based fundus cameras offers portability, user-friendliness, and affordability. Despite their potential, there is a lack of comprehensive review studies examining the clinical utilities of these handheld (e.g. Zeiss Visuscout 100, Volk Pictor Plus, Volk Pictor Prestige, Remidio NMFOP, FC161) and smartphone-based (e.g. D-EYE, iExaminer, Peek Retina, Volk iNview, Volk Vistaview, oDocs visoScope, oDocs Nun, oDocs Nun IR) fundus cameras. This review study aims to evaluate the feasibility and practicality of these available handheld and smartphone-based cameras in medical settings, emphasizing their advantages over traditional tabletop fundus cameras. By highlighting various clinical settings and use scenarios, this review aims to fill this gap by evaluating the efficiency, feasibility, cost-effectiveness, and remote capabilities of handheld and smartphone fundus cameras, ultimately enhancing the accessibility of ophthalmic services.

A Narrative Review of the Advances in Screening Methods for Diabetic Retinopathy: Enhancing Early Detection and Vision Preservation

Diabetes mellitus (DM) is putting a great burden worldwide. This rise in DM cases, both type 1 and 2, significantly impacts public health. India has grappled with a diabetes epidemic for several years, leading to many misdiagnosed and untreated diabetes cases. Diabetes remains a significant factor in adult-onset blindness despite improvements in diabetes management. This increases the danger of diabetic retinopathy (DR) with permanent loss of sight for those affected. The screening for DR aims to identify those persons with complications arising from diabetes or DR, which could potentially result in blindness, so that treatment can be started immediately and blindness can be avoided. A comprehensive health system approach is required to ensure that the public sector in India effectively screens for DR. Improving patient outcomes and avoiding visual loss depends significantly on early identification and treatment. This article discusses the actions that should be implemented to establish a national effort for systematic DR screening. It also highlights the importance of screening in DR and its impact on treatment effectiveness. Regular screenings enable the early detection of retinopathy, allowing for timely intervention and treatment. Early screening helps prevent complications associated with DR, such as macular edema or retinal detachment. Screening also assists healthcare providers in planning, optimizing treatment approaches, and monitoring treatment effectiveness. Meanwhile, early intervention is essential for enhancing treatment outcomes, thus enhancing the chances of preserving vision and preventing further progression of the disease. This helps in improving the overall management of this sight-threatening complication.

An Overview of Diabetic Cardiomyopathy

Diabetic cardiomyopathy (DCM) is a myocardial disorder that is characterised by structural and functional abnormalities of the heart muscle in the absence of hypertension, valvular heart disease, congenital heart defects, or coronary artery disease (CAD). After witnessing a particular form of cardiomyopathy in diabetic individuals, Rubler et al. came up with the moniker diabetic cardiomyopathy in 1972. Four stages of DCM are documented, and the American College of Cardiology/American Heart Association Stage and New York Heart Association Class for HF have some overlap. Diabetes is linked to several distinct forms of heart failure. Around 40% of people with heart failure with preserved ejection fraction (HFpEF) have diabetes, which is thought to be closely associated with the pathophysiology of HFpEF. Diabetes and HF are uniquely associated in a bidirectional manner. When compared to the general population without diabetes, those with diabetes have a risk of heart failure that is up to four times higher. A biomarker is a trait that is reliably measured and assessed as a predictor of healthy biological activities, pathological processes, or pharmacologic responses to a clinical treatment. Several biomarker values have been discovered to be greater in patients with diabetes than in control subjects among those who have recently developed heart failure. Myocardial fibrosis and hypertrophy are the primary characteristics of DCM, and structural alterations in the diabetic myocardium are often examined by non-invasive, reliable, and reproducible procedures. An invasive method called endomyocardial biopsy (EMB) is most often used to diagnose many cardiac illnesses.

Bridging the Camera Domain Gap With Image-to-Image Translation Improves Glaucoma Diagnosis

Purpose

The purpose of this study was to improve the automated diagnosis of glaucomatous optic neuropathy (GON), we propose a generative adversarial network (GAN) model that translates Optain images to Topcon images.

Methods

We trained the GAN model on 725 paired images from Topcon and Optain cameras and externally validated it using an additional 843 paired images collected from the Aravind Eye Hospital in India. An optic disc segmentation model was used to assess the disparities in disc parameters across cameras. The performance of the translated images was evaluated using root mean square error (RMSE), peak signal-to-noise ratio (PSNR), structural similarity index (SSIM), 95% limits of agreement (LOA), Pearson's correlations, and Cohen's Kappa coefficient. The evaluation compared the performance of the GON model on Topcon photographs as a reference to that of Optain photographs and GAN-translated photographs.

Results

The GAN model significantly reduced Optain false positive results for GON diagnosis, with RMSE, PSNR, and SSIM of GAN images being 0.067, 14.31, and 0.64, respectively, the mean difference of VCDR and cup-to-disc area ratio between Topcon and GAN images being 0.03, 95% LOA ranging from -0.09 to 0.15 and -0.05 to 0.10. Pearson correlation coefficients increased from 0.61 to 0.85 in VCDR and 0.70 to 0.89 in cup-to-disc area ratio, whereas Cohen's Kappa improved from 0.32 to 0.60 after GAN translation.

Conclusions

Image-to-image translation across cameras can be achieved by using GAN to solve the problem of disc overexposure in Optain cameras.

Translational Relevance

Our approach enhances the generalizability of deep learning diagnostic models, ensuring their performance on cameras that are outside of the original training data set.

Integration of systematic review, lipidomics with experiment verification reveals abnormal sphingolipids facilitate diabetic retinopathy by inducing oxidative stress on RMECs

OBJECTIVE

This study aims to explore the potential biomarkers in the development of diabetes mellitus (DM) into diabetic retinopathy (DR).

METHODS

Systematic review of diabetic metabolomics was used to screen the differential metabolites and related pathways during the development of DM. Non-targeted lipidomics of rat plasma was performed to explore the differential metabolites in the development of DM into DR in vivo. To verify the effects of differential metabolites in inducing retinal microvascular endothelial cells (RMECs) injury by increasing oxidative stress, high glucose medium containing differential metabolites was used to induce rat RMECs injury and cell viability, malondialdehyde (MDA) contents, superoxide dismutase (SOD) activities, reactive oxygen species (ROS) levels and mitochondrial membrane potential (MMP) were evaluated in vitro. Network pharmacology was performed to explore the potential mechanism of differential metabolites in inducing DR.

RESULTS

Through the systematic review, 148 differential metabolites were obtained and the sphingolipid metabolic pathway attracted our attention. Plasma non-targeted lipidomics found that sphingolipids were accompanied by the development of DM into DR. In vitro experiments showed sphinganine and sphingosine-1-phosphate aggravated rat RMECs injury induced by high glucose, further increased MDA and ROS levels, and further decreased SOD activities and MMP. Network pharmacology revealed sphinganine and sphingosine-1-phosphate may induce DR by regulating the AGE-RAGE and HIF-1 signaling pathways.

CONCLUSIONS

Integrated systematic review, lipidomics and experiment verification reveal that abnormal sphingolipid metabolism facilitates DR by inducing oxidative stress on RMECs. Our study could provide the experimental basis for finding potential biomarkers for the diagnosis and treatment of DR.

Transfer learning for accurate fetal organ classification from ultrasound images: a potential tool for maternal healthcare providers

Ultrasound imaging is commonly used to aid in fetal development. It has the advantage of being real-time, low-cost, non-invasive, and easy to use. However, fetal organ detection is a challenging task for obstetricians, it depends on several factors, such as the position of the fetus, the habitus of the mother, and the imaging technique. In addition, image interpretation must be performed by a trained healthcare professional who can take into account all relevant clinical factors. Artificial intelligence is playing an increasingly important role in medical imaging and can help solve many of the challenges associated with fetal organ classification. In this paper, we propose a deep-learning model for automating fetal organ classification from ultrasound images. We trained and tested the model on a dataset of fetal ultrasound images, including two datasets from different regions, and recorded them with different machines to ensure the effective detection of fetal organs. We performed a training process on a labeled dataset with annotations for fetal organs such as the brain, abdomen, femur, and thorax, as well as the maternal cervical part. The model was trained to detect these organs from fetal ultrasound images using a deep convolutional neural network architecture. Following the training process, the model, DenseNet169, was assessed on a separate test dataset. The results were promising, with an accuracy of 99.84%, which is an impressive result. The F1 score was 99.84% and the AUC was 98.95%. Our study showed that the proposed model outperformed traditional methods that relied on the manual interpretation of ultrasound images by experienced clinicians. In addition, it also outperformed other deep learning-based methods that used different network architectures and training strategies. This study may contribute to the development of more accessible and effective maternal health services around the world and improve the health status of mothers and their newborns worldwide.

Dynamic Graph Clustering Learning for Unsupervised Diabetic Retinopathy Classification

Diabetic retinopathy (DR) is a common complication of diabetes, which can lead to vision loss. Early diagnosis is crucial to prevent the progression of DR. In recent years, deep learning approaches have shown promising results in the development of an intelligent and efficient system for DR classification. However, one major drawback is the need for expert-annotated datasets, which are both time-consuming and costly. To address these challenges, this paper proposes a novel dynamic graph clustering learning (DGCL) method for unsupervised classification of DR, which innovatively deploys the Euclidean and topological features from fundus images for dynamic clustering. Firstly, a multi-structural feature fusion (MFF) module extracts features from the structure of the fundus image and captures topological relationships among multiple samples, generating a fused representation. Secondly, another consistency smoothing clustering (CSC) module combines network updates and deep clustering to ensure stability and smooth performance improvement during model convergence, optimizing the clustering process by iteratively updating the network and refining the clustering results. Lastly, dynamic memory storage is utilized to track and store important information from previous iterations, enhancing the training stability and convergence. During validation, the experimental results with public datasets demonstrated the superiority of our proposed DGCL network.

Mendelian Randomization Analysis Reveals Causal Effects of Polyunsaturated Fatty Acids on Subtypes of Diabetic Retinopathy Risk

Polyunsaturated fatty acids (PUFAs) affect several physiological processes, including visual acuity, but their relationship with diabetic retinopathy (DR) remains elusive. The aim of this study was to determine whether PUFAs have a causal effect on DR. PUFAs- (total and omega-3 [FAw3] and omega-6 [FAw6] fatty acids and their ratio) and DR-associated single nucleotide polymorphisms derived from genome-wide association studies; sample sizes were 114,999 for fatty acids and 216,666 for any DR (ADR), background DR (BDR), severe non-proliferative DR (SNPDR), and proliferative DR (PDR). We hypothesized that the intra-body levels of PUFAs have an impact on DR and conducted a two-sample Mendelian randomization (MR) study to assess the causality. Pleiotropy, heterogeneity, and sensitivity analyses were performed to verify result reliability. High levels of PUFAs were found to be associated with reduced risk of both ADR and PDR. Moreover, FAw3 was associated with a decreased risk of PDR, whereas FAw6 demonstrated an association with lowered risks of both BDR and PDR. Our findings provide genetic evidence, for the first time, for a causal relationship between PUFAs and reduced DR risk. Consequently, our comprehensive MR analysis strongly urges further investigation into the precise functions and long-term effects of PUFAs, FAw3, and FAw6 on DR.

Comprehensive retinal vascular measurements: time in range is associated with peripheral retinal venular calibers in type 2 diabetes in China

AIM

Retinal vascular parameters are biomarkers of diabetic microangiopathy. We aimed to investigate the relationship between time in range (TIR) assessed by continuous glucose monitoring (CGM) and retinal vascular parameters in patients with type 2 diabetes in China.

METHODS

The TIR assessed by CGM and retinal photographs were obtained at the same time from adults with type 2 diabetes who were recruited. Retinal vascular parameters were extracted from retinal photographs by a validated fully automated computer program, and TIR was defined as between 3.9-7.8 mmol/L over a 24-h period. The association between TIR and caliber of retinal vessels distributed in different zones were analyzed using multivariable linear regression analyses.

RESULTS

For retinal vascular parameters measurements, the peripheral arteriovenous and middle venular calibers widen with decreasing TIR quartiles (P < 0.05). Lower TIR was associated with wider peripheral venule after adjusting for potential confounders. Even after further correction for GV, there was still a significant correlation between TIR and peripheral vascular calibers (CV: β = - 0.015 [- 0.027, - 0.003], P = 0.013; MAGE: β = - 0.013 [- 0.025, - 0.001], P = 0.038) and SD: β = - 0.013 [- 0.026, - 0.001], P = 0.004). Similar findings were not found for the middle and central venular calibers or arterial calibers located in different zones.

CONCLUSIONS

The TIR was associated with adverse changes to peripheral retinal venules but not central and middle vessels in patients with type 2 diabetes, suggesting that peripheral retinal vascular calibers may be affected by glycemic fluctuations earlier.

Performance of artificial intelligence in diabetic retinopathy screening: a systematic review and meta-analysis of prospective studies

Aims

To systematically evaluate the diagnostic value of an artificial intelligence (AI) algorithm model for various types of diabetic retinopathy (DR) in prospective studies over the previous five years, and to explore the factors affecting its diagnostic effectiveness.

Materials and methods

A search was conducted in Cochrane Library, Embase, Web of Science, PubMed, and IEEE databases to collect prospective studies on AI models for the diagnosis of DR from January 2017 to December 2022. We used QUADAS-2 to evaluate the risk of bias in the included studies. Meta-analysis was performed using MetaDiSc and STATA 14.0 software to calculate the combined sensitivity, specificity, positive likelihood ratio, and negative likelihood ratio of various types of DR. Diagnostic odds ratios, summary receiver operating characteristic (SROC) plots, coupled forest plots, and subgroup analysis were performed according to the DR categories, patient source, region of study, and quality of literature, image, and algorithm.

Results

Finally, 21 studies were included. Meta-analysis showed that the pooled sensitivity, specificity, pooled positive likelihood ratio, pooled negative likelihood ratio, area under the curve, Cochrane Q index, and pooled diagnostic odds ratio of AI model for the diagnosis of DR were 0.880 (0.875-0.884), 0.912 (0.99-0.913), 13.021 (10.738-15.789), 0.083 (0.061-0.112), 0.9798, 0.9388, and 206.80 (124.82-342.63), respectively. The DR categories, patient source, region of study, sample size, quality of literature, image, and algorithm may affect the diagnostic efficiency of AI for DR.

Conclusion

AI model has a clear diagnostic value for DR, but it is influenced by many factors that deserve further study.

Systematic review registration

https://www.crd.york.ac.uk/prospero/, identifier CRD42023389687.

CD44 Drives M1 Macrophage Polarization in Diabetic Retinopathy

PURPOSE

Diabetic retinopathy is a typical complication of diabetes, which can facilitate the risk of blindness in severe cases. We sought to determine the function of CD44 in inflammatory responses of human retinal microvascular endothelial cells (HRMECs) and macrophage polarization during diabetic retinopathy (DR).

METHODS

The hub genes were tested based on two datasets from the Gene Expression Omnibus database. Gene Ontology and pathway enrichment analysis was conducted on the base of differentially expressed genes (DEGs). The infiltration score and infiltration of the immune cells were assessed, and the link between key genes and macrophages was analyzed. The role of CD44 in HRMECs and macrophage polarization was determined by quantitative reverse transcription polymerase chain reaction, western blot, cell counting kit-8, Enzyme-linked immunosorbent assay, flow cytometry, and immunofluorescence.

RESULTS

DEGs were enriched in several pathways linked to DR, such as cellular response to retinoic acid, retinol metabolic process, retina homeostasis, PI3K-AKT signaling pathway, and leukocyte transendothelial migration. A total of 144 DEGs were identified by up-regulation both in GSE102485 and GSE160306. Moreover, the infiltration of macrophages was greater in the DR group than that in the control group. We highlighted an obvious increase in the expression of CD44 and CD86 in patients with DR, and distinct positive associations were found between levels of macrophages and levels of CD44 and CD86. Furthermore, CD44 expression was substantially increased in HRMECs under high glucose (HG) conditions and CD44 knockdown markedly inhibited HG-induced inflammatory responses of HRMECs. HG-induced HRMECs remarkably influenced M1 polarization of macrophages, but CD44 knockdown significantly nullified this effect.

CONCLUSIONS

CD44 influenced the advancement of DR via meditating M1 polarization of macrophages. Our findings could enhance the understanding of the mechanism of DR, which might offer a therapeutic target for DR patients.

Application of Deep Learning Methods in a Moroccan Ophthalmic Center: Analysis and Discussion

Diabetic retinopathy (DR) remains one of the world's frequent eye illnesses, leading to vision loss among working-aged individuals. Hemorrhages and exudates are examples of signs of DR. However, artificial intelligence (AI), particularly deep learning (DL), is poised to impact nearly every aspect of human life and gradually transform medical practice. Insight into the condition of the retina is becoming more accessible thanks to major advancements in diagnostic technology. AI approaches can be used to assess lots of morphological datasets derived from digital images in a rapid and noninvasive manner. Computer-aided diagnosis tools for automatic detection of DR early-stage signs will ease the pressure on clinicians. In this work, we apply two methods to the color fundus images taken on-site at the Cheikh Zaïd Foundation's Ophthalmic Center in Rabat to detect both exudates and hemorrhages. First, we apply the U-Net method to segment exudates and hemorrhages into red and green colors, respectively. Second, the You Look Only Once Version 5 (YOLOv5) method identifies the presence of hemorrhages and exudates in an image and predicts a probability for each bounding box. The segmentation proposed method obtained a specificity of 85%, a sensitivity of 85%, and a Dice score of 85%. The detection software successfully detected 100% of diabetic retinopathy signs, the expert doctor detected 99% of DR signs, and the resident doctor detected 84%.

Clinical validation of a smartphone-based retinal camera for diabetic retinopathy screening

AIMS

This study aims to compare the performance of a handheld fundus camera (Eyer) and standard tabletop fundus cameras (Visucam 500, Visucam 540, and Canon CR-2) for diabetic retinopathy and diabetic macular edema screening.

METHODS

This was a multicenter, cross-sectional study that included images from 327 individuals with diabetes. The participants underwent pharmacological mydriasis and fundus photography in two fields (macula and optic disk centered) with both strategies. All images were acquired by trained healthcare professionals, de-identified, and graded independently by two masked ophthalmologists, with a third senior ophthalmologist adjudicating in discordant cases. The International Classification of Diabetic Retinopathy was used for grading, and demographic data, diabetic retinopathy classification, artifacts, and image quality were compared between devices. The tabletop senior ophthalmologist adjudication label was used as the ground truth for comparative analysis. A univariate and stepwise multivariate logistic regression was performed to determine the relationship of each independent factor in referable diabetic retinopathy.

RESULTS

The mean age of participants was 57.03 years (SD 16.82, 9-90 years), and the mean duration of diabetes was 16.35 years (SD 9.69, 1-60 years). Age (P = .005), diabetes duration (P = .004), body mass index (P = .005), and hypertension (P < .001) were statistically different between referable and non-referable patients. Multivariate logistic regression analysis revealed a positive association between male sex (OR 1.687) and hypertension (OR 3.603) with referable diabetic retinopathy. The agreement between devices for diabetic retinopathy classification was 73.18%, with a weighted kappa of 0.808 (almost perfect). The agreement for macular edema was 88.48%, with a kappa of 0.809 (almost perfect). For referable diabetic retinopathy, the agreement was 85.88%, with a kappa of 0.716 (substantial), sensitivity of 0.906, and specificity of 0.808. As for image quality, 84.02% of tabletop fundus camera images were gradable and 85.31% of the Eyer images were gradable.

CONCLUSIONS

Our study shows that the handheld retinal camera Eyer performed comparably to standard tabletop fundus cameras for diabetic retinopathy and macular edema screening. The high agreement with tabletop devices, portability, and low costs makes the handheld retinal camera a promising tool for increasing coverage of diabetic retinopathy screening programs, particularly in low-income countries. Early diagnosis and treatment have the potential to prevent avoidable blindness, and the present validation study brings evidence that supports its contribution to diabetic retinopathy early diagnosis and treatment.

Association between high-density lipoprotein cholesterol to apolipoprotein A ratio and diabetic retinopathy: A cross-sectional study

AIMS

Our study is aimed to investigate the relationship between high-density lipoprotein cholesterol to apolipoprotein A ratio (HDL-C/ApoA) and diabetic retinopathy (DR) in subjects with type 2 diabetes mellitus (T2DM).

METHODS

We retrospect the consecutive medical files of 1058 subjects with T2DM and recorded their clinical information and laboratory findings. Subjects with T2DM were divided into DR group (n = 522) and non-DR group (n = 536). We compared the lipids values of the two groups. Meanwhile we also observed the prevalence of DR at different HDL-C/ApoA levels. Binary logistic regression was used to correct confounding factors. Smooth curve fitting model and subgroup analysis were used to determine the correlation, non-linear relationship and threshold effect between HDL/ApoA and DR.

RESULTS

HDL-C/ApoA value of DR group was significantly higher than non-DR group (0.88 ± 0.17 vs 0.84 ± 0.13, P < 0.05). The prevalence of DR significantly increased as HDL-C/ApoA level increased. There was association between HDL/ApoA levels and DR in the adjusted models (OR 1.55, 95%CI 0.60 to 4.02). After full adjustments for other relevant clinical covariates, patients with HDL/ApoA values in quartile 3 (Q3) had 1.50 times (95 % CI 1.00 to 2.17) and in Q4 had 2.39 times (95%CI 1.65 to 3.47) as high as the risk of DR compared with patients in Q1. HDL/ApoA showed a non-linear relationship with DR, with an inflection point value of 0.759. When HDL/ApoA>0.759, HDL/ApoA was significantly positively associated with DR (HR = 26.508, 95 % CI 7.623-92.174; P < 0.0001). Compared to patients with age < 60, HDL/ApoA was obviously associated with DR when age ≥ 60 (OR = 38.05, 95 % CI 8.06-179.69; P < 0.001).

CONCLUSIONS

HDL-C/ApoA was found to be associated with the incidence of DR in patients with T2DM. After adjusting potential related factors HDL-C/ApoA OR value was 1.55 (95%CI 0.60 to 4.02). A non-linear association between HDL/ApoA and DR was observed in T2DM. Subgroup analysis showed that age could alter the relationship between HDL/ApoA and DR.

Research progress on diagnosing retinal vascular diseases based on artificial intelligence and fundus images

As the only blood vessels that can directly be seen in the whole body, pathological changes in retinal vessels are related to the metabolic state of the whole body and many systems, which seriously affect the vision and quality of life of patients. Timely diagnosis and treatment are key to improving vision prognosis. In recent years, with the rapid development of artificial intelligence, the application of artificial intelligence in ophthalmology has become increasingly extensive and in-depth, especially in the field of retinal vascular diseases. Research study results based on artificial intelligence and fundus images are remarkable and provides a great possibility for early diagnosis and treatment. This paper reviews the recent research progress on artificial intelligence in retinal vascular diseases (including diabetic retinopathy, hypertensive retinopathy, retinal vein occlusion, retinopathy of prematurity, and age-related macular degeneration). The limitations and challenges of the research process are also discussed.