Diabetic Retinopathy Fundus Images Research Articles

Literature Survey on Diabetic Retinopathy Classification Using Deep Learning

Abstract: Diabetic retinopathy (DR) is a medical condition that damages eye retinal tissues. Diabetic retinopathy leads to mild to complete blindness. It has been a leading cause of global blindness. The identification and categorization of DR take place through the segmentation of parts of the fundus image or the examination of the fundus image for the incidence of exudates, lesions, microaneurysms, and so on. This research aims to study and summarize various recent proposed techniques applied to automate the process of classification of diabetic retinopathy. In the current study, the researchers focused on the concept of classifying the DR fundus images based on their severity level. Emphasis is on studying papers that proposed models developed using transfer learning. Thus, it becomes vital to develop an automatic diagnosis system to support physicians in their work.

An interpretable multiple-instance approach for the detection of referable diabetic retinopathy in fundus images

Diabetic retinopathy (DR) is one of the leading causes of vision loss across the world. Yet despite its wide prevalence, the majority of affected people lack access to the specialized ophthalmologists and equipment required for monitoring their condition. This can lead to delays in the start of treatment, thereby lowering their chances for a successful outcome. Machine learning systems that automatically detect the disease in eye fundus images have been proposed as a means of facilitating access to retinopathy severity estimates for patients in remote regions or even for complementing the human expert’s diagnosis. Here we propose a machine learning system for the detection of referable diabetic retinopathy in fundus images, which is based on the paradigm of multiple-instance learning. Our method extracts local information independently from multiple rectangular image patches and combines it efficiently through an attention mechanism that focuses on the abnormal regions of the eye (i.e. those that contain DR-induced lesions), thus resulting in a final image representation that is suitable for classification. Furthermore, by leveraging the attention mechanism our algorithm can seamlessly produce informative heatmaps that highlight the regions where the lesions are located. We evaluate our approach on the publicly available Kaggle, Messidor-2 and IDRiD retinal image datasets, in which it exhibits near state-of-the-art classification performance (AUC of 0.961 in Kaggle and 0.976 in Messidor-2), while also producing valid lesion heatmaps (AUPRC of 0.869 in the 81 images of IDRiD that contain pixel-level lesion annotations). Our results suggest that the proposed approach provides an efficient and interpretable solution against the problem of automated diabetic retinopathy grading.

Diabetic Retinopathy Fundus Image Classification and Lesions Localization System Using Deep Learning.

Diabetic retinopathy (DR) is a disease resulting from diabetes complications, causing non-reversible damage to retina blood vessels. DR is a leading cause of blindness if not detected early. The currently available DR treatments are limited to stopping or delaying the deterioration of sight, highlighting the importance of regular scanning using high-efficiency computer-based systems to diagnose cases early. The current work presented fully automatic diagnosis systems that exceed manual techniques to avoid misdiagnosis, reducing time, effort and cost. The proposed system classifies DR images into five stages—no-DR, mild, moderate, severe and proliferative DR—as well as localizing the affected lesions on retain surface. The system comprises two deep learning-based models. The first model (CNN512) used the whole image as an input to the CNN model to classify it into one of the five DR stages. It achieved an accuracy of 88.6% and 84.1% on the DDR and the APTOS Kaggle 2019 public datasets, respectively, compared to the state-of-the-art results. Simultaneously, the second model used an adopted YOLOv3 model to detect and localize the DR lesions, achieving a 0.216 mAP in lesion localization on the DDR dataset, which improves the current state-of-the-art results. Finally, both of the proposed structures, CNN512 and YOLOv3, were fused to classify DR images and localize DR lesions, obtaining an accuracy of 89% with 89% sensitivity, 97.3 specificity and that exceeds the current state-of-the-art results.

A Fully Convolutional Neural Network for Recognition of Diabetic Retinopathy in Fundus Images

Background: Diabetic retinopathy is one of the complexities of diabetics and a major cause of vision loss worldwide which come into sight due to prolonged diabetes. For the automatic detection of diabetic retinopathy through fundus images several technical approaches have been proposed. The visual information processing by convolutional neural network makes itself more suitable due to its spatial arrangement of units. Convolutional neural networks are at their peak of development and best results can be gained by proper use of the technique. The local connectivity, parameter sharing and pooling of hidden units are advantageous for various predictions. Objective: Objective of this paper is to design a model for classification of diabetic retinopathy. Method: A fully convolutional neural network model is developed to classify the diseased and healthy fundus images. Here, proposed neural network consists of six convolutional layers along with rectified linear unit activations and max pooling layers. The absence of fully connected layer reduces the computational complexity of the model and trains faster as compared to traditional convolutional neural network models. Result and Conclusion: The validation of the proposed model is accomplished by training it with a publicly available High-Resolution Fundus image database. The model is also compared with various existing state-of-the-art methods which show competitive result as compared to these models. A behavioural study of different parameters of the network model is represented. The intelligence of our model lies in its ability to re-tune weight to overcome outliers encountered in future. The proposed model works well with satisfactory performance.

Multiple Lesions Insertion: boosting diabetic retinopathy screening through Poisson editing.

Deep neural networks have made incredible progress in many computer vision tasks, owing to access to a great amount of data. However, collecting ground truth for large medical image datasets is extremely inconvenient and difficult to implement in practical applications, due to high professional requirements. Synthesizing can generate meaningful supplement samples to enlarge the insufficient medical image dataset. In this study, we propose a new data augmentation method, Multiple Lesions Insertion (MLI), to simulate new diabetic retinopathy (DR) fundus images based on the healthy fundus images that insert real lesions, such as exudates, hemorrhages, microaneurysms templates, into new healthy fundus images with Poisson editing. The synthetic fundus images can be generated according to the clinical rules, i.e., in different DR grading fundus images, the number of exudates, hemorrhages, microaneurysms are different. The generated DR fundus images by our MLI method are realistic with the real texture features and rich details, without black spots, artifacts, and discontinuities. We first demonstrate the feasibility of this method in a DR computer-aided diagnosis (CAD) system, which judges whether the patient has transferred treatment or not. Our results indicate that the MLI method outperforms most of the traditional augmentation methods, i.e, oversampling, under-sampling, cropping, rotation, and adding other real sample methods in the DR screening task.

Evaluating the Viability of a Smartphone-Based Annotation Tool for Faster and Accurate Image Labelling for Artificial Intelligence in Diabetic Retinopathy

IntroductionDeep Learning (DL) and Artificial Intelligence (AI) have become widespread due to the advanced technologies and availability of digital data. Supervised learning algorithms have shown human-level performance or even better and are better feature extractor-quantifier than unsupervised learning algorithms. To get huge dataset with good quality control, there is a need of an annotation tool with a customizable feature set. This paper evaluates the viability of having an in house annotation tool which works on a smartphone and can be used in a healthcare setting.MethodsWe developed a smartphone-based grading system to help researchers in grading multiple retinal fundi. The process consisted of designing the flow of user interface (UI) keeping in view feedback from experts. Quantitative and qualitative analysis of change in speed of a grader over time and feature usage statistics was done. The dataset size was approximately 16,000 images with adjudicated labels by a minimum of 2 doctors. Results for an AI model trained on the images graded using this tool and its validation over some public datasets were prepared.ResultsWe created a DL model and analysed its performance for a binary referrable DR Classification task, whether a retinal image has Referrable DR or not. A total of 32 doctors used the tool for minimum of 20 images each. Data analytics suggested significant portability and flexibility of the tool. Grader variability for images was in favour of agreement on images annotated. Number of images used to assess agreement is 550. Mean of 75.9% was seen in agreement.ConclusionOur aim was to make Annotation of Medical imaging easier and to minimize time taken for annotations without quality degradation. The user feedback and feature usage statistics confirm our hypotheses of incorporation of brightness and contrast variations, green channels and zooming add-ons in correlation to certain disease types. Simulation of multiple review cycles and establishing quality control can boost the accuracy of AI models even further. Although our study aims at developing an annotation tool for diagnosing and classifying diabetic retinopathy fundus images but same concept can be used for fundus images of other ocular diseases as well as other streams of medical science such as radiology where image-based diagnostic applications are utilised.

Greedy segmentation based diabetic retinopathy identification using curvelet transform and scale invariant features

Diabetic retinopathy (DR) is the major reason of vision loss in the active population. It can usually be prevented by regulating the blood glucose and providing a timely treatment. DR has clinical features recognized by the experts including the blood vessel area, exudates, neovascularization, hemorrhages, and microaneurysm. Because DR has some varieties and complexities due to its geometrical and haemodynamic features, it is hard and time-consuming to detect DR in manual diagnosis. In Computer Aided Diagnosis (CAD) systems, the features of DR fundus images are detected using computer vision techniques. In this paper, a CAD system is proposed, which distinguishes automatically whether the fundus is normal or it suffers from diabetic retinopathy disease. As preprocess morphological operations like filtering, opening, and dilation are applied to the images firstly, then, Optic Disk (OD) segmentation is implemented using Greedy algorithm. Because of the intensity of an OD is similar to some DR intensities, OD regions are removed from the fundus images for an accurate feature extraction. The features extracted with Curvelet Transform (CT) and Scale Invariant Feature Transform (SIFT), respectively, are concatenated to provide a feature set that defines the fundus data optimally. Finally, the feature set is given to the Support Vector Machines (SVM), K-Nearest Neighborhood (KNN), and Naïve–Bayes (NB) classifiers for the DR identification purpose. The proposed method achieves the highest accuracy and sensitivity as 92.8% and 97.6%, respectively, with SVM and specificity as 92.5% with KNN classifier.

Synergic Adversarial Label Learning for Grading Retinal Diseases via Knowledge Distillation and Multi-Task Learning.

The need for comprehensive and automated screening methods for retinal image classification has long been recognized. Well-qualified doctors annotated images are very expensive and only a limited amount of data is available for various retinal diseases such as diabetic retinopathy (DR) and age-related macular degeneration (AMD). Some studies show that some retinal diseases such as DR and AMD share some common features like haemorrhages and exudation but most classification algorithms only train those disease models independently when the only single label for one image is available. Inspired by multi-task learning where additional monitoring signals from various sources is beneficial to train a robust model. We propose a method called synergic adversarial label learning (SALL) which leverages relevant retinal disease labels in both semantic and feature space as additional signals and train the model in a collaborative manner using knowledge distillation. Our experiments on DR and AMD fundus image classification task demonstrate that the proposed method can significantly improve the accuracy of the model for grading diseases by 5.91% and 3.69% respectively. In addition, we conduct additional experiments to show the effectiveness of SALL from the aspects of reliability and interpretability in the context of medical imaging application.

Evaluation of convolutional neural network variants for diagnosis of diabetic retinopathy

Diabetic Retinopathy (DR) is a long-term complication of Diabetes Mellitus (DM) that impairs vision. This stage occurs in visual impairment and blindness if treated late. DR identified through scanning fundus images. A technique on classifying DR in fundus images is the deep learning approach, one of the methods of implementing machine learning. In this study, the Convolutional Neural Networks (CNN) method applied with the ResNet-50 and DenseNet-121 architectures. The data adopted in this analysis was generated from DIARETDB1, an online database containing fundus images. Then, the pre-processing stage is carried out on the fundus image to improve model performance, such as selected the green channel from the images and inverted it, converted the images into grayscale images, and applied Contrast Limited Adaptive Histogram Equalization (CLAHE) for uniform contrast in the images. The outcome of this research indicates that the ResNet-50 model is better than DenseNet-121 in detecting DR. The most reliable results from the ResNet-50 model's case testing are accuracy, precision, and recall of 95%, 98%, and 96% respectively.

RF-GANs: A Method to Synthesize Retinal Fundus Images Based on Generative Adversarial Network.

Diabetic retinopathy (DR) is a diabetic complication affecting the eyes, which is the main cause of blindness in young and middle-aged people. In order to speed up the diagnosis of DR, a mass of deep learning methods have been used for the detection of this disease, but they failed to attain excellent results due to unbalanced training data, i.e., the lack of DR fundus images. To address the problem of data imbalance, this paper proposes a method dubbed retinal fundus images generative adversarial networks (RF-GANs), which is based on generative adversarial network, to synthesize retinal fundus images. RF-GANs is composed of two generation models, RF-GAN1 and RF-GAN2. Firstly, RF-GAN1 is employed to translate retinal fundus images from source domain (the domain of semantic segmentation datasets) to target domain (the domain of EyePACS dataset connected to Kaggle (EyePACS)). Then, we train the semantic segmentation models with the translated images, and employ the trained models to extract the structural and lesion masks (hereafter, we refer to it as Masks) of EyePACS. Finally, we employ RF-GAN2 to synthesize retinal fundus images using the Masks and DR grading labels. This paper verifies the effectiveness of the method: RF-GAN1 can narrow down the domain gap between different datasets to improve the performance of the segmentation models. RF-GAN2 can synthesize realistic retinal fundus images. Adopting the synthesized images for data augmentation, the accuracy and quadratic weighted kappa of the state-of-the-art DR grading model on the testing set of EyePACS increase by 1.53% and 1.70%, respectively.

Automated detection of diabetic retinopathy in fundus images using fused features.

Diabetic retinopathy (DR) is one of the severe eye conditions due to diabetes complication which can lead to vision loss if left untreated. In this paper, a computationally simple, yet very effective, DR detection method is proposed. First, a segmentation independent two-stage preprocessing based technique is proposed which can effectively extract DR pathognomonic signs; both bright and red lesions, and blood vessels from the eye fundus image. Then, the performance of Local Binary Patterns (LBP), Local Ternary Patterns (LTP), Dense Scale-Invariant Feature Transform (DSIFT) and Histogram of Oriented Gradients (HOG) as a feature descriptor for fundus images, is thoroughly analyzed. SVM kernel-based classifiers are trained and tested, using a 5-fold cross-validation scheme, on both newly acquired fundus image database from the local hospital and combined database created from the open-sourced available databases. The classification accuracy of 96.6% with 0.964 sensitivity and 0.969 specificity is achieved using a Cubic SVM classifier with LBP and LTP fused features for the local database. More importantly, in out-of-sample testing on the combined database, the model gives an accuracy of 95.21% with a sensitivity of 0.970 and specificity of 0.932. This indicates the proposed model is very well-fitted and generalized which is further corroborated by the presented train-test curves.

A Multi-Label Deep Learning Model with Interpretable Grad-CAM for Diabetic Retinopathy Classification.

The characteristics of diabetic retinopathy (DR) fundus images generally consist of multiple types of lesions which provided strong evidence for the ophthalmologists to make diagnosis. It is particularly significant to figure out an efficient method to not only accurately classify DR fundus images but also recognize all kinds of lesions on them. In this paper, a deep learning-based multi-label classification model with Gradient-weighted Class Activation Mapping (Grad-CAM) was proposed, which can both make DR classification and automatically locate the regions of different lesions. To reducing laborious annotation work and improve the efficiency of labeling, this paper innovatively considered different types of lesions as different labels for a fundus image so that this paper changed the task of lesion detection into that of image classification. A total of five labels were pre-defined and 3228 fundus images were collected for developing our model. The architecture of deep learning model was designed by ourselves based on ResNet. Through experiments on the test images, this method acquired a sensitive of 93.9% and a specificity of 94.4% on DR classification. Moreover, the corresponding regions of lesions were reasonably outlined on the DR fundus images.

Severity Level Detection in Retinal Fundus Images for Diabetic Retinopathy

Diabetic Retinopathy is the disease caused for diabetic people which doesn’t have symptoms in the first phase. As it progresses, it becomes symptomatic. This disease, sometimes, might lead to complete blindness. Red lesions contain microaneurysms, haemorrhages and exudates. This work focuses on detection of red lesions in fundus image. Ophthalmologists use pupil dilation of chemical solutions in order to detect the abnormality which takes time and also causes irritation to patients. Image processing techniques are used to avoid these limitations. Morphological operations are used to identify the pixels belonging to the red lesions. Gabor filter is used for separating the blood vessels. Some spots are formed near macular region because blood vessels become leaky which leads to exudates. The severity level of the disease is determined by finding the distance between lesions and macular region. The disease is considered as severe if the distance between them is closer and confirms as less if the distance is far.

Assessment of Training Outcomes of Nurse Readers for Diabetic Retinopathy Telescreening: Validation Study.

BackgroundWith the high prevalence of diabetic retinopathy and its significant visual consequences if untreated, timely identification and management of diabetic retinopathy is essential. Teleophthalmology programs have assisted in screening a large number of individuals at risk for vision loss from diabetic retinopathy. Training nonophthalmological readers to assess remote fundus images for diabetic retinopathy may further improve the efficiency of such programs.ObjectiveThis study aimed to evaluate the performance, safety implications, and progress of 2 ophthalmology nurses trained to read and assess diabetic retinopathy fundus images within a hospital diabetic retinopathy telescreening program.MethodsIn this retrospective interobserver study, 2 ophthalmology nurses followed a specific training program within a hospital diabetic retinopathy telescreening program and were trained to assess diabetic retinopathy images at 2 levels of intervention: detection of diabetic retinopathy (level 1) and identification of referable disease (level 2). The reliability of the assessment by level 1−trained readers in 266 patients and of the identification of patients at risk of vision loss from diabetic retinopathy by level 2−trained readers in 559 more patients were measured. The learning curve, sensitivity, and specificity of the readings were evaluated using a group consensus gold standard.ResultsAn almost perfect agreement was measured in identifying the presence of diabetic retinopathy in both level 1 readers (κ=0.86 and 0.80) and in identifying referable diabetic retinopathy by level 2 readers (κ=0.80 and 0.83). At least substantial agreement was measured in the level 2 readers for macular edema (κ=0.79 and 0.88) for all eyes. Good screening threshold sensitivities and specificities were obtained for all level readers, with sensitivities of 90.6% and 96.9% and specificities of 95.1% and 85.1% for level 1 readers (readers A and B) and with sensitivities of 86.8% and 91.2% and specificities of 91.7% and 97.0% for level 2 readers (readers A and B). This performance was achieved immediately after training and remained stable throughout the study.ConclusionsNotwithstanding the small number of trained readers, this study validates the screening performance of level 1 and level 2 diabetic retinopathy readers within this training program, emphasizing practical experience, and allows the establishment of an ongoing assessment clinic. This highlights the importance of supervised, hands-on experience and may help set parameters to further calibrate the training of diabetic retinopathy readers for safe screening programs.

Automated detection and classification of fundus diabetic retinopathy images using synergic deep learning model

In recent days, the incidence of Diabetic Retinopathy (DR)has become high, affecting the eyes because of drastic increase in the glucose level in blood. Globally, almost half of the people under the age of 70 gets severely affected by diabetes. In the absence of earlier recognition and proper medication, the DR patients tend to lose their vision. When the warning signs are tracked down, the severity level of the disease has to be validated so to take decisions regarding appropriate treatment further. The current research paper focuses on the concept of classification of DR fundus images on the basis of severity level using a deep learning model. This paper proposes a deep learning-based automated detection and classification model for fundus DR images. The proposed method involves various processes namely preprocessing, segmentation and classification. The methods begins with preprocessing stage in which unnecessary noise that exists in the edges is removed. Next, histogram-based segmentation takes place to extract the useful regions from the image. Then, Synergic Deep Learning (SDL) model was applied to classify the DR fundus images to various severity levels. The justification for the presented SDL model was carried out on Messidor DR dataset. The experimentation results indicated that the presented SDL model offers better classification over the existing models.

Detection of Proliferative Diabetic Retinopathy in Fundus Images Using Convolution Neural Network

Convolution Neural Network (CNN) is one of the techniques under Artificial Neural Network (ANN) used to develop a Deep Learning Neural Network (DLNN) algorithm for detection of Proliferative Diabetic Retinopathy (PDR) on the fundus images. About 116 PDR and 150 Non-Proliferative Diabetic Retinopathy (NPDR) of fundus images retrieved from the publicly available MESSIDOR database applied in this research. This study consisted three objectives that included the execution of two pre-processing techniques on the data-set which were resizing and normalizing the fundus images, developed deep learning operational Artificial Intelligence (AI) network of feature extraction algorithm for detection of PDR on the fundus images and determined the output classification of the network encompassing the accuracy, sensitivity and specificity. There were five different parameters carried out along this research. Here, Parameter 5 showed the best performance among the five parameters based on the value of accuracy, sensitivity, and specificity that was 73.81%, 76%, and 69% respectively.

Gaussian Kernel Prompted Fuzzy C Means Algorithm with Multi- Object Contouring Method for Segmenting NPDR Features in Diabetic Retinopathy Fundus Images

Diabetic retinopathy is an ophthalmic inflammation caused by diabetes which ends in visual defacement if not diagnosed earlier and that has two types namely Non-Proliferative Diabetic Retinopathy (NPDR) and Proliferative Diabetic Retinopathy (PDR). NPDR features are present in the earliest stage and systematic detection of these features can improve the diagnosis of the disease severity formerly. Several detection methods exists previously, still there is performance lack on large datasets. The objective of this study is detecting NPDR features from diabetic retinaopathy fundus images of large datasets with good performance level. The study has investigated different fuzzy based systems and to execute the objective; GK_FCM approach is proposed which integrates Gaussian Kernel function in conventional FCM. The execution has four phases, initially the input image undergoes preprocessing using green channel extraction, median filter to enhance the image quality and background removal is performed with extended minima transform technique, mathematical arithmetic operation and pixel replacement method to remove the outlier called Fovea (FV).

Detection of Diabetic Retinopathy using Convolutional Neural Network

Diabetic Retinopathy is a medical condition in which damage occurs to the retina due to diabetes mellitus. The diagnosis of Diabetic Retinopathy through colored fundus images stand in need of experienced clinicians to identify the presence and significance of many small features, which makes it a time consuming task. In this paper, we propose a CNN based approach to detect Diabetic Retinopathy in fundus images. Data used to train the model is prepocessed by a new segmentation technique using Gabor filters. Due to small dataset, data augmentation is done to get enough data to train the model. Our segmentation model detects intricate features in the fundus images and detect the presence of DR. A high-end Graphics Processor Unit (GPU) is used to train the model efficiently. The publicly available Kaggle Dataset is used to demonstrate impressive results, particularly for a high-level classification task. On the training dataset of 14,650 images, our proposed CNN achieves a specificity of 94% and an accuracy of 69% on 3,660 validation images.

ES08.07 System Approach to Screening Management

ES08.07 System Approach to Screening Management

Identifikasi Potensi Glaukoma dan Diabetes Retinopati Melalui Citra Fundus Menggunakan Jaringan Syaraf Tiruan

Identifikasi potensi glaukoma dan retinopati diabetes dapat dilakukan melalui citra fundus. Jaringan syaraf tiruan dapat digunakan untuk mengidentifikasi potensi glaukoma dan diabetes retinopati. Dataset yang digunakan berjumlah 60 citra fundus yang terdiri dari 20 citra fundus terjangkit glaukoma, 20 citra fundus terjangkit diabetes retinopati dan 20 citra fundus mata normal. Hasil penelitian menghasilkan rata-rata untuk recall sebesar 86,6%, precission sebesar 86,6%, dan untuk accuracy sebesar 91,06%.