This repository contains the implementation of Diabatic retinopathy Detection using Texture Features and Ensemble Learning, using the dataset available at https://www.kaggle.com/mohammadasimbluemoon/diabeticretinopathy-messidor-eyepac-preprocessed .
Diabetic Retinopathy is an eye-condition developed in patients with diabetics that gradually turn into blindness. This is a code implementation of the paper - "Diabetic Retinopathy Detection using Texture Features and Ensemble Learning" (https://doi.org/10.1109/TENSYMP50017.2020.9230600)
The modified dataset used in this project contains only 360 photos of the nomral eye and 360 images of proliferative DR condition. Thus, a total of 720 images to perform binary classification. I tried to make feature extraction method as similar as possible with the information given in the paper. The results are however better due to the varying components such as dataset, feature vector length, preprocessing, etc.