Diabetic-Retinopathy-Diseases-Classification-using-Deep-Learning

Diabetic Retinopathy Diseases Classification using Deep Learning

According to our research, we came to know that in Diabetic Retinopathy (DR), researchers are using deep convolutional neural networks to distinguish the different levels of discussed domain. Dataset being used in this area is publically available Diabetic Retinopathy Detection challenge dataset (www.kaggle.com) Proposed technique: Proposed technique workflow is presented in Figure 1. This technique will have the following phases.

Pre-Processing:
Data Augmentation:
Features Extraction using Pre-Trained Models:
Feature Fusion:
Feature Reduction:
Classification
Classification Performance Evaluation:

Figure 1 Proposed Technique Workflow DR Detection Challenge Dataset: The dataset consists distinct classes with five different levels like: 0 - No DR 1 - Mild 2 - Moderate 3 - Severe 4 - Proliferative DR