Vessel Segmentation

Retinal blood vessel assessment plays an important role in the diagnosis of ophthalmic pathologies. The use of digital images for this purpose enables the application of a computerized approach and has fostered the development of multiple methods for automated vascular tree segmentation. Metrics based on contingency tables for binary classification have been widely used for evaluating the performance of these algorithms. Metrics from this family are based on the measurement of a success or failure rate in the detected pixels, obtained by means of pixel-to-pixel comparison between the automated segmentation and a manually-labeled reference image. Therefore, vessel pixels are not considered as a part of a vascular structure with specific features. This paper contributes a function for the evaluation of global quality in retinal vessel segmentations. This function is based on the characterization of vascular structures as connected segments with measurable area and length. Thus, its design is meant to be sensitive to anatomical vascularity features. Comparison of results between the proposed function and other general quality evaluation functions shows that this proposal renders a high matching degree with human quality perception. Therefore, it can be used to enhance quality evaluation in retinal vessel segmentations, supplementing the existing functions. On the other hand, from a general point of view, the applied concept of measuring descriptive properties may be used to design specialized functions aimed at segmentation quality evaluation in other complex structures.

This paper presents an innovative methodology to detect the vessel tree in retinal angiographies. The automatic analysis of retinal vessel tree facilitates the computation of the arteriovenous index, which is essential for the diagnosis of a wide range of eye diseases. We have developed a system inspired in the classical snake but incorporating domain specific knowledge, such as blood vessels topological properties. It profites mainly from the automatic localization of the optic disc and from the extraction and enhancement of the vascular tree centerlines. Encouraging results in the detection of arteriovenous structures are efficiently achieved, as shown by the systems performance evaluation on the publicy available DRIVE database.

The blood flow dynamics in microcirculation depends strongly on the microvascular networks composed with short irregular vessel segments which are linked by numerous bifurcations. This paper presents the application of a confocal micro-PTV system to track RBCs through a rectangular polydimethysiloxane (PDMS) microchannel with a bifurcation. By using a confocal micro-PTV system, we have measured the effect of bifurcation on the flow behaviour of both fluorescent particles diluted in pure water and RBCs in concentrated suspensions. After performing simulations with the commercial finite element software package POLYFLOW®;, some experimental results were compared with the numerical results and the limitations of these simulations were outlined.

Hepatic-vessel trees are the key structures in the liver. Knowledge of the hepatic-vessel tree is required because it provides information for liver lesion detection in the computer-aided diagnosis (CAD) system. However, hepatic vessels cannot easily be distinguished from other liver tissues in plain CT images. Automated segmentation of hepatic vessels in plain (non-contrast) CT images is a challenging issue. In this paper, an approach to automatic segmentation of hepatic vessels is proposed. The approach consists of two processing steps: enhancement of hepatic vessels and hepatic-vessel extractions. Enhancement of the vessels was performed with two techniques: (1) histogram transformation based on a Gaussian function; (2) multi-scale line filtering based on eigenvalues of a Hessian matrix. After the enhancement of the vessels, candidates of hepatic vessels were extracted by a thresholding method. Small connected regions in the final results were considered as false positives and we...

Background—Coronary remodeling plays a significant role in lumen loss in transplant allograft vasculopathy (TxCAD), but the determinants of remodeling are unknown. We assessed the relationship between remodeling and plaque topography, coronary compliance, and blood flow in TxCAD. Methods and Results—One artery in each of 27 transplant patients was investigated with simultaneous intravascular ultrasound and coronary flow measurements (basal and hyperemic

Carotid artery stenosis is usually the bottleneck caused by atherosclerosis or carotid artery luminal narrowing. Carotid arteries are in close proximity to bone and bony structures as spongy. Contrast-enhanced computerized tomography angiography (CTA) is used to monitor and measure carotid arteries under the control of an expert. Recently, there is a strong and growing demand for improving the computer aided carotid segmentation process. Recently, there has been a strong and growing demand for computer aided carotid segmentation. In this study, segmentation of the vessels in the CTA images is performed by using region-based active contours method and classification of the segmented regions. The boundaries of the vessel-bone regions are found by region-based active contour segmentation. After segmentation of regions with high gray-scale level values, such as veins and bones, these regions should be separated from each other. In order to perform only carotid artery segmentation in CTA images, it is necessary to eliminate bone fragments and noisy vessel-like structures. For this purpose, in this study, a decision-making mechanism at the point of vein-bone separation is established to classify the segmented regions with a supervised learning system. The method is applied on different patients' CTA images and the performance evaluation is done with statistical and area-based metrics. In these experimental results; average of over 89% Dice similarity 99% accuracy are obtained.

Atherosclerosis disease is one of the most important causes of death in the world. Carotid artery stenosis causes narrowing of blood vessels and this forward results with stroke. The carotid arteries enter from the skull cavity and show close proximity to the bone and osteoid structures. Bone tissue and contrast enhanced carotid arteries generally cannot distinguish when vessel evaluation is performed. In this study, the segmentation of carotid arteries and extraction of bone regions are done with seeded region-growing and random walk segmentation methods. And, methods are compared. These methods are applied on different patients' CTA images and the performance evaluations are done with statistical, area and distance based metrics. Region growing and random walk methods in vessel segmentation give approximately similar results. In general, random walk is more successful according to average results in vessel segmentation. It is observed that region growing gives more successful results in bone segmentation and execution time is shorter than random walk method.

The internal carotid artery (ICA) segmentation is a complicated task at skull base in computed tomography angiography (CTA) images. The ICA enters into from skull cavity and shows close proximity to bone and surrounding soft tissues. For this reason, there exists a robust intensity overlap between vessels, bone and other surrounding tissues. Thus, these similar objects are not separated properly in images only according to the intensity level. In this paper, a texture-based 3D region growing approach is proposed and applied to the ICA through the skull base. The main contribution of this study is that the method does not ask for an extra computed tomography scan for bone masking. Moreover, the method dynamically sets the segmentation parameters according to texture knowledge. The proposed method was evaluated by the experiments on 15 actual clinical data. The performance evaluations were performed by comparing the automatic outputs with manual segmentations which are done by two radiologist observers. As a result, dice similarity rate of 89% was achieved together with 99% accuracy and 0.32 mm mean surface distance (Msd) for ICA segmentation through the skull base. The results show that the average overlap for the observers are similar. The proposed texture-based approach decreases significantly explosions, over-segmentations and increases rate of area overlap, sensitivity, precision at skull base. Therefore, the method is clinically useful and has potential to segment carotid arteries at skull base efficiently.

2D vessel segmentation algorithms working on 2D digital subtraction angiography (DSA) images suffer from inhomogeneous contrast agent distributions within the vessels. In this work, we present a novel semi-automatic vessel segmentation method based on local adaptive contrast enhancement. Either a forward projected 3D centerline or a set of manual selected seed points define the vessel branches to be segmented on the image. The algorithm uses bilateral filtering followed by local contrast enhancement to eliminate intensity inhomogeneity within the vessel region that is caused by unequally distributed contrast agent. Our segmentation algorithm is extensively evaluated on 45 different DSA images and exhibits an average Hausdorff distance of 22 pixels and sensitivity of 89 %.

Volumetric analysis of coronary arteries can be performed using intravascular ultrasound (IVUS) images selected at 1 mm intervals without ECG gating. However, there are few data regarding the influence of coronary pulsation on this volumetric analysis. We developed two models of consecutive area measurements consisting of duplicated area measurements from short coronary segments and virtual measurements based on a sine function. These models allowed the re-calculation of volumes using different sets of frames from the same simulated segments. The variability of the volume determinations was evaluated by its percent standard deviation [%SD = (SD/the mean value) × 100]. The relation of the variability to the extent of external elastic membrane (EEM) area change during the cardiac cycle (amplitude) and heart rates (frequency) were examined. In 58 short coronary segments of 15 patients, consecutive IVUS images were measured [%EEM area change: 12.3 ± 7.7 %, heart rate 78 ± 21 beats/min (...

Abstract Carotid artery stenosis is generally a constriction caused by atherosclerosis or carotid artery lumen bottleneck. Carotid arteries are located closely to bones and osteoid structures. Osteoid structures and carotid arteries are frequently confused with each other when performing vessel evaluations. This study provides a novel method for carotid artery lumen segmentation on CTA images by using automatic vessel segmentation with inverse approach, in which vessel segmentation is performed after bone region is segmented and eliminated. The region growing and random walk segmentation methods are utilized in the elimination of bone region and the vessel segmentation. The seed points in the mentioned methods are not manually determined by any starting point. In automatic segmentation, seeds are selected from the experimentally determined intervals according to the local histogram. The stages of preprocessing and post-processing are utilized for better segmentation. The tracking of vessel centers based on continuity is employed for 3D reconstruction and 3D imaging of the vessels. Experiments were conducted with different data sets including various CTA images by using the mentioned methods. As a result, dice similarity rate above 92% was achieved together with 0.16 mm Msd and 99% accuracy. It was concluded based on these results that the proposed method provides successful results in different points of common, internal, external, vertebral arteries, carotid bifurcation and locales close to osteoid structures which are deemed challenging regions for carotid artery lumen segmentation.

Current expert-recommended views for coronary angiography are based on heuristic experience and have not been scientifically studied. We sought to identify optimal viewing regions for first and second order vessel segments of the coronary arteries that provide optimal diagnostic value in terms of minimizing vessel foreshortening and overlap. Using orthogonal 2D images of the coronary tree, 3D models were created from which patient-specific optimal view maps (OVM) allowing quantitative assessment of vessel foreshortening and overlap were generated. Using a novel methodology that averages 3D-based optimal projection geometries, a universal OVM was created for each individual coronary vessel segment that minimized both vessel foreshortening and overlap. A universal OVM model for each coronary segment was generated based on data from 137 patients undergoing coronary angiography. We identified viewing regions for each vessel segment achieving a mean vessel foreshortening value of 5.8 ± 3.9% for the left coronary artery (LCA) and 5.6 ± 3.6% for the right coronary artery (RCA). The overall mean overlap values achieved were 8.7 ± 7.9% for the LCA and 4.6 ± 3.2% for the RCA. This scientifically-based OVM evaluation of coronary vessel segments provides the means to facilitate acquisitions during coronary angiography and interventions that minimize imaging inaccuracies related to foreshortening and overlap, improving the accuracy, efficiency, and safety of diagnostic and interventional coronary procedures.

This paper presents a comparative study on five feature selection heuristics applied to a retinal image database called DRIVE. Features are chosen from a feature vector (encoding local information, but as well information from structures and shapes available in the image) constructed for each pixel in the field of view (FOV) of the image. After selecting the most discriminatory features, an AdaBoost classifier is applied for training. The results of classifications are used to compare the effectiveness of the five feature selection methods.

The internal carotid artery (ICA) segmentation is a complicated task at skull base in computed tomography angiography (CTA) images. The ICA enters into from skull cavity and shows close proximity to bone and surrounding soft tissues. For this reason, there exists a robust intensity overlap between vessels, bone and other surrounding tissues. Thus, these similar objects are not separated properly in images only according to the intensity level. In this paper, a texture-based 3D region growing approach is proposed and applied to the ICA through the skull base. The main contribution of this study is that the method does not ask for an extra computed tomography scan for bone masking. Moreover, the method dynamically sets the segmentation parameters according to texture knowledge. The proposed method was evaluated by the experiments on 15 actual clinical data. The performance evaluations were performed by comparing the automatic outputs with manual segmentations which are done by two radiologist observers. As a result, dice similarity rate of 89% was achieved together with 99% accuracy and 0.32 mm mean surface distance (Msd) for ICA segmentation through the skull base. The results show that the average overlap for the observers are similar. The proposed texture-based approach decreases significantly explosions, over-segmentations and increases rate of area overlap, sensitivity, precision at skull base. Therefore, the method is clinically useful and has potential to segment carotid arteries at skull base efficiently.

Angiography (i.e. vessel imaging after the injection of a radiopaque substance) is a widely used procedure for vessel observation in both clinical routine and medical research. Often for the subsequent analysis of the vascu- lature it is needed to measure the angiogram area covered by vessels and/or the vessel length. For this purpose we need vessel enhancement and segmentation. While