ABSTRACT
Purpose: To examine the current provision and practice patterns of diabetic retinopathy screening (DRS) in Haryana. Methods: This was a descriptive cross?sectional survey in Haryana. All ophthalmologists registered with Haryana Ophthalmological Society in Haryana state were invited to participate on an online survey comprised of twenty questions exploring diabetic retinopathy screening provision, and barriers to screening services in Haryana. Results: The response rate was 82% (153/186). The majority (84%) of the eye care providers practiced in urban areas. Most ophthalmologists (89%, 136/153), considered diabetic retinopathy screening by non?ophthalmic human resource inappropriate because of technical feasibility issues (62%) followed by non?availability of trained staff (33%). Only half (54%) of the respondents had access to written protocols for the diagnosis and management of diabetic retinopathy in their practice. Barriers to optimize diabetic retinopathy screening were lack of knowledge or awareness among patients (95.5%), perception that eye complications were unlikely (76%) and cost of care (30%). Conclusion: Diabetic retinopathy screening practices are mainly opportunistic and urban?centric, likely delivering inequitable services for the rural populations in the state. The inclusion of other personnel in screening will require stakeholder engagement from all health professions and changing the perceptions of ophthalmologists about task shifting.
ABSTRACT
Purpose: Deep learning is a newer and advanced subfield in artificial intelligence (AI). The aim of our study is to validate a machine-based algorithm developed based on deep convolutional neural networks as a tool for screening to detect referable diabetic retinopathy (DR). Methods: An AI algorithm to detect DR was validated at our hospital using an internal dataset consisting of 1,533 macula-centered fundus images collected retrospectively and an external validation set using Methods to Evaluate Segmentation and Indexing Techniques in the field of Retinal Ophthalmology (MESSIDOR) dataset. Images were graded by two retina specialists as any DR, prompt referral (moderate nonproliferative diabetic retinopathy (NPDR) or above or presence of macular edema) and sight-threatening DR/STDR (severe NPDR or above) and compared with AI results. Sensitivity, specificity, and area under curve (AUC) for both internal and external validation sets for any DR detection, prompt referral, and STDR were calculated. Interobserver agreement using kappa value was calculated for both the sets and two out of three agreements for DR grading was considered as ground truth to compare with AI results. Results: In the internal validation set, the overall sensitivity and specificity was 99.7% and 98.5% for Any DR detection and 98.9% and 94.84%for Prompt referral respectively. The AUC was 0.991 and 0.969 for any DR detection and prompt referral respectively. The agreement between two observers was 99.5% and 99.2% for any DR detection and prompt referral with a kappa value of 0.94 and 0.96, respectively. In the external validation set (MESSIDOR 1), the overall sensitivity and specificity was 90.4% and 91.0% for any DR detection and 94.7% and 97.4% for prompt referral, respectively. The AUC was. 907 and. 960 for any DR detection and prompt referral, respectively. The agreement between two observers was 98.5% and 97.8% for any DR detection and prompt referral with a kappa value of 0.971 and 0.980, respectively. Conclusion: With increasing diabetic population and growing demand supply gap in trained resources, AI is the future for early identification of DR and reducing blindness. This can revolutionize telescreening in ophthalmology, especially where people do not have access to specialized health care.
ABSTRACT
PURPOSE: To identify factors associated with care of diabetic retinopathy. METHODS: The fourth Korea National Health and Nutrition Examination Survey (KNHANES IV) is a nationwide survey. This survey included 1, 257 people aged 40 years and older with a history of diabetic mellitus who answered questions, "Within one year, have you ever received eye examination (fundus photography) for screening diabetic retinopathy?" Factors that affect care of diabetic retinopathy were identified using multiple logistic regression analysis. RESULTS: Among the 1,257 people aged 40 years and older, 464 (36.9%) received screening for diabetic retinopathy. People aged 65 years and older (aOR = 0.7, 95% CI: 0.51-0.85) with university education (aOR = 0.5, 95% CI: 0.32-0.74) were more likely to undergo screening for diabetic retinopathy compared to those in the reference category (40-64 years old and those who had elementary school or lower education). People living in rural areas were less likely to undergo screening for diabetic retinopathy compared to those living in urban areas (aOR = 1.7, 95% CI: 1.32-2.24). Diabetic retinopathy screening was also associated with self-reported health status (ref: unhealthy [aOR = 1], fair [aOR = 1.7, 95% CI: 1.25-2.23], and healthy [aOR = 1.8, 95% CI: 1.30-2.44]). CONCLUSIONS: To increase nationwide screening rates for diabetic retinopathy, more attention should be given to underserved groups, particularly people aged between 40 and 64 years, those with a low education level, those living in rural areas, and those with a positive attitude toward self-reported health status. These issues highlight the need for a new emphasis in health education and public health policies aimed towards these underserved groups.