Deep Learning Approach to Diabetic Retinopathy Detection

Tymchenko, Borys; Marchenko, Philip; Spodarets, Dmitry

doi:10.5220/0008970805010509

Cited by 116 publications

(30 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The phases of diabetic retinopathy are with four stages, including micro aneurysms, hemorrhage, neovascularization, and venous leading. The four stages denote the ponderances of retinopathy [37]:…”

Section: A Diabetic Retinopathy Diagnosismentioning

confidence: 99%

Hybrid Graph Convolutional Network for Semi-Supervised Retinal Image Classification

Zhang¹,

Pan²,

Zhang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Diabetic Retinopathy (DR) causes a significant health threat to the patient's vision with diabetic disease, which may result in blindness in severe situations. Various automatic DR diagnosis models have been proposed along with the development of deep learning, while there always relies on a large scale annotated data to train the network. However, annotating medical fundus images is cost-expensive and requires well-trained professional doctors to identity the DR grades. To overcome this drawback, this paper focuses on utilizing the easily-obtained unlabeled data with the help of limited annotated data to identify DR grades accurately. Hence we proposes a semi-supervised retinal image classification method by a Hybrid Graph Convolutional Network (HGCN). This HGCN network designs a modularity-based graph learning module and integrates Convolutional Neural Network (CNN) features into the graph representation by graph convolutional network. The synthesized hybrid features are optimized by a semi-supervised classification task which is assisted by a similarity-based pseudo label estimator. Through the proposed HGCN method, the retinal image classification model can be trained efficiently by partially labeled samples and the complicated annotating work is not required for the most retinal images. The experimental results on MESSIDOR dataset demonstrate the favorable performance of HGCN on semi-supervised retinal image classification, and the fully labeled data training also achieves an obvious superiority to the state-of-the-art supervised learning methods.

show abstract

Section: A Diabetic Retinopathy Diagnosismentioning

confidence: 99%

Hybrid Graph Convolutional Network for Semi-Supervised Retinal Image Classification

Zhang¹,

Pan²,

Zhang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…To further demonstrate the advanced DR grading performance, we compare our method with three completely supervised methods, including SE-ResNeXt50 [31], EfficientNet [32], and EnsembleNet [33], which are proposed recently with same training and testing data. In detail, SE-ResNeXt50 [31] designed a squeeze-and-excitation (SE) block adaptively recalibrating channel-wise feature responses by explicitly modeling interdependencies between channels, which boost the representational power of a network.…”

Section: E Comparison With Supervised Modelsmentioning

confidence: 99%

Semi-Supervised Auto-Encoder Graph Network for Diabetic Retinopathy Grading

Zhang

Kang

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Diabetic Retinopathy (DR) causes quite a few blindness worldwide, which can be refrained by the timely diagnosis on retinal images. Recently, researches on deep learning-based retinal image classification have accelerated outstanding improvements in DR grading task. However, existing DR grading works are mostly limited to a supervised manner. They require accurately annotated data labeled by professional experts, and the annotating work is very laborious and time-consuming. We propose a Semisupervised Auto-encoder Graph Network (SAGN) for the challenging DR diagnosis to relax this constraint. Precisely, SAGN consists of three major modules: auto-encoder feature learning, neighbor correlation mining, and graph representation. Firstly, our model learns to extract representations from retinal images and reconstruct them as close to original inputs as possible. Then neighbor correlations among labeled and unlabeled samples are established by their similarities, calculated by the radial basis function. Finally, we operate Graph Convolutional Neural Network (GCN) to grade retinal samples from extracted features and their correlations. To evaluate the performance of SAGN, we conduct sufficient comparative experiments on APTOS 2019 dataset, trained from EyePACS. Results demonstrate that our SAGN model can achieve comparable performance with limited labeled retinal images with the help of large amounts of unlabeled data.

show abstract

“…In [20] they introduce a Diabetic Retinopathy Detection Deep Learning Approach. We propose an AI-based methodology for stage detection of DR in this analysis using single imagery of the person fundus.…”

Section: In [5] They Have Explored Implementing Machinementioning

confidence: 99%

A detailed survey on Prognostication of diabetes diagnosis on the basis of machine learning techniques and the detection approaches to diabetic retinopathy using Artificial Intelligence

2021

IJATCSE

View full text Add to dashboard Cite

The aim of the study is to compare, assess the optimum tools as well as the techniques and advanced features focused on prediction of diabetes diagnosis based on machine learning tactics and diabetic retinopathy using Artificial Intelligence. The literature on data science, Artificial Intelligence (AI) contains important knowledge and understanding of AI entities such as Data science, machine learning, deep learning, Medical image processing, feature extraction, classification techniques, etc. Diabetes diagnosis is a phenomenon that impacts individuals around the globe. Now, with diabetes impacting people from children to the elderly, the out-dated approaches to diabetes diagnosis should be replaced with new, time-saving technologies. There's several studies carried out by researchers to recognise and predict diabetes. Here plenty of classifiers in machine learning can be used, such as KNN, Random Tree, etc.They can save time and get more precise outcome when using these techniques to predict diabetes. Diabetic retinopathy (DR) is a typical disorder of diabetic disease that induces vision-impacting lesions in the retina. It also can turn to visual impairment if it is not addressed early. DR therapy only helps vision. Deep learning has in recent times being one of the most widely used approaches that has accomplished higher outcomes in so many fields, especially in the analysing and identification of medical image classification. In medical image processing, convolutional neural networks (CNN) using transfer learning are commonly used as a deep learning approach and they are incredibly beneficial. Key words: Diab

show abstract

Deep Learning Approach to Diabetic Retinopathy Detection

Cited by 116 publications

References 18 publications

Hybrid Graph Convolutional Network for Semi-Supervised Retinal Image Classification

Hybrid Graph Convolutional Network for Semi-Supervised Retinal Image Classification

Semi-Supervised Auto-Encoder Graph Network for Diabetic Retinopathy Grading

A detailed survey on Prognostication of diabetes diagnosis on the basis of machine learning techniques and the detection approaches to diabetic retinopathy using Artificial Intelligence

Contact Info

Product

Resources

About