Digital Imaging System for the Detection of Diabetic Retinopathy

Perez-Siguas R.; Matta-Solis E.; Remuzgo-Artezano A.; Matta-Solis H.; Matta-Perez H.; Perez-Siguas L. · 2023 3rd International Conference on Advances in Electrical, Computing, Communication and Sustainable Technologies, ICAECT 2023 (2023) · DOI: 10.1109/icaect57570.2023.10117640

Resumen

COVID-19 since its appearance caused serious problems to the health sector due to the increase in infected and deceased people by directly affecting their respiratory system, making it a primordial disease that led all countries to fight this virus, generating that other diseases go to the background such as diabetes mellitus, which is a disease caused by the neglect of people's lifestyles, that has been increasing over time and that has no cure but can be prevented by controlling your blood glucose level, this disease causes diabetic retinopathy in people that with the advance of it can cause loss of sight. In addition, to detect its stage the ophthalmologist relies on his experience, occupying a lot of time and being prone to make mistakes about the patient. In view of this problem, in this article a digital image processing system was performed for the detection of diabetic retinopathy and classified according to the characteristics obtained from the features by analyzing the fundus of the eye automatically and determining the stage in which the patient is. Through the development of this system, it was determined that it works in the best way, visualizing an efficiency of 95.78% in the detection of exudates, and an efficiency of 97.14% in the detection of hemorrhages and blood vessels, resulting in a reliable and safe system to detect diabetic retinopathy early in diabetic patients.