Diabetic Retinopathy Detection using Deep Learning Techniques

Abstract

Diabetic Retinopathy is a type of eye condition induced by diabetes, which damages the blood vessels in the retinal region and the area covered with lesions of varying magnitude determines the severity of the disease. It is one of the most leading causes of blindness amongst the employed community. A variety of factors are observed to play a role in a person to get this disease. Stress and prolonged diabetes are two of the most critical factors to top the list. This disease, if not predicted early, can lead to a permanent impairment of vision. If predicted in advance, the rate of impairment can be brought down or averted. However, it is not easy to detect the presence of this disease, given the time-consuming and tedious process of diagnosis. Presently, digital color photographs are evaluated manually by trained clinicians to observe the presence of lesions caused due to vascular abnormalities, which is the major effect of Diabetic Retinopathy. This method, although it is pretty accurate, proves to be costly. The delay brings out the need to automate the diagnosing, which will, in turn, have a significant positive impact on the health sector. In recent times, the adoption of AI in disease diagnosis has ensured promising and reliable results and this serves as the motivation for this journal. The paper employs Deep learning methodologies for automatic detection of Diabetic Retinopathy, resulting in a maximum accuracy of 80%, as compared to traditional Machine learning approaches giving only a maximum accuracy of 48% on the same IRDiR Disease Grading Dataset (413 images with 5 levels of DR-Training set; 103 images with 5 levels of DR-Test set). The data set contains digital fundus images of different levels of Diabetic Retinopathy in discrete frequency distributions.

A Diabetic Retinopathy Detection Using Customized Convolutional Neural Network

“…Till date many researchers are tried to apply many machine learning techniques on DR detection. However according to the previous results, the highest accuracy for diabetic retinopathy detection of existing papers was 80% [1]. We have increased the accuracy rate by using a customized convolutional neural network (CCNN) algorithm which achieved 97.24% accuracy.…”

“…Authors [1] presented co-learning is a machine learning (ML) model, to ensure an integrated approach. In the first instance, they were trying to simulate a database with the use of traditional ML techniques including Logic Regression, KNN, LDA, Random Forest, SVM, decision tree, and Naive Bayes Algorithm.…”

“…The simpler it is to develop a treatment, however, the sooner the sickness is discovered. High blood pressure and a high blood sugar level in the body both have an impact on diabetic retinopathy [1]. The majority of individuals have diabetic retinopathy (DR).…”

A Comparative Study of Deep Learning and Transfer Learning in Detection of Diabetic Retinopathy

“…Several researchers have developed models aimed at classifying Diabetic retinopathy. The models are trained using public datasets such: as the EyePACs dataset, Indian Diabetic Retinopathy Dataset, APTOS 2019 blindness dataset, and Messidor Dataset [7][8] [9]. Researchers have used a combination of two or more of these datasets or just one dataset in developing their models.…”

Diagnosis of Diabetic Retinopathy using Fundus Images and Image Processing Methods

“…Figure 1: A fundus image with diabetic retinopathy, (a) Healthy Retina b) Diabetic Retina [11] In addition, some people develop exudate [3][9] when fluid and lipids escape from damaged blood vessels onto the macula (Refer to figure1(b) of the retina). The macula swells as fluid enters it, resulting in impaired vision.…”