Automated detection and classification of diabetic retinopathy using deep learning

Project Title

Automated detection and classification of diabetic retinopathy using deep learning

Project Area of Specialization Mechatronics EngineeringProject Summary

Our project is related to automated detection and classification of diabetic retinopathy using deep learning. In this, we detect and classify the lesions. We use bench mark dataset that contain 1451 images. we use Mask RCNN model for training dataset, Faster RCNN for detection and SVM and Convolutional Neural Network for classification.  We find  the best accuracy and comparison this accuracy with previous ones.

Project Objectives

To predict the severity level of DR using retinal images.

To provide the capability to localize the exact lesion.

To enhance classification and detection classification on benchmark dataset with 6 classes.

To detect eye disease faster.

Project Implementation Method

Dataset preprocessing and data annotation.

Faster RCNN for detection.

Convolutional Neural Network for classsification.

Benefits of the Project

It can facilitate the fast processsing in eye disease.

This system helpful in health care centres.

Technical Details of Final Deliverable

MATLAB based application capable of classifying the given input  retinal images and also localize the lesion in retinaal images.

Final Deliverable of the Project Software SystemCore Industry MedicalOther IndustriesCore Technology Artificial Intelligence(AI)Other TechnologiesSustainable Development Goals Good Health and Well-Being for PeopleRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	60000
smartphone retinal camera with accessories	Equipment	1	60000	60000