Diabetes Retinopathy Detection

Diabetes Retinopathy Detection

Mayuresh Deore¹, Abhishek Kshirsagar², Vishvesh Patel³ , Vedanshu Patel⁴, Mrs. Vashudha Phaltankar⁵

^1,2,3,4,5 Dept. of Computer Science Engineering

^1,2,3,4,5 Dr. D.Y. Patil Institute of Technology, Pune, Maharashtra, India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Diabetes, often known as chronic sickness, is a collection of metabolic illnesses caused by a sustained high amount of sugar in the blood. Diabetes risk factors and severity can be considerably decreased if accurate early prognosis is achievable. Diabetic Retinopathy is a retinal condition that affects patients with diabetes mellitus and is the leading cause of blindness. It is a condition in which the blood vessels of the retina enlarge. If the level of diabetes is really high, this destroys the retina of the eye and may result in blindness. Diabetes prediction is extremely difficult due to the small number of labelled data and the existence of outliers (or missing values) in the data. In this paper, we provide a robust framework for diabetes prediction that includes outlier rejection, data normalisation, feature selection, and various Machine Learning (ML) classifiers (k-nearest Neighbours, Decision Trees, Random Forest, Naive Bayes, and XGBoost). Graphs and charts will be displayed using data visualisation tools to provide deeper insights.

Key Words: Machine Learning, Classification, SVM, KNN, Random Forest, Naive Bayes, KGBoost.