[The non-mydriatic camera as a screening method in diabetics in Tarragona]

Knowledge of Diabetic Retinopathy among Primary Care Nurses Performing Fundus Photography and Agreement with Ophthalmologists on Screening

Diabetic retinopathy (DR) is one of the complications of diabetes mellitus (DM), with macular oedema being one of the leading causes of avoidable blindness among individuals with DM worldwide. Fundus screening is the only method for early detection and treatment. High-quality training programmes for professionals performing primary care screening are essential to produce high-quality images that facilitate accurate lesion identification. This is a two-phase observational, descriptive, and cross-sectional study. The first phase analysed DR knowledge in a sample of nurses. The second phase explored agreement on DR screening between referral ophthalmologists in image assessment (gold standard) and a small group of nurses involved in the previous phase. In phase 1, the agreement rate for screening results was 90%. In phase 2, the overall raw agreement on the screening of fundus photography results between nurses and ophthalmologists was 75% (Cohen's kappa = 0.477; p < 0.001). Agreement on screening with ophthalmologists was moderate, suggesting that implementing a specific training programme for nurse-led imaging screening would help develop this competence among nurses, ensuring a good level of agreement and patient safety and adding value for users, and also for the sustainability of the healthcare system. This study was not registered.

Validation of an autonomous artificial intelligence-based diagnostic system for holistic maculopathy screening in a routine occupational health checkup context

PURPOSE

This study aims to evaluate the ability of an autonomous artificial intelligence (AI) system for detection of the most common central retinal pathologies in fundus photography.

METHODS

Retrospective diagnostic test evaluation on a raw dataset of 5918 images (2839 individuals) evaluated with non-mydriatic cameras during routine occupational health checkups. Three camera models were employed: Optomed Aurora (field of view - FOV 50º, 88% of the dataset), ZEISS VISUSCOUT 100 (FOV 40º, 9%), and Optomed SmartScope M5 (FOV 40º, 3%). Image acquisition took 2 min per patient. Ground truth for each image of the dataset was determined by 2 masked retina specialists, and disagreements were resolved by a 3rd retina specialist. The specific pathologies considered for evaluation were "diabetic retinopathy" (DR), "Age-related macular degeneration" (AMD), "glaucomatous optic neuropathy" (GON), and "Nevus." Images with maculopathy signs that did not match the described taxonomy were classified as "Other."

RESULTS

The combination of algorithms to detect any abnormalities had an area under the curve (AUC) of 0.963 with a sensitivity of 92.9% and a specificity of 86.8%. The algorithms individually obtained are as follows: AMD AUC 0.980 (sensitivity 93.8%; specificity 95.7%), DR AUC 0.950 (sensitivity 81.1%; specificity 94.8%), GON AUC 0.889 (sensitivity 53.6% specificity 95.7%), Nevus AUC 0.931 (sensitivity 86.7%; specificity 90.7%).

CONCLUSION

Our holistic AI approach reaches high diagnostic accuracy at simultaneous detection of DR, AMD, and Nevus. The integration of pathology-specific algorithms permits higher sensitivities with minimal impact on its specificity. It also reduces the risk of missing incidental findings. Deep learning may facilitate wider screenings of eye diseases.

Diabetic retinopathy screening programme utilising non-mydriatic fundus imaging in slum populations of New Delhi, India

OBJECTIVES

To develop and implement a community-based programme for screening of diabetic retinopathy (DR) in urban populations of Delhi.

METHODS

Known diabetics (KDs) aged 40 years and older were identified through house-to-house surveys, volunteers and publicity. All KDs were referred to DR screening camps organised locally where procedures included brief medical history, ocular examination and non-mydriatic fundus photography using portable handheld camera. Fundal images were graded on the spot by trained optometrists for DR. Patients with DR were referred to tertiary centre for management.

RESULTS

A total of 11 566 KDs were identified, of whom 9435 (81.6%) visited DR screening camps and 8432 (89.4%) had DR gradable images. DR was identified in 13.5% of subjects; 351 cases were mild NPDR, 567 moderate, 92 severe. Seventy-seven had PDR, and 49 had DME, and 2.7% of participants were blind (presenting visual acuity <3/60 in better eye). Non-use of lifestyle management, presence of systemic complications, BMI <18.5 kg/m² , disease duration of >5 years and uncontrolled diabetes were associated with increased odds of DR. All cases with DR were referred, and 420 (37%) successful referrals to base hospital were observed.

CONCLUSION

The programme of creating awareness about DR, identifying KDs and optometrist-led DR screening using non-mydriatic fundus camera based in slums was successful.

Five-year experience of tele-ophthalmology for diabetic retinopathy screening in a rural population

OBJECTIVE

To identify the prevalence and risk factors of diabetic retinopathy (DR) among rural inhabitants included in a tele-ophthalmology program. To analyse diagnostic accuracy among primary care physicians, concordance with ophthalmologists, and financial savings.

MATERIAL AND METHODS

An observational randomised study was conducted on 394 patients included in a tele-ophthalmology program (from January 2010 to January 2015). An analysis was performed on the clinical characteristics, DR findings in retinography images, and correspondence between the request for a second interpretation by an ophthalmologist, and previously established criteria for it: presence of moderate to severe DR, vision loss, poor image quality and/or intraocular pressure >22mmHg. Statistical analysis was performed using the SPSS program (Student t and χ(2) tests).

RESULTS

DR prevalence was 12.1%. Patients with glycosylated haemoglobin values >7.68% or those treated with a combination of insulin and oral antidiabetic drugs showed a higher risk of DR (P<.05). 43.3% of patients correctly referred to ophthalmologists showed moderate to severe DR. Unnecessary referrals to specialists were improved from 91.7% in 2010 to 98.6% in 2014. It is estimated that the program has made a total saving of €152,550.45.

CONCLUSIONS

Tele-ophthalmology programs are a useful tool in DR screening. Primary care physicians are able to distinguish patients who need specialist care, avoiding unnecessary referrals to ophthalmologists, and saving costs.