A screening programme for diabetic retinopathy

Deep Learning for Diabetic Retinopathy Analysis: A Review, Research Challenges, and Future Directions

Deep learning (DL) enables the creation of computational models comprising multiple processing layers that learn data representations at multiple levels of abstraction. In the recent past, the use of deep learning has been proliferating, yielding promising results in applications across a growing number of fields, most notably in image processing, medical image analysis, data analysis, and bioinformatics. DL algorithms have also had a significant positive impact through yielding improvements in screening, recognition, segmentation, prediction, and classification applications across different domains of healthcare, such as those concerning the abdomen, cardiac, pathology, and retina. Given the extensive body of recent scientific contributions in this discipline, a comprehensive review of deep learning developments in the domain of diabetic retinopathy (DR) analysis, viz., screening, segmentation, prediction, classification, and validation, is presented here. A critical analysis of the relevant reported techniques is carried out, and the associated advantages and limitations highlighted, culminating in the identification of research gaps and future challenges that help to inform the research community to develop more efficient, robust, and accurate DL models for the various challenges in the monitoring and diagnosis of DR.

Reduced forearm reactive hyperaemia in normoalbuminuric subjects with Type 1 diabetes and retinopathy

AIM

To determine whether the forearm vasodilatory response to reactive hyperaemia (RH) is reduced in normoalbuminuric subjects with Type 1 diabetes mellitus and retinopathy compared with subjects with no retinopathy.

METHODS

Forearm RH, an indicator of endothelial function, was measured, using strain-gauge plethysmography, in 39 normoalbuminuric subjects (22 with retinopathy) with long-standing Type 1 diabetes mellitus.

RESULTS

were evaluated in relation to conventional risk factors for atherosclerosis, and C-reactive protein (CRP), which we have recently determined to be an independent correlate of forearm RH.

RESULTS

Forearm RH was decreased in subjects with retinopathy compared with those with no retinopathy (219 +/- 182 vs. 473 +/- 355, P < 0.01). Both retinopathy and CRP proved to be independent and negative predictors, and explain 27% of the variance, in forearm RH.

CONCLUSION

Retinopathy in subjects with Type 1 diabetes mellitus may reflect a generalized process of endothelial dysfunction, even in the absence of microalbuminuria.

The quality of photographs produced by the non-mydriatic fundus camera in a screening programme for diabetic retinopathy: a 1 year prospective study

A prospective study was performed on the quality of photographs produced by the non-mydriatic fundus camera used in a hospital-based screening programme for diabetic retinopathy. In 1 year 981 binocular patients were photographed. A photograph of acceptable quality was obtained from 90.5% of eyes and 84.4% of patients had an acceptable photograph of both eyes. The photograph of the second eye was more often unacceptable than that of the first. This tendency was significant in females (p = 0.0196) and when considering the sexes together (p = 0.0044), but not significant in males (p = 0.1042). Photographs of unacceptable quality were obtained significantly more often in patients aged over 55 years for both right and left eyes (p = 0.001). An overall improvement in photographic quality might be achieved by allowing full recovery of flash-induced pupil constriction before taking the second photograph and by dilating those aged over 55 years.