Are we making good use of our public resources? The false-positive rate of screening by fundus photography for diabetic macular oedema

Diagnostic Utility of Swept-Source OCT-Based Biometry and Fundus Photographs Compared to Spectral Domain OCT in Center-Involving Diabetic Macular Edema

PURPOSE

This study was aimed to evaluate the agreement between the swept-source optical coherence tomography (SS-OCT)-based biometry, fundus photographs, and their combination, in comparison to the gold standard spectral-domain optical coherence tomography (SD-OCT) for the detection of center-involving diabetic macular edema (CI-DME).

METHODS

We conducted a retrospective cross-sectional study involving 55 subjects (78 eyes) diagnosed with diabetic macular edema (DME) detected clinically and on SD-OCT (Carl Zeiss Meditec AG). Post-mydriatic 45-degree color fundus photograph (Crystal-Vue NFC-700), 1 mm macular scan obtained from SS-OCT-based biometry (IOL-Master 700), and macula cube scan obtained from SD-OCT was used to detect and grade DME into CI-DME and NCI-DME.

RESULTS

Our findings revealed that SS-OCT-based biometry was noted to have a high sensitivity of 1 (0.94-1.00) and a specificity of 0.63 (0.31-0.89) in detecting CI-DME compared to the gold standard (SD-OCT). When combined with data from fundus photographs, specificity decreased to 0.32 (0.15-0.53). Fundus photographs alone exhibited a low sensitivity of 0.52 (0.38-0.64) and a specificity of 0.45 (0.16-0.76) in CI-DME detection.

CONCLUSION

In conclusion, SS-OCT-based biometry can be used as an effective tool for the detection of CI-DME in diabetic patients undergoing cataract surgery and can serve as a screening tool in centers without SD-OCT facilities.

Comparisons of Handheld Retinal Imaging with Optical Coherence Tomography for the Identification of Macular Pathology in Patients with Diabetes

INTRODUCTION

Handheld retinal imaging cameras are relatively inexpensive and highly portable devices that have the potential to significantly expand diabetic retinopathy (DR) screening, allowing a much broader population to be evaluated. However, it is essential to evaluate if these devices can accurately identify vision-threatening macular diseases if DR screening programs will rely on these instruments. Thus, the purpose of this study was to evaluate the detection of diabetic macular pathology using monoscopic macula-centered images using mydriatic handheld retinal imaging compared with spectral domain optical coherence tomography (SDOCT).

METHODS

Mydriatic 40°-60° macula-centered images taken with 3 handheld retinal imaging devices (Aurora [AU], SmartScope [SS], RetinaVue 700 [RV]) were compared with the Cirrus 6000 SDOCT taken during the same visit. Images were evaluated for the presence of diabetic macular edema (DME) on monoscopic fundus photographs adapted from Early Treatment Diabetic Retinopathy Study (ETDRS) definitions (no DME, noncenter-involved DME [non-ciDME], and center-involved DME [ciDME]). Sensitivity, specificity, positive predictive value, and negative predictive value were calculated for each device with SDOCT as gold standard.

RESULTS

Severity by ETDRS photos: no DR 33.3%, mild NPDR 20.4%, moderate 14.2%, severe 11.6%, proliferative 20.4%, and ungradable for DR 0%; no DME 83.1%, non-ciDME 4.9%, ciDME 12.0%, and ungradable for DME 0%. Gradable images by SDOCT (N = 217, 96.4%) showed no DME in 75.6%, non-ciDME in 9.8%, and ciDME in 11.1%. The ungradable rate for images (poor visualization in >50% of the macula) was AU: 0.9%, SS: 4.4%, and RV: 6.2%. For DME, sensitivity and specificity were similar across devices (0.5-0.64, 0.93-0.97). For nondiabetic macular pathology (ERM, pigment epithelial detachment, traction retinal detachment) across all devices, sensitivity was low to moderate (0.2-0.5) but highly specific (0.93-1.00).

CONCLUSIONS

Compared to SDOCT, handheld macular imaging attained high specificity but low sensitivity in identifying macular pathology. This suggests the importance of SDOCT evaluation for patients suspected to have DME on fundus photography, leading to more appropriate referral refinement.

Medical Staff and Resident Preferences for Using Deep Learning in Eye Disease Screening: Discrete Choice Experiment

Background

Deep learning–assisted eye disease diagnosis technology is increasingly applied in eye disease screening. However, no research has suggested the prerequisites for health care service providers and residents willing to use it.

Objective

The aim of this paper is to reveal the preferences of health care service providers and residents for using artificial intelligence (AI) in community-based eye disease screening, particularly their preference for accuracy.

Methods

Discrete choice experiments for health care providers and residents were conducted in Shanghai, China. In total, 34 medical institutions with adequate AI-assisted screening experience participated. A total of 39 medical staff and 318 residents were asked to answer the questionnaire and make a trade-off among alternative screening strategies with different attributes, including missed diagnosis rate, overdiagnosis rate, screening result feedback efficiency, level of ophthalmologist involvement, organizational form, cost, and screening result feedback form. Conditional logit models with the stepwise selection method were used to estimate the preferences.

Results

Medical staff preferred high accuracy: The specificity of deep learning models should be more than 90% (odds ratio [OR]=0.61 for 10% overdiagnosis; P<.001), which was much higher than the Food and Drug Administration standards. However, accuracy was not the residents’ preference. Rather, they preferred to have the doctors involved in the screening process. In addition, when compared with a fully manual diagnosis, AI technology was more favored by the medical staff (OR=2.08 for semiautomated AI model and OR=2.39 for fully automated AI model; P<.001), while the residents were in disfavor of the AI technology without doctors’ supervision (OR=0.24; P<.001).

Conclusions

Deep learning model under doctors’ supervision is strongly recommended, and the specificity of the model should be more than 90%. In addition, digital transformation should help medical staff move away from heavy and repetitive work and spend more time on communicating with residents.

Diabetic retinopathy screening in the emerging era of artificial intelligence

Diabetic retinopathy is a frequent complication in diabetes and a leading cause of visual impairment. Regular eye screening is imperative to detect sight-threatening stages of diabetic retinopathy such as proliferative diabetic retinopathy and diabetic macular oedema in order to treat these before irreversible visual loss occurs. Screening is cost-effective and has been implemented in various countries in Europe and elsewhere. Along with optimised diabetes care, this has substantially reduced the risk of visual loss. Nevertheless, the growing number of patients with diabetes poses an increasing burden on healthcare systems and automated solutions are needed to alleviate the task of screening and improve diagnostic accuracy. Deep learning by convolutional neural networks is an optimised branch of artificial intelligence that is particularly well suited to automated image analysis. Pivotal studies have demonstrated high sensitivity and specificity for classifying advanced stages of diabetic retinopathy and identifying diabetic macular oedema in optical coherence tomography scans. Based on this, different algorithms have obtained regulatory approval for clinical use and have recently been implemented to some extent in a few countries. Handheld mobile devices are another promising option for self-monitoring, but so far they have not demonstrated comparable image quality to that of fundus photography using non-portable retinal cameras, which is the gold standard for diabetic retinopathy screening. Such technology has the potential to be integrated in telemedicine-based screening programmes, enabling self-captured retinal images to be transferred virtually to reading centres for analysis and planning of further steps. While emerging technologies have shown a lot of promise, clinical implementation has been sparse. Legal obstacles and difficulties in software integration may partly explain this, but it may also indicate that existing algorithms may not necessarily integrate well with national screening initiatives, which often differ substantially between countries.

Application of artificial intelligence-based dual-modality analysis combining fundus photography and optical coherence tomography in diabetic retinopathy screening in a community hospital

Background

To assess the feasibility and clinical utility of artificial intelligence (AI)-based screening for diabetic retinopathy (DR) and macular edema (ME) by combining fundus photos and optical coherence tomography (OCT) images in a community hospital.

Methods

Fundus photos and OCT images were taken for 600 diabetic patients in a community hospital. Ophthalmologists graded these fundus photos according to the International Clinical Diabetic Retinopathy (ICDR) Severity Scale as the ground truth. Two existing trained AI models were used to automatically classify the fundus images into DR grades according to ICDR, and to detect concomitant ME from OCT images, respectively. The criteria for referral were DR grades 2–4 and/or the presence of ME. The sensitivity and specificity of AI grading were evaluated. The number of referable DR cases confirmed by ophthalmologists and AI was calculated, respectively.

Results

DR was detected in 81 (13.5%) participants by ophthalmologists and in 94 (15.6%) by AI, and 45 (7.5%) and 53 (8.8%) participants were diagnosed with referable DR by ophthalmologists and by AI, respectively. The sensitivity, specificity and area under the curve (AUC) of AI for detecting DR were 91.67%, 96.92% and 0.944, respectively. For detecting referable DR, the sensitivity, specificity and AUC of AI were 97.78%, 98.38% and 0.981, respectively. ME was detected from OCT images in 49 (8.2%) participants by ophthalmologists and in 57 (9.5%) by AI, and the sensitivity, specificity and AUC of AI were 91.30%, 97.46% and 0.944, respectively. When combining fundus photos and OCT images, the number of referrals identified by ophthalmologists increased from 45 to 75 and from 53 to 85 by AI.

Conclusion

AI-based DR screening has high sensitivity and specificity and may feasibly improve the referral rate of community DR.

Burden of Diabetic Retinopathy amongst People with Diabetes Attending Primary Care in Kerala: Nayanamritham Project

Background: The burden of diabetic retinopathy (DR) in people attending the public health sector in India is unclear. Thirty percent of the population in India is reliant on public healthcare. This study aimed to estimate the prevalence of DR and its risk factors in people with diabetes in the non-communicable disease registers who were attending the family health centres (FHCs) in the Thiruvananthapuram district in Kerala. Methods: This cross-sectional study was conducted over 12 months in 2019 within the framework of a pilot district-wide teleophthalmology DR screening programme. The age- and gender-adjusted prevalence of any DR and sight-threatening DR (STDR) in the whole sample, considering socio-demography, lifestyle and known clinical risk groups, are reported. Results: A total of 4527 out of 5307 (85.3%) screened in the FHCs had gradable retinal images in at least one eye. The age and gender standardised prevalence for any DR was 17.4% (95% CI 15.1, 19.7), and STDR was 3.3% (95% CI 2.1, 4.5). Ages 41–70 years, males, longer diabetes duration, hyperglycaemia and hypertension, insulin users and lower socio-economic status were associated with both DR outcomes. Conclusions: The burden of DR and its risk factors in this study highlights the need to implement DR screening programs within primary care to reduce health inequality.

Role of Oral Antioxidant Supplementation in the Current Management of Diabetic Retinopathy

Oxidative stress has been postulated as an underlying pathophysiologic mechanism of diabetic retinopathy (DR), the main cause of avoidable blindness in working-aged people. This review addressed the current daily clinical practice of DR and the role of antioxidants in this practice. A systematic review of the studies on antioxidant supplementation in DR patients was presented. Fifteen studies accomplished the inclusion criteria. The analysis of these studies concluded that antioxidant supplementation has a IIB level of recommendation in adult Type 1 and Type 2 diabetes mellitus subjects without retinopathy or mild-to-moderate nonproliferative DR without diabetic macular oedema as a complementary therapy together with standard medical care.

Fundamental principles of an effective diabetic retinopathy screening program

BACKGROUND

Diabetic retinopathy (DR) is the leading cause of blindness among working-age adults worldwide. Early detection and treatment are necessary to forestall vision loss from DR.

METHODS

A working group of ophthalmic and diabetes experts was established to develop a consensus on the key principles of an effective DR screening program. Recommendations are based on analysis of a structured literature review.

RESULTS

The recommendations for implementing an effective DR screening program are: (1) Examination methods must be suitable for the screening region, and DR classification/grading systems must be systematic and uniformly applied. Two-field retinal imaging is sufficient for DR screening and is preferable to seven-field imaging, and referable DR should be well defined and reliably identifiable by qualified screening staff; (2) in many countries/regions, screening can and should take place outside the ophthalmology clinic; (3) screening staff should be accredited and show evidence of ongoing training; (4) screening programs should adhere to relevant national quality assurance standards; (5) studies that use uniform definitions of risk to determine optimum risk-based screening intervals are required; (6) technology infrastructure should be in place to ensure that high-quality images can be stored securely to protect patient information; (7) although screening for diabetic macular edema (DME) in conjunction with DR evaluations may have merit, there is currently insufficient evidence to support implementation of programs solely for DME screening.

CONCLUSION

Use of these recommendations may yield more effective DR screening programs that reduce the risk of vision loss worldwide.

Health Economic and Safety Considerations for Artificial Intelligence Applications in Diabetic Retinopathy Screening

Systematic screening for diabetic retinopathy (DR) has been widely recommended for early detection in patients with diabetes to address preventable vision loss. However, substantial manpower and financial resources are required to deploy opportunistic screening and transition to systematic DR screening programs. The advent of artificial intelligence (AI) technologies may improve access and reduce the financial burden for DR screening while maintaining comparable or enhanced clinical effectiveness. To deploy an AI-based DR screening program in a real-world setting, it is imperative that health economic assessment (HEA) and patient safety analyses are conducted to guide appropriate allocation of resources and design safe, reliable systems. Few studies published to date include these considerations when integrating AI-based solutions into DR screening programs. In this article, we provide an overview of the current state-of-the-art of AI technology (focusing on deep learning systems), followed by an appraisal of existing literature on the applications of AI in ophthalmology. We also discuss practical considerations that drive the development of a successful DR screening program, such as the implications of false-positive or false-negative results and image gradeability. Finally, we examine different plausible methods for HEA and safety analyses that can be used to assess concerns regarding AI-based screening.

Predicting optical coherence tomography-derived diabetic macular edema grades from fundus photographs using deep learning

Center-involved diabetic macular edema (ci-DME) is a major cause of vision loss. Although the gold standard for diagnosis involves 3D imaging, 2D imaging by fundus photography is usually used in screening settings, resulting in high false-positive and false-negative calls. To address this, we train a deep learning model to predict ci-DME from fundus photographs, with an ROC-AUC of 0.89 (95% CI: 0.87-0.91), corresponding to 85% sensitivity at 80% specificity. In comparison, retinal specialists have similar sensitivities (82-85%), but only half the specificity (45-50%, p < 0.001). Our model can also detect the presence of intraretinal fluid (AUC: 0.81; 95% CI: 0.81-0.86) and subretinal fluid (AUC 0.88; 95% CI: 0.85-0.91). Using deep learning to make predictions via simple 2D images without sophisticated 3D-imaging equipment and with better than specialist performance, has broad relevance to many other applications in medical imaging.

Use of personalised risk-based screening schedules to optimise workload and sojourn time in screening programmes for diabetic retinopathy: A retrospective cohort study

BACKGROUND

National guidelines in most countries set screening intervals for diabetic retinopathy (DR) that are insufficiently informed by contemporary incidence rates. This has unspecified implications for interval disease risks (IDs) of referable DR, disparities in ID between groups or individuals, time spent in referable state before screening (sojourn time), and workload. We explored the effect of various screening schedules on these outcomes and developed an open-access interactive policy tool informed by contemporary DR incidence rates.

METHODS AND FINDINGS

Scottish Diabetic Retinopathy Screening Programme data from 1 January 2007 to 31 December 2016 were linked to diabetes registry data. This yielded 128,606 screening examinations in people with type 1 diabetes (T1D) and 1,384,360 examinations in people with type 2 diabetes (T2D). Among those with T1D, 47% of those without and 44% of those with referable DR were female, mean diabetes duration was 21 and 23 years, respectively, and mean age was 26 and 24 years, respectively. Among those with T2D, 44% of those without and 42% of those with referable DR were female, mean diabetes duration was 9 and 14 years, respectively, and mean age was 58 and 52 years, respectively. Individual probability of developing referable DR was estimated using a generalised linear model and was used to calculate the intervals needed to achieve various IDs across prior grade strata, or at the individual level, and the resultant workload and sojourn time. The current policy in Scotland-screening people with no or mild disease annually and moderate disease every 6 months-yielded large differences in ID by prior grade (13.2%, 3.6%, and 0.6% annually for moderate, mild, and no prior DR strata, respectively, in T1D) and diabetes type (2.4% in T1D and 0.6% in T2D overall). Maintaining these overall risks but equalising risk across prior grade strata would require extremely short intervals in those with moderate DR (1-2 months) and very long intervals in those with no prior DR (35-47 months), with little change in workload or average sojourn time. Changing to intervals of 12, 9, and 3 months in T1D and to 24, 9, and 3 months in T2D for no, mild, and moderate DR strata, respectively, would substantially reduce disparity in ID across strata and between diabetes types whilst reducing workload by 26% and increasing sojourn time by 2.3 months. Including clinical risk factor data gave a small but significant increment in prediction of referable DR beyond grade (increase in C-statistic of 0.013 in T1D and 0.016 in T2D, both p < 0.001). However, using this model to derive personalised intervals did not have substantial workload or sojourn time benefits over stratum-specific intervals. The main limitation is that the results are pertinent only to countries that share broadly similar rates of retinal disease and risk factor distributions to Scotland.

CONCLUSIONS

Changing current policies could reduce disparities in ID and achieve substantial reductions in workload within the range of IDs likely to be deemed acceptable. Our tool should facilitate more rational policy setting for screening.

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.

Retinal Telemedicine

PURPOSE OF REVIEW

An update and overview of the literature on current telemedicine applications in retina.

RECENT FINDINGS

The application of telemedicine to the field of Ophthalmology and Retina has been growing with advancing technologies in ophthalmic imaging. Retinal telemedicine has been most commonly applied to diabetic retinopathy and retinopathy of prematurity in adult and pediatric patients respectively. Telemedicine has the potential to alleviate the growing demand for clinical evaluation of retinal diseases. Subsequently, automated image analysis and deep learning systems may facilitate efficient processing of large, increasing numbers of images generated in telemedicine systems. Telemedicine may additionally improve access to education and standardized training through tele-education systems.

SUMMARY

Telemedicine has the potential to be utilized as a useful adjunct but not a complete replacement for physical clinical examinations. Retinal telemedicine programs should be carefully and appropriately integrated into current clinical systems.