Diabetic retinopathy is a complication of diabetes that affects the eyes. In the initial stages, it is generally asymptomatic. However, diabetic retinopathy is the leading cause of new blindness in persons aged 25-74 years in the United States. The exact mechanism by which diabetes originates retinopathy remains unclear, but experts suggest that it may be caused by high blood sugar levels damaging the blood vessels of the light-sensitive tissue at the back of the eye (retina).
On early stages of retinopathy, the damage is limited to tiny bulges (microaneurysms) in the blood vessel walls. Although these can leak blood and fluid, they do not usually affect the vision. Visual loss due to diabetes can be prevented by retinal laser treatment if retinopathy is spotted early. In patients with type 2 diabetes, retinopathy may be the first sign of diabetes and vision-threatening retinopathy may already have developed by the time of diagnosis. To protect the vision, one has to take prevention seriously by carefully controlling the blood sugar level and performing regular eye exams.
The goal of this thesis is to create a mobile-based solution that can provide an effective pre-diagnosis of Diabetic Retinopathy to be used in self-monitoring end-users. On the latest generation of smartphones, which exhibits significant improvements in terms of image acquisition, portable adapters are mounted to built-in cameras to provide optical magnification required to reach the interior surface of the fundus of the eye (including retina, optic disc and macula). Automated image processing for the detection of microaneurysms on early stages of diabetic retinopathy is the first target given its great importance, in the context of end-user self-exam. These users require a continuous monitoring solution with computer-aided alignment of the eye fundus on imaging exams. The self-exam is meant to be fast and simple, requiring the adapter to barely touch the face and without any discomfort, and also helping the user on the crucial alignment process of the eye in fundography.
> Smartphone application for computer-aided alignment of the eye fundus on imaging exams;
> Smartphone application for image recognition and automatic annotation of microaneurysms on eye fundus, for risk assessment of Diabetic Retinopathy.
For any additional information regarding this project, please contact us using the inquiries form.