A Systematic Review of Detection of Diabetic Retinopathy using Machine Learning Approach

Abstract

Diabetes Mellitus causes diabetic retinopathy (DR). It can cause blindness if not diagnosed early. Disease diagnosis is an essential and highly scrutinized biomedical field in which machine learning has been significantly used. Recently, machine learning has emerged as one of the most widely used approaches for improving performance in various areas, including medical image analysis and classification. This research compares several machine learning experiments based on the accuracy and sensitivity of retinal fundus pictures acquired by the fundus camera to assess several strategies for identifying Diabetic retinopathy. Inflammatory illnesses in the posterior portion of fundus photography are followed by retinal imaging. In particular, machine learning and deep learning are cutting-edge technologies well-suited for data analytics applications in the medical field. The results were compared to those of other approaches such as deep neural networks and other best practices. This work will be beneficial to researchers who want to apply their research in this field. During this research, we have gone through several research papers. This paper includes findings from other researcher’s studies, which have been summarized to present their pros and cons for disease diagnosis