Diabetic Eye Health: Deep Learning Classification

Abstract

In individuals around the world, Diabetic Retinopathy (DR) and Diabetic Macular Edema (DME) is the most prevalent consequence of diabetes and a major factor in visual loss. The Convolutional Neural Network (CNN) architecture shown in this research is intended to automatically identify Diabetic Macular Edema (DME) and Diabetic Retinopathy (DR) from retinal fundus images. After being trained on a sizable dataset made up of several classes, the CNN model used inception is capable of outperforming earlier methods by reliably diagnosing the presence and severity of specific diseases. Its ability to handle a wide range of image qualities and intricate pathological aspects makes it a solid instrument for improved patient outcomes and early intervention, which lessens the toll that Diabetic eye disease takes on society and healthcare systems. We give a thorough experimental assessment of our methodology on a benchmark dataset, illustrating its efficacy in precisely identifying various stages involves Diabetic Retinopathy and Diabetic Macular Edema. The obtained results demonstrate a good level of performance and highlight the potential of deep learning methods in diagnosis.