Image-based risk assessment for diabetic retinopathy

The challenge

Diabetic retinopathy (DR) is a microvascular complication of type 1 and type 2 diabetes that affects the small vessels of the retina.

DR can lead to complete vision loss if not detected early and is the major cause of preventable blindness in working-age adults.[1] When detected on time, the risk of DR can be reduced by 95%[2] but the disease is asymptomatic in its early stages and the risk of developing DR varies among patients.

Globally, DR is the fifth cause of blindness and visual impairment affecting between 3-4% of people in Europe.[3] DR affects 30% of people with diabetes[4] and it is estimated that 280 million people have lost their sight due to diabetes worldwide.[5]

DR can be prevented by controlling risk factors, such as glycemia or arterial hypertension. But existing healthcare systems are inefficient in detecting DR for several reasons. The volume of patients with diabetes means the cameras needed for screening are in short supply.

Most patients are screened at ophthalmology units, despite the option of telemedicine using a special type of camera. The fact that 29% of Europeans[6] live outside urban areas, where ophthalmologist units are located, also presents a challenge.

The solution

The RetinaReadRisk team have come up with a new platform for the early detection and diagnosis of DR in rural areas. The solution combines artificial intelligence (AI) models and 5G technologies, allowing healthcare professionals to reach people in urban and rural areas. It would also promote the efficiency of operations throughout the care cycle, empowering better patient outcomes.

RetinaReadRisk would enable the capture of images at medical offices closest to the population by smart phone. 5G technologies and cloud computing would assure real-time diagnosis.

The team plan to use deep learning models to automatically analyse DR images. They will enhance existing diagnostic systems, using AI techniques to model electronic health record data. This will help to predict the risk of a person developing DR.

Expected impact

The RetinaReadRisk platform promises to improve the efficiency of DR screening, boost the productivity of healthcare systems, and increase patient satisfaction. It mitigates the limitations of DR screening considering both economic costs and patient outcomes.

The project will facilitate the collection of retina photography by primary care professionals, especially for people living in rural areas, far from main hospitals that own image scanners. It will help optimise the entire diagnostic stage of the care pathway for patients at risk of DR.

By providing personalised screening, RetinaReadRisk will also improve patient outcomes and satisfaction. In addition, it will improve interaction between main hospitals and primary care centres. This will help them to provide better and more timely support, decreasing the burden on families.

Ultimately, RetinaReadRisk promises to help reduce the number of patients reaching the later stages of DR and decrease treatment costs.

References

[1] Wild, S. (2004). Global prevalence of diabetes. Estimates for the year 2000 and projections for 2030. Diabetes Care, 27(5), 1047-53.

[2] Sabanayagam, C. (2019). Incidence and progression of diabetic retinopathy: A systematic review”. Lancet Diabetes Endocrinology, 7(2), 140–149.

[3] Ogurtsova, K. (2017). IDF Diabetes Atlas: Global estimates for the prevalence of diabetes for 2015 and 2040. Diabetes Research and Clinical Practice, 128, 40-50.

[4] Yau, J.W. (2012). Global prevalence and major risk factors of diabetic retinopathy. Diabetes Care, 35 (3), 556– 564.

[5] Shaw, J.E. (2010). Global estimates of the prevalence of diabetes for 2010 and 2030. Diabetes Research and Clinical Practice, 87(1), 4–14.

[6] Costello. (2018). Archive: Statistics on rural areas in the EU. Eurostat: statistics explained. [online] Available at:  https://ec.europa.eu/eurostat/statistics-explained/index.php?title=Statistics_on_rural_areas_in_the_EU&oldid=391832 (Accessed: 25 March 2022)

Alba Marti Roig
| Head of the International Projects Unit | Fundació Institut d’Investigació Sanitària Pere Virgili (IISPV)
Contact
Xavier Cos Claramunt MD, PhD
| Innovation & Research Support Office | Institut Català de la Salut (ICS)
Contact
Prof. Pere Romero-Aroca, PhD
| Principal Investigator Ophthalmic Research Group -Retiprogram-, Professor of Ophthalmology Universitat Rovira & Virgili, Director of Ophthalmic Department of Ophthalmology of Hospital Universitari Sant Joan de Reus | Fundació Institut d’Investigació Sanitària Pere Virgili (IISPV)
Contact