Search

Diabetic retinopathy (DR) is one of the leading causes of blindness in a significant portion of the working population, and its damage on vision is irreversible. Therefore, rapid diagnosis on DR is crucial for saving the patient's eyesight. Since Transformer shows superior performance in the field of computer vision compared with Convolutional Neural Networks (CNNs), it has been proposed and applied in computer aided diagnosis of DR. However, a large number of images should be used for training due to the lack of inductive bias in Transformers. It has been demonstrated that the retinal vessels follow self-similar fractal scaling law, and the fractal dimension of DR patients shows an evident difference from that of normal people. Based on this, the fractal dimension is introduced as a prior into Transformers to mitigate the adverse influence of lack of inductive bias on model performance. A new Transformer method pretrained with Masked Autoencoders and fractal dimension (MAEFD) is developed and proposed in this paper. The experiments on the APTOS dataset show that the classification performance for DR by the proposed MAEFD can be substantially improved. Additionally, the present model pretrained with 100,000 retinal images outperforms that pretrained with 1 million natural images in terms of DR classification performance. [ABSTRACT FROM AUTHOR]

Retinal blood vessel segmentation is a valuable tool for clinicians to diagnose conditions such as atherosclerosis, glaucoma, and age-related macular degeneration. This paper presents a new framework for segmenting blood vessels in retinal images. The framework has two stages: a multi-layer preprocessing stage and a subsequent segmentation stage employing a U-Net with a multi-residual attention block. The multi-layer preprocessing stage has three steps. The first step is noise reduction, employing a U-shaped convolutional neural network with matrix factorization (CNN with MF) and detailed U-shaped U-Net (D_U-Net) to minimize image noise, culminating in the selection of the most suitable image based on the PSNR and SSIM values. The second step is dynamic data imputation, utilizing multiple models for the purpose of filling in missing data. The third step is data augmentation through the utilization of a latent diffusion model (LDM) to expand the training dataset size. The second stage of the framework is segmentation, where the U-Nets with a multi-residual attention block are used to segment the retinal images after they have been preprocessed and noise has been removed. The experiments show that the framework is effective at segmenting retinal blood vessels. It achieved Dice scores of 95.32, accuracy of 93.56, precision of 95.68, and recall of 95.45. It also achieved efficient results in removing noise using CNN with matrix factorization (MF) and D-U-NET according to values of PSNR and SSIM for (0.1, 0.25, 0.5, and 0.75) levels of noise. The LDM achieved an inception score of 13.6 and an FID of 46.2 in the augmentation step. [ABSTRACT FROM AUTHOR]

Eye diseases that are common and many diseases that result in visual ailments, such as diabetes and vascular disease, can be diagnosed through retinal imaging. The enhancement of retinal images often helps in diagnosing diseases related to retinal organ failure. However, today's image enhancement methods may lead to artificial boundaries, sudden color gradation, and the loss of image details. Therefore, to prevent these side effects, a new method of retinal image enhancement is proposed. In this work, we propose a new method for enhancing the overall contrast of colored retinal images. That is, we propose low-light image enhancement using a new retinex method based on a powerful semidecoupled retinex method. In particular, illumination layer I gradually approximates the S input image according to the file. This leads to a complete Gaussian transformation model, while the R-layer reflectance is estimated jointly by S and intermediary by I to suppress image noise simultaneously during R estimation on the publicly available Messidor database. From our assessment measurements (PSNR and SSIM), we show that this proposed method is effective in comparison with the relevant and recently proposed retinal imaging methods; moreover, the color, which is determined by the data, does not change the image structure. Finally, a technique is presented to improve the pronounced color of a retinal image, which is useful for ophthalmologists to screen for retinal disease more effectively. Moreover, this technique can be used in the development of robotics for imaging tests to search for clinical markers. [ABSTRACT FROM AUTHOR]

Retinal images play a crucial role in the clinical diagnosis and detection of many eye diseases. However, the retinal image is degraded with noise and suffers from low contrast, which makes it difficult to interpret the image precisely and accurately. So retinal image enhancement becomes a vital step before any further processing. Here, a new approach is suggested for fundus image enhancement, where the image is processed in the spatial domain through an optimal statistical feature‐based transformation function (SFTF). The proposed SFTF has four tunable parameters. An optimal set of these parameters is obtained through a newly suggested adaptive enhanced leader particle swarm optimization technique. The suggested method is investigated with freely accessible DRIVE, STARE, and IDRiD databases. The efficiency of the suggested technique is assessed through different validation indices. The results reveal superior performance compared with other popular methods used for vessel enhancement. [ABSTRACT FROM AUTHOR]

Decreased vision acuity and blindness due to retinal detachment (RD) are significant concerns, emphasizing the importance of early diagnosis and identification. However, manual screening of RD is labor-intensive and time-consuming and faces challenges such as poor quality and low accuracy. A novel hybrid machine learning(ML) algorithm incorporating Levy flight-based atom search optimization (LFB-ASO) is proposed to solve the above challenges. The dataset utilized for the experiment is the Retinal Fundus Multi-disease Image dataset (RFMiD). The data preprocessing pipeline involves image resizing, normalization, data augmentation, masking and segmentation. To ensure consistent dimensions, all retinal images are standardized through resizing. Performance and convergence are improved using normalization. The data augmentation technique enhances diversity, while segmentation focuses on the region of interest (ROI). Then the deep features are extracted from the preprocessed retinal images using a pre-trained ResNet18 model. LFB-ASO is employed to select the most discriminative deep features for RD classification. To achieve superior accuracy, hybrid ML algorithms, namely Support Vector Machine (SVM), Gradient Boosting Machine (GBM) and Random Forest (RF) are employed. The proposed model achieves remarkable results with accuracy, recall, F1 score and precision of 98.75%, 96.70%, 97.01% and 97.62%. These results outperform existing methods such as HOS-LSDA, LR, NB, PCA and RD-Light-Net. [ABSTRACT FROM AUTHOR]

In this paper, we have proposed a systematic retinal image enhancement and classification method. The proposed method deals with balancing all the visual and technical aspects of the image for glaucoma diagnosis. Initially, similar 3D image blocks are obtained for each retinal image using novel block-matching and grouping techniques. The proposed enhancement technique emphasizes these blocks, which constitute careful estimation of low frequency for each 3D block followed by enhancement using the new alpha-rooting method with adaptive alpha value. During this, the image may over-enhance in some areas, which can be corrected through the image polishing phase that uses cumulative distribution function (CDF) transformations. The enhanced retinal images are qualitatively compared with the outcomes of some existing methods and are employed in glaucoma classification using principal component analysis (PCA) and its variants using discrete wavelet transformations (DWT). We have carried out a deep investigation to find the best combination of DWT and PCA variants. The results obtained from the implementation proved that the performance of the proposed method is highly satisfactory. [ABSTRACT FROM AUTHOR]

The performance of diagnostic Computer-Aided Design (CAD) systems for retinal diseases depends on the quality of the retinal images being screened. Thus, many studies have been developed to evaluate and assess the quality of such retinal images. However, most of them did not investigate the relationship between the accuracy of the developed models and the quality of the visualization of interpretability methods for distinguishing between gradable and non-gradable retinal images. Consequently, this paper presents a novel framework called "FGR-Net" to automatically assess and interpret underlying fundus image quality by merging an autoencoder network with a classifier network. The FGR-Net model also provides an interpretable quality assessment through visualizations. In particular, FGR-Net uses a deep autoencoder to reconstruct the input image in order to extract the visual characteristics of the input fundus images based on self-supervised learning. The extracted features by the autoencoder are then fed into a deep classifier network to distinguish between gradable and ungradable fundus images. FGR-Net is evaluated with different interpretability methods, which indicates that the autoencoder is a key factor in forcing the classifier to focus on the relevant structures of the fundus images, such as the fovea, optic disk, and prominent blood vessels. Additionally, the interpretability methods can provide visual feedback for ophthalmologists to understand how our model evaluates the quality of fundus images. The experimental results showed the superiority of FGR-Net over the state-of-the-art quality assessment methods, with an accuracy of > 89 % and an F1-score of > 87 %. The code is publicly available at https://github.com/saifalkh/FGR-Net. • FGR-NET is a deep learning model to assess fundus image gradeability. • FGR-NET consists of two parallel deep networks: autoencoder and classifier. • Using self-supervised learning, autoencoder is to extract the images' key features. • A classifier is to distinguish between gradable and ungradable fundus images. • Interpretability is performed to understand how FGR-NET evaluates the gradeability. [ABSTRACT FROM AUTHOR]

Dust particle artifacts are present in all imaging modalities but have more adverse consequences in medical images like retinal images. They could be mistaken as small lesions, such as microaneurysms. We propose a method for detecting and accurately segmenting dust artifacts in retinal images based on multi-scale template-matching on several input images and an iterative segmentation via an inpainting approach. The inpainting is done through dictionary learning and sparse-based representation. The artifact segmentation is refined by comparing the original image to the initial restoration. On average, 90% of the dust artifacts were detected in the test images, with state-of-theart restoration results. All detected artifacts were accurately segmented and removed. Even the most challenging artifacts located on top of blood vessels were removed. Thus, ensuring the continuity of the retinal structures. The proposed method successfully detects and removes dust artifacts in retinal images, which could be used to avoid false-positive lesion detections or as an image quality criterion. An implementation of the proposed algorithm can be accessed and executed through a Code Ocean compute capsule. [ABSTRACT FROM AUTHOR]

In Mexico, chronic degenerative diseases are the leading cause of morbidity, which has frequent retina complications, being the leading cause of blindness in our population. Unfortunately, the detection of the pathology is usually late, resulting in more significant disability. To propose the detection of different pathologies with different artificial intelligence algorithms have been used for the images taken from the fundus of the eye. Objective. Evaluate different machine learning algorithms for the detection of retinal alterations in the Mexican population. Methodology. Evaluate two types of models to estimate artificial intelligence tools' screening capacity, one based on transfer learning and ensemble methods against one based only on convolutional networks. Results. We obtained good values to differentiate between healthy and sick but not to diagnose different pathologies. Conclusions. It is necessary to enlarge the imaging sample and to improve the screening models. [ABSTRACT FROM AUTHOR]

Purpose: To compare the quality of images viewed through an extended depth of focus (EDF) intraocular lens (IOL) to that through multifocal or monofocal IOL. Study design: Experimental study Methods: EDF IOL (Symfony®, ZXR00V, Johnson & Johnson Vision), diffractive multifocal (ZLB00, + 3.25D or ZMB00, + 4.0D), or monofocal (ZCB00V) IOL with a spherical power of + 20.0 diopter (D) was placed in a fluid-filled model eye with corneal aberrations similar to those of human eyes. A United States Air Force Resolution Grating Target was glued to the posterior surface of the model eye and viewed through a flat contact lens, a 60D or 128D wide-angle non-contact lens (Resight®) or wide-angle contact lens (MiniQuad®). The contrast of the grating images recorded with the EDF and multifocal IOLs were compared to those through the monofocal IOL. Results: The grating images viewed through the flat contact lens were slightly blurred when viewed through the EDF IOL but clearer than those through the multifocal IOLs with very blurred images in the periphery. The contrast of the images viewed through the EDF and multifocal IOLs through the flat contact lens was significantly lower than through the monofocal IOL (P < 0.02). The contrast of the images viewed through the EDF IOL with 60D or 128D wide-angle non-contact lens was significantly lower than through the monofocal IOL (P < 0.05) but not with wide-angle contact lens. Conclusion: Our results suggest that vitreous surgeons can accomplish a clearer view during vitrectomy in EDF IOL-implanted eyes with a wide-angle viewing contact lens and a flat contact lens than in multifocal IOL-implanted eyes. [ABSTRACT FROM AUTHOR]

Fundus images are broadly used by medical ophthalmologists to detect and assess any customary abnormalities. Fundus imaging sensors capture the eye's rigid portion, which characteristically covers the core, tangential retina, optic disc, and macula. Existing state‐of‐the‐art fundus sensors have the drawback of producing low contrast and noisy information, which makes scientific and algorithmic evaluation very complicated. This article proposes an Adaptive Histogram Equalization—Tuned with Nonsimilar Grouping Curvelet (HET‐NOSCU), which works through a joint denoising enhancement approach. The proposed algorithm's main contribution includes (i) use of curvelet features to better preserve edges during denoising. (ii) Adaptive enhancement using the histogram to prevent halo ringing and specular artifacts, which yields superior results than the very recently established state‐of ‐the‐art methods, using similar performing parameters such as peak signal to noise ratio (PSNR), structural similarity index (SSIM), and correlation coefficient (CoC). We observe an improvement of 17.66%, 0.93%, and 0.24%, respectively, for the above parameters. [ABSTRACT FROM AUTHOR]

In this work, we present a new methodology to simultaneously segment anatomical structures in medical images. Additionally, this methodology is instantiated in a method that is used to simultaneously segment the optic disc (OD) and fovea in retinal images. The OD and fovea are important anatomical structures that must be previously identified in any image-based computer-aided diagnosis system dedicated to diagnosing retinal pathologies that cause vision problems. Basically, the simultaneous segmentation method uses an OD-fovea model and an evolutionary algorithm. On the one hand, the model is built using the intra-structure relational knowledge, associated with each structure, and the inter-structure relational knowledge existing between both and other retinal structures. On the other hand, the evolutionary algorithm (differential evolution) allows us to automatically adjust the instance parameters that best approximate the OD-fovea model in a given retinal image. The method is evaluated in the MESSIDOR public database. Compared with other recent segmentation methods in the related literature, competitive segmentation results are achieved. In particular, a sensitivity and specificity of 0.9072 and 0.9995 are respectively obtained for the OD. Considering a success when the distance between the detected and actual center is less than or equal to η times the OD radius, the success rates obtained for the fovea are 97.3% and 99.0% for η = 1 / 2 and η = 1 , respectively. The segmentation average time per image is 29.35 s. [ABSTRACT FROM AUTHOR]

In order to improve the security of images exchanged in telemedicine, we propose in this paper, 2 watermarking schemes for retinal image protection; each scheme will be declined into 2 variants. The integration of the watermark will be performed by combining the parity of the successive values; each variant will represent a different combination. These approaches will be implemented in the three insertion domains: spatial, frequency and multi-resolution domain. For the spatial domain, the watermark will be integrated into the R, G and B values of the image. In the frequency domain, the watermark bits will be substituted to the DCT coefficient's least significant bit. For the multi-resolution domain insertion, after calculating a DWT, the obtained LL sub-band coefficients will be used for the integration process. After comparing our approaches to the different recent works in the three domains; the obtained results demonstrate that our first proposed approach offers a good imperceptibility for the frequency and spatial domains insertion. However, using 512 × 512 images for our experiments considerably reduces the capacity of our approach in the frequency domain. [ABSTRACT FROM AUTHOR]

Medical image registration can be used for studying longitudinal and cross-sectional data, quantitatively monitoring disease progression and guiding computer assisted diagnosis and treatments. However, deformable registration which enables more precise and quantitative comparison has not been well developed for retinal optical coherence tomography (OCT) images. This paper proposes a new 3D registration approach for retinal OCT data called OCTRexpert. To the best of our knowledge, the proposed algorithm is the first full 3D registration approach for retinal OCT images which can be applied to longitudinal OCT images for both normal and serious pathological subjects. In this approach, a pre-processing method is first performed to remove eye motion artifact and then a novel design-detection-deformation strategy is applied for the registration. In the design step, a couple of features are designed for each voxel in the image. In the detection step, active voxels are selected and the point-to-point correspondences between the subject and template images are established. In the deformation step, the image is hierarchically deformed according to the detected correspondences in multi-resolution. The proposed method is evaluated on a dataset with longitudinal OCT images from 20 healthy subjects and 4 subjects diagnosed with serious Choroidal Neovascularization (CNV). Experimental results show that the proposed registration algorithm consistently yields statistically significant improvements in both Dice similarity coefficient and the average unsigned surface error compared with the other registration methods. [ABSTRACT FROM AUTHOR]

Proper contrast and sufficient illuminance are important in clearly identifying the retinal structures, while the required quality cannot always be guaranteed due to major reasons like acquisition process and diseases. To ensure the effectiveness of enhancement, two solutions are developed for blurry retinal images with sufficient illuminance and insufficient illuminance, respectively. The proposed contrast stretching and intensity transfer are main steps in both of the two solutions. The contrast stretching is based on base-intensity removal and non-uniform addition. We assume that a base-intensity exists in an image, which mainly supports the basic illuminance but has less contribution to texture information. The base-intensity is estimated by the constrained Gaussian function and then removed. The non-uniform addition using compressed Gamma map is further developed to improve the contrast. Additionally, an effective intensity transfer strategy is introduced, which can provide required illuminance for a single channel after contrast stretching. The color correction can be achieved if the intensity transfer is performed on three channels. Results show that the proposed solutions can effectively improve the contrast and illuminance, and good visual perception for quality degraded retinal images is obtained. Illustration of contrast stretching based on a signal colour channel. [ABSTRACT FROM AUTHOR]

Cataract is the clouding of lens, which affects vision and it is the leading cause of blindness in the world’s population. Accurate and convenient cataract detection and cataract severity evaluation will improve the situation. Automatic cataract detection and grading methods are proposed in this paper. With prior knowledge, the improved Haar features and visible structure features are combined as features, and multilayer perceptron with discrete state transition (DST-MLP) or exponential DST (EDST-MLP) are designed as classifiers. Without prior knowledge, residual neural networks with DST (DST-ResNet) or EDST (EDST-ResNet) are proposed. Whether with prior knowledge or not, our proposed DST and EDST strategy can prevent overfitting and reduce storage memory during network training and implementation, and neural networks with these strategies achieve state-of-the-art accuracy in cataract detection and grading. The experimental results indicate that combined features always achieve better performance than a single type of feature, and classification methods with feature extraction based on prior knowledge are more suitable for complicated medical image classification task. These analyses can provide constructive advice for other medical image processing applications. [ABSTRACT FROM AUTHOR]

Registration of retinal images is significant for clinical diagnosis. Numerous methods have been proposed to evaluate registration performance. The available evaluation methods can work well in normal image pairs, but fair evaluation cannot be obtained for image pairs with anatomical changes. We propose an automatic method to quantitatively assess the registration of retinal images based on the extraction of similar vessel structures and modified Hausdorff distance. Firstly, vessel detection and skeletonization are performed to detect the vascular centerline. Secondly, the vessel segments having similar structures in the image pair are selected for assessment of registration. The bifurcation and terminal points are determined from the vascular centerline. Then, the Hungarian matching algorithm with a pruning process is employed to match the bifurcation and terminal points to detect similar vessel segments. Finally, a modified Hausdorff distance is employed to evaluate the performance of registration. Our experimental results show that the Pearson product-moment correlation coefficient can reach 0.76 and 0.63 in test set of normal image pairs and image pairs with anomalies respectively, which outperforms other methods. An accurate evaluation can not only compare the performance of different registration methods but also can facilitate the clinical diagnosis by screening out the inaccurate registration. Graphical abstract . [ABSTRACT FROM AUTHOR]

Background and objective: The detection of abnormalities such as lesions or leakage from retinal images is an important health informatics task for automated early diagnosis of diabetic and malarial retinopathy or other eye diseases, in order to prevent blindness and common systematic conditions. In this work, we propose a novel retinal lesion detection method by adapting the concepts of saliency. Methods: Retinal images are first segmented as superpixels, two new saliency feature representations: uniqueness and compactness, are then derived to represent the superpixels. The pixel level saliency is then estimated from these superpixel saliency values via a bilateral filter. These extracted saliency features form a matrix for low‐rank analysis to achieve saliency detection. The precise contour of a lesion is finally extracted from the generated saliency map after removing confounding structures such as blood vessels, the optic disk, and the fovea. The main novelty of this method is that it is an effective tool for detecting different abnormalities at the pixel level from different modalities of retinal images, without the need to tune parameters. Results: To evaluate its effectiveness, we have applied our method to seven public datasets of diabetic and malarial retinopathy with four different types of lesions: exudate, hemorrhage, microaneurysms, and leakage. The evaluation was undertaken at the pixel level, lesion level, or image level according to ground truth availability in these datasets. Conclusions: The experimental results show that the proposed method outperforms existing state‐of‐the‐art ones in applicability, effectiveness, and accuracy. [ABSTRACT FROM AUTHOR]

Oculopathy is a widespread disease among people of all ages around the world. Teleophthalmology can facilitate the ophthalmological diagnosis for less developed countries that lack medical resources. In teleophthalmology, the assessment of retinal image quality is of great importance. In this paper, we propose a no-reference retinal image assessment system based on DenseNet, a convolutional neural network architecture. This system classified fundus images into good and bad quality or five categories: adequate, just noticeable blur, inappropriate illumination, incomplete optic disc, and opacity. The proposed system was evaluated on different datasets and compared to the applications based on other two networks: VGG-16 and GoogLenet. For binary classification, the good-and-bad binary classifier achieves an AUC of 1.000, and the degradation-specified classifiers that distinguish one specified degradation versus the rest achieve AUC values of 0.972, 0.990, 0.982, 0.982 for four categories, respectively. The multi-classification based on DenseNet achieves an overall accuracy of 0.927, which is significantly higher than 0.549 and 0.757 obtained using VGG-16 and GoogLeNet, respectively. The experimental results indicate that the proposed approach produces outstanding performance in retinal image quality assessment and is worth applying in ophthalmological telemedicine applications. In addition, the proposed approach is robust to the image noise. This study fills the gap of multi-classification in retinal image quality assessment. [ABSTRACT FROM AUTHOR]

• A multilevel stimulated denoising strategy termed MUTE is developed based on the double-pass fundus reflection model to achieve single cataractous retinal image dehazing. • The estimation of cataract (haze) transmission function is converted into an image denoising problem which is solved by a special designed single-layered, fully connected network. • An unsupervised non-convex intensity-based cost function is designed for the network and is solved using gradient descent with momentum to guide the updating of model parameters adaptively. • Experimental results show the superiority of the MUTE over other state-of-art non-learning and learning-based methods in terms of restoration quality and implementation efficiency. • Clinical applications including retinal blood vessel segmentation, retinal image registration, and diagnosing using retinal images can benefit from the proposed MUTE method. In this research, we studied the duality between cataractous retinal image dehazing and image denoising and proposed that the dehazing task for cataractous retinal images can be achieved with the combination of image denoising and sigmoid function. To do so, we introduce the double-pass fundus reflection model in the YPbPr color space and developed a multilevel stimulated denoising strategy termed MUTE. The transmission matrix of the cataract layer is expressed as the superposition of denoised raw images of different levels weighted by pixel-wise sigmoid functions. We further designed an intensity-based cost function that can guide the updating of the model parameters. They are updated by gradient descent with adaptive momentum estimation, which gives us the final refined transmission matrix of the cataract layer. We tested our methods on cataract retinal images from both public and proprietary databases, and we compared the performance of our method with other state-of-the-art enhancement methods. Both visual assessments and objective assessments show the superiority of the proposed method. We further demonstrated three potential applications including blood vessel segmentation, retinal image registrations, and diagnosing with enhanced images that may largely benefit from our proposed methods. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Highlights • A method is proposed to measure vessel width from retinal images. • It is based on a parallel line model and k -means clustering algorithm. • It is compared with the state-of-the-art method and ground truth. • It is advantageous in measuring pathological vessels with central light reflex. Abstract Changes in retinal blood vessel width and morphology are closely related to the development of many diseases. However, some vessels (especially pathological vessels) often show central light reflex in retinal images, making it difficult to detect the vessel boundary and to measure the vessel width. In this study, a measuring methodology is proposed to improve the performance in measuring blood vessels with central light reflex, by use of centerline correction and k -means clustering algorithm. Firstly, the vessel tree in the retinal image is segmented using Isotropic Undecimated Wavelet Transform. Then, the centerline of each vessel is roughly extracted using morphological thinning algorithm, and corrected for vessels with central light reflex by use of a parallel line model. With known centerline, the direction and length of the measuring axis are determined, and the intensity profile of the pixels along the axis is obtained. By use of k -means clustering algorithm for the intensities, the pixels on the measuring axis are classified into vessel and background, based on which the vessel width is finally determined. The proposed method is tested on the public dataset REVIEW and our own dataset, and compared with ground truth and state-of-the-art method (ARIA method). It is shown that the measuring accuracy of proposed method is comparable to ARIA (average relative error: 3% vs. 4%), but it is more advantageous in measuring pathological vessels with central light reflex. It succeeds in measuring all the testing pathological vessels whereas ARIA is capable of measuring 20% of them. Result of the work has potential usage in quantitative evaluation of the progressing of retinal diseases. [ABSTRACT FROM AUTHOR]

Separation of the vascular tree into arteries and veins is a fundamental prerequisite in the automatic diagnosis of retinal biomarkers associated with systemic and neurodegenerative diseases. In this paper, we present a novel graph search metaheuristic approach for automatic separation of arteries/veins (A/V) from color fundus images. Our method exploits local information to disentangle the complex vascular tree into multiple subtrees, and global information to label these vessel subtrees into arteries and veins. Given a binary vessel map, a graph representation of the vascular network is constructed representing the topological and spatial connectivity of the vascular structures. Based on the anatomical uniqueness at vessel crossing and branching points, the vascular tree is split into multiple subtrees containing arteries and veins. Finally, the identified vessel subtrees are labeled with A/V based on a set of hand-crafted features trained with random forest classifier. The proposed method has been tested on four different publicly available retinal datasets with an average accuracy of 94.7%, 93.2%, 96.8%, and 90.2% across AV-DRIVE, CT-DRIVE, INSPIRE-AVR, and WIDE datasets, respectively. These results demonstrate the superiority of our proposed approach in outperforming the state-of-the-art methods for A/V separation. [ABSTRACT FROM AUTHOR]

Normally, the optic disc detection of retinal images is useful during the treatment of glaucoma and diabetic retinopathy. In this paper, the novel preprocessing of a retinal image with a bat algorithm (BA) optimization is proposed to detect the optic disc of the retinal image. As the optic disk is a bright area and the vessels that emerge from it are dark, these facts lead to the selected segments being regions with a great diversity of intensity, which does not usually happen in pathological regions. First, in the preprocessing stage, the image is fully converted into a gray image using a gray scale conversion, and then morphological operations are implemented in order to remove dark elements such as blood vessels, from the images. In the next stage, a bat algorithm (BA) is used to find the optimum threshold value for the optic disc location. In order to improve the accuracy and to obtain the best result for the segmented optic disc, the ellipse fitting approach was used in the last stage to enhance and smooth the segmented optic disc boundary region. The ellipse fitting is carried out using the least square distance approach. The efficiency of the proposed method was tested on six publicly available datasets, MESSIDOR, DRIVE, DIARETDB1, DIARETDB0, STARE, and DRIONS-DB. The optic disc segmentation average overlaps and accuracy was in the range of 78.5-88.2% and 96.6-99.91% in these six databases. The optic disk of the retinal images was segmented in less than 2.1 s per image. The use of the proposed method improved the optic disc segmentation results for healthy and pathological retinal images in a low computation time. Graphical abstract ᅟ. [ABSTRACT FROM AUTHOR]

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. [ABSTRACT FROM AUTHOR]

Purpose: To assess the effect of pupil dilation on measures of retinal vessel caliber. Design: Observational study with self-comparisons Methods: Retinal photographs were taken for both eyes of 25 subjects before and after pupil dilation. Three photographic graders, masked to pupil dilation status, measured the same set of images using a computer-assisted, semi-automatic method. We compared means (standard deviations) of retinal arteriolar caliber equivalent (CRAE) and retinal venular caliber equivalent (CRVE) of the same eyes between pre- and post-dilation images. We assessed concordance correlation coefficients (CC), Bland Altman limits-of-agreements, and used linear mixed models to assess CRAE and CRVE measures associated with pupil dilation (influencing image quality), graders (observers) and right-left eye variation. Results: We found high CCs for CRAE (0.82-0.94) and CRVE (0.87-0.94) between pre- and post-dilation images of the same eyes across the graders. Bland Altman plots showed that mean differences ranged from 0.55-3.42μm for CRAE and 1.56-2.29μm for CRVE. After adjusting for right-left eye random variation, a significant fixed effect of dilation was evident in mean CRAE in two of the three graders. There was no significant fixed effect of dilation in mean CRVE across all graders. In models including data of both eyes’ measures from pre- and post-dilation images by three graders, the fixed effect for dilation status contributed significantly to CRAE and CRVE variances whereas random effects for graders and dilation status contributed minimally. Conclusions: Contrary to our hypothesis, we found a systematic effect of pupil dilation on retinal vessel caliber measures. [ABSTRACT FROM AUTHOR]

Diabetic macular edema (DME), one of the leading causes of visual impairment and blindness, is usually diagnosed by the presence of exudates. However, exudate detection is challenging due to the large intraclass variation and high interclass similarity. To overcome these challenges, we propose the cascaded deep residual networks to recognize DME. Specifically, we first design a fully convolutional residual network that fuses multi-level hierarchical information to segment exudates accurately with a fast speed. Compared with previous methods, our approach avoids a wide range of preprocessing or postprocessing steps, reducing the impact of subjective factors. Then based on the segmentation results, the region centered on the pixel with the maximum probability is cropped and fed into the other deep residual network (for classification) to distinguish DME from its hard mimics. This makes the classification network to extract more representative features based on the segmentation results instead of the original images, further reducing the influence of complicated background. We evaluate the proposed method on two publicly available databases, the HEI-MED and e-ophtha EX databases. Extensive experiments demonstrate that our approach achieves better performance than the state-of-the-art methods with a fast processing speed, making it suitable for real-world clinical applications. [ABSTRACT FROM AUTHOR]

(Background and Objective): The occurrence of hard exudates is one of the early signs of diabetic retinopathy which is one of the leading causes of the blindness. Many patients with diabetic retinopathy lose their vision because of the late detection of the disease. Thus, this paper is to propose a novel method of hard exudates segmentation in retinal images in an automatic way. (Methods): The existing methods are based on either supervised or unsupervised learning techniques. In addition, the learned segmentation models may often cause miss-detection and/or fault-detection of hard exudates, due to the lack of rich characteristics, the intra-variations, and the similarity with other components in the retinal image. Thus, in this paper, the supervised learning based on the multilayer perceptron (MLP) is only used to identify initial seeds with high confidences to be hard exudates. Then, the segmentation is finalized by unsupervised learning based on the iterative graph cut (GC) using clusters of initial seeds. Also, in order to reduce color intra-variations of hard exudates in different retinal images, the color transfer (CT) is applied to normalize their color information, in the pre-processing step. (Results): The experiments and comparisons with the other existing methods are based on the two well-known datasets, e_ophtha EX and DIARETDB1. It can be seen that the proposed method outperforms the other existing methods in the literature, with the sensitivity in the pixel-level of 0.891 for the DIARETDB1 dataset and 0.564 for the e_ophtha EX dataset. The cross datasets validation where the training process is performed on one dataset and the testing process is performed on another dataset is also evaluated in this paper, in order to illustrate the robustness of the proposed method. (Conclusions): This newly proposed method integrates the supervised learning and unsupervised learning based techniques. It achieves the improved performance, when compared with the existing methods in the literature. The robustness of the proposed method for the scenario of cross datasets could enhance its practical usage. That is, the trained model could be more practical for unseen data in the real-world situation, especially when the capturing environments of training and testing images are not the same. [ABSTRACT FROM AUTHOR]

Objective: Many common eye diseases and cardiovascular diseases can be diagnosed through retinal imaging. However, due to uneven illumination, image blurring, and low contrast, retinal images with poor quality are not useful for diagnosis, especially in automated image analyzing systems. Here, we propose a new image enhancement method to improve color retinal image luminosity and contrast. Methods: A luminance gain matrix, which is obtained by gamma correction of the value channel in the HSV (hue, saturation, and value) color space, is used to enhance the R, G, and B (red, green and blue) channels, respectively. Contrast is then enhanced in the luminosity channel of L*a*b* color space by CLAHE (contrast-limited adaptive histogram equalization). Image enhancement by the proposed method is compared to other methods by evaluating quality scores of the enhanced images. Results: The performance of the method is mainly validated on a dataset of 961 poor-quality retinal images. Quality assessment (range 0–1) of image enhancement of this poor dataset indicated that our method improved color retinal image quality from an average of 0.0404 (standard deviation 0.0291) up to an average of 0.4565 (standard deviation 0.1000). Conclusion: The proposed method is shown to achieve superior image enhancement compared to contrast enhancement in other color spaces or by other related methods, while simultaneously preserving image naturalness. Significance: This method of color retinal image enhancement may be employed to assist ophthalmologists in more efficient screening of retinal diseases and in development of improved automated image analysis for clinical diagnosis. [ABSTRACT FROM PUBLISHER]

Diabetic Retinopathy is one complication of diabetes, which can cause blindness. Diabetic retinopathy can be divided into Non-Proliferative Diabetic Retinopathy (NPDR) and Proliferative Diabetic Retinopahy (PDR), and neovascularization is a key symbol to make diagnosis between them. An automatic detection of neovascularization in retinal images using extreme learning machine is proposed. Furthermore, we use a series of filter banks to get the features of neovascularization from retinal images. The detection framework is evaluated with images annotated by expert ophthalmologists based on the images selected from several public retinal image databases. The experimental results illustrate that the framework can mark and show the suspected neovascularization regions to ophthalmologists, and thus support for their decision making. [ABSTRACT FROM AUTHOR]

Automatic artery/vein (A/V) classification is one of the important topics in retinal image analysis. It allows the researchers to investigate the association between biomarkers and disease progression on a huge amount of data for arteries and veins separately. Recent proposed methods, which employ contextual information of vessels to achieve better A/V classification accuracy, still rely on the performance of pixel-wise classification, which has received limited attention in recent years. In this paper, we show that these classification methods can be markedly improved. We propose a new normalization technique for extracting four new features which are associated with the lightness reflection of vessels. The accuracy of a linear discriminate analysis classifier is used to validate these features. Accuracy rates of 85.1, 86.9 and 90.6% were obtained on three datasets using only local information. Based on the introduced features, the advanced graph-based methods will achieve a better performance on A/V classification. [ABSTRACT FROM AUTHOR]

(Background and objectives) Retinal cysts are formed by accumulation of fluid in the retina caused by leakages from inflammation or vitreous fractures. Analysis of the retinal cystic spaces holds significance in detection and treatment of several ocular diseases like age-related macular degeneration, diabetic macular edema etc. Thus, segmentation of intra-retinal cysts and quantification of cystic spaces are vital for retinal pathology and severity detection. In the recent years, automated segmentation of intra-retinal cysts using optical coherence tomography B-scans has gained significant importance in the field of retinal image analysis. The objective of this paper is to compare different intra-retinal cyst segmentation algorithms for comparative analysis and benchmarking purposes. (Methods) In this work, we employ a modular approach for standardizing the different segmentation algorithms. Further, we analyze the variations in automated cyst segmentation performances and method scalability across image acquisition systems by using the publicly available cyst segmentation challenge dataset (OPTIMA cyst segmentation challenge). (Results) Several key automated methods are comparatively analyzed using quantitative and qualitative experiments. Our analysis demonstrates the significance of variations in signal-to-noise ratio (SNR), retinal layer morphology and post-processing steps on the automated cyst segmentation processes. (Conclusion) This benchmarking study provides insights towards the scalability of automated processes across vendor-specific imaging modalities to provide guidance for retinal pathology diagnostics and treatment processes. [ABSTRACT FROM AUTHOR]

Automatic retinal image registration is still a great challenge in computer aided diagnosis and screening system. In this paper, a new retinal image registration method is proposed based on the combination of blood vessel segmentation and scale invariant feature transform (SIFT) feature. The algorithm includes two stages: retinal image segmentation and registration. In the segmentation stage, the blood vessel is segmented by using the guided filter to enhance the vessel structure and the bottom-hat transformation to extract blood vessel. In the registration stage, the SIFT algorithm is adopted to detect the feature of vessel segmentation image, complemented by using a random sample consensus (RANSAC) algorithm to eliminate incorrect matches. We evaluate our method from both segmentation and registration aspects. For segmentation evaluation, we test our method on DRIVE database, which provides manually labeled images from two specialists. The experimental results show that our method achieves 0.9562 in accuracy (Acc), which presents competitive performance compare to other existing segmentation methods. For registration evaluation, we test our method on STARE database, and the experimental results demonstrate the superior performance of the proposed method, which makes the algorithm a suitable tool for automated retinal image analysis. [ABSTRACT FROM AUTHOR]

Retinal fundus cameras suffer from dust particles attaching to the sensor and lens, which manifest as small artifacts on the images. We propose a new strategy for the detection and removal of dust particle artifacts in retinal images. We consider as input two or more color fundus images acquired within the same session, in which we assume the artifacts remain in the same position. Our method consists in detecting candidate artifacts via normalized cross correlation with an artifact template, performing segmentation via region growing, and comparing the segmentations in all images. This guarantees that all detections are consistent for all images. The removal stage consists in an inpainting procedure so that the new region does not stand out from the neighboring regions. Encouraging experimental results show the localization of artifacts is effective and the artifacts are successfully removed, while not introducing new artifacts in the color retinal images. [ABSTRACT FROM AUTHOR]

Background and objective : The anatomical structure detection in retinal images is an open problem. However, most of the works in the related literature are oriented to the detection of each structure individually or assume the previous detection of a structure which is used as a reference. The objective of this paper is to obtain simultaneous detection of the main retinal structures (optic disc, macula, network of vessels and vascular bundle) in a fast and robust way. Methods : We propose a new methodology oriented to accomplish the mentioned objective. It consists of two stages. In an initial stage, a set of operators is applied to the retinal image. Each operator uses intra-structure relational knowledge in order to produce a set of candidate blobs that belongs to the desired structure. In a second stage, a set of tuples is created, each of which contains a different combination of the candidate blobs. Next, filtering operators, using inter-structure relational knowledge, are used in order to find the winner tuple. A method using template matching and mathematical morphology is implemented following the proposed methodology. Results : A success is achieved if the distance between the automatically detected blob center and the actual structure center is less than or equal to one optic disc radius. The success rates obtained in the different public databases analyzed were: MESSIDOR (99.33%, 98.58%, 97.92%), DIARETDB1 (96.63%, 100%, 97.75%), DRIONS (100%, n/a, 100%) and ONHSD (100%, 98.85%, 97.70%) for optic disc (OD), macula (M) and vascular bundle (VB), respectively. Finally, the overall success rate obtained in this study for each structure was: 99.26% (OD), 98.69% (M) and 98.95% (VB). The average time of processing per image was 4.16 ± 0.72 s. Conclusions : The main advantage of the use of inter-structure relational knowledge was the reduction of the number of false positives in the detection process. The implemented method is able to simultaneously detect four structures. It is fast, robust and its detection results are competitive in relation to other methods of the recent literature. [ABSTRACT FROM AUTHOR]

This paper presents a supervised retinal vessel segmentation by incorporating vessel filtering and wavelet transform features from orientation scores (OSs), and green intensity. Through an anisotropic wavelet-type transform, a 2D image is lifted to a 3D orientation score in the Lie-group domain of positions and orientations R 2 ⋊ S 1 . Elongated structures are disentangled into their corresponding orientation planes and enhanced via multi-orientation vessel filtering. In addition, scale-selective OSs (in the domain of positions, orientations and scales R 2 ⋊ S 1 × R + ) are obtained by adding a scale adaptation to the wavelet transform. Features are optimally extracted by taking maximum orientation responses at multiple scales, to represent vessels of changing calibers. Finally, we train a Random Forest classifier for vessel segmentation. Extensive validations show that our method achieves a competitive segmentation, and better vessel preservation with less false detections compared with the state-of-the-art methods. [ABSTRACT FROM AUTHOR]

• A luminosity rectified Richardson-Lucy deconvolution model is proposed to achieve single-image blind deconvolution and illumination correction for retinal images. • A non-convex cost function based on the double-pass fundus reflection model is designed and is solved using gradient descent with Nesterov accelerated adaptive momentum. • The proposed model estimates the latent image, the illumination pattern, and the blurry kernel simultaneously. • The proposed model is compared against the state-of-the-art methods, showing its superiority in terms of restoration quality and implementation efficiency. • The proposed model improves the clarity of retinal images which benefits clinical diagnosing. Due to imperfect imaging conditions, retinal images can be degraded by uneven/insufficient illumination, blurriness caused by optical aberrations and unintentional motions. Degraded images reduce the effectiveness of diagnosis by an ophthalmologist. To restore the image quality, in this research we propose the luminosity rectified Richardson-Lucy (LRRL) blind deconvolution framework for single retinal image restoration. We established an image formation model based on the double-pass fundus reflection feature and developed a differentiable non-convex cost function that jointly achieves illumination correction and blind deconvolution. To solve this non-convex optimization problem, we derived the closed-form expression of the gradients and used gradient descent with Nesterov-accelerated adaptive momentum estimation to accelerate the optimization, which is more efficient than the traditional half quadratic splitting method. The LRRL was tested on 1719 images from three public databases. Four image quality matrixes including image definition, image sharpness, image entropy, and image multiscale contrast were used for objective assessments. The LRRL was compared against the state-of-the-art retinal image blind deconvolution methods. Our LRRL corrects the problematic illumination and improves the clarity of the retinal image simultaneously, showing its superiority in terms of restoration quality and implementation efficiency. The MATLAB code is available on Github. [ABSTRACT FROM AUTHOR]

Optical systems that emit radiation between the visible and near-infrared wavelength region pose a potential hazard to human vision as the radiation is imaged on the retina. The radiation interacts with the retinal tissue in a photomechanical, photothermal, and photochemical manner that can result in irreversible injuries. To ensure an eye safe system, it is important to correctly apply the laser safety standard IEC 60825-1:2014 or the lamp safety standard IEC 62471:2006. We aim to provide a general calculation procedure for both coherent and incoherent sources, which are compliant with the respective safety standards. An air-equivalent eye model generates retinal images. Two software-based calculation methods are introduced, which are referred to as image analysis. The first method calculates the angular subtense of the apparent source needed for photomechanical and photothermal limits. The second method applies to photochemical limits. Using exemplary optical systems, the image analysis is investigated. The proposed image analysis gives guidance for missing aspects in the standards and reduces ambiguity and complexity. The proposed method can be used for eye safety evaluations since it follows the concept of the safety standards with conservative approaches. [ABSTRACT FROM AUTHOR]

• We propose a novel retinal image quality assessment method that shows promising performance comparing with the state-of-the-art and latest works. • We are the first to introduce the idea of label smoothing and cost-sensitive regularization to RIQA, which is beneficial to train and practical to use. • A dual-path network structure with novel attention blocks are designed to interpret CNN decision bases and improve the performance. Background and Objective: The assessment of the image quality is crucial before the computer-aided diagnosis of fundus images. This task is very challenging. Firstly, the subjective judgments of graders on image quality lead to ambiguous labels. Secondly, despite being treated as classification in existing works, grading has regression properties that cannot be ignored. Solving the ambiguity problem and regression problem in the label space, and extracting discriminative features, have become the keys to quality assessment. Methods: In this paper, we proposed a framework that can assess the quality of fundus images accurately and reasonably based on deep convolutional neural networks. Drawing on the experience of human graders, a dual-path convolutional neural network with attention blocks is designed to better extract discriminative features and present the bases of decision. Label smoothing and cost-sensitive regularization are designed to solve the label ambiguity problem and the potential regression problem respectively. Besides, a large number of images are annotated by us to further improve the results. Results: We conducted our experiments on the largest retinal image quality assessment dataset with 28,792 retinal images. Our approach achieves 0.8868 precision, 0.8786 recall, 0.8820 F1, and 0.9138 Kappa score. Results show that our approach outperforms state-of-the-art methods. Conclusions: The promising performances reveal that our methods are beneficial to retinal image quality assessment and have potential in other grading tasks. [ABSTRACT FROM AUTHOR]

Glaucoma is an eye disease that usually occurs with the increased Intra-Ocular Pressure (IOP), which damages the vision of eyes. So, detecting and classifying Glaucoma is an important and demanding task in recent days. For this purpose, some of the clustering and segmentation techniques are proposed in the existing works. But, it has some drawbacks that include inefficient, inaccurate and estimates only the affected area. In order to solve these issues, a Neighboring Differential Clustering (NDC) - Intensity Variation Masking (IVM) are proposed in this paper. The main intention of this work is to extract and diagnose the abnormal retinal image by identifying the optic disc. This work includes three stages such as, preprocessing, clustering and segmentation. At first, the given retinal image is preprocessed by using the Gaussian Mask Updated (GMU) model for eliminating the noise and improving the quality of the image. Then, the cluster is formed by extracting the threshold and patterns with the help of NDC technique. In the segmentation stage, the weight is calculated for pixel matching and ROI extraction by using the proposed IVM method. Here, the novelty is presented in the clustering and segmentation processes by developing NDC and IVM algorithms for accurate Glaucoma identification. In experiments, the results of both existing and proposed techniques are evaluated in terms of sensitivity, specificity, accuracy, Hausdorff distance, Jaccard and dice metrics. [ABSTRACT FROM AUTHOR]

Goal: Reliable recognition of microaneurysms (MAs) is an essential task when developing an automated analysis system for diabetic retinopathy (DR) detection. In this study, we propose an integrated approach for automated MA detection with high accuracy. Methods: Candidate objects are first located by applying a dark object filtering process. Their cross-section profiles along multiple directions are processed through singular spectrum analysis. The correlation coefficient between each processed profile and a typical MA profile is measured and used as a scale factor to adjust the shape of the candidate profile. This is to increase the difference in their profiles between true MAs and other non-MA candidates. A set of statistical features of those profiles is then extracted for a K-nearest neighbor classifier. Results: Experiments show that by applying this process, MAs can be separated well from the retinal background, the most common interfering objects and artifacts. Conclusion: The results have demonstrated the robustness of the approach when testing on large scale datasets with clinically acceptable sensitivity and specificity. Significance: The approach proposed in the evaluated system has great potential when used in an automated DR screening tool or for large scale eye epidemiology studies. [ABSTRACT FROM AUTHOR]

(Background and objectives) Retinal artery and vein classification is an important task for the automatic computer-aided diagnosis of various eye diseases and systemic diseases. This paper presents an improved supervised artery and vein classification method in retinal image. (Methods) Intra-image regularization and inter-subject normalization is applied to reduce the differences in feature space. Novel features, including first-order and second-order texture features, are utilized to capture the discriminating characteristics of arteries and veins. (Results) The proposed method was tested on the DRIVE dataset and achieved an overall accuracy of 0.923. (Conclusion) This retinal artery and vein classification algorithm serves as a potentially important tool for the early diagnosis of various diseases, including diabetic retinopathy and cardiovascular diseases. [ABSTRACT FROM AUTHOR]

Almost all existing 3D visual discomfort prediction models are based, at least in part, on features that are extracted from computed disparity maps. These include such estimated quantities such as the maximum disparity, disparity range, disparity energy and other measures of the disparity distribution. A common first step when implementing a 3D visual discomfort model is some form of disparity calculation, whence the accuracy of prediction largely depends on the accuracy of the disparity result. Unfortunately, most algorithms that compute disparity maps are expensive, and are not guaranteed to deliver sufficiently accurate or perceptually relevant disparity data. This raises the question of whether it is possible to build a 3D discomfort prediction model without explicit disparity calculation. Towards this possibility, we have developed a new feature map, called the percentage of un-linked pixels (PUP), that is descriptive of the presence of disparity, and which can be used to accurately predict experienced 3D visual discomfort without the need for actually calculating disparity values. Instead, PUP features are extracted by predicting the percentage of un-linked pixels in corresponding retinal patches of image pairs. The un-linked pixels are determined by feature classification on orientation and luminance distributions. Calculation of PUP maps is much faster than traditional disparity computation, and the experimental results demonstrate that the predictive power attained using the PUP map is highly competitive with prior models that rely on computed disparity maps. [ABSTRACT FROM AUTHOR]

Medical digital imaging has become a key element of modern health care procedures. It provides visual documentation and a permanent record for the patients, and most important the ability to extract quantitative information about many diseases. Modern ophthalmology relies on the advances in digital imaging and computing power. In this paper we present an overview of the results from the doctoral dissertation by Andrés G. Marrugo. This dissertation contributes to the digital analysis of retinal images and the problems that arise along the imaging pipeline of fundus photography, a field that is commonly referred to as retinal image analysis. We have dealt with and proposed solutions to problems that arise in retinal image acquisition and longitudinal monitoring of retinal disease evolution. Specifically, non-uniform illumination compensation, poor image quality, automated focusing, image segmentation, change detection, space-invariant (SI) and space-variant (SV) blind deconvolution (BD). Digital retinal image analysis can be effective and cost-efficient for disease management, computeraided diagnosis, screening and telemedicine and applicable to a variety of disorders such as glaucoma, macular degeneration, and retinopathy. [ABSTRACT FROM AUTHOR]

• Developed an improved generative adversarial network. • Designed a novel residual attention block. • Use the Charbonnier loss function instead of the MSE loss function. • Remove the BN layer and add multiple updated residual blocks. The retina is the only organ in the body that can use visible light for non-invasive observation. By analyzing retinal images, we can achieve early screening, diagnosis and prevention of many ophthalmological and systemic diseases, helping patients avoid the risk of blindness. Due to the powerful feature extraction capabilities, many deep learning super-resolution reconstruction networks have been applied to retinal image analysis and achieved excellent results. Given the lack of high-frequency information and poor visual perception in the current reconstruction results of super-resolution reconstruction networks under large-scale factors, we present an improved generative adversarial network (IGAN) algorithm for retinal image super-resolution reconstruction. Firstly, we construct a novel residual attention block, improving the reconstruction results lacking high-frequency information and texture details under large-scale factors. Secondly, we remove the Batch Normalization layer that affects the quality of image generation in the residual network. Finally, we use the more robust Charbonnier loss function instead of the mean square error loss function and the TV regular term to smooth the training results. Experimental results show that our proposed method significantly improves objective evaluation indicators such as peak signal-to-noise ratio and structural similarity. The obtained image has rich texture details and a better visual experience than the state-of-the-art image super-resolution methods. Our proposed method can better learn the mapping relationship between low-resolution and high-resolution retinal images. This method can be effectively and stably applied to the analysis of retinal images, providing an effective basis for early clinical treatment. [ABSTRACT FROM AUTHOR]

This paper presents a new supervised method for vessel segmentation in retinal images. This method remolds the task of segmentation as a problem of cross-modality data transformation from retinal image to vessel map. A wide and deep neural network with strong induction ability is proposed to model the transformation, and an efficient training strategy is presented. Instead of a single label of the center pixel, the network can output the label map of all pixels for a given image patch. Our approach outperforms reported state-of-the-art methods in terms of sensitivity, specificity and accuracy. The result of cross-training evaluation indicates its robustness to the training set. The approach needs no artificially designed feature and no preprocessing step, reducing the impact of subjective factors. The proposed method has the potential for application in image diagnosis of ophthalmologic diseases, and it may provide a new, general, high-performance computing framework for image segmentation. [ABSTRACT FROM AUTHOR]

Diabetic retinopathy (DR) is a disease with an increasing prevalence and the main cause of blindness among working-age population. The risk of severe vision loss can be significantly reduced by timely diagnosis and treatment. Systematic screening for DR has been identified as a cost-effective way to save health services resources. Automatic retinal image analysis is emerging as an important screening tool for early DR detection, which can reduce the workload associated to manual grading as well as save diagnosis costs and time. Many research efforts in the last years have been devoted to developing automatic tools to help in the detection and evaluation of DR lesions. However, there is a large variability in the databases and evaluation criteria used in the literature, which hampers a direct comparison of the different studies. This work is aimed at summarizing the results of the available algorithms for the detection and classification of DR pathology. A detailed literature search was conducted using PubMed. Selected relevant studies in the last 10 years were scrutinized and included in the review. Furthermore, we will try to give an overview of the available commercial software for automatic retinal image analysis. [ABSTRACT FROM AUTHOR]

• Developed an end-to-end residual attention mechanism framework. • Designed the multi-scale feature extraction module. • Based on the encoding-decoding structure. • Achieved good subjective and objective evaluation results, and the effect of retinal dehazing is better. Cataract is one of the most common causes of vision loss. Light scattering due to clouding of the lens in cataract patients makes it extremely difficult to image the retina of cataract patients with fundus cameras, resulting in a serious decrease in the quality of the retinal images taken. Furthermore, the age of cataract patients is generally too old, in addition to cataracts, the patients often have other retinal diseases, which brings great challenges to experts in the clinical diagnosis of cataract patients using retinal imaging. In this paper, we present the End-to-End Residual Attention Mechanism (ERAN) for Cataractous Retinal Image Dehazing, which it includes four modules: encoding module, multi-scale feature extraction module, feature fusion module, and decoding module. The encoding module encodes the input cataract haze image into an image, facilitating subsequent feature extraction and reducing memory usage. The multi-scale feature extraction module includes a hole convolution module, a residual block, and an adaptive skip connection, which can expand the receptive field and extract features of different scales through weighted screening for fusion. The feature fusion module uses adaptive skip connections to enhance the network's ability to extract haze density images to make haze removal more thorough. Furthermore, the decoding module performs non-linear mapping on the fused features to obtain the haze density image, and then restores the haze-free image. The experimental results show that the proposed method has achieved better objective and subjective evaluation results, and has a better dehazing effect. We proposed ERAN method not only provides visually better images, but also helps experts better diagnose other retinal diseases in cataract patients, leading to better care and treatment. [ABSTRACT FROM AUTHOR]

• Dilated convolutions ensure better feature transfer and accurate classification that result in a sensitivity boost. • Dilated convolutions serve as the main constituents of decoder, encoder blocks and skip connections. • Dilated convolutions of different rates are more effective at capturing dynamically complex blood vessels. • Proposed Weighted multi-output fusion retrieves high fidelity vascular map by extracting critical features from each output block. Retinal image segmentation can help clinicians detect pathological disorders by studying changes in retinal blood vessels. This early detection can help prevent blindness and many other vision impairments. So far, several supervised and unsupervised methods have been proposed for the task of automatic blood vessel segmentation. However, the sensitivity and the robustness of these methods can be improved by correctly classifying more vessel pixels. We proposed an automatic, retinal blood vessel segmentation method based on the U-net architecture. This end-to-end framework utilizes preprocessing and a data augmentation pipeline for training. The architecture utilizes multiscale input and multioutput modules with improved skip connections and the correct use of dilated convolutions for effective feature extraction. In multiscale input, the input image is scaled down and concatenated with the output of convolutional blocks at different points in the encoder path to ensure the feature transfer of the original image. The multioutput module obtains upsampled outputs from each decoder block that are combined to obtain the final output. Skip paths connect each encoder block with the corresponding decoder block, and the whole architecture utilizes different dilation rates to improve the overall feature extraction. The proposed method achieved an accuracy: of 0.9680, 0.9694, and 0.9701; sensitivity of 0.8837, 0.8263, and 0.8713; and Intersection Over Union (IOU) of 0.8698, 0.7951, and 0.8184 with three publicly available datasets: DRIVE, STARE, and CHASE, respectively. An ablation study is performed to show the contribution of each proposed module and technique. The evaluation metrics revealed that the performance of the proposed method is higher than that of the original U-net and other U-net-based architectures, as well as many other state-of-the-art segmentation techniques, and that the proposed method is robust to noise. [ABSTRACT FROM AUTHOR]

Geometrical optics predicts the formation of an inverted image of the world on the retina. In this video, we confirm that this affirmation is true. The movie shows a light on the left, and an eye from a death cow on the right. We removed the superior part of the sclera in order to see the eye fundus. It is possible to see a reflex of the light on the cornea and the image of the light on the retina. We move the light up and down, and the image of the lamp on the back of the eye moves in the contrary sense. So the eye inverts the image. [ABSTRACT FROM AUTHOR]