An improved algorithm for vessel centerline tracking in coronary angiograms

Prior skeleton based online deep reinforcement learning for coronary artery centerline extraction

Coronary centerline extraction is an essential technique for X-ray coronary angiography (XCA) image analysis, which provides qualitative and quantitative guidance for percutaneous coronary intervention (PCI). In this paper, an online deep reinforcement learning method for coronary centerline extraction is proposed based on the prior vascular skeleton. Firstly, with XCA image preprocessing (foreground extraction and vessel segmentation) results, the improved ZhangSuen image thinning algorithm is used to rapidly extract the preliminary vascular skeleton network. On this basis, according to the spatial-temporal and morphological continuity of the angiography image sequence, the connectivity of different branches is determined using k-means clustering, and the vessel segments are then grouped, screened, and reconnected to obtain the aorta and its major branches. Finally, using the previous results as prior information, an online Deep Q-Network (DQN) reinforcement learning method is proposed to optimize each branch simultaneously. It comprehensively considers grayscale intensity and eigenvector continuity to achieve the combination of data-driven and model-driven without pre-training. Experimental results on clinical images and the third-party dataset demonstrate that the proposed method can accurately extract, restructure, and optimize the centerline of XCA images with a higher overall accuracy than the existing state-of-the-art methods.

X-ray coronary centerline extraction based on C-UNet and a multifactor reconnection algorithm

BACKGROUND AND OBJECTIVE

Accurate extraction of the coronary artery centerline is crucial in the processes of coronary artery reconstruction, coronary artery stenosis or lesion detection, and surgical navigation. Furthermore, in clinical medicine, the complex background of angiography, low signal-to-noise ratio, and complex vascular structure make coronary artery centerline extraction challenging. In this study, a direct centerline extraction method is proposed that automatically and accurately extracts vascular centerlines from X-ray coronary angiography images based on deep learning and conventional methods.

METHODS

In this study, a coronary artery centerline extraction method is proposed that comprises two parts: the preliminary centerline extraction network based on U-Net with a residual network, called C-UNet, and the multifactor centerline reconnection algorithm based on the geometric characteristics of blood vessels.

RESULTS

The qualitative and quantitative results demonstrate the effectiveness of the presented method. In this study, three widely used evaluation indices were adopted to evaluate the performance of the method: precision, recall, and F1_Score. The experimental results show that this method can accurately extract coronary artery centerlines.

CONCLUSIONS

The proposed centerline extraction method accurately extracts centerlines from X-ray coronary angiography images and improves both the accuracy and continuity of centerline extraction.

Automatic construction of coronary artery tree structure based on vessel blood flow tracking

We sought to propose an innovative vessel blood flow tracking (VBFT) method to extract coronary artery tree (CAT) and to assess the effectiveness of this VBFT versus the single-frame method. Construction of a CAT from a segmented artery is the basis of artificial intelligence-aided angiographic diagnosis. However, construction of a CAT using a single frame remains challenging, due to bifurcations and overlaps in two-dimensional angiograms. Overall, 13,222 angiograms, including 28,539 vessels, were retrospectively collected from 3275 patients and were then annotated. Coronary arteries were automatically segmented by a previously established deep neural networks (DNNs), and the skeleton lines were then extracted from segmentation images to construct CAT using the single-frame method and the VBFT method. Additionally, 1322 angiograms with 2201 vessels were used to test these two methods. Compared to the single-frame method, the VBFT method can significantly improve the accuracy of CAT as (84.3% vs. 72.3%; p < 0.001). Overlap (OV) was higher in the VBFT group than that in the Single-Frame group (91.1% vs. 87.5%; p < 0.001). The VBFT method significantly reduced the incidence of the lack of branching (7.30% vs. 13.9%, p < 0.001), insufficient length (6.70% vs. 11.0%, p < 0.001), and redundant branches (1.60% vs. 3.10%, p < 0.001). The VBFT method improved the extraction of a CAT structure, which will facilitate the development of artificial intelligence-aided angiographic diagnosis. Cardiologists can efficiently diagnose CAD using this method.

Automated coronary artery tree segmentation in coronary CTA using a multiobjective clustering and toroidal model-guided tracking method

BACKGROUND AND OBJECTIVE

Accurate coronary artery tree segmentation can now be developed to assist radiologists in detecting coronary artery disease. In clinical medicine, the noise, low contrast, and uneven intensity of medical images along with complex shapes and vessel bifurcation structures make coronary artery segmentation challenging. In this work, we propose a multiobjective clustering and toroidal model-guided tracking method that can accurately extract coronary arteries from computed tomography angiography (CTA) imagery.

METHODS

Utilizing integrated noise reduction, candidate region detection, geometric feature extraction, and coronary artery tracking techniques, a new segmentation framework for 3D coronary artery trees is presented. The candidate regions are extracted using a multiobjective clustering method, and the coronary arteries are tracked by a toroidal model-guided tracking method.

RESULTS

The qualitative and quantitative results demonstrate the effectiveness of the presented framework, which achieves better performance than the compared segmentation methods in three widely used evaluation indices: the Dice similarity coefficient (DSC), Jaccard index and Recall across the CTA data. The proposed method can accurately identify the coronary artery tree with a mean DSC of 84%, a Jaccard index of 74%, and a Recall of 93%.

CONCLUSIONS

The proposed segmentation framework effectively segments the coronary tree from the CTA volume, which improves the accuracy of 3D vascular tree segmentation.

Diabetic retinopathy detection through novel tetragonal local octa patterns and extreme learning machines

Diabetic retinopathy (DR) is an eye disease that victimize the people suffering from diabetes from many years. The severe form of DR results in form of the blindness that can initially be controlled by the DR-screening oriented treatment. The effective screening programs require the trained human resource that manually grade the fundus images to understand the severity of the disease. But due to the complexity of this process, and the insufficient number of the trained workers, the precise manual grading is an expensive process. The CAD-based solutions try to address these limitations but most of the existing DR detection systems are as evaluated over small sets and become ineffective when applied in real scenarios. Therefore, in this paper we proposed a novel technique to precisely detect the various stages of the DR by extending the research of the content-based image retrieval domain. To achieve the human-level performance over the large-scale DR-datasets (i.e. Kaggle-DR), the fundus images are represented by the novel tetragonal local octa pattern (T-LOP) features, that are then classified through the extreme learning machine (ELM). To justify the significance of the method, the proposed scheme is compared against several state-of-the-art methods including the deep learning-based methods over four DR-datasets of variational lengths (i.e. Kaggle-DR, DRIVE, Review-DB, STARE). The experimental results confirm the significance of the DR-detection scheme to serve as a stand-alone solution for providing the precise information of the severity of the DR in an efficient manner.

Vascular tree tracking and bifurcation points detection in retinal images using a hierarchical probabilistic model

BACKGROUND AND OBJECTIVE

Retinal vascular tree extraction plays an important role in computer-aided diagnosis and surgical operations. Junction point detection and classification provide useful information about the structure of the vascular network, facilitating objective analysis of retinal diseases.

METHODS

In this study, we present a new machine learning algorithm for joint classification and tracking of retinal blood vessels. Our method is based on a hierarchical probabilistic framework, where the local intensity cross sections are classified as either junction or vessel points. Gaussian basis functions are used for intensity interpolation, and the corresponding linear coefficients are assumed to be samples from class-specific Gamma distributions. Hence, a directed Probabilistic Graphical Model (PGM) is proposed and the hyperparameters are estimated using a Maximum Likelihood (ML) solution based on Laplace approximation.

RESULTS

The performance of proposed method is evaluated using precision and recall rates on the REVIEW database. Our experiments show the proposed approach reaches promising results in bifurcation point detection and classification, achieving 88.67% precision and 88.67% recall rates.

CONCLUSIONS

This technique results in a classifier with high precision and recall when comparing it with Xu's method.

Tracking and diameter estimation of retinal vessels using Gaussian process and Radon transform

Extraction of blood vessels in retinal images is an important step for computer-aided diagnosis of ophthalmic pathologies. We propose an approach for blood vessel tracking and diameter estimation. We hypothesize that the curvature and the diameter of blood vessels are Gaussian processes (GPs). Local Radon transform, which is robust against noise, is subsequently used to compute the features and train the GPs. By learning the kernelized covariance matrix from training data, vessel direction and its diameter are estimated. In order to detect bifurcations, multiple GPs are used and the difference between their corresponding predicted directions is quantified. The combination of Radon features and GP results in a good performance in the presence of noise. The proposed method successfully deals with typically difficult cases such as bifurcations and central arterial reflex, and also tracks thin vessels with high accuracy. Experiments are conducted on the publicly available DRIVE, STARE, CHASEDB1, and high-resolution fundus databases evaluating sensitivity, specificity, and Matthew's correlation coefficient (MCC). Experimental results on these datasets show that the proposed method reaches an average sensitivity of 75.67%, specificity of 97.46%, and MCC of 72.18% which is comparable to the state-of-the-art.

Topology adaptive vessel network skeleton extraction with novel medialness measuring function

Vessel tree skeleton extraction is widely applied in vascular structure segmentation, however, conventional approaches often suffer from the adjacent interferences and poor topological adaptability. To avoid these problems, a robust, topology adaptive tree-like structure skeleton extraction framework is proposed in this paper. Specifically, to avoid the adjacent interferences, a local message passing procedure called Gaussian affinity voting (GAV) is proposed to realize adaptive scale-growing of vessel voxels. Then the medialness measuring function (MMF) based on GAV, namely GAV-MMF, is constructed to extract medialness patterns robustly. In order to improve topological adaptability, a level-set graph embedded with GAV-MMF is employed to build initial curve skeletons without any user interaction. Furthermore, the GAV-MMF is embedded in stretching open active contours (SOAC) to drive the initial curves to the expected location, maintaining smoothness and continuity. In addition, to provide an accurate and smooth final skeleton tree topology, topological checks and skeleton network reconfiguration is proposed. The continuity and scalability of this method is validated experimentally on synthetic and clinical images for multi-scale vessels. Experimental results show that the proposed method achieves acceptable topological adaptability for skeleton extraction of vessel trees.

3D multimodal cardiac data reconstruction using angiography and computerized tomographic angiography registration

BACKGROUND

Computerized tomographic angiography (3D data representing the coronary arteries) and X-ray angiography (2D X-ray image sequences providing information about coronary arteries and their stenosis) are standard and popular assessment tools utilized for medical diagnosis of coronary artery diseases. At present, the results of both modalities are individually analyzed by specialists and it is difficult for them to mentally connect the details of these two techniques. The aim of this work is to assist medical diagnosis by providing specialists with the relationship between computerized tomographic angiography and X-ray angiography.

METHODS

In this study, coronary arteries from two modalities are registered in order to create a 3D reconstruction of the stenosis position. The proposed method starts with coronary artery segmentation and labeling for both modalities. Then, stenosis and relevant labeled artery in X-ray angiography image are marked by a specialist. Proper control points for the marked artery in both modalities are automatically detected and normalized. Then, a geometrical transformation function is computed using these control points. Finally, this function is utilized to register the marked artery from the X-ray angiography image on the computerized tomographic angiography and get the 3D position of the stenosis lesion.

RESULTS

The result is a 3D informative model consisting of stenosis and coronary arteries' information from the X-ray angiography and computerized tomographic angiography modalities. The results of the proposed method for coronary artery segmentation, labeling and 3D reconstruction are evaluated and validated on the dataset containing both modalities.

CONCLUSIONS

The advantage of this method is to aid specialists to determine a visual relationship between the correspondent coronary arteries from two modalities and also set up a connection between stenosis points from an X-ray angiography along with their 3D positions on the coronary arteries from computerized tomographic angiography. Moreover, another benefit of this work is that the medical acquisition standards remain unchanged, which means that no calibration in the acquisition devices is required. It can be applied on most computerized tomographic angiography and angiography devices.

Automatic vasculature identification in coronary angiograms by adaptive geometrical tracking

As the uneven distribution of contrast agents and the perspective projection principle of X-ray, the vasculatures in angiographic image are with low contrast and are generally superposed with other organic tissues; therefore, it is very difficult to identify the vasculature and quantitatively estimate the blood flow directly from angiographic images. In this paper, we propose a fully automatic algorithm named adaptive geometrical vessel tracking (AGVT) for coronary artery identification in X-ray angiograms. Initially, the ridge enhancement (RE) image is obtained utilizing multiscale Hessian information. Then, automatic initialization procedures including seed points detection, and initial directions determination are performed on the RE image. The extracted ridge points can be adjusted to the geometrical centerline points adaptively through diameter estimation. Bifurcations are identified by discriminating connecting relationship of the tracked ridge points. Finally, all the tracked centerlines are merged and smoothed by classifying the connecting components on the vascular structures. Synthetic angiographic images and clinical angiograms are used to evaluate the performance of the proposed algorithm. The proposed algorithm is compared with other two vascular tracking techniques in terms of the efficiency and accuracy, which demonstrate successful applications of the proposed segmentation and extraction scheme in vasculature identification.

A simple framework to generate 3D patient-specific model of coronary artery bifurcation from single-plane angiographic images

Although computer-based simulations, such as structural finite element analysis, have proven their usefulness to support procedural planning of coronary stenting, the link between the clinical practice and these engineering techniques is still limited to research test-cases. A key point to further promote such an interaction is to generate in a fast and effective manner the computational grids from the medical images. Hence, the present study proposes a simple framework to generate 3D meshes of coronary bifurcations from a pair of planar angiographic images obtained by X-ray angiography, which is the gold standard technique for the diagnosis of coronary artery stenosis.

Coronary artery center-line extraction using second order local features

Of interest is the accurate and robust delineation of vessel center-lines for complete arterial tree structure in coronary angiograms which is an imperative step towards 3D reconstruction of coronary tree and feature-based registration of multiple view angiograms. Most existing center-line tracking methods encounter limitations in coping with abrupt variations in local artery direction and sudden changes of lumen diameter that occur in the vicinity of arterial lesions. This paper presents an improved center-line tracing algorithm for automatic extraction of coronary arterial tree based on robust local features. The algorithm employs an improved scanning schema based on eigenvalues of Hessian matrix for reliable identification of true vessel points as well as an adaptive look-ahead distance schema for calculating the magnitude of scanning profile. In addition to a huge variety of clinical examples, a well-established vessel simulation tool was used to create several synthetic angiograms for objective comparison and performance evaluation. The experimental results on the accuracy and robustness of the proposed algorithm and its counterparts under difficult situations such as poor image quality and complicated vessel geometry are presented.

An automated computational framework for retinal vascular network labeling and branching order analysis

Changes in retinal vascular morphology are well known as predictive clinical signs of many diseases such as hypertension, diabetes and so on. Computer-aid image processing and analysis for retinal vessels in fundus images are effective and efficient in clinical diagnosis instead of tedious manual labeling and measurement. An automated computational framework for retinal vascular network labeling and analysis is presented in this work. The framework includes 1) detecting and locating the optic disc; 2) tracking the vessel centerline from detected seed points and linking the breaks after tracing; 3) extracting all the retinal vascular trees and identifying all the significant points; and 4) classifying terminal points into starting points and ending points based on the information of optic disc location, and finally assigning branch order for each extracted vascular tree in the image. All the modules in the framework are fully automated. Based on the results, morphological analysis is then applied to achieve geometrical and topological features based on branching order for one individual vascular tree or for the vascular network through the retinal vascular network in the images. Validation and experiments on the public DRIVE database have demonstrated that the proposed framework is a novel approach to analyze and study the vascular network pattern, and may offer new insights to the diagnosis of retinopathy.

Robust adaptive 3-D segmentation of vessel laminae from fluorescence confocal microscope images and parallel GPU implementation

This paper presents robust 3-D algorithms to segment vasculature that is imaged by labeling laminae, rather than the lumenal volume. The signal is weak, sparse, noisy, nonuniform, low-contrast, and exhibits gaps and spectral artifacts, so adaptive thresholding and Hessian filtering based methods are not effective. The structure deviates from a tubular geometry, so tracing algorithms are not effective. We propose a four step approach. The first step detects candidate voxels using a robust hypothesis test based on a model that assumes Poisson noise and locally planar geometry. The second step performs an adaptive region growth to extract weakly labeled and fine vessels while rejecting spectral artifacts. To enable interactive visualization and estimation of features such as statistical confidence, local curvature, local thickness, and local normal, we perform the third step. In the third step, we construct an accurate mesh representation using marching tetrahedra, volume-preserving smoothing, and adaptive decimation algorithms. To enable topological analysis and efficient validation, we describe a method to estimate vessel centerlines using a ray casting and vote accumulation algorithm which forms the final step of our algorithm. Our algorithm lends itself to parallel processing, and yielded an 8 x speedup on a graphics processor (GPU). On synthetic data, our meshes had average error per face (EPF) values of (0.1-1.6) voxels per mesh face for peak signal-to-noise ratios from (110-28 dB). Separately, the error from decimating the mesh to less than 1% of its original size, the EPF was less than 1 voxel/face. When validated on real datasets, the average recall and precision values were found to be 94.66% and 94.84%, respectively.

A novel method for blood vessel detection from retinal images

Background

The morphological changes of the retinal blood vessels in retinal images are important indicators for diseases like diabetes, hypertension and glaucoma. Thus the accurate segmentation of blood vessel is of diagnostic value.

Methods

In this paper, we present a novel method to segment retinal blood vessels to overcome the variations in contrast of large and thin vessels. This method uses adaptive local thresholding to produce a binary image then extract large connected components as large vessels. The residual fragments in the binary image including some thin vessel segments (or pixels), are classified by Support Vector Machine (SVM). The tracking growth is applied to the thin vessel segments to form the whole vascular network.

Results

The proposed algorithm is tested on DRIVE database, and the average sensitivity is over 77% while the average accuracy reaches 93.2%.

Conclusions

In this paper, we distinguish large vessels by adaptive local thresholding for their good contrast. Then identify some thin vessel segments with bad contrast by SVM, which can be lengthened by tracking. This proposed method can avoid heavy computation and manual intervention.