In the realm of diabetic retinopathy, extensive research has been conducted by numerous scholars, who have explored and implemented a variety of machine learning techniques, contributing significantly to both healthcare and data science domains. With the increasing availability of deep learning procedures, packages, and libraries, these advancements have become pivotal in enhancing model performance. Consequently, this study embraces a novel methodology and platform in contrast to existing approaches to diabetic retinopathy, taking into account the outcomes and discoveries. For the deep learning model, the researcher leveraged the GPU provided by Google Colab. The dataset was sourced from a Kaggle competition hosted on the Kaggle website. Subsequently, the image data was stored in the researcher's personal server memory, and the URLs for each image were documented in an Excel sheet.
