JOINEDTrans: Prior guided multi-task transformer for joint optic disc/cup segmentation and fovea detection

Deep learning-based image segmentation and detection models have largely improved the efficiency of analyzing retinal landmarks such as optic disc (OD), optic cup (OC), and fovea. However, factors including ophthalmic disease-related lesions and low image quality issues may severely complicate automatic OD/OC segmentation and fovea detection. Most existing works treat the identification of each landmark as a single task, and take into account no prior information. To address these issues, we propose a prior guided multi-task transformer framework for joint OD/OC segmentation and fovea detection, named JOINEDTrans. JOINEDTrans effectively combines various spatial features of the fundus images, relieving the structural distortions induced by lesions and other imaging issues. It contains a segmentation branch and a detection branch. To be noted, we employ an encoder with prior-learning in a vessel segmentation task to effectively exploit the positional relationship among vessel, OD/OC, and fovea, successfully incorporating spatial prior into the proposed JOINEDTrans framework. There are a coarse stage and a fine stage in JOINEDTrans. In the coarse stage, OD/OC coarse segmentation and fovea heatmap localization are obtained through a joint segmentation and detection module. In the fine stage, we crop regions of interest for subsequent refinement and use predictions obtained in the coarse stage to provide additional information for better performance and faster convergence. Experimental results demonstrate that JOINEDTrans outperforms existing state-of-the-art methods on the publicly available GAMMA, REFUGE, and PALM fundus image datasets. We make our code available at https://github.com/HuaqingHe/JOINEDTrans.

Global Transformer and Dual Local Attention Network via Deep-Shallow Hierarchical Feature Fusion for Retinal Vessel Segmentation

Clinically, retinal vessel segmentation is a significant step in the diagnosis of fundus diseases. However, recent methods generally neglect the difference of semantic information between deep and shallow features, which fail to capture the global and local characterizations in fundus images simultaneously, resulting in the limited segmentation performance for fine vessels. In this article, a global transformer (GT) and dual local attention (DLA) network via deep-shallow hierarchical feature fusion (GT-DLA-dsHFF) are investigated to solve the above limitations. First, the GT is developed to integrate the global information in the retinal image, which effectively captures the long-distance dependence between pixels, alleviating the discontinuity of blood vessels in the segmentation results. Second, DLA, which is constructed using dilated convolutions with varied dilation rates, unsupervised edge detection, and squeeze-excitation block, is proposed to extract local vessel information, consolidating the edge details in the segmentation result. Finally, a novel deep-shallow hierarchical feature fusion (dsHFF) algorithm is studied to fuse the features in different scales in the deep learning framework, respectively, which can mitigate the attenuation of valid information in the process of feature fusion. We verified the GT-DLA-dsHFF on four typical fundus image datasets. The experimental results demonstrate our GT-DLA-dsHFF achieves superior performance against the current methods and detailed discussions verify the efficacy of the proposed three modules. Segmentation results of diseased images show the robustness of our proposed GT-DLA-dsHFF. Implementation codes will be available on https://github.com/YangLibuaa/GT-DLA-dsHFF.

An improved U-net based retinal vessel image segmentation method

Diabetic retinopathy is not just the most common complication of diabetes but also the leading cause of adult blindness. Currently, doctors determine the cause of diabetic retinopathy primarily by diagnosing fundus images. Large-scale manual screening is difficult to achieve for retinal health screen. In this paper, we proposed an improved U-net network for segmenting retinal vessels. Firstly, due to the lack of retinal data, pre-processing of the raw data is required. The data processed by grayscale transformation, normalization, CLAHE, gamma transformation. Data augmentation can prevent overfitting in the training process. Secondly, the basic network structure model U-net is built, and the Bi-FPN network is fused based on U-net. Datasets from a public challenge are used to evaluate the performance of the proposed method, which is able to detect vessel SP of 0.8604, SE of 0.9767, ACC of 0.9651, and AUC of 0.9787.

A Multichannel Deep Neural Network for Retina Vessel Segmentation via a Fusion Mechanism

Changes in fundus blood vessels reflect the occurrence of eye diseases, and from this, we can explore other physical diseases that cause fundus lesions, such as diabetes and hypertension complication. However, the existing computational methods lack high efficiency and precision segmentation for the vascular ends and thin retina vessels. It is important to construct a reliable and quantitative automatic diagnostic method for improving the diagnosis efficiency. In this study, we propose a multichannel deep neural network for retina vessel segmentation. First, we apply U-net on original and thin (or thick) vessels for multi-objective optimization for purposively training thick and thin vessels. Then, we design a specific fusion mechanism for combining three kinds of prediction probability maps into a final binary segmentation map. Experiments show that our method can effectively improve the segmentation performances of thin blood vessels and vascular ends. It outperforms many current excellent vessel segmentation methods on three public datasets. In particular, it is pretty impressive that we achieve the best F1-score of 0.8247 on the DRIVE dataset and 0.8239 on the STARE dataset. The findings of this study have the potential for the application in an automated retinal image analysis, and it may provide a new, general, and high-performance computing framework for image segmentation.

Review of Various Tasks Performed in the Preprocessing Phase of a Diabetic Retinopathy Diagnosis System

Diabetic Retinopathy (DR) is a major cause of blindness in diabetic patients. The increasing population of diabetic patients and difficulty to diagnose it at an early stage are limiting the screening capabilities of manual diagnosis by ophthalmologists. Color fundus images are widely used to detect DR lesions due to their comfortable, cost-effective and non-invasive acquisition procedure. Computer Aided Diagnosis (CAD) of DR based on these images can assist ophthalmologists and help in saving many sight years of diabetic patients. In a CAD system, preprocessing is a crucial phase, which significantly affects its performance. Commonly used preprocessing operations are the enhancement of poor contrast, balancing the illumination imbalance due to the spherical shape of a retina, noise reduction, image resizing to support multi-resolution, color normalization, extraction of a field of view (FOV), etc. Also, the presence of blood vessels and optic discs makes the lesion detection more challenging because these two artifacts exhibit specific attributes, which are similar to those of DR lesions. Preprocessing operations can be broadly divided into three categories: 1) fixing the native defects, 2) segmentation of blood vessels, and 3) localization and segmentation of optic discs. This paper presents a review of the state-of-the-art preprocessing techniques related to three categories of operations, highlighting their significant aspects and limitations. The survey is concluded with the most effective preprocessing methods, which have been shown to improve the accuracy and efficiency of the CAD systems.

An enhanced deep image model for glaucoma diagnosis using feature-based detection in retinal fundus

This paper proposes a deep image analysis-based model for glaucoma diagnosis that uses several features to detect the formation of glaucoma in retinal fundus. These features are combined with most extracted parameters like inferior, superior, nasal, and temporal region area, and cup-to-disc ratio that overall forms a deep image analysis. This proposed model is exercised to investigate the various aspects related to the prediction of glaucoma in retinal fundus images that help the ophthalmologist in making better decisions for the human eye. The proposed model is presented with the combination of four machine learning algorithms that provide the classification accuracy of 98.60% while other existing models like support vector machine (SVM), K-nearest neighbors (KNN), and Naïve Bayes provide individually with accuracies of 97.61%, 90.47%, and 95.23% respectively. These results clearly demonstrate that this proposed model offers the best methodology to an early diagnosis of glaucoma in retinal fundus.

A review of machine learning methods for retinal blood vessel segmentation and artery/vein classification

The eye affords a unique opportunity to inspect a rich part of the human microvasculature non-invasively via retinal imaging. Retinal blood vessel segmentation and classification are prime steps for the diagnosis and risk assessment of microvascular and systemic diseases. A high volume of techniques based on deep learning have been published in recent years. In this context, we review 158 papers published between 2012 and 2020, focussing on methods based on machine and deep learning (DL) for automatic vessel segmentation and classification for fundus camera images. We divide the methods into various classes by task (segmentation or artery-vein classification), technique (supervised or unsupervised, deep and non-deep learning, hand-crafted methods) and more specific algorithms (e.g. multiscale, morphology). We discuss advantages and limitations, and include tables summarising results at-a-glance. Finally, we attempt to assess the quantitative merit of DL methods in terms of accuracy improvement compared to other methods. The results allow us to offer our views on the outlook for vessel segmentation and classification for fundus camera images.

Analysis on diagnosing diabetic retinopathy by segmenting blood vessels, optic disc and retinal abnormalities

The main intention of mass screening programmes for Diabetic Retinopathy (DR) is to detect and diagnose the disorder earlier than it leads to vision loss. Automated analysis of retinal images has the likelihood to improve the efficacy of screening programmes when compared over the manual image analysis. This article plans to develop a framework for the detection of DR from the retinal fundus images using three evaluations based on optic disc, blood vessels and retinal abnormalities. Initially, the pre-processing steps like green channel conversion and Contrast Limited Adaptive Histogram Equalisation is done. Further, the segmentation procedure starts with optic disc segmentation by open-close watershed transform, blood vessel segmentation by grey level thresholding and abnormality segmentation (hard exudates, haemorrhages, Microaneurysm and soft exudates) by top hat transform and Gabor filtering mechanisms. From the three segmented images, the feature like local binary pattern, texture energy measurement, Shanon's and Kapur's entropy are extracted, which is subjected to optimal feature selection process using the new hybrid optimisation algorithm termed as Trial-based Bypass Improved Dragonfly Algorithm (TB - DA). These features are given to hybrid machine learning algorithm with the combination of NN and DBN. As a modification, the same hybrid TB - DA is used to enhance the training of hybrid classifier, which outputs the categorisation as normal, mild, moderate or severe images based on three components.

Automated segmentation of optic disc using statistical region merging and morphological operations

Accurate Optic Disc (OD) segmentation is vital in designing systems that aid the diagnosis and evaluation of early phases of retinal diseases. However, in many images, the OD boundary is ambiguous, which makes the automated OD segmentation process very challenging. A method to segment OD based on statistical region merging and morphological operations is proposed in this paper. The proposed method is tested on standard databases MESSIDOR, DIARETDB1, DIARETDB0, and DRIONS-DB. The average overlap ratios are found to be 91.35% for DIARETDB1 images, 88.80% for DRIONS-DB images, 86.60% for DIARETDB0 images and 89.68% for MESSIDOR images, with average accuracies of 99.68%, 99.89%, 99.69%, and 99.93% respectively. A comparison with alternative methods showed that the proposed algorithm in OD segmentation is better than existing ones.

HG-SVNN: HARMONIC GENETIC-BASED SUPPORT VECTOR NEURAL NETWORK CLASSIFIER FOR THE GLAUCOMA DETECTION

Glaucoma has emerged as the one of the leading causes of blindness. Even though the diagnosis of this disease has not yet been found, the early detection can cure the glaucoma disease. Various works presented for the glaucoma detection have many disadvantages such as increased run time, complex architecture, etc., during the real-time implementations. This work introduces the glaucoma detection system based on the proposed harmonic genetic-based support vector neural network (HG-SVNN) classifier. The proposed system detects glaucoma in the database through four major steps, (1) pre-processing, (2) proposed hybrid feature extraction, (3) segmentation and (4) classification through the proposed HG-SVNN classifier. The proposed model uses both the statistical and the vessel features from the segmented and the pre-processed images to construct the feature vector. The proposed HG-SVNN classifier uses both the harmonic operator and the genetic algorithm (GA) for the neural network training. From the simulation results, it is evident that the proposed glaucoma detection system has better performance than the existing works with the values of 0.945, 0.9, 0.9333 and 0.86667 for the segmentation accuracy, accuracy, sensitivity and specificity metric.

Automatic Annotation of Retinal Layers in Optical Coherence Tomography Images

Early diagnosis of retinal OCT images has been shown to curtail blindness and visual impairments. However, the advancement of ophthalmic imaging technologies produces an ever-growing scale of retina images, both in volume and variety, which overwhelms the ophthalmologist ability to segment these images. While many automated methods exist, speckle noise and intensity inhomogeneity negatively impacts the performance of these methods. We present a comprehensive and fully automatic method for annotation of retinal layers in OCT images comprising of fuzzy histogram hyperbolisation (FHH) and graph cut methods to segment 7 retinal layers across 8 boundaries. The FHH handles speckle noise and inhomogeneity in the preprocessing step. Then the normalised vertical image gradient, and it’s inverse to represent image intensity in calculating two adjacency matrices and then the FHH reassigns the edge-weights to make edges along retinal boundaries have a low cost, and graph cut method identifies the shortest-paths (layer boundaries). The method is evaluated on 150 B-Scan images, 50 each from the temporal, foveal and nasal regions were used in our study. Promising experimental results have been achieved with high tolerance and adaptability to contour variance and pathological inconsistency of the retinal layers in all (temporal, foveal and nasal) regions. The method also achieves high accuracy, sensitivity, and Dice score of 0.98360, 0.9692 and 0.9712, respectively in segmenting the retinal nerve fibre layer. The annotation can facilitate eye examination by providing accurate results. The integration of the vertical gradients into the graph cut framework, which captures the unique characteristics of retinal structures, is particularly useful in finding the actual minimum paths across multiple retinal layer boundaries. Prior knowledge plays an integral role in image segmentation.

A review on computer-aided recent developments for automatic detection of diabetic retinopathy

Diabetic retinopathy is a serious microvascular disorder that might result in loss of vision and blindness. It seriously damages the retinal blood vessels and reduces the light-sensitive inner layer of the eye. Due to the manual inspection of retinal fundus images on diabetic retinopathy to detect the morphological abnormalities in Microaneurysms (MAs), Exudates (EXs), Haemorrhages (HMs), and Inter retinal microvascular abnormalities (IRMA) is very difficult and time consuming process. In order to avoid this, the regular follow-up screening process, and early automatic Diabetic Retinopathy detection are necessary. This paper discusses various methods of analysing automatic retinopathy detection and classification of different grading based on the severity levels. In addition, retinal blood vessel detection techniques are also discussed for the ultimate detection and diagnostic procedure of proliferative diabetic retinopathy. Furthermore, the paper elaborately discussed the systematic review accessed by authors on various publicly available databases collected from different medical sources. In the survey, meta-analysis of several methods for diabetic feature extraction, segmentation and various types of classifiers have been used to evaluate the system performance metrics for the diagnosis of DR. This survey will be helpful for the technical persons and researchers who want to focus on enhancing the diagnosis of a system that would be more powerful in real life.

Extraction of Blood Vessels in Fundus Images of Retina through Hybrid Segmentation Approach

A hybrid segmentation algorithm is proposed is this paper to extract the blood vesselsfrom the fundus image of retina. Fundus camera captures the posterior surface of the eye and thecaptured images are used to diagnose diseases, like Diabetic Retinopathy, Retinoblastoma, Retinalhaemorrhage, etc. Segmentation or extraction of blood vessels is highly required, since the analysisof vessels is crucial for diagnosis, treatment planning, and execution of clinical outcomes in the fieldof ophthalmology. It is derived from the literature review that no unique segmentation algorithm issuitable for images of different eye-related diseases and the degradation of the vessels differ frompatient to patient. If the blood vessels are extracted from the fundus images, it will make thediagnosis process easier. Hence, this paper aims to frame a hybrid segmentation algorithmexclusively for the extraction of blood vessels from the fundus image. The proposed algorithm ishybridized with morphological operations, bottom hat transform, multi-scale vessel enhancement(MSVE) algorithm, and image fusion. After execution of the proposed segmentation algorithm, thearea-based morphological operator is applied to highlight the blood vessels. To validate theproposed algorithm, the results are compared with the ground truth of the High-Resolution Fundus(HRF) images dataset. Upon comparison, it is inferred that the proposed algorithm segments theblood vessels with more accuracy than the existing algorithms.

Retinal Fundus Image for Glaucoma Detection: A Review and Study

Glaucoma is one of the severe visual diseases that lead to damage the eyes irreversibly by affecting the optic nerve fibers and astrocytes. Consequently, the early detection of glaucoma plays a virtual role in the medical field. The literature presents various techniques for the early detection of glaucoma. Among the various techniques, retinal image-based detection plays a major role as it comes under noninvasive methods of detection. While detecting glaucoma disorder using retinal images, various medical features of the eyes, such as retinal nerve fiber layer, cup-to-disc ratio, apex point, optic disc, and optic nerve head, and image features, such as Haralick texture, higher-order spectra, and wavelet energy, are used. In this paper, a review and study were conducted for the different techniques of glaucoma detection using retinal fundus images. Accordingly, 45 research papers were reviewed and the analysis was provided based on the extracted features, classification accuracy, and the usage of different data sets, such as DIARETDB1 data set, MESSIDOR data set, IPN data set, ZEISS data set, local data set, and real data set. Finally, we present the various research issues and solutions that can be useful for the researchers to accomplish further research on glaucoma detection.

Mean curvature and texture constrained composite weighted random walk algorithm for optic disc segmentation towards glaucoma screening

Accurate optic disc (OD) segmentation is an important step in obtaining cup-to-disc ratio-based glaucoma screening using fundus imaging. It is a challenging task because of the subtle OD boundary, blood vessel occlusion and intensity inhomogeneity. In this Letter, the authors propose an improved version of the random walk algorithm for OD segmentation to tackle such challenges. The algorithm incorporates the mean curvature and Gabor texture energy features to define the new composite weight function to compute the edge weights. Unlike the deformable model-based OD segmentation techniques, the proposed algorithm remains unaffected by curve initialisation and local energy minima problem. The effectiveness of the proposed method is verified with DRIVE, DIARETDB1, DRISHTI-GS and MESSIDOR database images using the performance measures such as mean absolute distance, overlapping ratio, dice coefficient, sensitivity, specificity and precision. The obtained OD segmentation results and quantitative performance measures show robustness and superiority of the proposed algorithm in handling the complex challenges in OD segmentation.

Automatic Detection of Blood Vessel in Retinal Images Using Vesselness Enhancement Filter and Adaptive Thresholding

Retinal blood vessels detection and measurement of morphological attributes, such as length, width, sinuosity and corners are very much important for the diagnosis and treatment of different ocular diseases including diabetic retinopathy (DR), glaucoma, and hypertension. This paper presents a integration method for blood vessels detection in fundus retinal images. The proposed method consists of two main steps. The first step is pre-processing of retinal image to improve the retinal images by evaluation of several image enhancement techniques. The second step is vessels detection, the vesselness filter is usually used to enhance the blood vessels. The enhancement filter is designed from the adaptive thresholding of the output of the vesselness filter for vessels detection. The algorithms performance is compared and analyzed on three publicly available databases (DRIVE, STARE and CHASE_DB) of retinal images using a number of measures, which include accuracy, sensitivity, and specificity.

A Novel Adaptive Deformable Model for Automated Optic Disc and Cup Segmentation to Aid Glaucoma Diagnosis

This paper proposes a novel Adaptive Region-based Edge Smoothing Model (ARESM) for automatic boundary detection of optic disc and cup to aid automatic glaucoma diagnosis. The novelty of our approach consists of two aspects: 1) automatic detection of initial optimum object boundary based on a Region Classification Model (RCM) in a pixel-level multidimensional feature space; 2) an Adaptive Edge Smoothing Update model (AESU) of contour points (e.g. misclassified or irregular points) based on iterative force field calculations with contours obtained from the RCM by minimising energy function (an approach that does not require predefined geometric templates to guide auto-segmentation). Such an approach provides robustness in capturing a range of variations and shapes. We have conducted a comprehensive comparison between our approach and the state-of-the-art existing deformable models and validated it with publicly available datasets. The experimental evaluation shows that the proposed approach significantly outperforms existing methods. The generality of the proposed approach will enable segmentation and detection of other object boundaries and provide added value in the field of medical image processing and analysis.

Blood vessel extraction of diabetic retinopathy using optimized enhanced images and matched filter

Accurate extraction of structural changes in the blood vessels of the retina is an essential task in diagnosis of retinopathy. Matched filter (MF) technique is the effective way to extract blood vessels, but the effectiveness is reduced due to noisy images. The concept of MF and MF with first-order derivative of Gaussian (MF-FDOG) has been implemented for retina images of the DRIVE database. The optimized particle swarm optimization (PSO) algorithm is used for enhancing the images by edgels to improve the performance of filters. The vessels were detected by the response of thresholding to the MF, whereas the threshold is adjusted in response to the FDOG. The PSO-based enhanced MF response significantly improved the performances of filters to extract fine blood vessels structures. Experimental results show that the proposed method based on enhanced images improved the accuracy to 91.1%, which is higher than that of MF and MF-FDOG, respectively. The peak signal-to-noise ratio was also found to be higher with low mean square error values in enhanced MF response. The accuracy, sensitivity, and specificity values are significantly improved among MF, MF-FDOG, and PSO-enhanced images ([Formula: see text]).

Retinal blood vessel extraction using tunable bandpass filter and fuzzy conditional entropy

BACKGROUND AND OBJECTIVES

Extraction of blood vessels on retinal images plays a significant role for screening of different opthalmologic diseases. However, accurate extraction of the entire and individual type of vessel silhouette from the noisy images with poorly illuminated background is a complicated task. To this aim, an integrated system design platform is suggested in this work for vessel extraction using a sequential bandpass filter followed by fuzzy conditional entropy maximization on matched filter response.

METHODS

At first noise is eliminated from the image under consideration through curvelet based denoising. To include the fine details and the relatively less thick vessel structures, the image is passed through a bank of sequential bandpass filter structure optimized for contrast enhancement. Fuzzy conditional entropy on matched filter response is then maximized to find the set of multiple optimal thresholds to extract the different types of vessel silhouettes from the background. Differential Evolution algorithm is used to determine the optimal gain in bandpass filter and the combination of the fuzzy parameters. Using the multiple thresholds, retinal image is classified as the thick, the medium and the thin vessels including neovascularization.

RESULTS

Performance evaluated on different publicly available retinal image databases shows that the proposed method is very efficient in identifying the diverse types of vessels. Proposed method is also efficient in extracting the abnormal and the thin blood vessels in pathological retinal images. The average values of true positive rate, false positive rate and accuracy offered by the method is 76.32%, 1.99% and 96.28%, respectively for the DRIVE database and 72.82%, 2.6% and 96.16%, respectively for the STARE database. Simulation results demonstrate that the proposed method outperforms the existing methods in detecting the various types of vessels and the neovascularization structures.

CONCLUSIONS

The combination of curvelet transform and tunable bandpass filter is found to be very much effective in edge enhancement whereas fuzzy conditional entropy efficiently distinguishes vessels of different widths.

Localization and segmentation of optic disc in retinal images using circular Hough transform and grow-cut algorithm

Automated retinal image analysis has been emerging as an important diagnostic tool for early detection of eye-related diseases such as glaucoma and diabetic retinopathy. In this paper, we have presented a robust methodology for optic disc detection and boundary segmentation, which can be seen as the preliminary step in the development of a computer-assisted diagnostic system for glaucoma in retinal images. The proposed method is based on morphological operations, the circular Hough transform and the grow-cut algorithm. The morphological operators are used to enhance the optic disc and remove the retinal vasculature and other pathologies. The optic disc center is approximated using the circular Hough transform, and the grow-cut algorithm is employed to precisely segment the optic disc boundary. The method is quantitatively evaluated on five publicly available retinal image databases DRIVE, DIARETDB1, CHASE_DB1, DRIONS-DB, Messidor and one local Shifa Hospital Database. The method achieves an optic disc detection success rate of 100% for these databases with the exception of 99.09% and 99.25% for the DRIONS-DB, Messidor, and ONHSD databases, respectively. The optic disc boundary detection achieved an average spatial overlap of 78.6%, 85.12%, 83.23%, 85.1%, 87.93%, 80.1%, and 86.1%, respectively, for these databases. This unique method has shown significant improvement over existing methods in terms of detection and boundary extraction of the optic disc.

MAT-Based Thinning for Line Patterns

Reducing branching effect and increasing boundary noise immunity are of great importance for thinning patterns. An approach based on medial axis transform (MAT) to obtain a connected 1-pixel wide skeleton with few redundant branches is presented in this paper. Though the obtained skeleton by MAT is isotropic with few redundant branches, however, the skeleton points are usually disconnected. In order to rend the merits of the MAT and avoid its disadvantages, the proposed approach is composed of distance-map generation, grouping, ridge-path linking, and refining to obtain the connected 1-pixel wide thin line. The ridge-path linking strategy can guarantee the skeletons connected, whereas the refining process can be readily performed by a conventional thinning process to obtain the 1-pixel wide thinned pattern. The performances investigated by branching effect, signal-to-noise ratio (SNR), and measurement of skeleton deviation (MSD) confirm the feasibility of the proposed MAT-based thinning for line patterns.