Diabetic retinopathy: the unmet needs for screening and a review of potential solutions

Testing the performance of risk prediction models to determine progression to referable diabetic retinopathy in an Irish type 2 diabetes cohort

Background /Aims

To evaluate the performance of existing prediction models to determine risk of progression to referable diabetic retinopathy (RDR) using data from a prospective Irish cohort of people with type 2 diabetes (T2D).

Methods

A cohort of 939 people with T2D followed prospectively was used to test the performance of risk prediction models developed in Gloucester, UK, and Iceland. Observed risk of progression to RDR in the Irish cohort was compared with that derived from each of the prediction models evaluated. Receiver operating characteristic curves assessed models’ performance.

Results

The cohort was followed for a total of 2929 person years during which 2906 screening episodes occurred. Among 939 individuals followed, there were 40 referrals (4%) for diabetic maculopathy, pre-proliferative DR and proliferative DR. The original Gloucester model, which includes results of two consecutive retinal screenings; a model incorporating, in addition, systemic biomarkers (HbA1c and serum cholesterol); and a model including results of one retinopathy screening, HbA1c, total cholesterol and duration of diabetes, had acceptable discriminatory power (area under the curve (AUC) of 0.69, 0.76 and 0.77, respectively). The Icelandic model, which combined retinopathy grading, duration and type of diabetes, HbA1c and systolic blood pressure, performed very similarly (AUC of 0.74).

Conclusion

In an Irish cohort of people with T2D, the prediction models tested had an acceptable performance identifying those at risk of progression to RDR. These risk models would be useful in establishing more personalised screening intervals for people with T2D.

Artificial intelligence-based screening for diabetic retinopathy at community hospital

OBJECTIVES

The purpose of this study is to assess the accuracy of artificial intelligence (AI)-based screening for diabetic retinopathy (DR) and to explore the feasibility of applying AI-based technique to community hospital for DR screening.

METHODS

Nonmydriatic fundus photos were taken for 889 diabetic patients who were screened in community hospital clinic. According to DR international classification standards, ophthalmologists and AI identified and classified these fundus photos. The sensitivity and specificity of AI automatic grading were evaluated according to ophthalmologists' grading.

RESULTS

DR was detected by ophthalmologists in 143 (16.1%) participants and by AI in 145 (16.3%) participants. Among them, there were 101 (11.4%) participants diagnosed with referable diabetic retinopathy (RDR) by ophthalmologists and 103 (11.6%) by AI. The sensitivity, specificity and area under the curve (AUC) of AI for detecting DR were 90.79% (95% CI 86.4-94.1), 98.5% (95% CI 97.8-99.0) and 0.946 (95% CI 0.935-0.956), respectively. For detecting RDR, the sensitivity, specificity and AUC of AI were 91.18% (95% CI 86.4-94.7), 98.79% (95% CI 98.1-99.3) and 0.950 (95% CI 0.939-0.960), respectively.

CONCLUSION

AI has high sensitivity and specificity in detecting DR and RDR, so it is feasible to carry out AI-based DR screening in outpatient clinic of community hospital.

Diabetic Retinopathy-An Underdiagnosed and Undertreated Inflammatory, Neuro-Vascular Complication of Diabetes

Diabetes mellitus is a world-wide epidemic and diabetic retinopathy, a devastating, vision-threatening condition, is one of the most common diabetes-specific complications. Diabetic retinopathy is now recognized to be an inflammatory, neuro-vascular complication with neuronal injury/dysfunction preceding clinical microvascular damage. Importantly, the same pathophysiologic mechanisms that damage the pancreatic β-cell (e.g., inflammation, epigenetic changes, insulin resistance, fuel excess, and abnormal metabolic environment), also lead to cell and tissue damage causing organ dysfunction, elevating the risk of all complications, including diabetic retinopathy. Viewing diabetic retinopathy within the context whereby diabetes and all its complications arise from common pathophysiologic factors allows for the consideration of a wider array of potential ocular as well as systemic treatments for this common and devastating complication. Moreover, it also raises the importance of the need for methods that will provide more timely detection and prediction of the course in order to address early damage to the neurovascular unit prior to the clinical observation of microangiopathy. Currently, treatment success is limited as it is often initiated far too late and after significant neurodegeneration has occurred. This forward-thinking approach of earlier detection and treatment with a wider array of possible therapies broadens the physician's armamentarium and increases the opportunity for prevention and early treatment of diabetic retinopathy with preservation of good vision, as well the prevention of similar destructive processes occurring among other organs.

Crowdsourcing as a novel technique for retinal fundus photography classification: analysis of images in the EPIC Norfolk cohort on behalf of the UK Biobank Eye and Vision Consortium

Aim

Crowdsourcing is the process of outsourcing numerous tasks to many untrained individuals. Our aim was to assess the performance and repeatability of crowdsourcing for the classification of retinal fundus photography.

Methods

One hundred retinal fundus photograph images with pre-determined disease criteria were selected by experts from a large cohort study. After reading brief instructions and an example classification, we requested that knowledge workers (KWs) from a crowdsourcing platform classified each image as normal or abnormal with grades of severity. Each image was classified 20 times by different KWs. Four study designs were examined to assess the effect of varying incentive and KW experience in classification accuracy. All study designs were conducted twice to examine repeatability. Performance was assessed by comparing the sensitivity, specificity and area under the receiver operating characteristic curve (AUC).

Results

Without restriction on eligible participants, two thousand classifications of 100 images were received in under 24 hours at minimal cost. In trial 1 all study designs had an AUC (95%CI) of 0.701(0.680–0.721) or greater for classification of normal/abnormal. In trial 1, the highest AUC (95%CI) for normal/abnormal classification was 0.757 (0.738–0.776) for KWs with moderate experience. Comparable results were observed in trial 2. In trial 1, between 64–86% of any abnormal image was correctly classified by over half of all KWs. In trial 2, this ranged between 74–97%. Sensitivity was ≥96% for normal versus severely abnormal detections across all trials. Sensitivity for normal versus mildly abnormal varied between 61–79% across trials.

Conclusions

With minimal training, crowdsourcing represents an accurate, rapid and cost-effective method of retinal image analysis which demonstrates good repeatability. Larger studies with more comprehensive participant training are needed to explore the utility of this compelling technique in large scale medical image analysis.

Screening and public health strategies for diabetic retinopathy in the Eastern Mediterranean region

Diabetic retinopathy (DR) is a complication of diabetes mellitus that can cause blindness. As the prevalence of diabetes increases globally and patients live longer, the cases of DR are increasing. To address the visual disabilities due to DR, screening of all diabetics is suggested for early detection. The rationale and principles of DR screening are discussed. Based on the available evidence, the magnitude of DR in countries in the Eastern Mediterranean region (EMR) is presented. Public health strategies to control visual disabilities due to DR are discussed. These include generating evidence for planning, implementing standard operating procedures, periodic DR screening, focusing on primary prevention of DR, strengthening DR management, health information management and retrieval systems for DR, rehabilitating DR visually disabled, using low-cost technologies, adopting a comprehensive approach by integrating DR care into the existing health systems, health promotion/counseling, and involving the community. Although adopting the public health approach for DR has been accepted as a priority by member countries of EMR, challenges in implementation remain. These include limitations in the public health approach for DR compared to that for cataract, few skilled workers, poor health systems and lack of motivation in affecting health-related lifestyle changes in diabetics.Visual disabilities due to DR are likely to increase in the coming years. An organized public health approach must be adopted and all stakeholders must work together to control severe visual disabilities due to DR.

Effect of smoking on circulating angiogenic factors in high risk pregnancies

Objective

Changes in maternal concentrations of the anti-angiogenic factors, soluble fms-like tyrosine kinase 1 (sFlt1) and soluble endoglin (sEng), and the pro-angiogenic placental growth factor (PlGF) precede the development of preeclampsia in healthy women. The risk of preeclampsia is reduced in women who smoke during pregnancy. The objective of this study was to investigate whether smoking affects concentrations of angiogenic factors (sFlt1, PlGF, and sEng) in women at high risk for developing preeclampsia.

Study Design

We performed a secondary analysis of serum samples from 993 high-risk women (chronic hypertension, diabetes, multifetal gestation, and previous preeclampsia) in a preeclampsia prevention trial. sFlt1, sEng and PlGF were measured in serum samples obtained at study entry, which was prior to initiation of aspirin (median 19.0 weeks' [interquartile range of 16.0–22.6 weeks']). Smoking status was determined by self-report.

Results

sFlt1 was not significantly different in smokers from any high-risk groups compared to their nonsmoking counterparts. PlGF was higher among smokers compared to nonsmokers among diabetic women (142.7 [77.4–337.3] vs 95.9 [48.5–180.7] pg/ml, p = 0.005) and women with a history of preeclampsia (252.2 [137.1–486.0] vs 152.2 [73.6–253.7] pg/ml, p = 0.001). sEng was lower in smokers with multifetal gestations (5.8 [4.6–6.5] vs 6.8 [5.5–8.7] ng/ml, p = 0.002) and trended lower among smokers with diabetes (4.9 [3.8–5.6] vs 5.3 [4.3–6.3] ng/ml, p = 0.05). Smoking was not associated with a lower incidence of preeclampsia in any of these groups.

Conclusions

In certain high-risk groups, smoking is associated with changes in the concentrations of these factors towards a pro-angiogenic direction during early pregnancy; however, there was no apparent association between smoking and the development of preeclampsia in our cohort.

Statistics of optical coherence tomography data from human retina

Optical coherence tomography (OCT) has recently become one of the primary methods for noninvasive probing of the human retina. The pseudoimage formed by OCT (the so-called B-scan) varies probabilistically across pixels due to complexities in the measurement technique. Hence, sensitive automatic procedures of diagnosis using OCT may exploit statistical analysis of the spatial distribution of reflectance. In this paper, we perform a statistical study of retinal OCT data. We find that the stretched exponential probability density function can model well the distribution of intensities in OCT pseudoimages. Moreover, we show a small, but significant correlation between neighbor pixels when measuring OCT intensities with pixels of about 5 microm. We then develop a simple joint probability model for the OCT data consistent with known retinal features. This model fits well the stretched exponential distribution of intensities and their spatial correlation. In normal retinas, fit parameters of this model are relatively constant along retinal layers, but varies across layers. However, in retinas with diabetic retinopathy, large spikes of parameter modulation interrupt the constancy within layers, exactly where pathologies are visible. We argue that these results give hope for improvement in statistical pathology-detection methods even when the disease is in its early stages.

Eye disease and the older diabetic

Elderly diabetic persons are 1.5 times more likely than age-matched nondiabetic persons to develop vision loss and blindness. Annually, between 12,000 and 24,000 diabetic patients in the United States become legally blind because of complications caused by diabetic retinopathy. Even more diabetic persons experience vision loss caused by comorbid ocular and periocular conditions such as dry eye syndrome, cataracts, macular degeneration, and glaucoma. This article discusses the synergy between these conditions and diabetes. Standards of care that slow the progression of vision loss and exciting new research on new strategies of care that may reverse vision loss are presented.