Search

Modern Medical diagnosis has a significant reliance on systematic image analysis for rational prognosis. With the advancement of computer vision methods, supported by deep learning algorithms, the non-invasive and early-stage detection of various diseases is a significant area of research. Amongst various image-based disease identification scopes, in this paper it is focused on Diabetic Retinopathy (DR), which is an eye disease caused by diabetes that can even lead to blindness. Therefore, early detection is critical to prevent visual disturbances from such diseases. In this study, the objective is to identify diabetic retinopathy by analysing fundus images using deep learning methods. Identifying diabetic retinopathy from fundus images, requires considerable improvement of image quality. In this work the retina images obtained by the fundus camera is improved by Generative Adversarial Network(GAN) for specific analysis. This study aimed to detect lesions using mask-RCNN techniques, where pretrained models such as R-50, R-101, and X-101 are utilized for robust segmentation and masking of the damaged regions of the fundus images. As the lesions are pathologically classified into two categories: 'exudates' and 'microaneurysms', similarly the input images are also mapped into two categories. Evaluation is based on the mean average precision (MAP) and the findings infer the conclusion. The comprehensive set of real-life retina images obtained from the clinical dataset benchmarked with the proposed mechanism and it is found that the proper segmentation and improvisation of fundus images by applying GAN have considerably enhanced the degree of accuracy in the Region-of-interest(ROI) bounding box for exact identification of Diabetic Retinopathy, along with reporting on stage of the disease. The aim of the work is well established in implementing the algorithm and finding the result for mapping it into the automated classification of fundus images in batches where the X-101 model seen to perform better in terms of finding Bounding Box Average Precision for both exudates (75.20%) and microaneurysms (67.202%) followed by Segmented Region Average Precision for exudates (62.363%) and microaneurysms (57.690%). The obtained results may reduce human intervention and classify a large number of input images in field-level examination in a faster manner. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy stands out as one of the highly prevalent causes of vision loss in working people worldwide. In computer vision, deep learning based strategies are seen as a viable solution for efficient diabetic retinopathy detection. We present a UNet-based deep learning architecture for diabetic retinopathy segmentation of blood vessels, exudates, and microaneurysms. Traditional methods often consider the features only from the last convolution unit and discard the remaining features, resulting in low-quality feature maps. However, boundary information plays important role in medical image segmentation. To overcome this, we introduce a skip connection mechanism to concatenate all attributes from each layer. Additionally, we utilize an upsampling layer to aggregate the features at the final sigmoid layer. Finally, we apply channel and spatial attention mechanisms to generate the semantic feature map. Therefore, the proposed approach overcomes the issues of existing methods by incorporating dense skip connection along with channel and spatial attention mechanism which helps to retain the substantial information of image. We tested proposed approach on several publicly available datasets such as IDRiD, DIARETDB1, STARE, ChaseDB1, DRIVE, and HRF datasets. The comparative analysis shows that the proposed approach achieves superior performance, with an average accuracy of 98.10%, average sensitivity of 97.60%, and average specificity of 98.2% for segmentation. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy causes blindness in diabetics. Early identification and frequent screening of diabetic retinopathy can slow disease progression and visual loss. Retinal lesions result from diabetic retinopathy. Dark and brilliant retinal lesions predominate. Color, shape, and size distinguish lesions. Exudates are bright, while microaneurysms (MAs) and hemorrhages (HEMs) are dark. This study presents a retinal lesion screening method for diabetic retinopathy. The data is saturated at low and high intensities; picture intensity values are adjusted to enhance contrast. This study presents a unique rhombus multilevel retinal image segmentation method. In the proposed study, preprocessing, segmentation algorithms, morphological operation,median filter and gradient are all designed as parts of an effective automated system. With 40 photos, the recommended method produced accuracy and specificity of 99.9% and 99.5%, respectively. [ABSTRACT FROM AUTHOR]

Introduction. Diabetic retinopathy is a complication of diabetes mellitus caused by high blood sugar levels damaging the retina. Diabetic retinopathy leads to changes in ocular blood vessels and the appearance of solid exudates and microaneurysms. When diagnosed and treated in the late stages, this disease can cause blindness. The most common diagnostic method for diabetic retinopathy is based on ocular fundus imaging. However, the background interference and low contrast of such images significantly hinders the timely detection of vascular lesions. Therefore, the development of a method for detecting signs of diabetic retinopathy, particularly in its early stages, presents a relevant research task.Aim. Development of a method for diagnosing diabetic retinopathy based on an analysis of ocular fundus images using the decision-tree approach.Materials and methods. Methods based on image segmentation with identifying characteristic features and their binary classification were used. A verified database was used to access the accuracy of the proposed method for detecting diabetic retinopathy.Results. A method for detecting signs of diabetic retinopathy was developed, which includes the segmentation of vessels, exudates and microaneurysms based on digital processing of ocular vascular images using binary classification. The developed method showed a high level of diagnostic accuracy. Thus, the sensitivity, specificity and accuracy of diabetic retinopathy detection comprised 87.14, 88.50 and 87.81 %, respectively.Conclusion. The developed method allows diabetic retinopathy to be diagnosed with sufficiently high accuracy. The method can also be used for supporting decision making when managing patients with diabetic retinopathy.

Diabetic retinopathy (DR) is a leading cause of visual blindness. However if DR can be diagnosed and treated early, 90% of DR causing blindness can be prevented significantly. Microaneurysms (MAs) and exudates (EXs), as signs of DR, can be used for early DR diagnosis. However, MAs and EXs segmentation is a challenging task due to the low contrast of the lesions, the interference of noises, and the imbalance between the lesion areas and the background. In this paper, an enhanced residual U-Net (ERU-Net) for MAs and EXs segmentation is proposed. ERU-Net obtains three U-paths, which are composed by three upsampling paths together with one downsampling path. With such three U-paths structure, ERU-Net can enhance the corresponding features fusion and capture more details of fundus images. Also, a residual block is constructed in ERU-Net to extract more representative features. In the experiments, we evaluate the performance of ERU-Net for MAs and EXs segmentation on three public datasets, E-Ophtha, IDRiD, and DDR. The ERU-Net obtains the AUC values of 0.9956, 0.9962, 0.9801, 0.9866, 0.9679, 0.9609 for MAs and EXs segmentation on these three datasets, respectively, which are greater than that of the original U-Net. Compared with some traditional methods, convolutional neural networks and other recent U-Nets, ERU-Net also performs competitively. Besides, we have applied ERU-Net to segment optic disc (OD) on the DRISHTI-GS1 dataset, achieving the highest Jaccard index of 0.994 compared with the existing methods. The numerical results indicate that ERU-Net is a promising network for medical image segmentation.

This article proposes two methodologies for the detection of lesions in the retina, which may indicate the presence of diabetic retinopathy (DR). Through the use of digital image processing techniques, it is possible to isolate the pixels that correspond to a lesion of RD, to achieve segmenting microaneurysms, the edges of the objects contained in the image are highlighted in order to detect the contours of the objects to select by size those that meet an area of 15 to 25 pixels in the case of 512x512 images and identify the objects as possible microaneurysms, while for the detection of exudates the green channel is selected to contrast the luminous objects in the retinography and from the conversion to gray scale, a histogram is graphed to identify the ideal threshold for the segmentation of the pixels that belong to the exudates at the end of the optical disk previously identified by a specialist. A confusion matrix supervised by an ophthalmologist was created to quantify the results obtained by the two methodologies, obtaining a specificity of 0.94 and a sensitivity of 0.97, values that are outstanding to proceed with the classification stage. [ABSTRACT FROM AUTHOR]

Diabetic Retinopathy (DR) is a type of disease in eyes as a result of a diabetic condition that ends up damaging the retina, leading to blindness or loss of vision. Morphological and physiological retinal variations involving slowdown of blood flow in the retina, elevation of leukocyte cohesion, basement membrane dystrophy, and decline of pericyte cells, develop. As DR in its initial stage has no symptoms, early detection and automated diagnosis can prevent further visual damage. In this research, using a Deep Neural Network (DNN), segmentation methods are proposed to detect the retinal defects such as exudates, hemorrhages, microaneurysms from digital fundus images and then the conditions are classified accurately to identify the grades as mild, moderate, severe, no PDR, PDR in DR. Initially, saliency detection is applied on color images to detect maximum salient foreground objects from the background. Next, structure tensor is applied powerfully to enhance the local patterns of edge elements and intensity changes that occur on edges of the object. Finally, active contours approximation is performed using gradient descent to segment the lesions from the images. Afterwards, the output images from the proposed segmentation process are subjected to evaluate the ratio between the total contour area and the total true contour arc length to label the classes as mild, moderate, severe, No PDR and PDR. Based on the computed ratio obtained from segmented images, the severity levels were identified. Meanwhile, statistical parameters like the mean and the standard deviation of pixel intensities, mean of hue, saturation and deviation clustering, are estimated through K-means, which are computed as features from the output images of the proposed segmentation process. Using these derived feature sets as input to the classifier, the classification of DR was performed. Finally, a VGG-19 deep neural network was trained and tested using the derived feature sets from the KAGGLE fundus image dataset containing 35,126 images in total. The VGG-19 is trained with features extracted from 20,000 images and tested with features extracted from 5,000 images to achieve a sensitivity of 82% and an accuracy of 96%. The proposed system was able to label and classify DR grades automatically. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy (DR) is a leading cause of preventable blindness caused by damaged blood vessels in the eye, if not treated early on. The aim of this research work was to develop a method for the automatic detection of Diabetic Retinopathy and proposing a model for deciding the progression/severity using fundus images. The method was developed so that DR can be detected in an effective and efficient manner before causing damage to the eye, without the presence of an ophthalmologist. The manual screening requires the presence of an ophthalmologist and the resource of time. Detecting exudates is important for the diagnosis of DR. The approach adopted was two-fold: i. extracting features of interest from the images i.e. the blood vessels, optic disc (OD), exudates and microaneurysms by using morphological operations and ii. classifying its progression/severity as either mild or moderate by using the support vector machine (SVM) classifier for helping Ophthalmologists. The performance of the proposed method has been assessed by an ophthalmologist and approved. This paper contributes towards the field of automatic detection of anomalous structures and their severity. [ABSTRACT FROM AUTHOR]

Purpose: To describe the distribution of Type 2 DM retinal lesions and determine whether it is symmetrical between the two eyes, is random or follows a certain pattern. Methods: Cross‐sectional study of Type 2 DM patients who had been referred for an outpatients' ophthalmology visit for diabetic retinopathy screening in primary health care. Retinal photographic images were taken using central projection non‐mydriatic retinography. The lesions under study were microaneurysms/haemorrhages, and hard and soft exudates. The lesions were placed numerically along the x‐ and y‐axes obtained, with the fovea as the origin. Results: From among the 94 patients included in the study, 4770 lesions were identified. The retinal lesions were not distributed randomly, but rather followed a determined pattern. The left eye exhibited more microaneurysms/haemorrhages and hard exudates of a greater density in the central retina than was found in the right eye. Furthermore, more cells containing lesions were found in the upper temporal quadrants, (especially in the left eye), and tended to be more central in the left eye than in the right, while the hard exudates were more central than the microaneurysms/haemorrhages. Conclusion: The distribution of DR lesions is neither homogeneous nor random but rather follows a determined pattern for both microaneurysms/haemorrhages and hard exudates. This distribution means that the areas of the retina most vulnerable to metabolic alteration can be identified. The results may be useful for automated DR detection algorithms and for determining the underlying vascular and non‐vascular physiopathological mechanisms that can explain these differences. [ABSTRACT FROM AUTHOR]

Diabetic mellitus is a major reason of visual impairment an around the world. Early automatic diagnosis of diabetic retinopathy (DR) may avoid vision loss and blindness. The goal of this paper is to automatically detect retinal image as Non DR or DR based on radial basis function (RBF) neural network classifier. This experiment address to explore ophthalmic features such as blood vessels, exudates & microaneurysms and it's segmented from retinal background using A-IFS histon based segmentation method. This obtained feature set delivers to train RBF neural network. The Receiver operation characteristics (ROC) curve is plotted based on evaluated result. The projected experiment has been done on 130 DIARETDB0 & 89 DIARETDB1 retinal images database by using RBF neural network. The experiment perceive the accuracy of 71.2%, Sensitivity 0.83 & Specificity 0.043 for DIARETDB0 and the accuracy of 89.4% Sensitivity 0.94 & Specificity 0.16 for DIARETDB1. [ABSTRACT FROM AUTHOR]

Estimated blind people in the world will exceed 40 million by 2025. To develop novel algorithms based on fundus image descriptors that allow the automatic classification of retinal tissue into healthy and pathological in early stages is necessary. In this paper, we focus on one of the most common pathologies in the current society: diabetic retinopathy. The proposed method avoids the necessity of lesion segmentation or candidate map generation before the classification stage. Local binary patterns and granulometric profiles are locally computed to extract texture and morphological information from retinal images. Different combinations of this information feed classification algorithms to optimally discriminate bright and dark lesions from healthy tissues. Through several experiments, the ability of the proposed system to identify diabetic retinopathy signs is validated using different public databases with a large degree of variability and without image exclusion.

Objective: Diabetic retinopathy (DR) is characterized by the progressive deterioration of retina with the appearance of different types of lesions that include microaneurysms, hemorrhages, exudates, etc. Detection of these lesions plays a significant role for early diagnosis of DR. Methods: To this aim, this paper proposes a novel and automated lesion detection scheme, which consists of the four main steps: vessel extraction and optic disc removal, preprocessing, candidate lesion detection, and postprocessing. The optic disc and the blood vessels are suppressed first to facilitate further processing. Curvelet-based edge enhancement is done to separate out the dark lesions from the poorly illuminated retinal background, while the contrast between the bright lesions and the background is enhanced through an optimally designed wideband bandpass filter. The mutual information of the maximum matched filter response and the maximum Laplacian of Gaussian response are then jointly maximized. Differential evolution algorithm is used to determine the optimal values for the parameters of the fuzzy functions that determine the thresholds of segmenting the candidate regions. Morphology-based postprocessing is finally applied to exclude the falsely detected candidate pixels. Results and Conclusions: Extensive simulations on different publicly available databases highlight an improved performance over the existing methods with an average accuracy of $97.71\%$ and robustness in detecting the various types of DR lesions irrespective of their intrinsic properties. [ABSTRACT FROM PUBLISHER]

Non-proliferative Diabetic Retinopathy is a health problem which is prevalent in diabetic patients. It is caused by the expansion of untwisted blood vessels that over time, result in microaneurysms, haemorrhages, and exudates in the retina. This is the preliminary reason for visual impairment and eventual blindness in adults. These retinal anomalies can be identified using fundus images. This paper proposes a novel approach for classifying microaneurysms and exudates of Nonproliferative Diabetic Retinopathy (NPDR) patients, using digital image processing and support vector machine. The datasets used for the study were gathered from Vision Care (Pvt) Ltd, National Eye Hospital and publicly available catalogues such as DRIVE and STARE. User level evaluation conducted with 40 candidates showed a 100% and 84.2% success rates for detecting NPDR retinal abnormalities and exudates respectively. In addition to that through evaluation results indicate that the proposed approach microaneurysms lesion identification yields much better accuracy than the exudates. [ABSTRACT FROM AUTHOR]

The presence of microaneurysms (MAs) is usually the early sign of diabetic retinopathy (DR) disease. In this study, an effective contrast enhancement scheme has been incorporated with a diabetic screening system to improve the overall quality of MAs identification technique. The proposed scheme is compared with the existing retinal image contrast enhancement methods and a better result is established. Public datasets and also low-contrast noisy images collected from hospitals are used for MA's detection, and satisfactory results are found. The results on low-contrast noisy image diagnosis established the significance of the study. A total of 587 (training = 340, testing = 247) retinal images are examined. The mean sensitivity and specificity are scored as 95.94% and 99.21%, respectively. The value of AUC is 0.932 (SD = 0.013) from the ROC analysis. [ABSTRACT FROM PUBLISHER]

Diabetic retinopathy (DR) is a leading cause of vision loss, caused by damage to the retina from complications of diabetes. Analysis of the retinal photographs for key characteristics of DR can result in early diagnosis and better management of DR. This paper presents a method for automated analysis and classification of the retina as DR or non-DR using two-field mydriatic fundus photography. The optic disc region is located by multi-level wavelet decomposition and recursive region growing from an automatically identified seed point. Blood vessels are extracted by applying histogram analysis on the two median filtered images. Red lesions are detected using three stage intensity transformation and white lesions from multi-level histogram analysis. The final classification of the retina as DR or non-DR is based on an aggregate of the lesions extracted from each image. The proposed method has been validated against diagnosis by a panel of expert ophthalmologists on images from 368 patients. The observed sensitivity and specificity were 80% and 50% respectively. The results show that automated screening based on two-field photography can be applied in routine screening. [ABSTRACT FROM AUTHOR]

Abstract: Microaneurysm (MA), small dark round dots on retinal fundus image is the earliest clinical sign of Diabetic Retinopathy (DR) disease. MA detection at early stage can help to reduce the blindness. In such cases the retinal fundus images produced by fluorescent oscilloscope are often noisy and low in contrast. Detecting the Microaneurysms using those fundus images is very difficult for ophthalmologist. In the present paper, we propose a method using location based contrast enhancement process, popularly known as Contrast Limited Adaptive Histogram Equalization (CLAHE) for the detection of retinal changes in DR images. CLAHE is an adaptive extension of Histogram Equalization which helps in dynamic preservation of the local contrast characteristics of an image. The proposed algorithm divides the retinal fundus image into a number of small, non-overlapping contextual tiles. Following CLAHE at each tile separately, median filtering of DR images is carried out in order to smooth the background noise. Results of the proposed algorithm show a considerable improvement in the enhancement of DR image quality. The average sensitivity and specificity of this Diabetes Screening System revealed as 82.64% and 99.98% respectively. [Copyright &y& Elsevier]

Diabetes mellitus is a heterogeneous clinical syndrome characterized by hyperglycaemia and the long-term complications are retinopathy, neuropathy, nephropathy, and cardiomyopathy. It is a leading cause of blindness. Diabetic retinopathy is the progressive pathological alterations in the retinal microvasculature, leading to areas of retinal nonperfusion, increased vascular permeability, and the pathological proliferation of retinal vessels. Hence, it is beneficial to have regular cost-effective eye screening for diabetes subjects. Nowadays, different stages of diabetes retinopathy are detected by retinal examination using indirect biomicroscopy by senior ophthalmologists. In this work, morphological image processing and support vector machine (SVM) techniques were used for the automatic diagnosis of eye health. In this study, 331 fundus images were analysed. Five groups were identified: normal retina, mild non-proliferative diabetic retinopathy, moderate non-proliferative diabetic retinopathy, severe non-proliferative diabetic retinopathy, and proliferative diabetic retinopathy. Four salient features blood vessels, microaneurysms, exudates, and haemorrhages were extracted from the raw images using image-processing techniques and fed to the SVM for classification. A sensitivity of more than 82 per cent and specificity of 86 per cent was demonstrated for the system developed. [ABSTRACT FROM AUTHOR]