Convolutional Neural Networks for Diabetic Retinopathy

By Harry Pratt, Frans Coenen, Deborah M. Broadbent, Simon P. Harding, Yalin Zheng. 2016

872 citations recorded by the source corpus.

Abstract The diagnosis of diabetic retinopathy (DR) through colour fundus images requires experienced clinicians to identify the presence and significance of many small features which, along with a complex grading system, makes this a difficult and time consuming task. In this paper, we propose a CNN approach to diagnosing DR from digital fundus images and accurately classifying its severity. We develop a network with CNN architecture and data augmentation which can identify the intricate features involved in the classification task such as micro-aneurysms, exudate and haemorrhages on the retina and consequently provide a diagnosis automatically and without user input. We train this network using a high-end graphics processor unit (GPU) on the publicly available Kaggle dataset and demonstrate impressive results, particularly for a high-level classification task. On the data set of 80,000 images used our proposed CNN achieves a sensitivity of 95% and an accuracy of 75% on 5,000 validation images.

Subfields: Computer Vision and Pattern Recognition, Ophthalmology, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Radiology, Nuclear Medicine and Imaging

Citations in the corpus

Leveraging uncertainty information from deep neural networks for disease detection (2016, neutral): “We want to point out, that the quality of the former dataset was questioned previously - albeit informally, both by competition participants 46 as well as by clinicians 54 .”
Convolutional neural networks in medical image understanding: a survey (2021, background): “This was proposed for classifying diabetic retinopathy (DR) in fundus imagery in [57].”
Convolutional neural networks based transfer learning for diabetic retinopathy fundus image classification (2017, methodology): “Authors in paper [35] present a CNN method to implement five class classification of DR from digital fundus images and accurately classify its severity, where the five class classification corresponds to five levels of DR.”
Deep Learning and Medical Diagnosis: A Review of Literature (2018, positive): “[51] CNN Fundus images Diagnosis of diabetic retinopathy; on a dataset of 80,000 images the accuracy was 75% and the sensitivity is 95%.”
Deep learning and medical image processing for coronavirus (COVID-19) pandemic: A survey (2020, positive): “In Pratt, Coenen, Broadbent, Harding, and Zheng (2016), the authors developed a model by training 10-layer CNN with three completely integrated layers on around 90,000 fundus images to diagnose Diabetic Retinopathy (DR).”
CANet: Cross-Disease Attention Network for Joint Diabetic Retinopathy and Diabetic Macular Edema Grading (2019, background): “In the last few years, deep learning algorithms have become popular for DR grading [22]–[28].”
Fundus Image Classification Using VGG-19 Architecture with PCA and SVD (2018, methodology): “…of 54 American Furthermore, Sankar et al. [9] classified the fundus images into non-DR, mild DR, and severe DR. ophthalmologists and other medical researchers Pratt et al. [10] proposed a robust technique to classify the severity levels of DR followed by the standards of the severity scale [11].”
Deep Learning for Medical Image Processing: Overview, Challenges and Future (2017, methodology): “Picture Archive Communication System (EyePACS-1) dataset and Messidor-2 dataset for classification and detection of moderate and worse referable [21].”
Influence-Directed Explanations for Deep Convolutional Networks (2018, methodology): “Building on the work of (Pratt et al., 2016; Gulshan et al., 2016), we trained an Inception network (Szegedy et al.”
A Deep Learning Ensemble Approach for Diabetic Retinopathy Detection (2019, methodology): “[21] proposed a CNN architecture used for classifying five stages but could not classify the mild stage accurately, due to the nature of architecture.”