Images of eye's retina can be used for the diagnosis of several diseases. A prominent example is diabetic retinopathy, which is a leading cause of vision loss in many developed countries. Early screening of these diseases can simplify treatment and prevent blindness. Currently this type of screening is done at hospitals or other specialized institutions which use expensive and stationary fundus cameras to take a high definition retina images. This procedure usually requires the use of eye drops to dilate the pupil.
In a countries where this type of healthcare services are not available for the majority of the population, the screening is simply not possible. Thus a compact, handheld, low-cost and easy-to-use device which enables early screening, is essential for widespread early diagnosis.
The project aims to develop a handheld and easy-to-use retina camera with fully automated image capture and diagnostics of diabetic retinopathy through use of machine learning. It will require minimal training of the primary care physician or doctor's assistant. No dilating eye drops are required and the retinal scan will be performed in less than 30 seconds. This will enable early screening of diabetic retinopathy for millions of people worldwide. Project owner, Oivi AS, targets India market at initial stage, which is a huge market with 73 millions diabetics. SINTEF contributes to the R&D of novel optical design for the camera and methods for the automated image capture.
