Blood vessel segmentation in color fundus images based on regional and Hessian features

Colorimetric Analyses of the Optic Nerve Head and Retina Indicate Increased Blood Flow After Vitrectomy

Purpose

The purpose of this study was to evaluate the impact of vitrectomy and posterior hyaloid (PH) peeling on color alteration of optic nerve head (ONH) and retina as a surrogate biomarker of induced perfusion changes.

Methods

Masked morphometric and colorimetric analyses were conducted on preoperative (<1 month) and postoperative (<18 months) color fundus photographs of 54 patients undergoing vitrectomy, either with (44) or without (10) PH peeling and 31 years of age and gender-matched control eyes. Images were calibrated according to the hue and saturation values of the parapapillary venous blood column. Chromatic spectra of the retinal pigment epithelium and choroid were subtracted to avoid color aberrations. Red, green, and blue (RGB) bit values over the ONH and retina were plotted within the constructed RGB color space to analyze vitrectomy-induced color shift. Vitrectomy-induced parapapillary vein caliber changes were also computed morphometrically.

Results

A significant post-vitrectomy red hue shift was noted on the ONH (37.1 degrees ± 10.9 degrees vs. 4.1 degrees ± 17.7 degrees, P < 0.001), which indicates a 2.8-fold increase in blood perfusion compared to control (2.6 ± 1.9 vs. 0.9 ± 1.8, P < 0.001). A significant post-vitrectomy increase in the retinal vein diameter was also noticed (6.8 ± 6.4% vs. 0.1 ± 0.3%, P < 0.001), which was more pronounced with PH peeling (7.9 ± 6.6% vs. 3.1 ± 4.2%, P = 0.002).

Conclusions

Vitrectomy and PH peeling increase ONH and retinal blood flow. Colorimetric and morphometric analyses offer valuable insights for future artificial intelligence and deep learning applications in this field.

Translational Relevance

The methodology described herein can easily be applied in different clinical settings and may enlighten the beneficial effects of vitrectomy in several retinal vascular diseases.

Segmentation and Classification Approaches of Clinically Relevant Curvilinear Structures: A Review

Detection of curvilinear structures from microscopic images, which help the clinicians to make an unambiguous diagnosis is assuming paramount importance in recent clinical practice. Appearance and size of dermatophytic hyphae, keratitic fungi, corneal and retinal vessels vary widely making their automated detection cumbersome. Automated deep learning methods, endowed with superior self-learning capacity, have superseded the traditional machine learning methods, especially in complex images with challenging background. Automatic feature learning ability using large input data with better generalization and recognition capability, but devoid of human interference and excessive pre-processing, is highly beneficial in the above context. Varied attempts have been made by researchers to overcome challenges such as thin vessels, bifurcations and obstructive lesions in retinal vessel detection as revealed through several publications reviewed here. Revelations of diabetic neuropathic complications such as tortuosity, changes in the density and angles of the corneal fibers have been successfully sorted in many publications reviewed here. Since artifacts complicate the images and affect the quality of analysis, methods addressing these challenges have been described. Traditional and deep learning methods, that have been adapted and published between 2015 and 2021 covering retinal vessels, corneal nerves and filamentous fungi have been summarized in this review. We find several novel and meritorious ideas and techniques being put to use in the case of retinal vessel segmentation and classification, which by way of cross-domain adaptation can be utilized in the case of corneal and filamentous fungi also, making suitable adaptations to the challenges to be addressed.

Recent trends and advances in fundus image analysis: A review

Automated retinal image analysis holds prime significance in the accurate diagnosis of various critical eye diseases that include diabetic retinopathy (DR), age-related macular degeneration (AMD), atherosclerosis, and glaucoma. Manual diagnosis of retinal diseases by ophthalmologists takes time, effort, and financial resources, and is prone to error, in comparison to computer-aided diagnosis systems. In this context, robust classification and segmentation of retinal images are primary operations that aid clinicians in the early screening of patients to ensure the prevention and/or treatment of these diseases. This paper conducts an extensive review of the state-of-the-art methods for the detection and segmentation of retinal image features. Existing notable techniques for the detection of retinal features are categorized into essential groups and compared in depth. Additionally, a summary of quantifiable performance measures for various important stages of retinal image analysis, such as image acquisition and preprocessing, is provided. Finally, the widely used in the literature datasets for analyzing retinal images are described and their significance is emphasized.

Retinal Vessel Segmentation With Skeletal Prior and Contrastive Loss

The morphology of retinal vessels is closely associated with many kinds of ophthalmic diseases. Although huge progress in retinal vessel segmentation has been achieved with the advancement of deep learning, some challenging issues remain. For example, vessels can be disturbed or covered by other components presented in the retina (such as optic disc or lesions). Moreover, some thin vessels are also easily missed by current methods. In addition, existing fundus image datasets are generally tiny, due to the difficulty of vessel labeling. In this work, a new network called SkelCon is proposed to deal with these problems by introducing skeletal prior and contrastive loss. A skeleton fitting module is developed to preserve the morphology of the vessels and improve the completeness and continuity of thin vessels. A contrastive loss is employed to enhance the discrimination between vessels and background. In addition, a new data augmentation method is proposed to enrich the training samples and improve the robustness of the proposed model. Extensive validations were performed on several popular datasets (DRIVE, STARE, CHASE, and HRF), recently developed datasets (UoA-DR, IOSTAR, and RC-SLO), and some challenging clinical images (from RFMiD and JSIEC39 datasets). In addition, some specially designed metrics for vessel segmentation, including connectivity, overlapping area, consistency of vessel length, revised sensitivity, specificity, and accuracy were used for quantitative evaluation. The experimental results show that, the proposed model achieves state-of-the-art performance and significantly outperforms compared methods when extracting thin vessels in the regions of lesions or optic disc. Source code is available at https://www.github.com/tyb311/SkelCon.

Retinal Vessel Segmentation, a Review of Classic and Deep Methods

Retinal illnesses such as diabetic retinopathy (DR) are the main causes of vision loss. In the early recognition of eye diseases, the segmentation of blood vessels in retina images plays an important role. Different symptoms of ocular diseases can be identified by the geometric features of ocular arteries. However, due to the complex construction of the blood vessels and their different thicknesses, segmenting the retina image is a challenging task. There are a number of algorithms that helped the detection of retinal diseases. This paper presents an overview of papers from 2016 to 2022 that discuss machine learning and deep learning methods for automatic vessel segmentation. The methods are divided into two groups: Deep learning-based, and classic methods. Algorithms, classifiers, pre-processing and specific techniques of each group is described, comprehensively. The performances of recent works are compared based on their achieved accuracy in different datasets in inclusive tables. A survey of most popular datasets like DRIVE, STARE, HRF and CHASE_DB1 is also given in this paper. Finally, a list of findings from this review is presented in the conclusion section.

Automated Image Threshold Method Comparison for Conjunctival Vessel Quantification on Optical Coherence Tomography Angiography

Purpose

To determine the impact of image binarization and the best thresholding method for conjunctival optical coherence tomography angiography (OCTA).

Methods

Vessel density (VD) of 14 OCTA conjunctival images (nine nasal and five temporal conjunctivas, and eight right and six left eyes) from normal subjects was analyzed. The binarization of gold-standard images, created by removing pixels that do not represent vessels on ImageJ software, was assessed by three masked graders to determine consistency of VD for images. Various thresholding methods on ImageJ, including manual, 1-, 2- and 3-step processes, were performed on unprocessed images for comparison. Interclass correlation coefficient (ICC) ≥0.750 were classified as good reliability and selected for calculation of the performance of the pixel location in the binarized images of each method.

Results

Analysis of the gold-standard threshold method achieved an ICC of 0.816 with excellent agreement (R² = 0.965, P < 0.001). From a total 28 different methods and variations performed, only nine methods performed with good reliability, including two 1-step thresholds, six 2-step thresholds, and one 3-step threshold method. Overall, 2-step threshold methods were more reliable than 3-step threshold methods. The 2-step method of Bandpass filter + Phansalkar local threshold (LT) showed the best performance with mean pixel accuracy of 86.9% ± 6.8%, area under the curve of 0.826, sensitivity of 79.0%, and specificity 86.1%.

Conclusions

Bandpass filter + Phansalkar LT was the best method for VD measurement in conjunctival OCTA. Most commonly reported threshold methods showed unsatisfactory agreement. There is a need in the OCTA field for a standardized method to allow comparison between different studies.

Translational Relevance

The proposed threshold method using a widely accessible and commonly used software provides an accurate VD measurement for future OCTA studies.

Multifilters-Based Unsupervised Method for Retinal Blood Vessel Segmentation

Fundus imaging is one of the crucial methods that help ophthalmologists for diagnosing the various eye diseases in modern medicine. An accurate vessel segmentation method can be a convenient tool to foresee and analyze fatal diseases, including hypertension or diabetes, which damage the retinal vessel’s appearance. This work suggests an unsupervised approach for vessels segmentation out of retinal images. The proposed method includes multiple steps. Firstly, from the colored retinal image, green channel is extracted and preprocessed utilizing Contrast Limited Histogram Equalization as well as Fuzzy Histogram Based Equalization for contrast enhancement. To expel geometrical articles (macula, optic disk) and noise, top-hat morphological operations are used. On the resulted enhanced image, matched filter and Gabor wavelet filter are applied, and the outputs from both is added to extract vessels pixels. The resulting image with the now noticeable blood vessel is binarized using human visual system (HVS). A final image of segmented blood vessel is obtained by applying post-processing. The suggested method is assessed on two public datasets (DRIVE and STARE) and showed comparable results with regard to sensitivity, specificity and accuracy. The results we achieved with respect to sensitivity, specificity together with accuracy on DRIVE database are 0.7271, 0.9798 and 0.9573, and on STARE database these are 0.7164, 0.9760, and 0.9560, respectively, in less than 3.17 s on average per image.

R2U++: a multiscale recurrent residual U-Net with dense skip connections for medical image segmentation

U-Net is a widely adopted neural network in the domain of medical image segmentation. Despite its quick embracement by the medical imaging community, its performance suffers on complicated datasets. The problem can be ascribed to its simple feature extracting blocks: encoder/decoder, and the semantic gap between encoder and decoder. Variants of U-Net (such as R2U-Net) have been proposed to address the problem of simple feature extracting blocks by making the network deeper, but it does not deal with the semantic gap problem. On the other hand, another variant UNET++ deals with the semantic gap problem by introducing dense skip connections but has simple feature extraction blocks. To overcome these issues, we propose a new U-Net based medical image segmentation architecture R2U++. In the proposed architecture, the adapted changes from vanilla U-Net are: (1) the plain convolutional backbone is replaced by a deeper recurrent residual convolution block. The increased field of view with these blocks aids in extracting crucial features for segmentation which is proven by improvement in the overall performance of the network. (2) The semantic gap between encoder and decoder is reduced by dense skip pathways. These pathways accumulate features coming from multiple scales and apply concatenation accordingly. The modified architecture has embedded multi-depth models, and an ensemble of outputs taken from varying depths improves the performance on foreground objects appearing at various scales in the images. The performance of R2U++ is evaluated on four distinct medical imaging modalities: electron microscopy, X-rays, fundus, and computed tomography. The average gain achieved in IoU score is 1.5 ± 0.37% and in dice score is 0.9 ± 0.33% over UNET++, whereas, 4.21 ± 2.72 in IoU and 3.47 ± 1.89 in dice score over R2U-Net across different medical imaging segmentation datasets.

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.

Hybrid deep learning network for vascular segmentation in photoacoustic imaging

Photoacoustic (PA) technology has been used extensively on vessel imaging due to its capability of identifying molecular specificities and achieving high optical-diffraction-limited lateral resolution down to the cellular level. Vessel images carry essential medical information that provides guidelines for a professional diagnosis. Modern image processing techniques provide a decent contribution to vessel segmentation. However, these methods suffer from under or over-segmentation. Thus, we demonstrate both the results of adopting a fully convolutional network and U-net, and propose a hybrid network consisting of both applied on PA vessel images. Comparison results indicate that the hybrid network can significantly increase the segmentation accuracy and robustness.

Supervised Retinal Vessel Segmentation Based Average Filter and Iterative Self Organizing Data Analysis Technique

Retinal fundus is the inner surface of the eye associated with the lens. The identification of disease needs some parts of retinal fundus, such as blood vessel. Blood vessels are part of circulation system which functions to supply blood to retina area. This research proposed a method for segmentation of blood vessel in retinal image with Average Filter and Iterative Self-Organizing Data Analysis (ISODATA) Technique. The first step with the input image changed to Gamma Correction, increasing contrast with Contrast Limited Adaptive Histogram Equalization (CLAHE), the filtering process with Average Filter. The segmentation is used for ISODATA. Region of Interest was applied to take the center of a vessel object and remove the background. In the final stage, the process of noise reduction and removal of small pixel values with Median Filter and Closing Morphology. Datasets used in this research were DRIVE and STARE. The average result was obtained for STARE dataset with an accuracy of 94.41%, Sensitivity of 55.57%, Specification of 98.31%, F1 Score of 64.81% while for the DRIVE dataset with accuracy of 94.78%, Sensitivity of 43.46%, Specification of 99.81%, and F1 Score of 59.39%.

Retinal Vessel Segmentation Using an Entropy-Based Optimization Algorithm

This article presents an algorithm for the segmentation of retinal blood vessels for the detection of diabetic retinopathy eye diseases. This disease occurs in patients with untreated diabetes for a long time. Since this disease is related to the retina, it can eventually lead to vision impairment. The proposed algorithm is a supervised learning method of blood vessels segmentation in which the classification system is trained with the features that are extracted from the images. The proposed system is implemented on the images of DRIVE, STARE and CHASE_DB1 databases. The segmentation is done by forming clusters with the features of patterns. The features were extracted using independent component analysis and the classification is performed by support vector machines (SVM). The results of the parameters are grouped by accuracy, sensitivity, specificity, positive predictive value, false positive rate and are compared with particle swarm optimization (PSO), the firefly optimization algorithm (FA) and the lion optimization algorithm (LOA).

An accurate interactive segmentation and volume calculation of orbital soft tissue for orbital reconstruction after enucleation

Background

Accurate measurement and reconstruction of orbital soft tissue is important to diagnosis and treatment of orbital diseases. This study applied an interactive graph cut method to orbital soft tissue precise segmentation and calculation in computerized tomography (CT) images, and to estimate its application in orbital reconstruction.

Methods

The interactive graph cut method was introduced to segment extraocular muscle and intraorbital fat in CT images. Intra- and inter-observer variability of tissue volume measured by graph cut segmentation was validated. Accuracy and reliability of the method was accessed by comparing with manual delineation and commercial medical image software. Intraorbital structure of 10 patients after enucleation surgery was reconstructed based on graph cut segmentation and soft tissue volume were compared within two different surgical techniques.

Results

Both muscle and fat tissue segmentation results of graph cut method showed good consistency with ground truth in phantom data. There were no significant differences in muscle calculations between observers or segmental methods (p > 0.05). Graph cut results of fat tissue had coincidental variable trend with ground truth which could identify 0.1cm³ variation. The mean performance time of graph cut segmentation was significantly shorter than manual delineation and commercial software (p < 0.001). Jaccard similarity and Dice coefficient of graph cut method were 0.767 ± 0.045 and 0.836 ± 0.032 for human normal extraocular muscle segmentation. The measurements of fat tissue were significantly better in graph cut than those in commercial software (p < 0.05). Orbital soft tissue volume was decreased in post-enucleation orbit than that in normal orbit (p < 0.05).

Conclusion

The graph cut method was validated to have good accuracy, reliability and efficiency in orbit soft tissue segmentation. It could discern minor volume changes of soft tissue. The interactive segmenting technique would be a valuable tool for dynamic analysis and prediction of therapeutic effect and orbital reconstruction.

Blood Vessel Extraction in Color Retinal Fundus Images with Enhancement Filtering and Unsupervised Classification

Retinal blood vessels have a significant role in the diagnosis and treatment of various retinal diseases such as diabetic retinopathy, glaucoma, arteriosclerosis, and hypertension. For this reason, retinal vasculature extraction is important in order to help specialists for the diagnosis and treatment of systematic diseases. In this paper, a novel approach is developed to extract retinal blood vessel network. Our method comprises four stages: (1) preprocessing stage in order to prepare dataset for segmentation; (2) an enhancement procedure including Gabor, Frangi, and Gauss filters obtained separately before a top-hat transform; (3) a hard and soft clustering stage which includes K-means and Fuzzy C-means (FCM) in order to get binary vessel map; and (4) a postprocessing step which removes falsely segmented isolated regions. The method is tested on color retinal images obtained from STARE and DRIVE databases which are available online. As a result, Gabor filter followed by K-means clustering method achieves 95.94% and 95.71% of accuracy for STARE and DRIVE databases, respectively, which are acceptable for diagnosis systems.