Project leaderPere Romero Aroca
Participating organizationsPere Virgili Institute (IISPV), Rovira i Virgili University (URV)
To develop a support system for clinical diagnosis based on using artificial intelligence and computer-vision techniques to classify patients by their risk of developing diabetic retinopathy (DR), one of the most common comorbidities and the main cause of blindness.
RETIPROGRAM is led by the Ophthalmology Research Group-HUSJR-IISPV (Pere Virgili Institute) headed up by Dr. Pere Romero, with collaboration from the ITAKA Research Group (Intelligent Technologies for Advanced Knowledge Acquisition) headed up by Dr. Antoni Moreno and Dr. Aïda Valls, and the IRCV Research Group (Intelligent Robotics and Computer Vision) headed up by Dr. Domènec Puig, both at Rovira i Virgili University. The project came out of two projects funded by the Institute of Health Carlos III.
Most significant challenges
The initial challenge of RETIPROGRAM was to find funding to subcontract the IRCV and ITAKA research groups for computer development, as well as seeking collaborations with industry. In terms of the transfer of the project to clinical practice, the challenges were to boost the cost/efficacy of the current DR screening system, improve care of patients at high risk of developing the disease and to decrease the number of unnecessary referrals through early detection, lowering the social and healthcare burden.
Screening with non-mydriatic retinal cameras in primary care centers has helped improve detection of DR. According to the Catalan Health Service (ICS), eye exams should be given to people with type-2 diabetes when they are diagnosed and at least once a year after that. The RETIPROGRAM study on 28,000 diabetics followed over 8 years has proven more effective with one screening every 2.5 years. Nevertheless, there's no comparison for its effectiveness if it can be applied periodically according to the patient's personalized needs.
The project expects to apply the new system by making it compatible with computer clinical history systems in the Camp de Tarragona Healthcare Region. Plus, and automatic image-reading system will be incorporated to classify DR by its seriousness. The software has been registered and the project has been sent to the Innòbics platform for assessment.