Machine Learning-Based Automated Detection of Diabetic Retinopathy Using Retinal fundus images.

Abstract

Diabetic retinopathy is the most well-known side effect of diabetes (DR). People with diabetes experience it and can observe how it affects human sight. Patients with DR have damaged blood vessels in the retina, the delicate layer at the back of the eyes. Even though it may not initially show symptoms or cause mild vision problems, DR can cause blindness if not treated. In this study, the classification of the retina based on texture analysis is used to examine the various phases of DR, including mild, moderate, non-proliferative, proliferative, and regular human eye. Stages of DR show misunderstandings about the body. As a result, it might be difficult for a doctor to tell which stage of DR a patient is going through. In order to recognize and classify DR phases, this work suggests an automated approach that combines machine learning (ML) and image processing (IP) techniques. The m has been created for texture analysis utilizing a data fusion approach. A multi-feature dataset and an ML classifier were used to create the model (using cross-validation 10). The multi-layer perceptron (MLP) has shown an extremely high degree of performance with a classification accuracy of 98.53%.

Machine Learning Based Classification of Lung Cancer Using CT Scan Images

“…After creating a text file, this file is converted into a CSV file, and then an arff file is created from weka software to generate other results. Weka files are opened, and two main search methods are applied for optimization to extract the best features[10]. ANN Classifier Summary…”