Sistema Diagnóstico Asistido por Computadora para la detección de la Retinopatía Diabética No Proliferativa usando la Red Neuronal de Retropropagación

Abstract

One of the most serious complications of Diabetes Mellitus is the Diabetic Retinopathy, a disease worldwide considered as the leading cause of blindness in working age population. Diabetic Retinopathy is a silent disease and is only recognized when the changes on the retina have progressed to a level at which treatment turns complicate, so an early diagnosis and referral to an ophthalmologist or optometrist for the management of this disease can prevent 98% of severe visual loss. Nowadays, one of the areas of research that is attracting the interest of scientists and physicians is the medical image analysis. This analysis consist in the study of digital images in order to provide computational tools that assist the medical specialist in taking decisions based on the quantification and visualization of the pathology of interest and their anatomical structures. Under the preceding context, this master degree thesis shows a development of a Computer Aided Diagnosis System for the detection of Non Proliferative Diabetic Retinopathy through the implementation of Digital Image Processing and Back-Propagation Neural Networks algorithms. The Diabetic Retinopathy images, as well as the medical support were supplied by the “General Ignacio Zaragoza” Hospital of the ISSSTE and Ophthalmology Service of the High Specialized Medical Unit of the National Medical Center “La Raza” of the IMSS. Experimental results show that the proposed Computer Aided Diagnosis System presents optimal results in terms of right detection of absence or presence of the disease and their stages, obtaining a sensibility and specificity of 95% in comparison with other similar systems, whose, sensibility and specificity are in the range of 83.4% to 94%.