Vessel Segmentation In Retinal Images Research Articles

Denoised Non-Local Means with BDDU-Net Architecture for Robust Retinal Blood Vessel Segmentation

Retinal blood vessels can be obtained by image segmentation. This study proposes combining image enhancement and segmentation to obtain retinal blood vessels. The image enhancement stages use CLAHE and Denoised Non-Local Means to increase contrast and reduce noise on the original image, and Bottom-Hat (BTH) filtering is used to lighten dark features in the image so the features become lighter and darken the bright features in the image. Bottom Hat is applied to make the features of the blood vessels in the retinal image more visible. The segmentation architecture proposes BDDU-Net architecture which combines U-Net in the encoder part, DenseNet in the bridge part, and Bi-ConvLSTM in the decoder part. Image enhancement performance results are PSNR and SSIM. The PSNR is more than 40 dB on both the DRIVE and STARE datasets. The SSIM results are close to 1 on the DRIVE and STARE datasets. These results show that the image enhancement stages in the proposed method can enhance the quality of the original image. The segmentation performance results of BDDU-Net architecture are measured based on accuracy, sensitivity, specificity, IoU, and F1-Score. The DRIVE dataset obtained 95.578% for accuracy, 85.75% for sensitivity, 96.75% for specificity, 67.407% for IoU, and 80.53% for F1-Score. The STARE dataset obtained 97.63% for accuracy, 84.33% for sensitivity, 98.66% for specificity, 75.67% for IoU, and 86.15% for F1-Score. Based on the image enhancement and image segmentation results, these results show that the proposed method is great for enhancing image quality and excellent for blood vessel segmentation in retinal images, although IoU results on the DRIVE dataset need to be improved.

Optic Disc and Vessel Segmentation in Retinal Images using Correlation Texture

Globally, glaucoma has been recognized as the secondmost common reason for blindness. This affects the anatomy of the eyes, including blood vessels (BVs) and optic disc (OD). State-of-the-art literature has revealed that this disease is the root cause of degeneration in the OD, with some effects on the morphology of the BVs in the retina. These can result in total blindness if left untreated. Previous studies have shown that to effectively diagnose a disease, it is necessary to automatically detect the OD and the BVs with high accuracy in the fundus image. Segmentation of OD and BVs has, however, been challenging, due to the complex and poor segmentation performances. Such performances arise from the intricacies of vascular network width variation, illumination of images, and the structure of retinal images. Examples are retinal image boundaries on ODs and retinal lesions, due to disease. Hence the need for an investigation of the appropriate image feature(s) that will facilitate improvements to OD and BV segmentation performances. Texture is an important indicator used in image analysis for object segmentation. This work proposed the use of image-correlation texture information computational approaches for the segmentation of BVs, using variant window sizes; and OD using grayscale and green channel in the retinal images. This study revealed that the segmentation of BVs using variant window sizes (15x15) and (17x17) achieved high accuracy values of 95.11 per cent and 94.93 per cent; with accuracy values of 94.21 per cent and 93.98 per cent revealed in the DRIVE and STARE databases, respectively. Usually, sensitivity values of 74.84 per cent and 75.94 per cent, with average sensitivity values of 74.58 percent and 78.48 per cent, are achieved BV segmentation on the window sizes (15x15) and (17x17), indicated in DRIVE and STARE databases, respectively. Also, an average specificity value of 97.08 per cent and 96.77 per cent with average specificity values of 95.77 per cent and 95.18 per cent are achieved for BV segmentation on the window sizes (15x15) and (17x17), on DRIVE and STARE databases, respectively. Further, the segmentation of OD using grayscale achieved a slightly greater accuracy rate (98.59%) compared to the accuracy rate (98.36%) realized on green channel. This study also showed that the proposed approaches for BV and OD segmentation in this research achieved better results compared with several existing methods in the literature.

The Combination of Black Hat Transform and U-Net in Image Enhancement and Blood Vessel Segmentation in Retinal Images

Diabetic Retinopathy (DR) is a disorder of the eye caused by damage to blood vessels in the retina. Damage to the retinal blood vessels can be analyzed by segmenting the blood vessels on the image. This study proposes a combination of image enhancement and blood vessel segmentation in retinal images. Retinal image enhancement is carried out using the black hat transform method to obtain a detailed view of blood vessels in retinal images. Segmentation of blood vessels in retinal images is carried out using the U-Net architecture. The results of image enhancement are measured using MSE and PSNR. This study has an MSE value below 0.05 and a PSNR above 90dB. The MSE and PSNR values obtained show that the black hat transform method is very good at image enhancement. Segmentation has an accuracy value above 0.95 and a sensitivity value above 0.85. In addition, the specificity value and f1-score are above 0.8. This shows that the proposed stages of image enhancement and blood vessel segmentation are able to accurately recognize blood vessel features in retinal images.

SEGMENTATION AND IDENTIFICATION OF DISEASES IN RETINAL BLOOD VESSEL USING WIDE AND DEEP NEURAL NETWORK SYSTEM

A new supervised method for vessel segmentation in retinal images is studied here. Proposed 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. Key Words- new supervised method, cross-modality, cross-training evaluation.

DeepInfusion: A dynamic infusion based-neuro-symbolic AI model for segmentation of intracranial aneurysms

The detection and segmentation of cerebral aneurysms is a crucial step in the development of a clinical decision support system for estimating aneurysm rupture risk. However, accurately identifying and segmenting regions of interest in two-dimensional (2D) medical images is often challenging, particularly when using deep learning (DL) methods on small datasets with limited annotated data. The accuracy of DL approaches is often affected by the availability of large, annotated training datasets that are required for effective deep learning. Additionally, when using DL to differentiate aneurysms from arterial loops in 2D DSA images, DL can fail to detect aneurysms in areas where dye concentration is low. To address these issues and enhance the reliability and accuracy of aneurysm detection and segmentation methods, incorporating medical expert-advised, hand-crafted features can provide a clinical perspective to DL methods. This approach can help to improve the performance of DL methods by providing additional information that is not captured in the data. To this end, a novel Neuro-symbolic AI-based DeepInfusion model is proposed which allows for the infusion of human intellect through hand-crafted features into deep neural networks (DNNs), thus combining the strengths of DL with the knowledge and expertise of medical professionals. The proposed approach includes a novel technique for dynamic layer selection and feature weight adjustment during the model infusion process. The performance of the DeepInfusion model is evaluated on an in-house prepared dataset of 409 DSA images, and experimental results demonstrate the effectiveness of the proposed method for the segmentation of cerebral aneurysms. The model achieves an IOU score of 96.76% and an F1-score of 94.15% on unseen DSA images. The model is also tested on two publicly available datasets of Kvasir-SEG polyp and DRIVE for vessel segmentation of retinal images. The results show a significant improvement compared to existing methods, which indicates the generalizability of the approach in medical segmentation. The complete code for DeepInfusion is available on our GitHub repository at https://github.com/smileslab/deep-infusion/blob/main/deepinfusion.ipynb.

Automatic Detection of Diabetic Hypertensive Retinopathy in Fundus Images Using Transfer Learning

Diabetic retinopathy (DR) is a complication of diabetes that affects the eyes. It occurs when high blood sugar levels damage the blood vessels in the retina, the light-sensitive tissue at the back of the eye. Therefore, there is a need to detect DR in the early stages to reduce the risk of blindness. Transfer learning is a machine learning technique where a pre-trained model is used as a starting point for a new task. Transfer learning has been applied to diabetic retinopathy classification with promising results. Pre-trained models, such as convolutional neural networks (CNNs), can be fine-tuned on a new dataset of retinal images to classify diabetic retinopathy. This manuscript aims at developing an automated scheme for diagnosing and grading DR and HR. The retinal image classification has been performed using three phases that include preprocessing, segmentation and feature extraction techniques. The pre-processing methodology has been proposed for reducing the noise in retinal images. A-CLAHE, DNCNN and Wiener filter techniques have been applied for the enhancement of images. After pre-processing, blood vessel segmentation in retinal images has been performed utilizing OTSU thresholding and mathematical morphology. Feature extraction and classification have been performed using transfer learning models. The segmented images were then classified using Modified ResNet 101 architecture. The performance for enhanced images has been evaluated on PSNR and shows better results as compared to the existing literature. The network is trained on more than 6000 images from MESSIDOR and ODIR datasets and achieves the classification accuracy of 98.72%.

A hierarchical image matting model for blood vessel segmentation in retinal images

A hierarchical image matting model for blood vessel segmentation in retinal images

An Enhanced Retinal vessel segmentation using Deep Convolution Neural Network

Accurate vessel segmentation in retinal images plays a vital role for retinopathy diagnosis and analysis. The presence of very thin vessels in low image contrast, on the other hand makes the segmentation task difficult. In the proposed method retinal vessels are segmented using multiscale Fully Convolved Convolutional Neural Network (FCCNN) architecture. The proposed architecture is trained for pixel classification to cope with the varying width and direction of the vessel structure in the retina. Green channel extraction gives better contrast difference between vessels and background. The skeletonization process is done which prevents the change in structure, thus the vasculature remains unchanged. In addition, an improved class balanced cross entropy loss function is included to avoid misclassification and imbalance problem. The proposed method is verified on public retinal vessel segmentation database (DRIVE). The accuracy of DRIVE FCCNN after 90 epochs was attained as 92.73% and the loss was attained as 0.0632. The experimental results show that segmentation results and accuracy obtained for FCNN are greater compared to the other architecture.

A new deep learning method for blood vessel segmentation in retinal images based on convolutional kernels and modified U-Net model

Background and Objective: Automatic monitoring of retinal blood vessels proves very useful for the clinical assessment of ocular vascular anomalies or retinopathies. This paper presents an efficient and accurate deep learning-based method for vessel segmentation in eye fundus images.Methods: The approach consists of a convolutional neural network based on a simplified version of the U-Net architecture that combines residual blocks and batch normalization in the up- and downscaling phases. The network receives patches extracted from the original image as input and is trained with a novel loss function that considers the distance of each pixel to the vascular tree. At its output, it generates the probability of each pixel of the input patch belonging to the vascular structure. The application of the network to the patches in which a retinal image can be divided allows obtaining the pixel-wise probability map of the complete image. This probability map is then binarized with a certain threshold to generate the blood vessel segmentation provided by the method.Results: The method has been developed and evaluated in the DRIVE, STARE and CHASE_Db1 databases, which offer a manual segmentation of the vascular tree by each of its images. Using this set of images as ground truth, the accuracy of the vessel segmentations obtained for an operating point proposal (established by a single threshold value for each database) was quantified. The overall performance was measured using the area of its receiver operating characteristic curve. The method demonstrated robustness in the face of the variability of the fundus images of diverse origin, being capable of working with the highest level of accuracy in the entire set of possible points of operation, compared to those provided by the most accurate methods found in literature.Conclusions: The analysis of results concludes that the proposed method reaches better performance than the rest of state-of-art methods and can be considered the most promising for integration into a real tool for vascular structure segmentation.

Hybrid Learning of Vessel Segmentation in Retinal Images

This paper aims to develop a technique of vessel segmentation in retinal images. Interpreting the segmented vessels is necessary for the automatic detection of the severe stage of the diabetic retinopathy. Thus, it is important to have the technique for segmenting vessels in an automatic way with high performance, for the sake of further analysis. In this paper, the proposed method is developed based on the double layer combining supervised and non-supervised learning aspects. The first layer is to detect the initial seeds of vessels using the supervised learning. It learns based on three types of features including green intensity, line operators, and Gabor filters. Then, the support vector machine (SVM) is applied as the classification tool. In the second layer, the segmentation results from the first layer is further revised and completed using the non-supervised learning. The morphological operations with the watershed technique are applied on the results obtained from the first layer, to remain with the segmented pixels with high confidential to be vessels. Then, these pixels are used as the initial seeds of foreground in the iterative graph cut. As the result, the more completed and comprehensive foreground (i.e. vessels) can be obtained. The proposed method is evaluated using two well-known datasets including DRIVE and STARE. The experimental results show the promising performance of the proposed method when compared with other existing methods in the literature.

A vessel segmentation technique for retinal images

AbstractSegmentation of the human eye retinal image is an essential step for proper examination and diagnosis by the ophthalmologists or eye care specialists. A technique for vessel segmentation of retinal images is proposed. Retinal images are mostly low‐light images, which are first processed for enhancement of light as well as for detail amplification. Illumination of low‐light images is enhanced, and details are amplified using content‐adaptive filters. For extraction of vessels from retinal images, after low‐light and detail enhancement, the B‐cosfire filter is modified by including extraction of details and small elements, which may otherwise be ignored. A modified B‐cosfire filter is used to extract vessels while minimizing false edges and halo artifacts. The morphological opening is performed to crop vessels that are falsely segmented. The technique is contrasted with other existing methods in terms of accuracy using publicly available datasets. The proposed technique is tested on STARE, CHASE‐DB1, and DRIVE databases. The outcome of the proposed procedure has better accuracy, preserved edges, minimum noise, and artifacts than the state‐of‐the‐art techniques.

Computer-aided retinal vessel segmentation in retinal images: convolutional neural networks

In medicine, diagnosis is as important as treatment. Retinal blood vessels are the most easily visible vessels in the whole body, and therefore, play a key role in the diagnosis of numerous diseases and eye disorders. Systematic and eye diseases cause morphologic variations, such as the growing, narrowing or branching of retinal blood vessels. Imaging-based screening of retinal blood vessels plays an important role in the identification and follow-up of eye diseases. Therefore, automatic retinal vessel segmentation can be used to diagnose and monitor those diseases. Computer-aided algorithms are required for the analysis of progression of eye diseases. This study proposes a hybrid method that provides a combination of pre-processing and data augmentation methods with a deep learning model. Pre-processing was used to solve the irregular clarification problems and to form a contrast between the background and retinal blood vessels. After pre-processing step, a convolutional neural network (CNN) was designed and then trained for the extraction of retinal blood vessels. In the training phase, data augmentation was performed to improve training performance. The CNN was trained and tested in the DRIVE database, which is commonly used in retinal blood vessel segmentation and publicly available for studies in this area. Results showed that the proposed system extracted vessels with a sensitivity of 77.78%, specificity of 97,84%, precision of 84.17% and accuracy of 95.27%. This study also compared the results to those of previous studies. The comparison showed that the proposed method is an efficient and successful method for extracting retinal blood vessels. Moreover, the pre-processing phases improved the system performance. We believe that the proposed method and results will make contribution to the literature.

Blood vessel segmentation and extraction using H-minima method based on image processing techniques

In this paper, the H-minima transform is used for blood vessel segmentation. The aim of this study is to get the high accuracy of blood vessel segmentation in retinal images. In this study the good result and good performance were got. We compared our result with other methods. Also for simulation result we implemented on DRIVE and STARE database. The proposed method shows very remarkable performance on pathological retinal images. For the implementing of the proposed method MATLAB 2019a software is used. The running time of this method was 1 s for each image and the average accuracy for STARE dataset and DRIVE dataset achieved to 0.9591 and 0.9672 respectively.

Semi-Supervised Learning Method of U-Net Deep Learning Network for Blood Vessel Segmentation in Retinal Images

Blood vessel segmentation methods based on deep neural networks have achieved satisfactory results. However, these methods are usually supervised learning methods, which require large numbers of retinal images with high quality pixel-level ground-truth labels. In practice, the task of labeling these retinal images is very costly, financially and in human effort. To deal with these problems, we propose a semi-supervised learning method which can be used in blood vessel segmentation with limited labeled data. In this method, we use the improved U-Net deep learning network to segment the blood vessel tree. On this basis, we implement the U-Net network-based training dataset updating strategy. A large number of experiments are presented to analyze the segmentation performance of the proposed semi-supervised learning method. The experiment results demonstrate that the proposed methodology is able to avoid the problems of insufficient hand-labels, and achieve satisfactory performance.

Echo state network‐based feature extraction for efficient color image segmentation

SummaryImage segmentation plays a crucial role in many image processing and understanding applications. Despite the huge number of proposed image segmentation techniques, accurate segmentation remains a significant challenge in image analysis. This article investigates the viability of using echo state network (ESN), a biologically inspired recurrent neural network, as features extractor for efficient color image segmentation. First, an ensemble of initial pixel features is extracted from the original images and injected into the ESN reservoir. Second, the internal activations of the reservoir neurons are used as new pixel features. Third, the new features are classified using a feed forward neural network as a readout layer for the ESN. The quality of the pixel features produced by the ESN is evaluated through extensive series of experiments conducted on real world image datasets. The optimal operating range of different ESN setup parameters for producing competitive quality features is identified. The performance of the proposed ESN‐based framework is also evaluated on a domain‐specific application, namely, blood vessel segmentation in retinal images where experiments are conducted on the widely used digital retinal images for vessel extraction (DRIVE) dataset. The obtained results demonstrate that the proposed method outperforms state‐of‐the‐art general segmentation techniques in terms of performance with an F‐score of 0.92 ± 0.003 on the segmentation evaluation dataset. In addition, the proposed method achieves a comparable segmentation accuracy (0.9470) comparing with reported techniques of segmentation of blood vessels in images of retina and outperform them in terms of processing time. The average time required by our technique to segment one retinal image from DRIVE dataset is 8 seconds. Furthermore, empirically derived guidelines are proposed for adequately setting the ESN parameters for effective color image segmentation.

Chaining a U-Net With a Residual U-Net for Retinal Blood Vessels Segmentation

Retina images are the only non-invasive way of accessing the cardiovascular system, offering us a means of observing patterns such as microaneurysms, hemorrhages and the vasculature structure which can be used to diagnose a variety of diseases. The main goal of this paper is to automate retinal blood vessel segmentation with a good tradeoff between blood vessel classification and training time in the presence of high unbalanced classes. In this work, a novel methodology is proposed using two convolutional neural networks (CNN’s), chained to each other. The second CNN has been designed with residual network blocks, which joined to the information flow from the first, give us metrics like recall and F1-Score, which are, in most cases, superior to state of the art in vessel segmentation task. We tested this work on two public datasets for blood vessel segmentation in retinal images showing that this work outperforms many of other contributions by other authors.

A Composite Feature Set Based Blood Vessel Segmentation in Retinal Images through Supervised Learning

A Composite Feature Set Based Blood Vessel Segmentation in Retinal Images through Supervised Learning

Deep Retinal Image Segmentation with Regularization Under Geometric Priors.

Vessel segmentation of retinal images is a key diagnostic capability in ophthalmology. This problem faces several challenges including low contrast, variable vessel size and thickness, and presence of interfering pathology such as micro-aneurysms and hemorrhages. Early approaches addressing this problem employed hand-crafted filters to capture vessel structures, accompanied by morphological post-processing. More recently, deep learning techniques have been employed with significantly enhanced segmentation accuracy. We propose a novel domain enriched deep network that consists of two components: 1) a representation network that learns geometric features specific to retinal images, and 2) a custom designed computationally efficient residual task network that utilizes the features obtained from the representation layer to perform pixel-level segmentation. The representation and task networks are jointly learned for any given training set. To obtain physically meaningful and practically effective representation filters, we propose two new constraints that are inspired by expected prior structure on these filters: 1) orientation constraint that promotes geometric diversity of curvilinear features, and 2) a data adaptive noise regularizer that penalizes false positives. Multi-scale extensions are developed to enable accurate detection of thin vessels. Experiments performed on three challenging benchmark databases under a variety of training scenarios show that the proposed prior guided deep network outperforms state of the art alternatives as measured by common evaluation metrics, while being more economical in network size and inference time.

Recurrent residual U-Net for medical image segmentation.

Deep learning (DL)-based semantic segmentation methods have been providing state-of-the-art performance in the past few years. More specifically, these techniques have been successfully applied in medical image classification, segmentation, and detection tasks. One DL technique, U-Net, has become one of the most popular for these applications. We propose a recurrent U-Net model and a recurrent residual U-Net model, which are named RU-Net and R2U-Net, respectively. The proposed models utilize the power of U-Net, residual networks, and recurrent convolutional neural networks. There are several advantages to using these proposed architectures for segmentation tasks. First, a residual unit helps when training deep architectures. Second, feature accumulation with recurrent residual convolutional layers ensures better feature representation for segmentation tasks. Third, it allows us to design better U-Net architectures with the same number of network parameters with better performance for medical image segmentation. The proposed models are tested on three benchmark datasets, such as blood vessel segmentation in retinal images, skin cancer segmentation, and lung lesion segmentation. The experimental results show superior performance on segmentation tasks compared to equivalent models, including a variant of a fully connected convolutional neural network called SegNet, U-Net, and residual U-Net.

Comparative Analysis of Vessel Segmentation Techniques in Retinal Images

The blood vessels are the primary anatomical structure that can be visible in retinal images. The segmentation of retinal blood vessels has been accepted worldwide for the diagnosis of both cardiovascular (CVD) and retinal diseases. Thus, it requires an appropriate vessel segmentation method for automatic detection of retinal diseases such as diabetic retinopathy and cataract. The detection of retinal diseases using computer-aided diagnosis (CAD) can help people to avoid the risks of visual impairment and save medical resources. This survey presents a comparative analysis of various machine learning and deep learning-based methods for automated blood vessel segmentation in retinal images. This paper briefly describes fundus photography, publicly available retinal databases, pre-processing and post-processing techniques for retinal vessels segmentation. A comprehensive review of the state of the art supervised and unsupervised blood vessel segmentation methodologies are presented in this paper. The objective of this study is to establish a professional structure to familiarize an individual with up-to-date vessel segmentation techniques. Moreover, we compared these approaches to the dataset, evaluation metrics, pre-processing and post-processing steps, feature extraction, segmentation methods, and induced results.