• Aditi Panchal
• Kaustubh Naik
• Ankush Athawale

In 2010, the International Diabetes Federation (IDF) stated that around 50.8 million individuals in India were affected by diabetes, and this figure is projected to reach 87.0 million by 2030. Diabetic Retinopathy is a prominent worry when it comes to Type II diabetes complications. It frequently results in loss of vision, especially in people between 20 and 64 years old. Over a period of time, Diabetic Retinopathy interferes with the regular drainage of fluid from the eye, causing a rise in pressure within the eyeball and possibly harming the nerves, ultimately resulting in the development of glaucoma. Prompt detection and intervention for Diabetic Retinopathy are essential in order to prevent the loss of vision. Yet, diagnosing manually by ophthalmologists is time-consuming, demanding in labor, and expensive, while also posing a risk of misdiagnosis without the use of computer-aided diagnostic systems. Lately, Deep Learning has become a potent tool, especially in the field of medical image analysis and classification. This research focuses on the difficulty of forecasting diabetic retinopathy to avoid additional issues. A model was created utilising the compact MobileNet architecture and was trained on retinal fundus pictures from the Aptos 2019 competition dataset. The suggested model attained an impressive 96% accuracy, along with precision, recall, and f-1 scores of 0.95, 0.98, and 0.97 respectively. These findings suggest the possibility of using this model in clinical settings. Nevertheless, it is crucial to emphasise that the purpose of this project is not to supplant ophthalmologists but rather to support them in detecting diabetic retinopathy complications at an early stage.

Diabetic Retinopathy, Convolutional neural network, Deep learning, MobileNet Architecture

"Diabetic Retinopathy Binary Classifier", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.11, Issue 4, page no.h723-h727, April-2024, Available :http://www.jetir.org/papers/JETIR2404783.pdf

"Diabetic Retinopathy Binary Classifier", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.11, Issue 4, page no. pph723-h727, April-2024, Available at : http://www.jetir.org/papers/JETIR2404783.pdf