Diabetic Retinopathy Database Research Articles

Level-set based adaptive-active contour segmentation technique with long short-term memory for diabetic retinopathy classification.

Diabetic Retinopathy (DR) is a major type of eye defect that is caused by abnormalities in the blood vessels within the retinal tissue. Early detection by automatic approach using modern methodologies helps prevent consequences like vision loss. So, this research has developed an effective segmentation approach known as Level-set Based Adaptive-active Contour Segmentation (LBACS) to segment the images by improving the boundary conditions and detecting the edges using Level Set Method with Improved Boundary Indicator Function (LSMIBIF) and Adaptive-Active Counter Model (AACM). For evaluating the DR system, the information is collected from the publically available datasets named as Indian Diabetic Retinopathy Image Dataset (IDRiD) and Diabetic Retinopathy Database 1 (DIARETDB 1). Then the collected images are pre-processed using a Gaussian filter, edge detection sharpening, Contrast enhancement, and Luminosity enhancement to eliminate the noises/interferences, and data imbalance that exists in the available dataset. After that, the noise-free data are processed for segmentation by using the Level set-based active contour segmentation technique. Then, the segmented images are given to the feature extraction stage where Gray Level Co-occurrence Matrix (GLCM), Local ternary, and binary patterns are employed to extract the features from the segmented image. Finally, extracted features are given as input to the classification stage where Long Short-Term Memory (LSTM) is utilized to categorize various classes of DR. The result analysis evidently shows that the proposed LBACS-LSTM achieved better results in overall metrics. The accuracy of the proposed LBACS-LSTM for IDRiD and DIARETDB 1 datasets is 99.43% and 97.39%, respectively which is comparably higher than the existing approaches such as Three-dimensional semantic model, Delimiting Segmentation Approach Using Knowledge Learning (DSA-KL), K-Nearest Neighbor (KNN), Computer aided method and Chronological Tunicate Swarm Algorithm with Stacked Auto Encoder (CTSA-SAE).

An Efficient DenseNet for Diabetic Retinopathy Screening

This study aims to propose a novel deep learning framework, i.e., efficient DenseNet, for identifying diabetic retinopathy severity levels in retinal images. Diabetic retinopathy is an eye condition that damages blood vessels in the retina. Detecting diabetic retinopathy at the early stage can avoid retinal detachment and effects leading to blindness in diabetic adults. A thin-layered efficient DenseNet model has been proposed with fewer training learnable parameters, leading to higher classification accuracy than the other deep learning models. The proposed deep learning framework for diabetic retinopathy severity level detection has an inbuilt automatic pre-processing module. Afterward, the efficient DenseNet model and classifier will provide data augmentation and higher-level feature extraction. The proposed efficient DenseNet framework is trained and tested using 13000 retinal fundus images within the diabetic retinopathy database and combined with the k-nearest neighbor classifier demonstrating the best classification accuracy of 98.40%.

Automatic Optic Disc Detection in Retinal Images Using FKMT‒MOPDF

In recent days, detecting Optic Disc (OD) in retinal images has been challenging and very important to the early diagnosis of eye diseases. The process of detecting the OD is challenging due to the diversity of color, intensity, brightness and shape of the OD. Moreover, the color similarities of the neighboring organs of the OD create difficulties during OD detection. In the proposed Fuzzy K‒Means Threshold (FKMT) and Morphological Operation with Pixel Density Feature (MOPDF), the input retinal images are coarsely segmented by fuzzy K‒means clustering and thresholding, in which the OD is classified from its neighboring organs with intensity similarities. Then, the segmented images are given as the input to morphological operation with pixel density feature calculations, which reduce the false detection in the small pixel of the OD. Finally, the OD area is detected by applying the Sobel edge detection method, which accurately detects the OD from the retinal images. After detection optimization, the proposed method achieved Sensitivity (SEN), Specificity (SPEC) and Accuracy (ACC), with 96.74%, 96.78% and 96.92% in DiaretDB0 (Standard Diabetic Retinopathy Database Calibration level 0), 97.12%, 97.10% and 97.75% in DiaretDB1 (Standard Diabetic Retinopathy Database Calibration level 1) and 97.19%, 97.47% and 97.43% in STARE (Structured Analysis of the Retina) dataset respectively. The experimental results demonstrated the proposed method’s efficiency for segmenting and detecting OD areas.

A Novel Soft Clustering Method for Detection of Exudates

One of the earliest indications of diabetes consequence is Diabetic Retinopathy (DR), the main contributor to blindness worldwide. Recent studies have proposed that Exudates (EXs) are the hallmark of DR severity. The present study aims to accurately and automatically detect EXs that are difficult to detect in retinal images in the early stages. An improved Fusion of Histogram–Based Fuzzy C–Means Clustering (FHBFCM) by a New Weight Assignment Scheme (NWAS) and a set of four selected features from stages of pre-processing to evolve the detection method is proposed. The features of DR train the optimal parameter of FHBFCM for detecting EXs diseases through a stepwise enhancement method through the coarse segmentation stage. The histogram-based is applied to find the color intensity in each pixel and performed to accomplish Red, Green, and Blue (RGB) color information. This RGB color information is used as the initial cluster centers for creating the appropriate region and generating the homogeneous regions by Fuzzy C–Means (FCM). Afterward, the best expression of NWAS is used for the delicate detection stage. According to the experiment results, the proposed method successfully detects EXs on the retinal image datasets of DiaretDB0 (Standard Diabetic Retinopathy Database Calibration level 0), DiaretDB1 (Standard Diabetic Retinopathy Database Calibration level 1), and STARE (Structured Analysis of the Retina) with accuracy values of 96.12%, 97.20%, and 93.22%, respectively. As a result, this study proposes a new approach for the early detection of EXs with competitive accuracy and the ability to outperform existing methods by improving the detection quality and perhaps significantly reducing the segmentation of false positives.

MULTI-SCALE DIRECTED ACYCLIC GRAPH-CNN FOR AUTOMATED CLASSIFICATION OF DIABETIC RETINOPATHY FROM OCT IMAGES

Diabetic Retinopathy (DR) is the principal cause of vision loss that interrupts the regular interaction of vascular, neural, and retinal constituents leading to impaired neuronal function and retinal abnormalities. Diagnosis of DR from Optical Coherence Tomography (OCT) image is difficult and time-consuming because several small features must be identified and graded, which results in a strenuous diagnosis when integrated with the complexity of the grading system. This study focuses on classifying DR from normal Spectral Domain-OCT (SD-OCT) images using the Directed Acyclic Graph (DAG) network without any pre-processing techniques. The proposed DAG-CNN model comprises 16 convolutional blocks, which learns multi-scale features automatically from multiple layers in the convolutional network and combines them effectively for the DR and normal prediction. The proposed model is tested on the public OCTID_DR and private LFH_DR SD-OCT databases containing DR and healthy OCT images. The model achieved an accuracy, precision, recall, F1-score, and AUC on OCTID_DR database of 0.9841, 0.9727, 0.9818, 0.9772, and 0.9836, respectively; and on LFH_DR database the respective values are 0.9988, 1, 0.9976, 0.9988, and 0.9988 with only 0.1569 Million of learnable parameters. This method significantly reduces the number of learnable parameters and the model’s computational complexity in terms of memory required and FLoating point OPerations (FLOPs). Guided Gradient-weighted Class Activation Mapping (Grad-CAM) is performed to highlight the regions of SD-OCT images that contribute to the decision of the classifier. Our model significantly surpasses the accuracy of the existing models with lower resource consumption and higher real-time performance.

Recognition of Diabetic Retinopathy with Ground Truth Segmentation Using Fundus Images and Neural Network Algorithm.

Diabetes problems can lead to a condition called diabetic retinopathy (DR), which permanently damages the blood vessels in the retina. If not treated, DR is a significant cause of blindness. The only DR treatments currently accessible are those that block or delay vision loss, which emphasizes the value of routine scanning with high-efficiency computer-based technologies to identify patients early. The major goal of this study is to employ a deep learning neural network to identify diabetic retinopathy in the retina's blood vessels. The NN classifier is put to the test using the input fundus image and DR database. It effectively contrasts retinal images and distinguishes between classes when there is a legitimate edge. For the resolution of the problems in the photographs, it is particularly useful. Here, it will be tested to see if the classification of diabetic retinopathy is normal or abnormal. Modifying the existing study's conclusion strategy, existing diabetic retinopathy techniques have sensitivity, specificity, and accuracy levels that are much lower than what is required for this research.

Modelling of Retinal Images for Analysis of Diabetic Retinopathy Severity Levels

Synthetic data by various algorithms that resemble actual data in terms of statistical features. Computer-aided medical applications have been extensively applied to model specific scenarios, such as medical imaging of retinal images for diabetic retinopathy (DR) detection. The available data and annotated medical data are typically rare and costly due to the difficulties of conducting medical screening and rely on highly trained doctors to review and diagnose. The modelling of retinal images for DR analysis is essential since it will provide a model to guide and test DR detection algorithms. This paper aims to model normal retina and non-proliferative diabetic retinopathy (NPDR) stages (mild, moderate, and severe) data models with the variation of dynamic models. The Digital Retinal Images for Vessel Extraction (DRIVE), The Standard Diabetic Retinopathy Database, Calibration Level 1 (DIARETDB1), and E-OPHTHA datasets are analyzed to obtain the specification of the human retina and DR lesions. In the data modelling phases, the model includes the bright and dark retinal lesions with the variation of dynamic parameters. 4100 synthetic images are used where 200 normal images and 3900 NPDR images to test the performance of DR detection algorithms over the full range of parameters.

Diagnosis of Diabetic Retinopathy Utilizing Computer-Aided Diagnosis System

Purpose: Diabetes is considered one of most diseases spread among people; blindness is considered the most resulted effect. Diabetes can damage the retinal blood vessels and cause severe problems to the eyes, which may end with sight loss. Such medical condition is known as "diabetic retinopathy" (DR). In such a diagnosis, the retinal microvascular go through several stages of change threat. In the early stages of the DR, detecting the formation that happened to the retinal blood vessels helps prevent the disease's dangerous effects. Therefore, producing a method to diagnose the disease in the early stages is helpful. So, this work aimed to develop a system of detecting and classifying the retina formation, trying to avoid relevant effects.Methodology: The current method depends on vascular edges map extracted from images of retina captured by a fundus camera. Such a map been utilized to extract quantitative texture features. The system tested two independent groups of the region of interest in normal and abnormal images. The two sets were extracted from ground truth images of the 89 fundus images. Fundus images were annotated images from the Standard Diabetic Retinopathy Database (DIARETDB1).Findings: The system provided an accuracy of 71% with a sensitivity of 75%.Recommendation: The current work may open an opportunity for improvement for future work. Other methods may be reached to raise the accuracy and the sensitivity of the system. Besides, the current system may be tested on a larger sample size to study such effects. Finally, the ease of the current method makes it faster in adoption in the appropriate diagnosis especially in the early stages.

An Integrated System for Detection Exudates and Severity Quantification for Diabetic Macular Edema

The prevalence of vision loss because of Diabetic macular edema (DME), which is a complication caused by diabetes mellitus, is increasing. Seven per cent of Indian diabetic population suffer from DME or damage to macula. The study proposes an integrated model to contribute to the prevention of loss of visual acuity. For detection of exudates and visualization of severity of DME image processing, machine learning and data analysis strategies are used. Experimentation on Indian population is done using Indian Diabetic Eye Diseases Dataset (IDEDD), acquired from metropolitan region of Mumbai and publicly available Indian Diabetic Retinopathy Image Dataset (IDRiD, Nanded) and a standard public database DIARETDB1- Standard Diabetic Retinopathy Database. To negate the effect of increase in the dimensionality and overfitting of the dataset parameters, a tree-based forward search algorithm for optimal feature subset identification is proposed. This provides global insight of effect of the statistical and spatial features on model behaviour. Machine learning algorithms of random forest and support vector machine are used for classification. The sensitivity and specificity of 97.1% and 100% is achieved for IDEDD data. Severity grading is done using image processing for segmenting the vessel mesh, optic disk, and macula from the abnormal retinal scans for detecting clinically significant macular edema by visualizing exudates. The distance between optic disc—fovea (OF) and the ratio of distance (OF) to that of disc diameter is formulated for localization of macula. Region-based and distance-based grading methods for quantification of the severity and visualization are implemented. The proposed algorithm enables the building of a predictive model for primordial and secondary prevention of DME with high classification accuracy and visualization of severity grading.

Diabetic retinopathy detection using red lesion localization and convolutional neural networks

Detecting the early signs of diabetic retinopathy (DR) is essential, as timely treatment might reduce or even prevent vision loss. Moreover, automatically localizing the regions of the retinal image that might contain lesions can favorably assist specialists in the task of detection.In this study, we designed a lesion localization model using a deep network patch-based approach. Our goal was to reduce the complexity of the model while improving its performance. For this purpose, we designed an efficient procedure (including two convolutional neural network models) for selecting the training patches, such that the challenging examples would be given special attention during the training process. Using the labeling of the region, a DR decision can be given to the initial image, without the need for special training.The model is trained on the Standard Diabetic Retinopathy Database, Calibration Level 1 (DIARETDB1) database and is tested on several databases (including Messidor) without any further adaptation. It reaches an area under the receiver operating characteristic curve of 0.912−95%CI(0.897−0.928) for DR screening, and a sensitivity of 0.940−95%CI(0.921−0.959). These values are competitive with other state-of-the-art approaches.

A Novel Approach for Detection of Hard Exudates Using Random Forest Classifier.

Diabetic Retinopathy is the major cause of blindness for diabetics in which the retina is damaged. Regular screening system help in detecting the early symptoms like exudates, which are due to the leakage of blood pressure of vessels. The significant role of proposed system is detecting the hard exudates in prevention of visual loss and blindness. Many researchers studied and investigated about detecting the exudates region but not satisfied with their results. Fundamental medical image processing steps with different techniques are implemented by the proposed system. Random Forest is a novel classification which is applied on color retinal images able to classify cluster of data with high accuracy. The performance of the proposed system is obtained by analyzing the accuracy obtained from the Random Forest classifier. These images are obtained from Diabetic Retinopathy Database (DIARETDB) database. The simulation results are obtained with the help of MATLAB 2018. By applying novel classification techniques improves the automatic detection of hard exudates from color retinal images. The achieved accuracy is compared with existing classifiers since the proposed Random Forest classifier provides the accuracy of 99.89% applied on color retinal images.

Cotton wool spots detection in diabetic retinopathy based on adaptive thresholding and ant colony optimization coupling support vector machine

Diabetic retinopathy is the major issue of diabetes‐induced blindness worldwide but is curable if detected in time. Cotton wool spots (CWSs) are the critical lesions of diabetic retinopathy, which indicate not only advanced nonproliferative but also preproliferative diabetic retinopathy. It is crucial to detect CWSs for grading the severity of diabetic retinopathy. By grading the severity of diabetic retinopathy accurately, the eye specialist can make an effective treatment plan to protect the patient's vision against blindness. CWSs detection remains challenging because of their uneven appearance, in which some CWSs are not clearly visible and some resemble hard exudates. This paper proposed an automatic CWS detection method based on adaptive thresholding and ant colony optimization (ACO) coupled with support vector machine (SVM). One‐hundred and sixty‐two features from five feature sets, namely morphologies, first‐order statistics, gray‐level co‐occurrence matrix, gray‐level run length matrix, and lacunarity, are extracted, and then four feature selection methods,namely genetic algorithm, particle swarm optimization, stepwise method, and ACO, are coupled with SVM classifiers. The evaluation results of the proposed methods on local, standard diabetic retinopathy database calibration level 1, and high‐resolution fundus image database datasets containing 319 images indicate that ACO coupling cubic SVM performs better than the other pairs with sensitivity 90.16%, specificity 97.92%, accuracy 96.96%, and area under receiver operating characteristic curve 97.19%. © 2019 Institute of Electrical Engineers of Japan. Published by John Wiley & Sons, Inc.

A novel diagnostic information based framework for super-resolution of retinal fundus images

Advancements in tele-medicine have led to the development of portable and cheap hand-held retinal imaging devices. However, the images obtained from these devices have low resolution (LR) and poor quality that may not be suitable for retinal disease diagnosis. Therefore, this paper proposes a novel framework for the super-resolution (SR) of the LR fundus images. The method takes into consideration the diagnostic information in the fundus images during the SR process. In this work, SR is performed on the zone of interest of the fundus images. Clinical information of the selected zone is captured using the Shannon entropy, the contrast sensitivity index (CSI), the multi-resolution (MR) intra-band energy and the MR inter-band eigen features. The support vector machine (SVM) classifier is used to decide the clinical significance of the zone. Highly accurate learning based SR method or the bicubic interpolation is applied to the selected zone based on the classification output. The method is tested on the Standard Diabetic Retinopathy Database Calibration level 1 (DIARETDB1) and the Digital Retinal Images for Vessel Extraction (DRIVE) databases. Classification accuracy of 85.22% and 85.77% is achieved for the DIARETDB1 and DRIVE databases respectively. The SR performance of the algorithm is quantified in terms of peak signal to noise ratio (PSNR), structural similarity index measure (SSIM) and computational time. The proposed diagnostic information based SR achieves computational time efficiency without compromising with the high resolution (HR) reconstruction accuracy of the fundus image zones.

United Kingdom Diabetic Retinopathy Electronic Medical Record (UK DR EMR) Users Group: report 4, real-world data on the impact of deprivation on the presentation of diabetic eye disease at hospital services.

AimTo assess the impact of deprivation on diabetic retinopathy presentation and related treatment interventions, as observed within the UK hospital eye service.MethodsThis is a multicentre, national diabetic retinopathy database study with anonymised data extraction across 22 centres from an electronic medical record system. The following were the inclusion criteria: all patients with diabetes and a recorded, structured diabetic retinopathy grade. The minimum data set included, for baseline, age and Index of Multiple Deprivation, based on residential postcode; and for all time points, visual acuity, ETDRS grading of retinopathy and maculopathy, and interventions (laser, intravitreal therapies and surgery). The main outcome measures were (1) visual acuity and binocular visual state, and (2) presence of sight-threatening complications and need for early treatment.Results79 775 patients met the inclusion criteria. Deprivation was associated with later presentation in patients with diabetic eye disease: the OR of being sight-impaired at entry into the hospital eye service (defined as 6/18 to better than 3/60 in the better seeing eye) was 1.29 (95% CI 1.20 to 1.39) for the most deprived decile vs 0.77 (95% CI 0.70 to 0.86) for the least deprived decile; the OR for being severely sight-impaired (3/60 or worse in the better seeing eye) was 1.17 (95% CI 0.90 to 1.55) for the most deprived decile vs 0.88 (95% CI 0.61 to 1.27) for the least deprived decile (reference=fifth decile in all cases). There is also variation in sight-threatening complications at presentation and treatment undertaken: the least deprived deciles had lower chance of having a tractional retinal detachment (OR=0.48 and 0.58 for deciles 9 and 10, 95% CI 0.24 to 0.90 and 0.29 to 1.09, respectively); in terms of accessing treatment, the rate of having a vitrectomy was lowest in the most deprived cohort (OR=0.34, 95% CI 0.19 to 0.58).ConclusionsThis large real-world study suggests that first presentation at a hospital eye clinic with visual loss or sight-threatening diabetic eye disease is associated with deprivation. These initial hospital visits represent the first opportunities to receive treatment and to formally engage with support services. Such patients are more likely to be sight-impaired or severely sight-impaired at presentation, and may need additional resources to engage with the hospital eye services over complex treatment schedules.

Automatic detection and segmentation of optic disc and fovea in retinal images

Feature extraction from retinal images is gaining popularity worldwide as many pathologies are proved having connections with these features. Automatic detection of these features makes it easier for the specialist ophthalmologists to analyse them without spending exhaustive time to segment them manually. The proposed method automatically detects the optic disc (OD) using histogram-based template matching combined with the maximum sum of vessel information in the retinal image. The OD region is segmented by using the circular Hough transform. For detecting fovea, the retinal image is uniformly divided into three horizontal strips and the strip including the detected OD is selected. Contrast of the horizontal strip containing the OD region is then enhanced using a series of image processing steps. The macula region is first detected in the OD strip using various morphological operations and connected component analysis. The fovea is located inside this detected macular region. The proposed method achieves an OD detection accuracy over 95% upon testing on seven public databases and on our locally developed database, University of Auckland Diabetic Retinopathy database (UoA-DR). The average OD boundary segmentation overlap score, sensitivity and fovea detection accuracy achieved are 0.86, 0.968 and 97.26% respectively.

Microaneurysm Detection Using Principal Component Analysis and Machine Learning Methods

Diabetic retinopathy (DR) is an eye abnormality caused by long-term diabetes and it is the most common cause of blindness before the age of 50. Microaneurysms (MAs), resulting from leakage from retinal blood vessels, are early indicators of DR. In this paper, we analyzed MA detectability using small 25 by 25 pixel patches extracted from fundus images in the DIAbetic RETinopathy DataBase - Calibration Level 1 (DIARETDB1). Raw pixel intensities of extracted patches served directly as inputs into the following classifiers: random forest (RF), neural network, and support vector machine. We also explored the use of two techniques (principal component analysis and RF feature importance) for reducing input dimensionality. With traditional machine learning methods and leave-10-patients-out cross validation, our method outperformed a deep learning-based MA detection method, with AUC performance improved from 0.962 to 0.985 and F-measure improved from 0.913 to 0.926, using the same DIARETDB1 database. Furthermore, we validated our method on a different dataset-retinopathy online challenge (ROC) data set. The performance of the three classifiers and the pattern with different percentage of principal components are consistent on the two data sets. Especially, we trained the RF on DIARETDB1 and applied it to ROC; the performance is very similar to that of the RF trained and tested using cross validation on ROC data set. This result indicates that our method has the potential to generalize to different datasets.

Optic Disc Detection from Fundus Photography via Best-Buddies Similarity

Robust and effective optic disc (OD) detection is a necessary processing step in the research work of the automatic analysis of fundus images. In this paper, we propose a novel and robust method for the automated detection of ODs from fundus photographs. It is essentially carried out by performing template matching using the Best-Buddies Similarity (BBS) measure between the hand-marked OD region and the small parts of target images. For well characterizing the local spatial information of fundus images, a gradient constraint term was introduced for computing the BBS measurement. The performance of the proposed method is validated with Digital Retinal Images for Vessel Extraction (DRIVE) and Standard Diabetic Retinopathy Database Calibration Level 1 (DIARETDB1) databases, and quantitative results were obtained. Success rates/error distances of 100%/10.4 pixel and of 97.7%/12.9 pixel, respectively, were achieved. The algorithm has been tested and compared with other commonly used methods, and the results show that the proposed method shows superior performance.

Automatic detection and recognition of optic disk with maker-controlled watershed segmentation and mathematical morphology in color retinal images

Optic disk (OD) detection and recognition is an important stage for developing automatic screening applications of diabetic retinopathy disease in color retinal image. However, the retinal image has a low resolution and was influenced by salt-and-paper noise. Therefore, a retinal image needs a preprocessing procedure (i.e., color image normalization, image enhancement and noise removal) prior to the use of the retinal images. Afterward, a combination of a maker-controlled watershed segmentation and mathematical morphology exiting that was applied to detect of OD. These two methods have complementary drawbacks and advantages, and this is the motivation for the hybrid method presented. These modifications enable the proposed methods to become more robust and accurate to detection of the OD regions. The methods were evaluated by applying to two-color retinal dataset [local dataset in Thailand and a public available diabetic retinopathy database (STARE)]. Although the retinal images in this paper are fairly low, the results showed the proposed method has the performance of the OD detection about 99.33% on 300 images from the local dataset and 95.06% of 81 images from the STARE dataset, taking an average computational time of 3.4 s per image. These results show effectiveness in both detections of the OD regions and boundary.

The UK Diabetic Retinopathy Electronic Medical Record (UK DR EMR) Users Group, Report 2: real-world data for the impact of cataract surgery on diabetic macular oedema

AimTo assess the rate of ‘treatment-requiring diabetic macular oedema (DMO)’ in eyes for the two years before and after cataract surgery.MethodsMulticentre national diabetic retinopathy (DR) database study with anonymised data...

Tsallis entropy and sparse reconstructive dictionary learning for exudate detection in diabetic retinopathy.

Computer-assisted automated exudate detection is crucial for large-scale screening of diabetic retinopathy (DR). The motivation of this work is robust and accurate detection of low contrast and isolated hard exudates using fundus imaging. Gabor filtering is first performed to enhance exudate visibility followed by Tsallis entropy thresholding. The obtained candidate exudate pixel map is useful for further removal of falsely detected candidates using sparse-based dictionary learning and classification. Two reconstructive dictionaries are learnt using the intensity, gradient, local energy, and transform domain features extracted from exudate and background patches of the training fundus images. Then, a sparse representation-based classifier separates the true exudate pixels from false positives using least reconstruction error. The proposed method is evaluated on the publicly available e-ophtha EX and standard DR database calibration level 1 (DIARETDB1) databases and high exudate detection performance is achieved. In the e-ophtha EX database, mean sensitivity of 85.80% and positive predictive value of 57.93% are found. For the DIARETDB1 database, an area under the curve of 0.954 is obtained.