Automated System for the Classification of Diabetic Retinopathy

Perez-Siguas R.; Matta-Solis E.; Matta-Solis H.; Matta-Zamudio L. · International Journal of Emerging Technology and Advanced Engineering (2022) · DOI: 10.46338/ijetae0622_07

Resumen

In this current situation of COVID-19, this virus directly affected the respiratory system, all this generated an alert in all countries making it primordial unlike other diseases, including one of them diabetes mellitus. This disease over time has been increasing and currently has no cure, but it can be prevented since if not treated in time it can complicate the patient's health, even reaching diabetic retinopathy that can generate blindness. Diabetic retinopathy has stages and is evaluated by the ophthalmologist according to his experience and requires a lot of time to determine the stage of the patient who in many cases this evaluation has been erroneous. In view of this problem, this research work purpose the implementation of an automated system for the classification of diabetic retinopathy and to be able to detect the characteristics of diabetic retinopathy in those patients with diabetes, with this would help the patient with diabetic retinopathy to avoid complications, among them blindness, since this system will perform the analysis of the fundus of the eye automatically and determine the stage in which it is the diabetic patient. Through the development of the system, it was observed that the system worked at its best, with a percentage of efficiency of 94% in detecting exudates and 96% in detecting hemorrhages and blood vessels.

Instituciones: Universidad María Auxiliadora