Comparison of two reference standards in validating two field mydriatic digital photography as a method of screening for diabetic retinopathy

Diagnostic Accuracy of Hand-Held Fundus Camera and Artificial Intelligence in Diabetic Retinopathy Screening

Our study aimed to assess the role of a hand-held fundus camera and artificial intelligence (AI)-based grading system in diabetic retinopathy (DR) screening and determine its diagnostic accuracy in detecting DR compared with clinical examination and a standard fundus camera. This cross-sectional instrument validation study, as a part of the International Diabetes Federation (IDF) Diabetic Retinopathy Screening Project, included 160 patients (320 eyes) with type 2 diabetes (T2DM). After the standard indirect slit-lamp fundoscopy, each patient first underwent fundus photography with a standard 45° camera VISUCAM Zeiss and then with a hand-held camera TANG (Shanghai Zhi Tang Health Technology Co., Ltd.). Two retina specialists independently graded the images taken with the standard camera, while the images taken with the hand-held camera were graded using the DeepDR system and an independent IDF ophthalmologist. The three screening methods did not differ in detecting moderate/severe nonproliferative and proliferative DR. The area under the curve, sensitivity, specificity, positive predictive value, negative predictive value, positive likelihood ratio, negative likelihood ratio, kappa (ĸ) agreement, diagnostic odds ratio, and diagnostic effectiveness for a hand-held camera compared to clinical examination were 0.921, 89.1%, 100%, 100%, 91.4%, infinity, 0.11, 0.86, 936.48, and 94.9%, while compared to the standard fundus camera were 0.883, 83.2%, 100%, 100%, 87.3%, infinity, 0.17, 0.78, 574.6, and 92.2%. The results of our study suggest that fundus photography with a hand-held camera and AI-based grading system is a short, simple, and accurate method for the screening and early detection of DR, comparable to clinical examination and fundus photography with a standard camera.

Teleophthalmology and retina: a review of current tools, pathways and services

Telemedicine, the use of telecommunication and information technology to deliver healthcare remotely, has evolved beyond recognition since its inception in the 1970s. Advances in telecommunication infrastructure, the advent of the Internet, exponential growth in computing power and associated computer-aided diagnosis, and medical imaging developments have created an environment where telemedicine is more accessible and capable than ever before, particularly in the field of ophthalmology. Ever-increasing global demand for ophthalmic services due to population growth and ageing together with insufficient supply of ophthalmologists requires new models of healthcare provision integrating telemedicine to meet present day challenges, with the recent COVID-19 pandemic providing the catalyst for the widespread adoption and acceptance of teleophthalmology. In this review we discuss the history, present and future application of telemedicine within the field of ophthalmology, and specifically retinal disease. We consider the strengths and limitations of teleophthalmology, its role in screening, community and hospital management of retinal disease, patient and clinician attitudes, and barriers to its adoption.

Prevalence of Retinopathy in Prediabetic Populations: A Systematic Review and Meta-Analysis

Among the leading causes of vision impairment and blindness globally, diabetic retinopathy (DR) is one of the most important causes. There is increasing evidence of DR prevalence in the prediabetic population. This systematic review presents collective data on retinopathy in the prediabetic population. This review article aimed to estimate the reported prevalence of retinopathy in prediabetes, impaired glucose tolerance test (GTT) without diabetes mellitus, and the risk factors involved and to summarize it. Literature searches were done using the Web of Science, CINAHL, Google Scholar, Cochrane, EMBASE, and PubMed databases from inception to April 2023. Our search included the words prediabetes, DR, and risk factors. All searches were looked at for methodological quality and evidence. Thirty-one studies were included after the screening. Population-based data were used in 23 studies (82.1%). The prediabetic population screened was 10,539. The prevalence of retinopathy ranged between 0.3% and 20.9%, showing a median of 8.1% with an interquartile range (IQR) of 4.2-11%, showing great variance in estimates due to the use of different screening methods, methods used for retinopathy grading, and study populations. Several studies compared the population with normal GTT with impaired glucose tolerance (IGT) and inferred that there was a lower prevalence of retinopathy in the normal GTT population (3.0%, IQR 0.3-7.4%) than prediabetes (6.7%, IQR 1.9-10.1%). According to this data, a greater retinopathy prevalence was found in prediabetic populations.

EVALUATION OF THE INCLUSION OF SPECTRAL DOMAIN OPTICAL COHERENCE TOMOGRAPHY IN A TELEMEDICINE DIABETIC RETINOPATHY SCREENING PROGRAM: A Real Clinical Practice

PURPOSE

To evaluate whether combining spectral domain optical coherence tomography with monoscopic fundus photography using a nonmydriatic camera (MFP-NMC) improves the accuracy of diabetic macular edema (DME) referrals in a teleophthalmology diabetic retinopathy screening program.

METHODS

We conducted a cross-sectional study with all diabetic patients aged 18 years or older who attended screening from September 2016 to December 2017. We assessed DME according to the three MFP-NMC and the four spectral domain optical coherence tomography criteria. The sensitivity and specificity obtained for each criterion were estimated by comparing them with the ground truth of DME.

RESULTS

This study included 3,918 eyes (1,925 patients; median age, 66 years; interquartile range, 58-73; females, 40.7%; once-screened, 68.1%). The prevalence of DME ranged from 1.22% to 1.83% and 1.54% to 8.77% on MFP-NMC and spectral domain optical coherence tomography, respectively. Sensitivity barely reached 50% in MFP-NMC and less for the quantitative criteria of spectral domain optical coherence tomography. When macular thickening and anatomical signs of DME were considered, sensitivity increased to 88.3% and the false DMEs and non-gradable images were reduced.

CONCLUSION

Macular thickening and anatomical signs showed the highest suitability for screening, with a sensitivity of 88.3% and a specificity of 99.8%. Notably, MFP-NMC alone missed half of the true DMEs that lacked indirect signs.

Performance of a Support Vector Machine Learning Tool for Diagnosing Diabetic Retinopathy in Clinical Practice

PURPOSE

To examine the real-world performance of a support vector machine learning software (RetinaLyze) in order to identify the possible presence of diabetic retinopathy (DR) in patients with diabetes via software implementation in clinical practice.

METHODS

1001 eyes from 1001 patients-one eye per patient-participating in the Danish National Screening Programme were included. Three independent ophthalmologists graded all eyes according to the International Clinical Diabetic Retinopathy Disease Severity Scale with the exact level of disease being determined by majority decision. The software detected DR and no DR and was compared to the ophthalmologists' gradings.

RESULTS

At a clinical chosen threshold, the software showed a sensitivity, specificity, positive predictive value and negative predictive value of 84.9% (95% CI: 81.8-87.9), 89.9% (95% CI: 86.8-92.7), 92.1% (95% CI: 89.7-94.4), and 81.0% (95% CI: 77.2-84.7), respectively, when compared to human grading. The results from the routine screening were 87.0% (95% CI: 84.2-89.7), 85.3% (95% CI: 81.8-88.6), 89.2% (95% CI: 86.3-91.7), and 82.5% (95% CI: 78.5-86.0), respectively. AUC was 93.4%. The reference graders Conger's Exact Kappa was 0.827.

CONCLUSION

The software performed similarly to routine grading with overlapping confidence intervals, indicating comparable performance between the two groups. The intergrader agreement was satisfactory. However, evaluating the updated software alongside updated clinical procedures is crucial. It is therefore recommended that further clinical testing before implementation of the software as a decision support tool is conducted.

Artificial Intelligence Detection of Diabetic Retinopathy: Subgroup Comparison of the EyeArt System with Ophthalmologists' Dilated Exams

Objective

To compare general ophthalmologists, retina specialists, and the EyeArt Artificial Intelligence (AI) system to the clinical reference standard for detecting more than mild diabetic retinopathy (mtmDR).

Design

Prospective, pivotal, multicenter trial conducted from April 2017 to May 2018.

Participants

Participants were aged ≥ 18 years who had diabetes mellitus and underwent dilated ophthalmoscopy. A total of 521 of 893 participants met these criteria and completed the study protocol.

Testing

Participants underwent 2-field fundus photography (macula centered, disc centered) for the EyeArt system, dilated ophthalmoscopy, and 4-widefield stereoscopic dilated fundus photography for reference standard grading.

Main Outcome Measures

For mtmDR detection, sensitivity and specificity of EyeArt gradings of 2-field, fundus photographs and ophthalmoscopy grading versus a rigorous clinical reference standard comprising Reading Center grading of 4-widefield stereoscopic dilated fundus photographs using the ETDRS severity scale. The AI system provided automatic eye-level results regarding mtmDR.

Results

Overall, 521 participants (999 eyes) at 10 centers underwent dilated ophthalmoscopy: 406 by nonretina and 115 by retina specialists. Reading Center graded 207 positive and 792 eyes negative for mtmDR. Of these 999 eyes, 26 eyes were ungradable by the EyeArt system, leaving 973 eyes with both EyeArt and Reading Center gradings. Retina specialists correctly identified 22 of 37 eyes as positive (sensitivity 59.5%) and 182 of 184 eyes as negative (specificity 98.9%) for mtmDR versus the EyeArt AI system that identified 36 of 37 as positive (sensitivity 97%) and 162 of 184 eyes as negative (specificity of 88%) for mtmDR. General ophthalmologists correctly identified 35 of 170 eyes as positive (sensitivity 20.6%) and 607 of 608 eyes as negative (specificity 99.8%) for mtmDR compared with the EyeArt AI system that identified 164 of 170 as positive (sensitivity 96.5%) and 525 of 608 eyes as negative (specificity 86%) for mtmDR.

Conclusions

The AI system had a higher sensitivity for detecting mtmDR than either general ophthalmologists or retina specialists compared with the clinical reference standard. It can potentially serve as a low-cost point-of-care diabetic retinopathy detection tool and help address the diabetic eye screening burden.

Incidence and progression of diabetic retinopathy in a multi-ethnic US cohort: the Multi-Ethnic Study of Atherosclerosis

AIM

To provide contemporary longitudinal data on the incidence and progression of diabetic retinopathy (DR) in a multi-ethnic population of whites, African Americans, Chinese and Hispanics in the United States.

METHODS

A prospective, multi-region, multi-ethnic population-based cohort study that included 498 participants with diabetes, aged 45-84 years at baseline, from the Multi-Ethnic Study of Atherosclerosis with retinal images obtained twice, on average 8 years apart. Presence and severity of DR were graded from these retinal images according to the modified Airlie House classification system. Main outcome measures were 8-year incidence, progression and improvement of DR, and their associated risk factors.

RESULTS

Over the 8 years, the cumulative rates were 19.2% for incident DR, 17.3% for DR progression, 23.3% for DR improvement, 2.7% for incident vision-threatening DR, 1.8% for incident proliferative DR and 2.2% for incident macular oedema. In multivariate analysis, significant risk factors associated with incident DR were higher glycosylated haemoglobin (relative risk (RR) 1.28; 95% CI: 1.16 to 1.41) and higher systolic blood pressure (RR 1.14; 95% CI: 1.04 to 1.25). Significant factors associated with DR progression were higher glycosylated haemoglobin (RR 1.20; 95% CI: 1.00 to 1.43) and higher low-density lipoprotein cholesterol (RR 1.01; 95% CI: 1.00 to 1.03).

CONCLUSION

Over an 8-year period, approximately one in five participants with diabetes developed DR, while almost a quarter of those with DR at baseline showed improvement, possibly reflecting the positive impact of clinical and public health efforts in improving diabetes care in the United States over the last two decades.

Assessment of image quality on color fundus retinal images using the automatic retinal image analysis

Image quality assessment is essential for retinopathy detection on color fundus retinal image. However, most studies focused on the classification of good and poor quality without considering the different types of poor quality. This study developed an automatic retinal image analysis (ARIA) method, incorporating transfer net ResNet50 deep network with the automatic features generation approach to automatically assess image quality, and distinguish eye-abnormality-associated-poor-quality from artefact-associated-poor-quality on color fundus retinal images. A total of 2434 retinal images, including 1439 good quality and 995 poor quality (483 eye-abnormality-associated-poor-quality and 512 artefact-associated-poor-quality), were used for training, testing, and 10-ford cross-validation. We also analyzed the external validation with the clinical diagnosis of eye abnormality as the reference standard to evaluate the performance of the method. The sensitivity, specificity, and accuracy for testing good quality against poor quality were 98.0%, 99.1%, and 98.6%, and for differentiating between eye-abnormality-associated-poor-quality and artefact-associated-poor-quality were 92.2%, 93.8%, and 93.0%, respectively. In external validation, our method achieved an area under the ROC curve of 0.997 for the overall quality classification and 0.915 for the classification of two types of poor quality. The proposed approach, ARIA, showed good performance in testing, 10-fold cross validation and external validation. This study provides a novel angle for image quality screening based on the different poor quality types and corresponding dealing methods. It suggested that the ARIA can be used as a screening tool in the preliminary stage of retinopathy grading by telemedicine or artificial intelligence analysis.

Diagnostic accuracy of teleretinal screening for detection of diabetic retinopathy and age-related macular degeneration: a systematic review and meta-analysis

Objective

To evaluate the diagnostic accuracy of teleretinal screening compared with face-to-face examination for detection of diabetic retinopathy (DR) and age-related macular degeneration (AMD).

Methods and analysis

This study adhered to the Preferred Reporting Items for a Systematic Review and Meta-analysis of Diagnostic Test Accuracy Studies (PRISMA-DTA). A comprehensive search of OVID MEDLINE, EMBASE and Cochrane CENTRAL was performed from January 2010 to July 2021. QUADAS-2 tool was used to assess methodological quality and applicability of the studies. A bivariate random effects model was used to perform the meta-analysis. Referrable DR was defined as any disease severity equal to or worse than moderate non-proliferative DR or diabetic macular oedema (DMO).

Results

28 articles were included. Teleretinal screening achieved a sensitivity of 0.91 (95% CI: 0.82 to 0.96) and specificity of 0.88 (0.74 to 0.95) for any DR (13 studies, n=7207, Grading of Recommendations, Assessment, Development and Evaluation (GRADE) low). Accuracy for referrable DR (10 studies, n=6373, GRADE moderate) was lower with a sensitivity of 0.88 (0.81 to 0.93) and specificity of 0.86 (0.79 to 0.90). After exclusion of ungradable images, the specificity for referrable DR increased to 0.95 (0.90 to 0.98), while the sensitivity remained nearly unchanged at 0.85 (0.76 to 0.91). Teleretinal screening achieved a sensitivity of 0.71 (0.49 to 0.86) and specificity of 0.88 (0.85 to 0.90) for detection of AMD (three studies, n=697, GRADE low).

Conclusion

Teleretinal screening is highly accurate for detecting any DR and DR warranting referral. Data for AMD screening is promising but warrants further investigation.

PROSPERO registration number

CRD42020191994.

Imaging diabetic retinal disease: clinical imaging requirements

Diabetic retinopathy (DR) is a sight-threatening complication of diabetes mellitus (DM) and it contributes substantially to the burden of disease globally. During the last decades, the development of multiple imaging modalities to evaluate DR, combined with emerging treatment possibilities, has led to the implementation of large-scale screening programmes resulting in improved prevention of vision loss. However, not all patients are able to participate in such programmes and not all are at equal risk of DR development and progression. In this review, we discuss the relevance of the currently available imaging modalities for the evaluation of DR: colour fundus photography (CFP), ultrawide-field photography (UWFP), fundus fluorescein angiography (FFA), optical coherence tomography (OCT), OCT angiography (OCTA) and functional testing. Furthermore, we suggest where a particular imaging technique of DR may aid the evaluation of the disease in different clinical settings. Combining information from various imaging modalities may enable the design of more personalized care including the initiation of treatment and understanding the progression of disease more adequately.

Outreach screening to address demographic and economic barriers to diabetic retinopathy care in rural China

IMPORTANCE

Poor access to existing care for diabetic retinopathy (DR) limits effectiveness of proven treatments.

OBJECTIVES

We examined whether outreach screening in rural China improves equity of access.

DESIGN, SETTING AND PARTICIPANTS

We compared prevalence of female sex, age > = 65 years, primary education or below, and requiring referral care for DR between three cohorts with diabetes examined for DR in neighboring areas of Guangdong, China: passive case detection at secondary-level hospitals (n = 193); persons screened during primary-level DR outreach (n = 182); and individuals with newly- or previously-diagnosed diabetes in a population survey (n = 579). The latter reflected the "ideal" reach of a screening program.

RESULTS

Compared to the population cohort, passive case detection reached fewer women (50·8% vs. 62·3%, p = 0·006), older adults (37·8% vs. 51·3%, p < 0·001), and less-educated persons (39·9% vs. 89·6%, p < 0·001). Outreach screening, compared to passive case detection, improved representation of the elderly (49·5% vs. 37·8%, p = 0·03) and less-educated (70·3% vs. 39·9%, p<0·001). The proportion of women (59.8% vs 62.3%, P>0.300) and persons aged > = 65 years (49.5% vs 51.3%, p = 0.723) in the outreach screening and population cohorts did not differ significantly. Prevalence of requiring referral care for DR was significantly higher in the outreach screening cohort (28·0%) than the population (14·0%) and passive case detection cohorts (7·3%, p<0·001 for both).

CONCLUSIONS AND RELEVANCE

Primary-level outreach screening improves access for the poorly-educated and elderly, and removes gender inequity in access to DR care in this setting, while also identifying more severely-affected patients than case finding in hospital.

Effectiveness of quality of care for patients with type 2 diabetes in China: findings from the Shanghai Integration Model (SIM)

This cross-sectional study aimed to investigate the quality of care of diabetes in Shanghai, China. A total of 173 235 patients with type 2 diabetes in 2017 were included in the analysis. Profiles of risk factors and intermediate outcomes were determined. The patients had a mean age of 66.43 ± 8.12 (standard deviation (SD)) years and a mean diabetes duration of 7.95 ± 5.53 (SD) years. The percentage of patients who achieved the target level for HbA_1c (< 7.0%) was 48.6%. Patients who achieved the target levels for blood pressure (BP) < 130/80 mmHg and low-density lipoprotein-cholesterol (LDL-c) < 2.6 mmol/L reached 17.5% and 34.0%, respectively. A total of 3.8% achieved all three target levels, and the value increased to 6.8% with an adaptation of the BP target level (< 140/90 mmHg) for those over 65 years. Multivariable analysis identified the factors associated with a great likelihood of achieving all three target levels: male, young age, short diabetes duration, low body mass index, macrovascular complications, no microvascular complications, prescribed with lipid-lowering medication, and no prescription of antihypertensive medication. In conclusion, nearly 50% and one-third of the patients with diabetes met the target levels for HbA_1c and LDL-c, respectively, with a low percentage achieving the BP target level. The percentage of patients who achieved all three target levels needs significant improvement.

Two-field non-mydriatic fundus photography for diabetic retinopathy screening: a protocol for a systematic review and meta-analysis

INTRODUCTION

Diabetic retinopathy (DR) is one of the most prevalent microvascular complications of diabetes mellitus. Guidelines for DR screening in different countries vary greatly, including fundus photography, slit-lamp biomicroscopy, indirect ophthalmoscopy, Optical Coherence Tomography (OCT), OCT-A and Fundus Fluorescein Angiography (FFA). Two-field non-mydriatic fundus photography (NMFP) is an effective screening method due to its low cost and less time-consuming process. However, it is controversial due to the sensitivity and specificity of two-field NMFP. This review intends to evaluate the performance of the two-field NMFP in diagnosing DR and helps clinicians determine the most optimal screening method.

METHODS AND ANALYSIS

Two reviewers will independently search on the Medline, Embase, Cochrane databases, ProQuest, Opengrey, Chinese National Knowledge Infrastructure, Wanfang Data, VIP China Science and Technology Journal Database, Chinese BioMedical Literature Database, ISRCTN, ClinicalTrials.gov and the WHO ICTRP to identify relevant studies. There is no restriction posed on the language of the study. Included studies focus on the performance of two-field NMFP in detecting DR in diabetes patients. Analysis and evaluation of the studies will be examined by two reviewers independently using the Quality Assessment for Diagnostic Accuracy Studies-2 tool and later evaluated using the Population, Intervention, Comparison, Outcome, Study design criteria. A random-effect model will calculate the diagnostic indicators, including the sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic OR, area under the curve and 95% CIs. We will also develop a summary receiver operating characteristic curve. We anticipate analysing subgroups according to the factors, which may lead to heterogeneity, including DR levels of patients, the reference standards, camera models, the interpretation criteria. The data will be analysed by STATA software. This study was registered with PROSPERO.

ETHICS AND DISSEMINATION

This review will analyse the published data. Patients/the public were not involved in this research. The results of this study will be published in peer-reviewed journals.

PROSPERO REGISTRATION NUMBER

CRD42020203608.

Imaging Modalities Employed in Diabetic Retinopathy Screening: A Review and Meta-Analysis

INTRODUCTION

Urbanization has caused dramatic changes in lifestyle, and these rapid transitions have led to an increased risk of noncommunicable diseases, such as type 2 diabetes. In terms of cost-effectiveness, screening for diabetic retinopathy is a critical aspect in diabetes management. The aim of this study was to review the imaging modalities employed for retinal examination in diabetic retinopathy screening.

METHODS

The PubMed and Web of Science databases were the main sources used to investigate the medical literature. An extensive search was performed to identify relevant articles concerning "imaging", "diabetic retinopathy" and "screening" up to 1 June 2021. Imaging techniques were divided into the following: (i) mydriatic fundus photography, (ii) non-mydriatic fundus photography, (iii) smartphone-based imaging, and (iv) ultrawide-field imaging. A meta-analysis was performed to analyze the performance and technical failure rate of each method.

RESULTS

The technical failure rates for mydriatic and non-mydriatic digital fundus photography, smartphone-based and ultrawide-field imaging were 3.4% (95% CI: 2.3-4.6%), 12.1% (95% CI: 5.4-18.7%), 5.3% (95% CI: 1.5-9.0%) and 2.2% (95% CI: 0.3-4.0%), respectively. The rate was significantly different between all analyzed techniques (p < 0.001), and the overall failure rate was 6.6% (4.9-8.3%; I² = 97.2%). The publication bias factor for smartphone-based imaging was significantly higher than for mydriatic digital fundus photography and non-mydriatic digital fundus photography (b = -8.61, b = -2.59 and b = -7.03, respectively; p < 0.001). Ultrawide-field imaging studies were excluded from the final sensitivity/specificity analysis, as the total number of patients included was too small.

CONCLUSIONS

Regardless of the type of the device used, retinal photographs should be taken on eyes with dilated pupils, unless contraindicated, as this setting decreases the rate of ungradable images. Smartphone-based and ultrawide-field imaging may become potential alternative methods for optimized DR screening; however, there is not yet enough evidence for these techniques to displace mydriatic fundus photography.

The validity of diabetic retinopathy screening using nonmydriatic fundus camera and optical coherence tomography in comparison to clinical examination

PURPOSE:

To evaluate the use of non-mydriatic fundus camera as a screening modality for diabetic retinopathy in a sample of population in Riyadh, Saudi Arabia.

METHODS:

Patients coming, from April 2015 till September 2018, for their diabetic check up at the diabetic center clinics in King Abdul-Aziz University hospital were screened using a non-mydriatic fundus camera (NMFC). Photos were graded by retina specialist and compared to the findings of dilated fundus examination (DFE) by retina specialists.

RESULTS:

The grading results of NMFC and DFE were compared and the overall sensitivity and specificity for detection of diabetic retinopathy within one grade of retinopathy was 98.7% and 80% respectively. The sensitivity for detection of sight threatening conditions such as proliferative diabetic retinopathy, severe non-proliferative diabetic retinopathy, and diabetic macular edema (by Ocular Coherence Tomography) was found to be 86.7%, 90.3% and 100% respectively; while the specificity was found to be 96.5%, 93%, and 100% respectively.

CONCLUSION:

Non-mydriatic fundus camera has a high sensitivity and specificity in screening for diabetic retinopathy. It is a great screening tool, which is user friendly and can be operated by trained nurses in primary clinics during patient's regular routine diabetic checkups. It aids in early detection of sight threatening conditions which need urgent referral to ophthalmologists.

[Diabetic retinopathy]

Diabetic retinopathy (DR) is a vision-threatening microvascular complication of diabetes and the leading cause of blindness in working-age people. At the beginning of the metabolic disorder and in early stages of DR the patient's eyesight is often not affected. Depending on the duration of diabetes and in more advanced stages of DR the vision is compromised through the presence of diabetic macular edema (DME) and/or proliferative retinal complications. The management of DR comprises regular ophthalmic examinations according to clinical guidelines, the targeted application of multimodal imaging, and the specific treatment of DME and proliferative DR including secondary disorders such as neovascular glaucoma or persistent vitreous haemorrhage. Innovative ocular imaging techniques like optical coherence tomography (OCT), OCT angiography (OCT-A) and ultrawide field imaging play an important role in the assessment of diabetic patients. Various non-invasive imaging modalities have become part of the routine clinical work-up and help to identify new biomarkers for early diagnosis and long-term prognosis. In early stages of DR, the multifactorial intervention including glucose level and blood pressure control as well as optimizing the patient's cardiovascular risk profile is essential. A specific ophthalmic therapy is available for DME and proliferative DR (PDR). In patients with PDR the treatment regime includes panretinal laser photocoagulation or alternatively intravitreal anti-VEGF (vascular endothelial growth factor)-injections accompanied by close-meshed clinical monitoring. In patients with both, DME and PDR, it is suggested to start with Anti-VEGF drugs. In severe PDR with persistent vitreous haemorrhage, tractional maculopathy or tractional retinal detachment vitreoretinal surgery is recommended.

Diabetische Retinopathie

Die diabetische Retinopathie (DR) ist die häufigste Ursache für schwere Sehbehinderung und Erblindung im erwerbstätigen Alter. Eine subjektive Beeinträchtigung des Sehvermögens tritt häufig erst in fortgeschrittenen Stadien der DR ein. Daher sind Screening-Maßnahmen asymptomatischer Patienten und eine stadiengerechte Behandlung essenziell. Dieser Beitrag gibt einen praxisbezogenen Überblick über diagnostische und therapeutische Prinzipien der nicht-proliferativen und proliferativen Form.

Evidence-based Canadian guidelines for tele-retina screening for diabetic retinopathy: recommendations from the Canadian Retina Research Network (CR2N) Tele-Retina Steering Committee

OBJECTIVE

The purpose of this report is to develop a consensus for Canadian national guidelines specific to a tele-medicine approach to screening for diabetic retinopathy (DR) using evidence-based and clinical data.

METHODS

Canadian Tele-Screening Grading Scales for DR and diabetic macular edema (DME) were created primarily based on severity grading scales outlined by the International Clinical Diabetic Retinopathy Disease Severity Scale (ICDR) and the Scottish DR Grading Scheme 2007. Other grading scales used in international screening programs and the clinical expertise of the Canadian Retina Research Network members and retina specialists nationwide were also used in the creation of the guidelines.

RESULTS

National Tele-Screening Guidelines for DR and DME with and without optical coherence tomography (OCT) images are proposed. These outline a diagnosis and management algorithm for patients presenting with different stages of DR and/or DME. General guidelines detailing the requirements for imaged retina fields, image quality, quality control, and follow-up care and the role of visual acuity, pupil dilation, OCT, ultra-wide-field imaging, and artificial intelligence are discussed.

CONCLUSIONS

Tele-retina screening can help to address the need for timely and effective screening for DR, whose prevalence continues to rise. A standardized and evidence-based national approach to DR tele-screening has been proposed, based on DR/DME grading using two 45° image fields or a single widefield or ultra-wide-field image, preferable use of OCT imaging, and a focus on local quality control measures.

A deep learning system for detecting diabetic retinopathy across the disease spectrum

Retinal screening contributes to early detection of diabetic retinopathy and timely treatment. To facilitate the screening process, we develop a deep learning system, named DeepDR, that can detect early-to-late stages of diabetic retinopathy. DeepDR is trained for real-time image quality assessment, lesion detection and grading using 466,247 fundus images from 121,342 patients with diabetes. Evaluation is performed on a local dataset with 200,136 fundus images from 52,004 patients and three external datasets with a total of 209,322 images. The area under the receiver operating characteristic curves for detecting microaneurysms, cotton-wool spots, hard exudates and hemorrhages are 0.901, 0.941, 0.954 and 0.967, respectively. The grading of diabetic retinopathy as mild, moderate, severe and proliferative achieves area under the curves of 0.943, 0.955, 0.960 and 0.972, respectively. In external validations, the area under the curves for grading range from 0.916 to 0.970, which further supports the system is efficient for diabetic retinopathy grading.

An automated microaneurysm detector as a tool for identification of diabetic retinopathy in rural optometric practice

BACKGROUND

With the increase in the prevalence of diabetes, rural optometric clinics stand to increase their patient load and assessment of diabetic eye disease. This study aimed to assess whether automated identification of diabetic retinopathy based on the presence of microaneurysms is an effective tool in clinical practice.

METHODS

We analysed 758 fundal images of 385 patients with diabetes attending the clinic obtained using a Canon CR5 with an EOS10 digital camera through a dilated pupil. Five optometrists employed in the clinic assessed the diabetic retinopathy using binocular indirect ophthalmoscopy. The sensitivity and specificity of the automated system used to analyse the retinal fundal images was determined by comparison with optometric and ophthalmologic assessment.

RESULTS

The optometrists achieved 97 per cent sensitivity at 88 per cent specificity with respect to the ophthalmic classification for detecting retinopathy.

CONCLUSION

The automated retinopathy detector achieved 85 per cent sensitivity at 90 per cent specificity at detecting retinopathy. The automated microaneurysm detector has a lower sensitivity compared to the optometrists but meets NHMRC guidelines. It may impact on the efficiency of rural optometric practices by early identification of diabetic retinopathy. Automated assessment can save time and be cost-effective, and provide a history of changes in the retinal fundus and the opportunity for instant patient education using the digital images.

Six-year incidence and systemic associations of retinopathy in a multi-ethnic Asian population without diabetes

PURPOSE

We described the 6-year incidence and changes of retinopathy, and their associated risk factors in a multi-ethnic Asian population without diabetes.

METHODS

We included 4374 participants with non-diabetes from a population-based cohort, the Singapore Epidemiology of Eye Disease Study, with gradable retinal photographs at baseline and 6-year follow-up visit. Retinopathy was assessed according to the modified Airlie House classification system.

RESULTS

Over the 6-year period, the cumulative rates were 2.5% (106/4279) for retinopathy incidence, 1.0% (1/95) for retinopathy progression and 68.4% (65/95) for retinopathy regression. In multivariable analysis, higher diastolic blood pressure (DBP) (risk ratio (RR)=1.02; 95% CI: 1.00 to 1.04; per 10 mm Hg increase in DBP) and wider retinal arteriolar calibre (RR=1.36; 95% CI: 1.13 to 1.63; per SD increase in central retinal artery equivalent) were associated with higher risk of incident retinopathy, while higher level of high-density lipoprotein (HDL) was associated with lower risk of incident retinopathy (RR=0.56; 95% CI: 0.32 to 0.99; per mmol/L increase in HDL). Compared with Chinese, Malays were more likely to have retinopathy regression (RR=1.63; 95% CI: 1.20 to 2.22), while overweight (RR=0.47; 95% CI: 0.26 to 0.84) and higher glycosylated haemoglobin (HbA1c) level (RR=0.58; 95% CI: 0.37 to 0.93; per per cent increase in HbA1c) were associated with lower likelihood of retinopathy regression.

CONCLUSION

Risk of developing retinopathy in Asians without diabetes is generally low. However, regression of retinopathy over time is common, suggesting that these retinopathy signs may reflect subclinical reversible microvascular dysfunction. Several metabolic risk factors are associated with incidence or regression of retinopathy, suggesting that good metabolic control may still be important in the management of non-diabetic retinopathy.

Irish National Diabetic RetinaScreen Programme: report on five rounds of retinopathy screening and screen-positive referrals. (INDEAR study report no. 1)

Objective

To study the uptake of annual diabetic retinopathy screening and study the 5-year trends in the detection of screen-positive diabetic retinopathy and non-diabetes-related eye disease in a cohort of annually screened individuals.

Design

Retrospective retinopathy screening attendance and retinopathy grading analysis.

Setting

Community-based retinopathy screening centres for the Diabetic RetinaScreen Programme.

Participants

171 557 were identified by the screening programme to be eligible for annual diabetic retinopathy screening. 120 048 individuals over the age of 12 consented to and attended at least one screening appointment between February 2013 to December 2018.

Main Outcome Measures

Detection rate per 100 000 of any retinopathy, screen-positive referrable retinopathy and nondiabetic eye disease.

Results

Uptake of screening had reached 67.2% in the fifth round of screening. Detection rate of screen-positive retinopathy reduced from 13 229 to 4237 per 100 000 screened over five rounds. Detection of proliferative disease had reduced from 2898 to 713 per 100 000 screened. Non-diabetic eye disease detection and referral to treatment centres increased almost eightfold from 393 in round 1 to 3225 per 100 000 screened. The majority of individuals referred to treatment centres for ophthalmologist assessment are over the age of 50 years.

Conclusions

Screening programme has seen a reduced detection rate both screen-positive retinopathy referral in Ireland over five rounds of screening. Management of nondiabetic eye diseases poses a significant challenge in improving visual outcomes of people living with diabetes in Ireland.

Accuracy of the smartphone-based nonmydriatic retinal camera in the detection of sight-threatening diabetic retinopathy

Purpose:

To evaluate the sensitivity and specificity of smartphone-based nonmydriatic (NM) retinal camera in the detection of diabetic retinopathy (DR) and sight-threatening DR (STDR) in a tertiary eye care facility.

Methods:

Patients with diabetes underwent retinal photography with a smartphone-based NM fundus camera before mydriasis and standard 7-field fundus photography with a desktop mydriatic fundus camera after mydriasis. DR was graded using the international clinical classification of diabetic retinopathy system by two retinal expert ophthalmologists masked to each other and to the patient's identity. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) to detect DR and STDR by NM retinal imaging were assessed.

Results:

245 people had gradable images in one or both eyes. DR and STDR were detected in 45.3% and 24.5%, respectively using NM camera, and in 57.6% and 28.6%, respectively using mydriatic camera. The sensitivity and specificity to detect any DR by NM camera was 75.2% (95% confidence interval (CI) 68.1–82.3) and 95.2% (95%CI 91.1–99.3). For STDR the values were 82.9% (95% CI 74.0–91.7) and 98.9% (95% CI 97.3–100), respectively. The PPV to detect any DR was 95.5% (95% CI 89.8–98.5) and NPV was 73.9% (95% CI 66.4–81.3); PPV for STDR detection was 96.7% (95% CI 92.1–100)) and NPV was 93.5% (95% CI 90.0–97.1).

Conclusion:

Smartphone-based NM retinal camera had fairly high sensitivity and specificity for detection of DR and STDR in this clinic-based study. Further studies are warranted in other settings.

Digital Ophthalmology in the UK - Diabetic Retinopathy Screening and Virtual Glaucoma Clinics in the National Health Service

The customary doctor and patient interactions are currently undergoing significant changes through technological advances in imaging and data processing and the need for reducing person-to person contacts during the COVID-19 crisis. There is a trend away from face-to-face examinations to virtual assessments and decision making. Ophthalmology is particularly amenable to such changes, as a high proportion of clinical decisions are based on routine tests and imaging results, which can be assessed remotely. The uptake of digital ophthalmology varies significantly between countries. Due to financial constraints within the National Health Service, specialized ophthalmology units in the UK have been early adopters of digital technology. For more than a decade, patients have been managed remotely in the diabetic retinopathy screening service and virtual glaucoma clinics. We describe the day-to-day running of such services and the doctor and patient experiences with digital ophthalmology in daily practice.

Review of retinal cameras for global coverage of diabetic retinopathy screening

The global burden of diabetes has resulted in an increase in the prevalence of diabetic retinopathy (DR), a microvascular complication of diabetes. Lifelong repetitive screening for DR is essential for early detection and timely management to prevent visual impairment due to the silent sight-threatening disorder. Colour fundus photography (CFP) is helpful for documentation of the retinopathy as well as for counselling the patient. CFP has established roles in DR screening, detection, progression and monitoring of treatment response. DR screening programmes use validated mydriatic or non-mydriatic fundus cameras for retinal imaging and trained image graders identify referable DR. Smartphone-based fundus cameras and handheld fundus cameras that are cost-effective, portable and easy to handle in remote places are gaining popularity in recent years. The images captured with these low-cost devices can be immediately sent to trained ophthalmologists for grading of DR. Recent increase in numbers of telemedicine programmes based on imaging with digital fundus cameras and remote interpretation has facilitated larger population coverage of DR screening and timely referral of those with sight-threatening DR to ophthalmologists. Good-quality retinal imaging and accurate diagnosis are essential to reduce inappropriate referrals. Advances in digital imaging such as ultra-wide field imaging and multi-modal imaging have opened new avenues for assessing DR. Fundus cameras with integrated artificial intelligence (AI)-based automated algorithms can also provide instant DR diagnosis and reduce the burden of healthcare systems. We review the different types of fundus cameras currently used in DR screening and management around the world.

Validation of handheld fundus camera with mydriasis for retinal imaging of diabetic retinopathy screening in China: a prospective comparison study

OBJECTIVES

To investigate the clinical validity of using a handheld fundus camera to detect diabetic retinopathy (DR) in China.

DESIGN AND SETTINGS

Prospective comparison study of the handheld fundus camera with a standard validated instrument in detection of DR in hospital and a community screening clinic in Guangdong Province, China.

PARTICIPANTS

Participants aged 18 years and over with diabetes who were able to provide informed consent and agreed to attend the dilated eye examination with handheld tests and a standard desktop camera.

PRIMARY AND SECONDARY OUTCOME MEASURES

Primary outcome was the proportion of those with referable DR (R2 and above) identified by the handheld fundus camera (the index test) compared with the standard camera. Secondary outcome was the comparison of proportion of gradable images obtained from each test.

RESULTS

In this study, we examined 304 people (608 eyes) with each of the two cameras under mydriasis. The handheld camera detected 119 eyes (19.5%) with some level of DR, 81 (13.3%) of them were referable, while the standard camera detected 132 eyes (21.7%) with some level of DR and 83 (13.7%) were referable. It seems that the standard camera found more eyes with referable DR, although McNemar's test detected no significant difference between the two cameras.Of the 608 eyes with images obtained by desktop camera, 598 (98.4%) images were of sufficient quality for grading, 12 (1.9%) images were not gradable. By the handheld camera, 590 (97.0%) were gradable and 20 (3.2%) images were not gradable.The two cameras reached high agreement on diagnosis of retinopathy and maculopathy at all the levels of retinopathy.

CONCLUSION

Although it could not take the place of standard desktop camera on clinic fundus examination, the handheld fundus camera showed promising role on preliminary DR screening at primary level in China. To ensure quality images, mydriasis is required.

The Evolution of Diabetic Retinopathy Screening Programmes: A Chronology of Retinal Photography from 35 mm Slides to Artificial Intelligence

As a third of people with diabetes mellitus (DM) will suffer the microvascular complications of diabetic retinopathy (DR) and therapeutic options can effectively prevent visual impairment, systematic screening has substantially reduced disease burden in developed countries. In an effort to tackle the rising incidence of DM, screening programmes have modernized in synchrony with technical and infrastructural advancements. Patient evaluation has shifted from face-to-face ophthalmologist-based review delivered through community grassroots to asynchronous store-and-forward modern telemedicine platforms commissioned on a nationwide scale. First pioneered with primitive 35-mm slide film retinal photography, the last decade has seen an emergence of high resolution and widefield imaging devices, which may reveal extents of DR indiscernible to the clinician but with implications of potential earlier identification. Similar progress has been seen in image analysis approaches - automated image analysis of retinal photographs of DR has evolved from qualitative feature detection to rules-based algorithms to autonomous artificial intelligence-powered classification. Such models have, relatively rapidly, been validated and are now receiving approval from health regulation authorities with deployment into the clinical sphere. In this review, we chart the evolution of global DR screening programmes since their inception highlighting major milestones in healthcare infrastructure, telemedicine approaches and imaging devices that have shaped the robust and effective frameworks recognised today. We also provide an outlook for the future of DR screening in the context of recent technological advancements with respect to their limitations in current times.

Application of Artificial Intelligence in Targeting Retinal Diseases

Retinal diseases affect an increasing number of patients worldwide because of the aging population. Request for diagnostic imaging in ophthalmology is ramping up, while the number of specialists keeps shrinking. Cutting-edge technology embedding artificial intelligence (AI) algorithms are thus advocated to help ophthalmologists perform their clinical tasks as well as to provide a source for the advancement of novel biomarkers. In particular, optical coherence tomography (OCT) evaluation of the retina can be augmented by algorithms based on machine learning and deep learning to early detect, qualitatively localize and quantitatively measure epi/intra/subretinal abnormalities or pathological features of macular or neural diseases. In this paper, we discuss the use of AI to facilitate efficacy and accuracy of retinal imaging in those diseases increasingly treated by intravitreal vascular endothelial growth factor (VEGF) inhibitors (i.e. anti-VEGF drugs), also including integration and interpretation features in the process. We review recent advances by AI in diabetic retinopathy, age-related macular degeneration, and retinopathy of prematurity that envision a potentially key role of highly automated systems in screening, early diagnosis, grading and individualized therapy. We discuss benefits and critical aspects of automating the evaluation of disease activity, recurrences, the timing of retreatment and therapeutically potential novel targets in ophthalmology. The impact of massive employment of AI to optimize clinical assistance and encourage tailored therapies for distinct patterns of retinal diseases is also discussed.

Ten-year incidence and assessment of safe screening intervals for diabetic retinopathy: the OPHDIAT study

BACKGROUND

To estimate the 10-year incidence of referable diabetic retinopathy (DR) in a French population with type 1 and 2 diabetes mellitus (DM). A secondary objective was the assessment of safe screening intervals in patients with diabetes without retinopathy.

METHODS

Observational, prospective and multicentric study between June 2004 and September 2017 based on a regional screening programme for DR in the Paris region. The incidence of referable DR in patients without retinopathy at baseline was calculated by the Turnbull survival estimator. A safe screening interval was defined as a 95% probability of remaining without referable DR.

RESULTS

Among the 25 745 participants with type 1 (n=6086) or type 2 (n=19 659) DM, the 10-year cumulative incidence of referable DR was 19.10% (95% CI 17.21% to 21.14%) and 17.03% (15.78% to 18.35%), median (IQR) follow-up=3.33 (4.24) years. The safe screening interval for patients without DR at the first examination for type 1 and 2 DM was 2.2 (95% CI 2.0 to 2.4) and 3.0 (2.9 to 3.1) years, respectively. In a subgroup of low-risk patients with type 2 DM, the safe screening interval was 4.2 (3.8 to 4.6) years.

CONCLUSIONS

These data suggest that in Paris area, a 2-year, 3-year and 4-year screening interval was considered safe for type 1 DM, type 2 DM and for low-risk patients with type 2 DM, respectively, without DR at the first examination. While these data might be used to support the consideration of extending screening intervals, a randomised clinical trial would be suitable to confirm the safety for patients with DM.

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

BACKGROUND

Photographic diabetic retinopathy screening requires labour-intensive grading of retinal images by humans. Automated retinal image analysis software (ARIAS) could provide an alternative to human grading. We compare the performance of an ARIAS using true-colour, wide-field confocal scanning images and standard fundus images in the English National Diabetic Eye Screening Programme (NDESP) against human grading.

METHODS

Cross-sectional study with consecutive recruitment of patients attending annual diabetic eye screening. Imaging with mydriasis was performed (two-field protocol) with the EIDON platform (CenterVue, Padua, Italy) and standard NDESP cameras. Human grading was carried out according to NDESP protocol. Images were processed by EyeArt V.2.1.0 (Eyenuk Inc, Woodland Hills, California). The reference standard for analysis was the human grade of standard NDESP images.

RESULTS

We included 1257 patients. Sensitivity estimates for retinopathy grades were: EIDON images; 92.27% (95% CI: 88.43% to 94.69%) for any retinopathy, 99% (95% CI: 95.35% to 100%) for vision-threatening retinopathy and 100% (95% CI: 61% to 100%) for proliferative retinopathy. For NDESP images: 92.26% (95% CI: 88.37% to 94.69%) for any retinopathy, 100% (95% CI: 99.53% to 100%) for vision-threatening retinopathy and 100% (95% CI: 61% to 100%) for proliferative retinopathy. One case of vision-threatening retinopathy (R1M1) was missed by the EyeArt when analysing the EIDON images, but identified by the human graders. The EyeArt identified all cases of vision-threatening retinopathy in the standard images.

CONCLUSION

EyeArt identified diabetic retinopathy in EIDON images with similar sensitivity to standard images in a large-scale screening programme, exceeding the sensitivity threshold recommended for a screening test. Further work to optimise the identification of 'no retinopathy' and to understand the differential lesion detection in the two imaging systems would enhance the use of these two innovative technologies in a diabetic retinopathy screening setting.

Screening for diabetic retinopathy: new perspectives and challenges

Although the prevalence of all stages of diabetic retinopathy has been declining since 1980 in populations with improved diabetes control, the crude prevalence of visual impairment and blindness caused by diabetic retinopathy worldwide increased between 1990 and 2015, largely because of the increasing prevalence of type 2 diabetes, particularly in low-income and middle-income countries. Screening for diabetic retinopathy is essential to detect referable cases that need timely full ophthalmic examination and treatment to avoid permanent visual loss. In the past few years, personalised screening intervals that take into account several risk factors have been proposed, with good cost-effectiveness ratios. However, resources for nationwide screening programmes are scarce in many countries. New technologies, such as scanning confocal ophthalmology with ultrawide field imaging and handheld mobile devices, teleophthalmology for remote grading, and artificial intelligence for automated detection and classification of diabetic retinopathy, are changing screening strategies and improving cost-effectiveness. Additionally, emerging evidence suggests that retinal imaging could be useful for identifying individuals at risk of cardiovascular disease or cognitive impairment, which could expand the role of diabetic retinopathy screening beyond the prevention of sight-threatening disease.

Practice Guidelines for Ocular Telehealth-Diabetic Retinopathy, Third Edition

Contributors The following document and appendices represent the third edition of the Practice Guidelines for Ocular Telehealth-Diabetic Retinopathy. These guidelines were developed by the Diabetic Retinopathy Telehealth Practice Guidelines Working Group. This working group consisted of a large number of subject matter experts in clinical applications for telehealth in ophthalmology. The editorial committee consisted of Mark B. Horton, OD, MD, who served as working group chair and Christopher J. Brady, MD, MHS, and Jerry Cavallerano, OD, PhD, who served as cochairs. The writing committees were separated into seven different categories. They are as follows: 1.Clinical/operational: Jerry Cavallerano, OD, PhD (Chair), Gail Barker, PhD, MBA, Christopher J. Brady, MD, MHS, Yao Liu, MD, MS, Siddarth Rathi, MD, MBA, Veeral Sheth, MD, MBA, Paolo Silva, MD, and Ingrid Zimmer-Galler, MD. 2.Equipment: Veeral Sheth, MD (Chair), Mark B. Horton, OD, MD, Siddarth Rathi, MD, MBA, Paolo Silva, MD, and Kristen Stebbins, MSPH. 3.Quality assurance: Mark B. Horton, OD, MD (Chair), Seema Garg, MD, PhD, Yao Liu, MD, MS, and Ingrid Zimmer-Galler, MD. 4.Glaucoma: Yao Liu, MD, MS (Chair) and Siddarth Rathi, MD, MBA. 5.Retinopathy of prematurity: Christopher J. Brady, MD, MHS (Chair) and Ingrid Zimmer-Galler, MD. 6.Age-related macular degeneration: Christopher J. Brady, MD, MHS (Chair) and Ingrid Zimmer-Galler, MD. 7.Autonomous and computer assisted detection, classification and diagnosis of diabetic retinopathy: Michael Abramoff, MD, PhD (Chair), Michael F. Chiang, MD, and Paolo Silva, MD.

Automated and Computer-Assisted Detection, Classification, and Diagnosis of Diabetic Retinopathy

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

Design, Development, and Evaluation of a Teleophthalmology System Using a Low-Cost Fundus Camera

Introduction:

The increasing prevalence of diabetic retinopathy in developing countries has become a worldwide concern. This problem is preventable by timely diagnosis and treatment; however, in the majority of cases, patients attend the eye clinics very late because of a lack of specialists and travel difficulties. Running a teleophthalmology system would significantly help to manage this disease.

Aim:

This study seeks to assess the accuracy of the teleophthalmology system and its effect on reducing unnecessary referrals in Iran.

Methods:

This study was conducted on 125 diabetic patients. First, the patients were examined by a retina specialist using a slit lamp and, then, single-field digital photos were captured by a portable, low-cost fundus camera. The images were uploaded onto a website and, after two months, were assessed by two retina specialists and two general practitioners (GPs). Finally, the diagnoses based on the digital photos were contrasted with the diagnoses established through face-to-face visits as a gold standard.

Results:

Out of 125 diabetic patients, eight (6.4%) were removed because of low-quality images and a total of 117 were evaluated. The sensitivity and specificity of each retina specialist presented with the photographs produced success rates of 90% and 97% respectively when judged against the gold standard of face-to-face visits. The rates of sensitivity for retinopathy referrals from the retina specialists were 92% and 85%. The sensitivity and specificity of their diagnoses of clinically significant macular edema (CSME) were calculated at 93% and 100%. The rates of sensitivity for each GP were 95% and 93% and the level of specificity was estimated to be approximately 98% for both GPs. The diagnosis rate for GPs when viewing the photographs as opposed to hosting face-to-face visits was more than 90%. Generally, with the implementation of this system, between 40% and 55% of referrals were calculated to have been avoidable.

Conclusion:

Our results from the first-ever research conducted on this topic in Iran showed that the teleophthalmology system is extremely accurate, that it can prevent unnecessary referrals and that it is useful for locating treatable patients. The results of this study could be of assistance in the running and expansion of such systems throughout Iran and Kerman Province to reduce eye damage arising from diabetes, decrease avoidable referrals to clinics, increase the availability of specialist visits for people in remote and rural areas and optimize the use of clinical infrastructures for patients in emergencies.

Update on Screening for Sight-Threatening Diabetic Retinopathy

PURPOSE

The aim of this article was to describe recent advances in the use of new technology in diabetic retinopathy screening by looking at studies that assessed the effectiveness and cost-effectiveness of these technologies.

METHODS

The author conducts an ongoing search for articles relating to screening or management of diabetic retinopathy utilising Zetoc with keywords and contents page lists from relevant journals.

RESULTS

The areas discussed in this article are reference standards, alternatives to digital photography, area of retina covered by the screening method, size of the device and hand-held cameras, mydriasis versus non-mydriasis or a combination, measurement of distance visual acuity, grading of images, use of automated grading analysis and cost-effectiveness of the new technologies.

CONCLUSIONS

There have been many recent advances in technology that may be adopted in the future by screening programmes for sight-threatening diabetic retinopathy but each device will need to demonstrate effectiveness and cost-effectiveness before more widespread adoption.

Diagnostic test accuracy of diabetic retinopathy screening by physician graders using a hand-held non-mydriatic retinal camera at a tertiary level medical clinic

Background

The evidence on diagnostic test accuracy (DTA) of diabetic retinopathy (DR) screening utilising photographic studies by non-ophthalmologist personnel in low and middle-income country (LMIC) settings is scarce. We aimed to assess DTA of DR screening using a nonmydriatic hand-held digital camera by trained general physicians in a non-ophthalmic setting.

Methods

This study is a validation of a screening intervention. We selected 700 people with diabetes (PwDM) > 18 years of age, not previously screened or treated for DR, presenting at a tertiary medical clinic in Sri Lanka. Two-field retinal imaging was used to capture fundus images before and after pupil dilatation, using a hand-held non-mydriatic (Visuscout 100®-Germany) digital retinal camera. The images were captured and graded by two trained, masked independent physician graders. The DTA of different levels of DR was assessed comparing physician’s grading with a retinologist’s clinical examination by mydriatic bio-microscopy, according to a locally adopted guideline.

Results

Seven hundred eligible PwDM were screened by physician graders. The mean age of participants was 60.8 years (SD ±10.08) and mean duration of DM was 9.9 years (SD ±8.09). Ungradable image proportion in non-mydriatic imaging was 43.4% (either eye-31.3%, both eyes 12.1%). This decreased to 12.8% (either eye-11.6%, both eyes-1.2%) following pupil dilatation. In comparison to detection of any level of DR, a referable level DR (moderate non-proliferative DR and levels above) showed a higher level of DTA. The sensitivity of the defined referable DR was 88.7% (95% CI 81.7–93.8%) for grader 1 (positive predictive value [PPV] 59.1%) and 92.5% (95% CI 86.4–96.5%) for grader 2 (PPV 68%), using mydriatic imaging, after including ungradable images as screen positives. The specificity was 94.9% (95% CI 93.6–96.0%) for grader 1 (negative predictive value [NPV] 99%) and 96.4% (95% CI 95.3–97.3%) for grader 2 (NPV 99.4%).

Conclusions

The Physicians grading of images from a digital hand-held non-mydriatic camera at a medical clinic, with dilatation of pupil of those who have ungradable images, provides a valid modality to identify referable level of DR. This could be a feasible alternative modality to the existing opportunistic screening to improve the access and coverage.

Trial registration

Current Controlled Trials ISRCTN47559703. Date of Registration 18th March 2019, Retrospectively registered.

Electronic supplementary material

The online version of this article (10.1186/s12886-019-1092-3) contains supplementary material, which is available to authorized users.

Combination of Global Features for the Automatic Quality Assessment of Retinal Images

Diabetic retinopathy (DR) is one of the most common causes of visual loss in developed countries. Computer-aided diagnosis systems aimed at detecting DR can reduce the workload of ophthalmologists in screening programs. Nevertheless, a large number of retinal images cannot be analyzed by physicians and automatic methods due to poor quality. Automatic retinal image quality assessment (RIQA) is needed before image analysis. The purpose of this study was to combine novel generic quality features to develop a RIQA method. Several features were calculated from retinal images to achieve this goal. Features derived from the spatial and spectral entropy-based quality (SSEQ) and the natural images quality evaluator (NIQE) methods were extracted. They were combined with novel sharpness and luminosity measures based on the continuous wavelet transform (CWT) and the hue saturation value (HSV) color model, respectively. A subset of non-redundant features was selected using the fast correlation-based filter (FCBF) method. Subsequently, a multilayer perceptron (MLP) neural network was used to obtain the quality of images from the selected features. Classification results achieved 91.46% accuracy, 92.04% sensitivity, and 87.92% specificity. Results suggest that the proposed RIQA method could be applied in a more general computer-aided diagnosis system aimed at detecting a variety of retinal pathologies such as DR and age-related macular degeneration.

Artificial intelligence and deep learning in ophthalmology

Artificial intelligence (AI) based on deep learning (DL) has sparked tremendous global interest in recent years. DL has been widely adopted in image recognition, speech recognition and natural language processing, but is only beginning to impact on healthcare. In ophthalmology, DL has been applied to fundus photographs, optical coherence tomography and visual fields, achieving robust classification performance in the detection of diabetic retinopathy and retinopathy of prematurity, the glaucoma-like disc, macular oedema and age-related macular degeneration. DL in ocular imaging may be used in conjunction with telemedicine as a possible solution to screen, diagnose and monitor major eye diseases for patients in primary care and community settings. Nonetheless, there are also potential challenges with DL application in ophthalmology, including clinical and technical challenges, explainability of the algorithm results, medicolegal issues, and physician and patient acceptance of the AI 'black-box' algorithms. DL could potentially revolutionise how ophthalmology is practised in the future. This review provides a summary of the state-of-the-art DL systems described for ophthalmic applications, potential challenges in clinical deployment and the path forward.

Advances in Retinal Imaging and Applications in Diabetic Retinopathy Screening: A Review

Rising prevalence of diabetes worldwide has necessitated the implementation of population-based diabetic retinopathy (DR) screening programs that can perform retinal imaging and interpretation for extremely large patient cohorts in a rapid and sensitive manner while minimizing inappropriate referrals to retina specialists. While most current screening programs employ mydriatic or nonmydriatic color fundus photography and trained image graders to identify referable DR, new imaging modalities offer significant improvements in diagnostic accuracy, throughput, and affordability. Smartphone-based fundus photography, macular optical coherence tomography, ultrawide-field imaging, and artificial intelligence-based image reading address limitations of current approaches and will likely become necessary as DR becomes more prevalent. Here we review current trends in imaging for DR screening and emerging technologies that show potential for improving upon current screening approaches.

Systematic review and meta-analysis of diagnostic accuracy of detection of any level of diabetic retinopathy using digital retinal imaging

BACKGROUND

Visual impairment from diabetic retinopathy (DR) is an increasing global public health concern, which is preventable with screening and early treatment. Digital retinal imaging has become a preferred choice as it enables higher coverage of screening. The aim of this review is to evaluate how different characteristics of the DR screening (DRS) test impact on diagnostic test accuracy (DTA) and its relevance to a low-income setting.

METHODS

We conducted a systematic literature search to identify clinic-based studies on DRS using digital retinal imaging of people with DM (PwDM). Summary estimates of different sub-groups were calculated using DTA values weighted according to the sample size. The DTA of each screening method was derived after exclusion of ungradable images and considering the eye as the unit of analysis. The meta-analysis included studies which measured DTA of detecting any level of DR. We also examined the effect on detection from using different combinations of retinal fields, pupil status, index test graders and setting.

RESULTS

Six thousand six hundred forty-six titles and abstracts were retrieved, and data were extracted from 122 potentially eligible full reports. Twenty-six studies were included in the review, and 21 studies, mostly from high-income settings (18/21, 85.7%), were included in the meta-analysis. The highest sensitivity was observed in the mydriatic greater than two field strategy (92%, 95% CI 90-94%). The highest specificity was observed in greater than two field methods (94%, 95% CI 93-96%) where mydriasis did not affect specificity. Overall, there was no difference in sensitivity between non-mydriatic and mydriatic methods (86%, 95% CI 85-87) after exclusion of ungradable images. The highest DTA (sensitivity 90%, 95% CI 88-91%; specificity 95%, 95% CI 94-96%) was observed when screening was delivered at secondary/tertiary level clinics.

CONCLUSIONS

Non-mydriatic two-field strategy could be a more pragmatic approach in starting DRS programmes for facility-based PwDM in low-income settings, with dilatation of the pupils of those who have ungradable images. There was insufficient evidence in primary studies to draw firm conclusions on how graders' background influences DTA. Conducting more context-specific DRS validation studies in low-income and non-ophthalmic settings can be recommended.

Smartphones, tele-ophthalmology, and VISION 2020

Telemedicine is an emerging field in recent medical achievements with rapid development. The "smartphone" availability has increased in both developed and developing countries even among people in rural and remotes areas. Tele-based services can be used for screening ophthalmic diseases and also monitoring patients with known diseases. Electronic ophthalmologic records of the patients including captured images by smartphones from anterior and posterior segments of the eye will be evaluated by ophthalmologists, and if patients require further evaluations, they will be referred to experts in the relevant field. Eye diseases such as cataract, glaucoma, age-related macular degeneration, diabetic retinopathy, and retinopathy of prematurity are the most common causes of blindness in many countries and beneficial use of teleophthalmology with smartphones will be a good way to achieve the aim of VISION 2020 all over the world. Numerous studies have shown that teleophthalmology is similar to the conventional eye care system in clinical outcomes and even provides more patient satisfaction as it saves time and cost. This review explains how teleophthalmology helps to improve patient outcomes through smartphones.

Action on diabetic macular oedema: achieving optimal patient management in treating visual impairment due to diabetic eye disease

This paper identifies best practice recommendations for managing diabetes and sight-threatening diabetic eye disease. The authors provide an update for ophthalmologists and allied healthcare professionals on key aspects of diabetes management, supported by a review of the pertinent literature, and recommend practice principles for optimal patient management in treating visual impairment due to diabetic eye disease. In people with diabetes, early optimal glycaemic control reduces the long-term risk of both microvascular and macrovascular complications. The authors propose more can and should be done to maximise metabolic control, promote appropriate behavioural modifications and encourage timely treatment intensification when indicated to ameliorate diabetes-related complications. All people with diabetes should be screened for sight-threatening diabetic retinopathy promptly and regularly. It is shown that attitudes towards treatment adherence in diabetic macular oedema appear to mirror patients' views and health behaviours towards the management of their own diabetes. Awareness of diabetic macular oedema remains low among people with diabetes, who need access to education early in their disease about how to manage their diabetes to delay progression and possibly avoid eye-related complications. Ophthalmologists and allied healthcare professionals play a vital role in multidisciplinary diabetes management and establishment of dedicated diabetic macular oedema clinics is proposed. A broader understanding of the role of the diabetes specialist nurse may strengthen the case for comprehensive integrated care in ophthalmic practice. The recommendations are based on round table presentations and discussions held in London, UK, September 2016.

The English National Screening Programme for diabetic retinopathy 2003-2016

The aim of the English NHS Diabetic Eye Screening Programme is to reduce the risk of sight loss amongst people with diabetes by the prompt identification and effective treatment if necessary of sight-threatening diabetic retinopathy, at the appropriate stage during the disease process. In order to achieve the delivery of evidence-based, population-based screening programmes, it was recognised that certain key components were required. It is necessary to identify the eligible population in order to deliver the programme to the maximum number of people with diabetes. The programme is delivered and supported by suitably trained, competent, and qualified, clinical and non-clinical staff who participate in recognised ongoing Continuous Professional Development and Quality Assurance schemes. There is an appropriate referral route for those with screen-positive disease for ophthalmology treatment and for assessment of the retinal status in those with poor-quality images. Appropriate assessment of control of their diabetes is also important in those who are screen positive. Audit and internal and external quality assurance schemes are embedded in the service. In England, two-field mydriatic digital photographic screening is offered annually to all people with diabetes aged 12 years and over. The programme commenced in 2003 and reached population coverage across the whole of England by 2008. Increasing uptake has been achieved and the current annual uptake of the programme in 2015-16 is 82.8% when 2.59 million people with diabetes were offered screening and 2.14 million were screened. The benefit of the programme is that, in England, diabetic retinopathy/maculopathy is no longer the leading cause of certifiable blindness in the working age group.

Risk of diabetic retinopathy at first screen in children at 12 and 13 years of age

AIMS

To investigate the relationships between age at diagnosis of diabetes, age at diabetic eye screening and severity of diabetic retinopathy at first and subsequent screenings in children aged 12 or 13 years.

METHODS

Data were extracted from four English screening programmes and from the Scottish, Welsh and Northern Irish programmes on all children with diabetes invited for their first and subsequent screening episodes from the age of 12 years. Retinopathy levels at first and subsequent screens, time from diagnosis of diabetes to first screening and age at diagnosis in years were calculated.

RESULTS

Data were available for 2125 children with diabetes screened for the first time at age 12 or 13 years. In those diagnosed with diabetes at 2 years of age or less, the proportion with retinopathy in one or both eyes was 20% and 11%, respectively, decreasing to 8% and 2% in those diagnosed between 2 and 12 years (P < 0.0001). Only three children (aged 8, 10 and 11 years at diagnosis of diabetes) had images graded with referable retinopathy and, of these, two had non-referable diabetic retinopathy at all subsequent screenings. Of 1703 children with subsequent images, 25 were graded with referable diabetic retinopathy over a mean follow-up of 3.1 years, an incidence rate of 4.7 (95% confidence interval, 3.1-7.0) per 1000 per year.

CONCLUSIONS

In this large cohort of children, the low prevalence and incidence rates of referable diabetic retinopathy suggest that screening earlier than age 12 is not necessary.

Sensitivity and specificity of nonmydriatic digital imaging in screening diabetic retinopathy in Indian eyes

Background:

Nonmydriatic digital imaging (NMDI) is ideal for screening diabetic retinopathy (DR), but its use in Indian eyes has not been evaluated.

Aim:

The aim was to evaluate the sensitivity and specificity of NMDI as a screening tool in detecting DR in Indian eyes.

Design:

A prospective, nonrandomized, noncomparative, noninterventional study.

Materials and Methods:

A total of 500 diabetic patients visiting the endocrinology clinic (September 2008-June 2010) underwent NMDI (Zeiss Procam), followed by routine dilated fundus photography (FP; Zeiss Visupac 450+) of 345° retinal fields (1) optic disc and macula, (2) superotemporal, and (3) nasal to optic disc. Two-masked retina specialists graded the images for quality and severity of DR, and compared between NMDI and dilated FP.

Statistical Analysis:

SPSS Windows 17 for version.

Results:

Mean age was 52.97 ± 13.46 years (306 males: 194 females). The rate of ungradable images was 30.6% and 31% by the two observers. By observer 1, the sensitivity and specificity of detecting any DR was 58.8% and 69.1%, respectively, (κ = 0.608) and sight-threatening DR (STDR) was 63.1% and 68.9%, respectively, (κ = 0.641). By observer 2, the sensitivity and specificity was 57.3% and 68.3%, respectively, for any DR (κ = 0.593) and 62.8% and 68.3%, respectively, for STDR (κ = 0.637). The level of agreement between two observers was high (κ = 0.96).

Conclusion:

A high rate of poor quality photographs and low sensitivity limited the use of NMDI as a perfect screening system, particularly in dark iris population with diabetes as seen in Indian eyes.

A review on automatic analysis techniques for color fundus photographs

In this paper, we give a review on automatic image processing tools to recognize diseases causing specific distortions in the human retina. After a brief summary of the biology of the retina, we give an overview of the types of lesions that may appear as biomarkers of both eye and non-eye diseases. We present several state-of-the-art procedures to extract the anatomic components and lesions in color fundus photographs and decision support methods to help clinical diagnosis. We list publicly available databases and appropriate measurement techniques to compare quantitatively the performance of these approaches. Furthermore, we discuss on how the performance of image processing-based systems can be improved by fusing the output of individual detector algorithms. Retinal image analysis using mobile phones is also addressed as an expected future trend in this field.

Effect of Visual Impairment on Teleretinal Imaging for Diabetic Retinopathy Screening

BACKGROUND AND OBJECTIVE

To evaluate the effect of decreased visual acuity on image quality obtained by non-mydriatic retinal photography in diabetic subjects. PATIENTS AND METHODS This case-control study comprised 422 eyes (211 cases with ungradable images after non-mydriatic retinal photography, and 211 controls). All subjects underwent complete ophthalmic examination. The association between ungradable image rate and several eye factors, such as corrected distance visual acuity (CDVA), spherical equivalent (SE), astigmatism, and cataracts, was evaluated using a generalized estimating equations model.

RESULTS

Visual impairment (Snellen CDVA worse than 20/40) was significantly associated with an increased likelihood of ungradable images. The odds ratio (OR) for this association was 7.79 (95% CI, 4.19-14.50; P < .0001). This relationship remained significant in the multivariable model (OR: 5.23; 95% CI, 2.82-9.71; P < .0001). Similarly, refractive error worse than -6.0 diopters (D) SE or +5.0 D SE was associated with increased risk of ungradable scans, with an OR of 13.21 (95% CI, 2.61-66.77; P = .002).

CONCLUSION

Decreased visual acuity may be a predictor of inaccurate image analysis in subjects screened for diabetic retinopathy by non-mydriatic retinal photography.