Evaluation framework for carotid bifurcation lumen segmentation and stenosis grading

Deep learning-based fully automatic screening of carotid artery plaques in computed tomography angiography: a multicenter study

AIM

To develop and validate a deep learning (DL) algorithm for the automated detection and classification of carotid artery plaques (CAPs) on computed tomography angiography (CTA) images.

MATERIALS AND METHODS

This retrospective study enrolled 400 patients (300 in the Center Ⅰ and 100 in Ⅱ). Three radiologists co-labeled CAPs, and their revised calcification status (noncalcified, mixed, and calcified) was regarded as ground truth. Center Ⅰ patients were randomly divided into training and internal validation datasets, while Center Ⅱ patients served as the external validation dataset. Carotid artery regions were segmented using a modified 3D-UNet network, followed by CAPs detection and classification using a ResUNet-based architecture in a two-step DL system. The DL model's detection and classification performance were evaluated on the validation dataset using precision-recall curve, free-response receiver operating characteristic (fROC) curve, Cohen's kappa, and ROC curve analysis.

RESULTS

The DL model had achieved 83.4% sensitivity at 3.0 false positives (FPs)/CTA scan in internal validation and 78.9% in external validation. F1-scores were 0.764 and 0.769 at the optimal threshold, and area under fROC curves were 0.756 and 0.738, respectively, indicating good overall accuracy for CAP detection. The DL model also showed good performance for the ternary classification of CAPs, with Cohen's kappa achieved 0.728 and 0.703 in both validation datasets.

CONCLUSION

This study demonstrated the feasibility of using a fully automated DL-based algorithm for the detection and ternary classification of CAPs, which could be helpful for the workloads of radiologists.

Automated anatomical labeling of the intracranial arteries via deep learning in computed tomography angiography

Background and purpose: Anatomical labeling of the cerebral vasculature is a crucial topic in determining the morphological nature and characterizing the vital variations of vessels, yet precise labeling of the intracranial arteries is time-consuming and challenging, given anatomical structural variability and surging imaging data. We present a U-Net-based deep learning (DL) model to automatically label detailed anatomical segments in computed tomography angiography (CTA) for the first time. The trained DL algorithm was further tested on a clinically relevant set for the localization of intracranial aneurysms (IAs). Methods: 457 examinations with varying degrees of arterial stenosis were used to train, validate, and test the model, aiming to automatically label 42 segments of the intracranial arteries [e.g., 7 segments of the internal carotid artery (ICA)]. Evaluation metrics included Dice similarity coefficient (DSC), mean surface distance (MSD), and Hausdorff distance (HD). Additionally, 96 examinations containing at least one IA were enrolled to assess the model's potential in enhancing clinicians' precision in IA localization. A total of 5 clinicians with different experience levels participated as readers in the clinical experiment and identified the precise location of IA without and with algorithm assistance, where there was a washout period of 14 days between two interpretations. The diagnostic accuracy, time, and mean interrater agreement (Fleiss' Kappa) were calculated to assess the differences in clinical performance of clinicians. Results: The proposed model exhibited notable labeling performance on 42 segments that included 7 anatomical segments of ICA, with the mean DSC of 0.88, MSD of 0.82 mm and HD of 6.59 mm. Furthermore, the model demonstrated superior labeling performance in healthy subjects compared to patients with stenosis (DSC: 0.91 vs. 0.89, p < 0.05; HD: 4.75 vs. 6.19, p < 0.05). Concurrently, clinicians with model predictions achieved significant improvements when interpreting the precise location of IA. The clinicians' mean accuracy increased by 0.04 (p = 0.003), mean time to diagnosis reduced by 9.76 s (p < 0.001), and mean interrater agreement (Fleiss' Kappa) increased by 0.07 (p = 0.029). Conclusion: Our model stands proficient for labeling intracranial arteries using the largest CTA dataset. Crucially, it demonstrates clinical utility, helping prioritize the patients with high risks and ease clinical workload.

LIVE-Net: Comprehensive 3D vessel extraction framework in CT angiography

The extraction of vessels from computed tomography angiography (CTA) is significant in diagnosing and evaluating vascular diseases. However, due to the anatomical complexity, wide intensity distribution, and small volume proportion of vessels, vessel extraction is laborious and time-consuming, and it is easy to lead to error-prone diagnostic results in clinical practice. This study proposes a novel comprehensive vessel extraction framework, called the Local Iterative-based Vessel Extraction Network (LIVE-Net), to achieve 3D vessel segmentation while tracking vessel centerlines. LIVE-Net contains dual dataflow pathways that work alternately: an iterative tracking network and a local segmentation network. The former can generate the fine-grain direction and radius prediction of a vascular patch by using the attention-embedded atrous pyramid network (aAPN), and the latter can achieve 3D vascular lumen segmentation by constructing the multi-order self-attention U-shape network (MOSA-UNet). LIVE-Net is trained and evaluated on two datasets: the MICCAI 2008 Coronary Artery Tracking Challenge (CAT08) dataset and head and neck CTA dataset from the clinic. Experimental results of both tracking and segmentation show that our proposed LIVE-Net exhibits superior performance compared with other state-of-the-art (SOTA) networks. In the CAT08 dataset, the tracked centerlines have an average overlap of 95.2%, overlap until first error of 91.2%, overlap with the clinically relevant vessels of 98.3%, and error distance inside of 0.21 mm. The corresponding tracking overlap metrics in the head and neck CTA dataset are 96.7%, 91.0%, and 99.8%, respectively. In addition, the results of the consistent experiment also show strong clinical correspondence. For the segmentation of bilateral carotid and vertebral arteries, our method can not only achieve better accuracy with an average dice similarity coefficient (DSC) of 90.03%, Intersection over Union (IoU) of 81.97%, and 95% Hausdorff distance (95%HD) of 3.42 mm , but higher efficiency with an average time of 67.25 s , even three times faster compared to some methods applied in full field view. Both the tracking and segmentation results prove the potential clinical utility of our network.

Simultaneous vessel segmentation and unenhanced prediction using self-supervised dual-task learning in 3D CTA (SVSUP)

BACKGROUND AND OBJECTIVE

The vessel segmentation in CT angiography (CTA) provides an important basis for automatic diagnosis and hemodynamics analysis. Virtual unenhanced (VU) CT images obtained by dual-energy CT can assist clinical diagnosis and reduce radiation dose by obviating true unenhanced imaging (UECT). However, accurate segmentation of all vessels in the head-neck CTA (HNCTA) remains a challenge, and VU images are currently not available from conventional single-energy CT imaging.

METHODS

In this paper, we proposed a self-supervised dual-task deep learning strategy to fully automatically segment all vessels and predict unenhanced CT images from single-energy HNCTA based on a developed iterative residual-sharing scheme. The underlying idea was to use the correlation between the two tasks to improve task performance while avoiding manual annotation for model training.

RESULTS

The feasibility of the strategy was verified using the data of 24 patients. For vessel segmentation task, the proposed model achieves a significantly higher average Dice coefficient (84.83%, P-values 10^-3 in paired t-test) than the state-of-the-art segmentation model, vanilla VNet (78.94%), and several popular 3D vessel segmentation models, including Hessian-matrix based filter (62.59%), optically-oriented flux (66.33%), spherical flux model (66.91%), and deep vessel net (66.47%). For the unenhanced prediction task, the average ROI-based error compared to the UECT in the artery tissue is 6.1±4.5 HU, similar to previously reported 6.4±5.1 HU for VU reconstruction.

CONCLUSIONS

Results show that the proposed dual-task framework can effectively improve the accuracy of vessel segmentation in HNCTA, and it is feasible to predict the unenhanced image from single-energy CTA, providing a potential new approach for radiation dose saving. Moreover, to our best knowledge, this is the first reported annotation-free deep learning-based full-image vessel segmentation for HNCTA.

Assessment of fully automatic segmentation of pulmonary artery and aorta on noncontrast CT with optimal surface graph cuts

Purpose

Accurate segmentation of the pulmonary arteries and aorta is important due to the association of the diameter and the shape of these vessels with several cardiovascular diseases and with the risk of exacerbations and death in patients with chronic obstructive pulmonary disease. We propose a fully automatic method based on an optimal surface graph‐cut algorithm to quantify the full 3D shape and the diameters of the pulmonary arteries and aorta in noncontrast computed tomography (CT) scans.

Methods

The proposed algorithm first extracts seed points in the right and left pulmonary arteries, the pulmonary trunk, and the ascending and descending aorta by using multi‐atlas registration. Subsequently, the centerlines of the pulmonary arteries and aorta are extracted by a minimum cost path tracking between the extracted seed points, with a cost based on a combination of lumen intensity similarity and multiscale medialness in three planes. The centerlines are refined by applying the path tracking algorithm to curved multiplanar reformatted scans and are then smoothed and dilated nonuniformly according to the extracted local vessel radius from the medialness filter. The resulting coarse estimates of the vessels are used as initialization for a graph‐cut segmentation. Once the vessels are segmented, the diameters of the pulmonary artery (PA) and the ascending aorta (AA) and the PA:AA ratio are automatically calculated both in a single axial slice and in a 10 mm volume around the automatically extracted PA bifurcation level. The method is evaluated on noncontrast CT scans from the Danish Lung Cancer Screening Trial (DLCST). Segmentation accuracy is determined by comparing with manual annotations on 25 CT scans. Intraclass correlation (ICC) between manual and automatic diameters, both measured in axial slices at the PA bifurcation level, is computed on an additional 200 CT scans. Repeatability of the automated 3D volumetric diameter and PA:AA ratio calculations (perpendicular to the vessel axis) are evaluated on 118 scan–rescan pairs with an average in‐between time of 3 months.

Results

We obtained a Dice segmentation overlap of 0.94 ± 0.02 for pulmonary arteries and 0.96 ± 0.01 for the aorta, with a mean surface distance of 0.62 ± 0.33 mm and 0.43 ± 0.07 mm, respectively. ICC between manual and automatic in‐slice diameter measures was 0.92 for PA, 0.97 for AA, and 0.90 for the PA:AA ratio, and for automatic diameters in 3D volumes around the PA bifurcation level between scan and rescan was 0.89, 0.95, and 0.86, respectively.

Conclusion

The proposed automatic segmentation method can reliably extract diameters of the large arteries in non‐ECG‐gated noncontrast CT scans such as are acquired in lung cancer screening.

Coarse-to-fine multiplanar D-SEA UNet for automatic 3D carotid segmentation in CTA images

PURPOSE

Carotid artery atherosclerotic stenosis accounts for 18-25% of ischemic stroke. In the evaluation of carotid atherosclerotic lesions, the automatic, accurate and rapid segmentation of the carotid artery is a priority issue that needs to be addressed urgently. However, the carotid artery area occupies a small target in computed tomography angiography (CTA) images, which affect the segmentation accuracy.

METHODS

We proposed a coarse-to-fine segmentation pipeline with the Multiplanar D-SEA UNet to achieve fully automatic carotid artery segmentation on the entire 3D CTA images, and compared with other four neural networks (3D-UNet, RA-UNet, Isensee-UNet, Multiplanar-UNet) by assessing Dice, Jaccard similarity coefficient, sensitivity, area under the curve and average hausdorff distance.

RESULTS

Our proposed method can achieve a mean Dice score of 91.51% on the 68 neck CTA scans from Beijing Hospital, which remarkably outperforms state-of-the-art 3D image segmentation methods. And the C2F segmentation pipeline can effectively improve segmentation accuracy while avoiding resolution loss.

CONCLUSION

The proposed segmentation method can realize the fully automatic segmentation of the carotid artery and has robust performance with segmentation accuracy, which can be applied into plaque exfoliation and interventional surgery services. In addition, our method is easy to extend to other medical segmentation tasks with appropriate parameter settings.

Learning-based algorithms for vessel tracking: A review

Developing efficient vessel-tracking algorithms is crucial for imaging-based diagnosis and treatment of vascular diseases. Vessel tracking aims to solve recognition problems such as key (seed) point detection, centerline extraction, and vascular segmentation. Extensive image-processing techniques have been developed to overcome the problems of vessel tracking that are mainly attributed to the complex morphologies of vessels and image characteristics of angiography. This paper presents a literature review on vessel-tracking methods, focusing on machine-learning-based methods. First, the conventional machine-learning-based algorithms are reviewed, and then, a general survey of deep-learning-based frameworks is provided. On the basis of the reviewed methods, the evaluation issues are introduced. The paper is concluded with discussions about the remaining exigencies and future research.

Fully automatic deep learning trained on limited data for carotid artery segmentation from large image volumes

BACKGROUND

The objectives of this study were to develop a 3D convolutional deep learning framework (CarotidNet) for fully automatic segmentation of carotid bifurcations in computed tomography angiography (CTA) images and to facilitate the quantification of carotid stenosis and risk assessment of stroke.

METHODS

Our pipeline was a two-stage cascade network that included a localization phase and a segmentation phase. The network framework was based on the 3D version of U-Net, but was refined in three ways: (I) by adding residual connections and a deep supervision strategy to cope with the vanishing problem in back-propagation; (II) by adopting dilated convolution in order to strengthen the capacity to capture contextual information; and (III) by establishing a hybrid objective function to address the extreme imbalance between foreground and background voxels.

RESULTS

We trained our networks on 15 cases and evaluated their performance based on 41 cases from the MICCAI Challenge 2009 dataset. A Dice similarity coefficient of 82.3% was achieved for the test cases.

CONCLUSIONS

We developed a carotid segmentation method based on U-Net that can segment tiny carotid bifurcation lumens from very large backgrounds with no manual intervention. This was the first attempt to use deep learning to achieve carotid bifurcation segmentation in 3D CTA images. Our results indicate that deep learning is a promising method for automatically extracting carotid bifurcation lumens.

Overcoming calcium blooming and improving the quantification accuracy of percent area luminal stenosis by material decomposition of multi-energy computed tomography datasets

Purpose: Conventional stenosis quantification from single-energy computed tomography (SECT) images relies on segmentation of lumen boundaries, which suffers from partial volume averaging and calcium blooming effects. We present and evaluate a method for quantifying percent area stenosis using multienergy CT (MECT) images. Approach: We utilize material decomposition of MECT images to measure stenosis based on the ratio of iodine mass between vessel locations with and without a stenosis, thereby eliminating the requirement for segmentation of iodinated lumen. The method was first assessed using simulated MECT images created with different spatial resolutions. To experimentally assess this method, four phantoms with different stenosis severity (30% to 51%), vessel diameters (5.5 to 14 mm), and calcification densities (700 to 1100    mgHA / cc ) were fabricated. Conventional SECT images were acquired using a commercial CT system and were analyzed with commercial software. MECT images were acquired using a commercial dual-energy CT (DECT) system and also from a research photon-counting detector CT (PCD-CT) system. Three-material-decomposition was performed on MECT data, and iodine density maps were used to quantify stenosis. Clinical radiation doses were used for all data acquisitions. Results: Computer simulation verified that this method reduced partial volume and blooming effects, resulting in consistent stenosis measurements. Phantom experiments showed accurate and reproducible stenosis measurements from MECT images. For DECT and two-threshold PCD-CT images, the estimation errors were 4.0% to 7.0%, 2.0% to 9.0%, 10.0% to 18.0%, and - 1.0 % to - 5.0 % (ground truth: 51%, 51%, 51%, and 30%). For four-threshold PCD-CT images, the errors were 1.0% to 3.0%, 4.0% to 6.0%, - 1.0 % to 9.0%, and 0.0% to 6.0%. Errors using SECT were much larger, ranging from 4.4% to 46%, and were especially worse in the presence of dense calcifications. Conclusions: The proposed approach was shown to be insensitive to acquisition parameters, demonstrating the potential to improve the accuracy and precision of stenosis measurements in clinical practice.

Patient-specific three-dimensional aortic arch modeling for automatic measurements: clinical validation in aortic coarctation

AIM

A validated algorithm for automatic aortic arch measurements in aortic coarctation (CoA) patients could standardize procedures for clinical planning.

METHODS

The model-based assessment of the aortic arch anatomy consisted of three steps: first, machine-learning-based algorithms were trained on 212 three-dimensional magnetic resonance (MR) data to automatically allocate the aortic arch position in patients and segment the aortic arch; second, for each CoA patient (N = 33), the min/max aortic arch diameters were measured using the proposed software, manually and automatically, from noncontrast-enhanced three-dimensional steady-state free precession MRI sequence at five selected sites and compared ('internal comparison' referring to the same environment); third, moreover, the same min/max aortic arch diameters were compared, obtaining them independently, manually from common MR management software (MR Viewforum) and automatically from the model (external comparison). The measured sites were: aortic sinus, sino-tubular junction, mid-ascending aorta, transverse arch and thoracoabdominal aorta at the level of the diaphragm.

RESULTS

Manual and software-assisted measurements showed a good agreement: the difference between diameter measurements was not statistically significant (at α = 0.05), with only one exception, for both internal and external comparison. A high coefficient of correlation was attained for both maximum and minimum diameters in each site (for internal comparison, R > 0.73 for every site, with P < 2 × 10). Notably, in tricuspid aortic valve patients external comparison showed no statistically significant difference at any measurement sites.

CONCLUSION

The automatically derived aortic arch model, starting from three-dimensional MR images, could be a support to take the measurements in CoA patients and to quickly provide a patient-specific model of aortic arch anomalies.

Automated 3D segmentation and diameter measurement of the thoracic aorta on non-contrast enhanced CT

Objectives

To develop and evaluate a fully automatic method to measure diameters of the ascending and descending aorta on non-ECG-gated, non-contrast computed tomography (CT) scans.

Material and methods

The method combines multi-atlas registration to obtain seed points, aorta centerline extraction, and an optimal surface segmentation approach to extract the aorta surface around the centerline. From the extracted 3D aorta segmentation, the diameter of the ascending and descending aorta was calculated at cross-sectional slices perpendicular to the extracted centerline, at the level of the pulmonary artery bifurcation, and at 1-cm intervals up to 3 cm above and below this level. Agreement with manual annotations was evaluated by dice similarity coefficient (DSC) for segmentation overlap, mean surface distance (MSD), and intra-class correlation (ICC) of diameters on 100 CT scans from a lung cancer screening trial. Repeatability of the diameter measurements was evaluated on 617 baseline-one year follow-up CT scan pairs.

Results

The agreement between manual and automatic segmentations was good with 0.95 ± 0.01 DSC and 0.56 ± 0.08 mm MSD. ICC between the diameters derived from manual and from automatic segmentations was 0.97, with the per-level ICC ranging from 0.87 to 0.94. An ICC of 0.98 for all measurements and per-level ICC ranging from 0.91 to 0.96 were obtained for repeatability.

Conclusion

This fully automatic method can assess diameters in the thoracic aorta reliably even in non-ECG-gated, non-contrast CT scans. This could be a promising tool to assess aorta dilatation in screening and in clinical practice.

Key Points

• Fully automatic method to assess thoracic aorta diameters.

• High agreement between fully automatic method and manual segmentations.

• Method is suitable for non-ECG-gated CT and can therefore be used in screening.

Electronic supplementary material

The online version of this article (10.1007/s00330-018-5931-z) contains supplementary material, which is available to authorized users.

Why rankings of biomedical image analysis competitions should be interpreted with care

International challenges have become the standard for validation of biomedical image analysis methods. Given their scientific impact, it is surprising that a critical analysis of common practices related to the organization of challenges has not yet been performed. In this paper, we present a comprehensive analysis of biomedical image analysis challenges conducted up to now. We demonstrate the importance of challenges and show that the lack of quality control has critical consequences. First, reproducibility and interpretation of the results is often hampered as only a fraction of relevant information is typically provided. Second, the rank of an algorithm is generally not robust to a number of variables such as the test data used for validation, the ranking scheme applied and the observers that make the reference annotations. To overcome these problems, we recommend best practice guidelines and define open research questions to be addressed in the future.

Image-based assessment of uncertainty in quantification of carotid lumen

Measurements of the vessel lumen diameter are often used to determine the degree of atherosclerotic disease in carotid arteries. However, quantification results vary with imaging technique and acquisition settings. We aim at providing a tool that quantifies the lumen diameter on different image datasets and gives an estimate of quantification uncertainties, so that they can be taken into consideration when evaluating and comparing measurements. For the segmentation of the vessel lumen, we present an algorithm using ray-casting techniques and partial volume correction. We furthermore propose a scheme for the analysis and exploration of the lumen diameter. Finally, we present a clinically relevant application scenario, in which we explore agreement between lumen diameter estimations in corresponding computed tomography angiography, contrast-enhanced magnetic resonance angiography, time-of-flight, and subtraction images of carotid vessels with severe carotid atherosclerotic plaques.

Cooperative carotid artery centerline extraction in MRI

Centerline extraction of the carotid artery in MRI is important to analyze the artery geometry and to provide input for further processing such as registration and segmentation. The centerline of the artery bifurcation is often extracted by means of two independent minimum cost paths ranging from the common to the internal and the external carotid artery. Often the cost is not well defined at the artery bifurcation, leading to centerline errors. To solve this problem, we developed a method to cooperatively extract both centerlines, where in the cost to extract each centerline, we integrate a constraint region derived from the estimated position of the neighbor centerline. This method avoids that both centerlines follow the same cheapest path after the bifurcation, which is a common error when the paths are extracted independently. We show that this method results in less error compared to extracting them independently: 10 failed centerlines Vs. 3 failures in a data set of 161 arteries with manual annotations. Additionally, we show that the new method improves the non-cooperative approach in 28 cases (p < 0.0001) in a data set of 3,904 arteries.

Robust Segmentation of the Full Cerebral Vasculature in 4D CT of Suspected Stroke Patients

A robust method is presented for the segmentation of the full cerebral vasculature in 4-dimensional (4D) computed tomography (CT). The method consists of candidate vessel selection, feature extraction, random forest classification and postprocessing. Image features include among others the weighted temporal variance image and parameters, including entropy, of an intensity histogram in a local region at different scales. These histogram parameters revealed to be a strong feature in the detection of vessels regardless of shape and size. The method was trained and tested on a large database of 264 patients with suspicion of acute ischemia who underwent 4D CT in our hospital in the period January 2014 to December 2015. Five subvolumes representing different regions of the cerebral vasculature were annotated in each image in the training set by medical assistants. The evaluation was done on 242 patients. A total of 16 (<8%) patients showed severe under or over segmentation and were reported as failures. One out of five subvolumes was randomly annotated in 159 patients and was used for quantitative evaluation. Quantitative evaluation showed a Dice coefficient of 0.91 ± 0.07 and a modified Hausdorff distance of 0.23 ± 0.22 mm. Therefore, robust vessel segmentation in 4D CT is feasible with good accuracy.

Semi-automated carotid lumen segmentation in computed tomography angiography images

Carotid artery stenosis causes narrowing of carotid lumens and may lead to brain infarction. The purpose of this study was to develop a semi-automated method of segmenting vessel walls, surrounding tissues, and more importantly, the carotid artery lumen by contrast computed tomography angiography (CTA) images and to define the severity of stenosis and present a three-dimensional model of the carotid for visual inspection. In vivo contrast CTA images of 14 patients (7 normal subjects and 7 patients undergoing endarterectomy) were analyzed using a multi-step segmentation algorithm. This method uses graph cut followed by watershed and Hessian based shortest path method in order to extract lumen boundary correctly without being corrupted in the presence of surrounding tissues. Quantitative measurements of the proposed method were compared with those of manual delineation by independent board-certified radiologists. The results were quantitatively evaluated using spatial overlap surface distance indices. A slightly strong match was shown in terms of dice similarity coefficient (DSC) = 0.87±0.08; mean surface distance (Dmsd) = 0.32±0.32; root mean squared surface distance (Drmssd) = 0.49±0.54 and maximum surface distance (Dmax) = 2.14±2.08 between manual and automated segmentation of common, internal and external carotid arteries, carotid bifurcation and stenotic artery, respectively. Quantitative measurements showed that the proposed method has high potential to segment the carotid lumen and is robust to the changes of the lumen diameter and the shape of the stenosis area at the bifurcation site. The proposed method for CTA images provides a fast and reliable tool to quantify the severity of carotid artery stenosis.

Aortic length measurements for pulse wave velocity calculation: manual 2D vs automated 3D centreline extraction

BACKGROUND

Pulse wave velocity (PWV) is a biomarker for the intrinsic stiffness of the aortic wall, and has been shown to be predictive for cardiovascular events. It can be assessed using cardiovascular magnetic resonance (CMR) from the delay between phase-contrast flow waveforms at two or more locations in the aorta, and the distance on CMR images between those locations. This study aimed to investigate the impact of different distance measurement methods on PWV. We present and evaluate an algorithm for automated centreline tracking in 3D images, and compare PWV calculations using distances derived from 3D images to those obtained from a conventional 2D oblique-sagittal image of the aorta.

METHODS

We included 35 patients from a twin cohort, and 20 post-coarctation repair patients. Phase-contrast flow was acquired in the ascending, descending and diaphragmatic aorta. A 3D centreline tracking algorithm is presented and evaluated on a subset of 30 subjects, on three CMR sequences: balanced steady-state free precession (SSFP), black-blood double inversion recovery turbo spin echo, and contrast-enhanced CMR angiography. Aortic lengths are subsequently compared between measurements from a 2D oblique-sagittal plane, and a 3D geometry.

RESULTS

The error in length of automated 3D centreline tracking compared with manual annotations ranged from 2.4 [1.8-4.3] mm (mean [IQR], black-blood) to 6.4 [4.7-8.9] mm (SSFP). The impact on PWV was below 0.5m/s (<5%). Differences between 2D and 3D centreline length were significant for the majority of our experiments (p < 0.05). Individual differences in PWV were larger than 0.5m/s in 15% of all cases (thoracic aorta) and 37% when studying the aortic arch only. Finally, the difference between end-diastolic and end-systolic 2D centreline lengths was statistically significant (p < 0.01), but resulted in small differences in PWV (0.08 [0.04 - 0.10]m/s).

CONCLUSIONS

Automatic aortic centreline tracking in three commonly used CMR sequences is possible with good accuracy. The 3D length obtained from such sequences can differ considerably from lengths obtained from a 2D oblique-sagittal plane, depending on aortic curvature, adequate planning of the oblique-sagittal plane, and patient motion between acquisitions. For accurate PWV measurements we recommend using 3D centrelines.

Contrast-enhanced micro-CT imaging in murine carotid arteries: a new protocol for computing wall shear stress

BACKGROUND

Wall shear stress (WSS) is involved in the pathophysiology of atherosclerosis. The correlation between WSS and atherosclerosis can be investigated over time using a WSS-manipulated atherosclerotic mouse model. To determine WSS in vivo, detailed 3D geometry of the vessel network is required. However, a protocol to reconstruct 3D murine vasculature using this animal model is lacking. In this project, we evaluated the adequacy of eXIA 160, a small animal contrast agent, for assessing murine vascular network on micro-CT. Also, a protocol was established for vessel geometry segmentation and WSS analysis.

METHODS

A tapering cast was placed around the right common carotid artery (RCCA) of ApoE-/- mice (n = 8). Contrast-enhanced micro-CT was performed using eXIA 160. An innovative local threshold-based segmentation procedure was implemented to reconstruct 3D geometry of the RCCA. The reconstructed RCCA was compared to the vessel geometry using a global threshold-based segmentation method. Computational fluid dynamics was applied to compute the velocity field and WSS distribution along the RCCA.

RESULTS

eXIA 160-enhanced micro-CT allowed clear visualization and assessment of the RCCA in all eight animals. No adverse biological effects were observed from the use of eXIA 160. Segmentation using local threshold values generated more accurate RCCA geometry than the global threshold-based approach. Mouse-specific velocity data and the RCCA geometry generated 3D WSS maps with high resolution, enabling quantitative analysis of WSS. In all animals, we observed low WSS upstream of the cast. Downstream of the cast, asymmetric WSS patterns were revealed with variation in size and location between animals.

CONCLUSIONS

eXIA 160 provided good contrast to reconstruct 3D vessel geometry and determine WSS patterns in the RCCA of the atherosclerotic mouse model. We established a novel local threshold-based segmentation protocol for RCCA reconstruction and WSS computation. The observed differences between animals indicate the necessity to use mouse-specific data for WSS analysis. For our future work, our protocol makes it possible to study in vivo WSS longitudinally over a growing plaque.

Noninvasive hemodynamic assessment, treatment outcome prediction and follow-up of aortic coarctation from MR imaging

PURPOSE

Coarctation of the aorta (CoA) is a congenital heart disease characterized by an abnormal narrowing of the proximal descending aorta. Severity of this pathology is quantified by the blood pressure drop (△P) across the stenotic coarctation lesion. In order to evaluate the physiological significance of the preoperative coarctation and to assess the postoperative results, the hemodynamic analysis is routinely performed by measuring the △P across the coarctation site via invasive cardiac catheterization. The focus of this work is to present an alternative, noninvasive measurement of blood pressure drop △P through the introduction of a fast, image-based workflow for personalized computational modeling of the CoA hemodynamics.

METHODS

The authors propose an end-to-end system comprised of shape and computational models, their personalization setup using MR imaging, and a fast, noninvasive method based on computational fluid dynamics (CFD) to estimate the pre- and postoperative hemodynamics for coarctation patients. A virtual treatment method is investigated to assess the predictive power of our approach.

RESULTS

Automatic thoracic aorta segmentation was applied on a population of 212 3D MR volumes, with mean symmetric point-to-mesh error of 3.00 ± 1.58 mm and average computation time of 8 s. Through quantitative evaluation of 6 CoA patients, good agreement between computed blood pressure drop and catheter measurements is shown: average differences are 2.38 ± 0.82 mm Hg (pre-), 1.10 ± 0.63 mm Hg (postoperative), and 4.99 ± 3.00 mm Hg (virtual stenting), respectively.

CONCLUSIONS

The complete workflow is realized in a fast, mostly-automated system that is integrable in the clinical setting. To the best of our knowledge, this is the first time that three different settings (preoperative--severity assessment, poststenting--follow-up, and virtual stenting--treatment outcome prediction) of CoA are investigated on multiple subjects. We believe that in future-given wider clinical validation-our noninvasive in-silico method could replace invasive pressure catheterization for CoA.

Carotid Artery Wall Segmentation in Multispectral MRI by Coupled Optimal Surface Graph Cuts

We present a new three-dimensional coupled optimal surface graph-cut algorithm to segment the wall of the carotid artery bifurcation from Magnetic Resonance (MR) images. The method combines the search for both inner and outer borders into a single graph cut and uses cost functions that integrate information from multiple sequences. Our approach requires manual localization of only three seed points indicating the start and end points of the segmentation in the internal, external, and common carotid artery. We performed a quantitative validation using images of 57 carotid arteries. Dice overlap of 0.86 ± 0.06 for the complete vessel and 0.89 ± 0.05 for the lumen compared to manual annotation were obtained. Reproducibility tests were performed in 60 scans acquired with an interval of 15 ± 9 days, showing good agreement between baseline and follow-up segmentations with intraclass correlations of 0.96 and 0.74 for the lumen and complete vessel volumes respectively.

Brain extraction in pediatric ADC maps, toward characterizing neuro-development in multi-platform and multi-institution clinical images

Apparent Diffusion Coefficient (ADC) maps can be used to characterize myelination and to detect abnormalities in the developing brain. However, given the normal variation in regional ADC with myelination, detection of abnormalities is difficult when based on visual assessment. Quantitative and automated analysis of pediatric ADC maps is thus desired but requires accurate brain extraction as the first step. Currently, most existing brain extraction methods are optimized for structural T1-weighted MR images of fully myelinated brains. Due to differences in age and image contrast, these approaches do not translate well to pediatric ADC maps. To address this problem, we present a multi-atlas brain extraction framework that has 1) specificity: designed and optimized specifically for pediatric ADC maps; 2) generality: applicable to multi-platform and multi-institution data, and to subjects at various neuro-developmental stages across the first 6 years of life; 3) accuracy: highly accurate compared to expert annotations; and 4) consistency: consistently accurate regardless of sources of data and ages of subjects. We show how we achieve these goals, via optimizing major components in a multi-atlas brain extraction framework, and via developing and evaluating new criteria for its atlas ranking component. Moreover, we demonstrate that these goals can be achieved with a fixed set of atlases and a fixed set of parameters, which opens doors for our optimized framework to be used in large-scale and multi-institution neuro-developmental and clinical studies. In a pilot study, we use this framework in a dataset containing scanner-generated ADC maps from 308 pediatric patients collected during the course of routine clinical care. Our framework leads to successful quantifications of the changes in whole-brain volumes and mean ADC values across the first 6 years of life.

A semi-automated method for measuring the evolution of both lumen area and blood flow in carotid from Phase Contrast MRI

Phase-Contrast (PC) velocimetry Magnetic Resonance Imaging (MRI) is a useful modality to explore cardiovascular pathologies, but requires the automatic segmentation of vessels and the measurement of both lumen area and blood flow evolutions. In this paper, we propose a semi-automated method for extracting lumen boundaries of the carotid artery and compute both lumen area and blood flow evolutions over the cardiac cycle. This method uses narrow band region-based active contours in order to correctly capture the lumen boundary without being corrupted by surrounding structures. This approach is compared to traditional edge-based active contours, considered in related works, which significantly underestimate lumen area and blood flow. Experiments are performed using both a sequence of a homemade phantom and sequences of 20 real carotids, including a comparison with manual segmentation performed by a radiologist expert. Results obtained on the phantom sequence show that the edge-based approach leads to an underestimate of carotid lumen area and related flows of respectively 18.68% and 4.95%. This appears significantly larger than weak errors obtained using the region-based approach (respectively 2.73% and 1.23%). Benefits appear even better on the real sequences. The edge-based approach leads to underestimates of 40.88% for areas and 13.39% for blood flows, compared to limited errors of 7.41% and 4.6% with our method. Experiments also illustrate the high variability and therefore the lack of reliability of manual segmentation.

The Multimodal Brain Tumor Image Segmentation Benchmark (BRATS)

In this paper we report the set-up and results of the Multimodal Brain Tumor Image Segmentation Benchmark (BRATS) organized in conjunction with the MICCAI 2012 and 2013 conferences. Twenty state-of-the-art tumor segmentation algorithms were applied to a set of 65 multi-contrast MR scans of low- and high-grade glioma patients-manually annotated by up to four raters-and to 65 comparable scans generated using tumor image simulation software. Quantitative evaluations revealed considerable disagreement between the human raters in segmenting various tumor sub-regions (Dice scores in the range 74%-85%), illustrating the difficulty of this task. We found that different algorithms worked best for different sub-regions (reaching performance comparable to human inter-rater variability), but that no single algorithm ranked in the top for all sub-regions simultaneously. Fusing several good algorithms using a hierarchical majority vote yielded segmentations that consistently ranked above all individual algorithms, indicating remaining opportunities for further methodological improvements. The BRATS image data and manual annotations continue to be publicly available through an online evaluation system as an ongoing benchmarking resource.

Diagnostic Accuracy of 4 Commercially Available Semiautomatic Packages for Carotid Artery Stenosis Measurement on CTA

BACKGROUND AND PURPOSE

Semiautomatic measurement of ICA stenosis potentially increases observer reproducibility. In this study, we assessed the diagnostic accuracy and interobserver reproducibility of a commercially available semiautomatic ICA stenosis measurement on CTA and estimated the agreement among different software packages.

MATERIALS AND METHODS

We analyzed 141 arteries from 90 patients with TIA or ischemic stroke. Manual stenosis measurements were performed by 2 neuroradiologists. Semiautomatic measurements by using 4 methods (3mensio and comparable software from Philips, TeraRecon, and Siemens) were performed by 2 observers. Diagnostic accuracy was estimated by comparing semiautomatic with manual measurements. Interobserver reproducibility and agreement between different packages was assessed by calculation of the intraclass correlation coefficient and Bland-Altman 95% limits of agreement. False-negative classifications were retrospectively inspected by a neuroradiologist.

RESULTS

There was no significant difference in the diagnostic performance of the 4 semiautomatic methods. The sensitivity for detecting ≥50% and ≥70% degree of stenosis was between 76% and 82% and 46% and 62%, respectively. Specificity and overall diagnostic accuracy were between 92% and 97% and 85% and 90%, respectively. The interobserver intraclass correlation coefficient was between 0.83 and 0.96 for semiautomatic measurements and 0.81 for manual measurement. The limits of agreement between each pair of semiautomatic packages ranged from -18%-24% to -33%-31%. False-negative classifications were caused by ulcerative plaques and observer variation in stenosis and reference measurements.

CONCLUSIONS

Semiautomatic methods have a low-to-good sensitivity and a good specificity and overall diagnostic accuracy. The high interobserver reproducibility makes semiautomatic stenosis measurement valuable for clinical practice, but semiautomatic measurements should be checked by an experienced radiologist.

Quantitative evaluation of noise reduction and vesselness filters for liver vessel segmentation on abdominal CTA images

Liver vessel segmentation in CTA images is a challenging task, especially in the case of noisy images. This paper investigates whether pre-filtering improves liver vessel segmentation in 3D CTA images. We introduce a quantitative evaluation of several well-known filters based on a proposed liver vessel segmentation method on CTA images. We compare the effect of different diffusion techniques i.e. Regularized Perona-Malik, Hybrid Diffusion with Continuous Switch and Vessel Enhancing Diffusion as well as the vesselness approaches proposed by Sato, Frangi and Erdt. Liver vessel segmentation of the pre-processed images is performed using a histogram-based region grown with local maxima as seed points. Quantitative measurements (sensitivity, specificity and accuracy) are determined based on manual landmarks inside and outside the vessels, followed by T-tests for statistic comparisons on 51 clinical CTA images. The evaluation demonstrates that all the filters make liver vessel segmentation have a significantly higher accuracy than without using a filter (p  <  0.05); Hybrid Diffusion with Continuous Switch achieves the best performance. Compared to the diffusion filters, vesselness filters have a greater sensitivity but less specificity. In addition, the proposed liver vessel segmentation method with pre-filtering is shown to perform robustly on a clinical dataset having a low contrast-to-noise of up to 3 (dB). The results indicate that the pre-filtering step significantly improves liver vessel segmentation on 3D CTA images.

Stability, structure and scale: improvements in multi-modal vessel extraction for SEEG trajectory planning

Purpose

Brain vessels are among the most critical landmarks that need to be assessed for mitigating surgical risks in stereo-electroencephalography (SEEG) implantation. Intracranial haemorrhage is the most common complication associated with implantation, carrying significantly associated morbidity. SEEG planning is done pre-operatively to identify avascular trajectories for the electrodes. In current practice, neurosurgeons have no assistance in the planning of electrode trajectories. There is great interest in developing computer-assisted planning systems that can optimise the safety profile of electrode trajectories, maximising the distance to critical structures. This paper presents a method that integrates the concepts of scale, neighbourhood structure and feature stability with the aim of improving robustness and accuracy of vessel extraction within a SEEG planning system.

Methods

The developed method accounts for scale and vicinity of a voxel by formulating the problem within a multi-scale tensor voting framework. Feature stability is achieved through a similarity measure that evaluates the multi-modal consistency in vesselness responses. The proposed measurement allows the combination of multiple images modalities into a single image that is used within the planning system to visualise critical vessels.

Results

Twelve paired data sets from two image modalities available within the planning system were used for evaluation. The mean Dice similarity coefficient was \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0.89\pm 0.04$$\end{document}0.89±0.04, representing a statistically significantly improvement when compared to a semi-automated single human rater, single-modality segmentation protocol used in clinical practice (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0.80 \pm 0.03$$\end{document}0.80±0.03).

Conclusions

Multi-modal vessel extraction is superior to semi-automated single-modality segmentation, indicating the possibility of safer SEEG planning, with reduced patient morbidity.

Volumetric assessment of the carotid bifurcation: an alternative concept to stenosis grading

BACKGROUND

To design a volumetric method for the assessment of carotid atheromatosis (CA) based on computed tomography (CT) angiography and three-dimensional (3D) reconstructions; to analyze the accuracy and optimal threshold values to differentiate between equivalent degrees of severity by duplex scanning and CT angiography (two-dimensional maximum intensity projection [2D MIP]; internal carotid artery stenosis [ICS] <50%; ICS >50%); and to assess the method's suitability to detect progression of CA.

METHODS

DESIGN

suitability and accuracy of a new diagnostic method.

POPULATION

90 carotid bifurcations (45 patients) were assessed with duplex scanning and CT angiography, and reevaluated after 12 ± 2 months follow-up. Determinations: Assessment of internal carotid artery (ICA) stenosis degree with duplex scanning and 2D MIP CT angiography projections. Volumetric assessment of carotid bifurcation by CT angiography (contrast volume [mm(3)] and density [Hounsfield units, H.U.] between 2-cm below and 1-cm above the anatomic bifurcation of the carotid artery [BifV], and its ratio with 1-cm segment of the common carotid artery [CCV]).

STATISTICAL ANALYSIS

descriptive statistics; intraobserver and interobserver agreement (Bland-Altman plot and intraclass correlation coefficient [ICC], accuracy of 3D volumetry and duplex scanning as referred to MIP 2D CT angiography as gold standard: sensitivity (Sens), specificity (Sp), kappa index, and receiver operating characteristic curves (ROCs).

RESULTS

Estimation of MIP 2D images (CT angiography) confirmed the findings of duplex scanning in 23 of 30 ICS <50% and 48 of 53 ICS >50% (Sens, 0.91; Sp, 0.77% kappa = 0.68). Three-dimensional volumetric assessment of carotid bifurcation showed an intraobserver and interobserver agreement with an ICC of 0.96 (95% confidence interval [CI], 0.904-0.985) and 0.94 (95% CI, 0.822-0.977), respectively. The BifV-to-CCV ratio was 5.2 ± 1.8 in the ICS <50% group versus 3.8 ± 1.3 in the ICS >50% group (P = 0.001). The optimal cutoff point of the BifV-to-CCV relationship was identified from the ROC curve in 4.1 (Sens, 0.75; Sp, 0.75; kappa, 0.46). At 12 months, a decrease of the average BifV with regard to the baseline value (475.45 [155.6] mm(3) × H.U. vs. 501.3 [171.9] mm(3) × H.U.; P = 0.04) was observed. CA progression was detected in 32 bifurcations (14 ICS <50%; 18 ICS> 50%), with a reduced bifurcation volume of 137.8 (71.4) mm(3) × H.U.; P < 0.001.

CONCLUSIONS

Volumetric assessment of carotid bifurcation is a new concept based on assessing plaque burden rather than its hemodynamic effect or maximum stenosis; thus, justifying its moderate accuracy with regard to ICS conventional ICA grading based on biplanar images. This method can be especially useful in plaque progression studies given its accuracy to detect minor changes in the arterial lumen.

Comparison and supervised learning of segmentation methods dedicated to specular microscope images of corneal endothelium

The cornea is the front of the eye. Its inner cell layer, called the endothelium, is important because it is closely related to the light transparency of the cornea. An in vivo observation of this layer is performed by using specular microscopy to evaluate the health of the cells: a high spatial density will result in a good transparency. Thus, the main criterion required by ophthalmologists is the cell density of the cornea endothelium, mainly obtained by an image segmentation process. Different methods can perform the image segmentation of these cells, and the three most performing methods are studied here. The question for the ophthalmologists is how to choose the best algorithm and to obtain the best possible results with it. This paper presents a methodology to compare these algorithms together. Moreover, by the way of geometric dissimilarity criteria, the algorithms are tuned up, and the best parameter values are thus proposed to the expert ophthalmologists.

Comparing algorithms for automated vessel segmentation in computed tomography scans of the lung: the VESSEL12 study

The VESSEL12 (VESsel SEgmentation in the Lung) challenge objectively compares the performance of different algorithms to identify vessels in thoracic computed tomography (CT) scans. Vessel segmentation is fundamental in computer aided processing of data generated by 3D imaging modalities. As manual vessel segmentation is prohibitively time consuming, any real world application requires some form of automation. Several approaches exist for automated vessel segmentation, but judging their relative merits is difficult due to a lack of standardized evaluation. We present an annotated reference dataset containing 20 CT scans and propose nine categories to perform a comprehensive evaluation of vessel segmentation algorithms from both academia and industry. Twenty algorithms participated in the VESSEL12 challenge, held at International Symposium on Biomedical Imaging (ISBI) 2012. All results have been published at the VESSEL12 website http://vessel12.grand-challenge.org. The challenge remains ongoing and open to new participants. Our three contributions are: (1) an annotated reference dataset available online for evaluation of new algorithms; (2) a quantitative scoring system for objective comparison of algorithms; and (3) performance analysis of the strengths and weaknesses of the various vessel segmentation methods in the presence of various lung diseases.

A semi-automatic method to extract canal pathways in 3D micro-CT images of Octocorals

The long-term goal of our study is to understand the internal organization of the octocoral stem canals, as well as their physiological and functional role in the growth of the colonies, and finally to assess the influence of climatic changes on this species. Here we focus on imaging tools, namely acquisition and processing of three-dimensional high-resolution images, with emphasis on automated extraction of canal pathways. Our aim was to evaluate the feasibility of the whole process, to point out and solve – if possible – technical problems related to the specimen conditioning, to determine the best acquisition parameters and to develop necessary image-processing algorithms. The pathways extracted are expected to facilitate the structural analysis of the colonies, namely to help observing the distribution, formation and number of canals along the colony. Five volumetric images of Muricea muricata specimens were successfully acquired by X-ray computed tomography with spatial resolution ranging from 4.5 to 25 micrometers. The success mainly depended on specimen immobilization. More than of the canals were successfully detected and tracked by the image-processing method developed. Thus obtained three-dimensional representation of the canal network was generated for the first time without the need of histological or other destructive methods. Several canal patterns were observed. Although most of them were simple, i.e. only followed the main branch or “turned” into a secondary branch, many others bifurcated or fused. A majority of bifurcations were observed at branching points. However, some canals appeared and/or ended anywhere along a branch. At the tip of a branch, all canals fused into a unique chamber. Three-dimensional high-resolution tomographic imaging gives a non-destructive insight to the coral ultrastructure and helps understanding the organization of the canal network. Advanced image-processing techniques greatly reduce human observer's effort and provide methods to both visualize and quantify the structures of interest.

A collaborative resource to build consensus for automated left ventricular segmentation of cardiac MR images

A collaborative framework was initiated to establish a community resource of ground truth segmentations from cardiac MRI. Multi-site, multi-vendor cardiac MRI datasets comprising 95 patients (73 men, 22 women; mean age 62.73±11.24years) with coronary artery disease and prior myocardial infarction, were randomly selected from data made available by the Cardiac Atlas Project (Fonseca et al., 2011). Three semi- and two fully-automated raters segmented the left ventricular myocardium from short-axis cardiac MR images as part of a challenge introduced at the STACOM 2011 MICCAI workshop (Suinesiaputra et al., 2012). Consensus myocardium images were generated based on the Expectation-Maximization principle implemented by the STAPLE algorithm (Warfield et al., 2004). The mean sensitivity, specificity, positive predictive and negative predictive values ranged between 0.63 and 0.85, 0.60 and 0.98, 0.56 and 0.94, and 0.83 and 0.92, respectively, against the STAPLE consensus. Spatial and temporal agreement varied in different amounts for each rater. STAPLE produced high quality consensus images if the region of interest was limited to the area of discrepancy between raters. To maintain the quality of the consensus, an objective measure based on the candidate automated rater performance distribution is proposed. The consensus segmentation based on a combination of manual and automated raters were more consistent than any particular rater, even those with manual input. The consensus is expected to improve with the addition of new automated contributions. This resource is open for future contributions, and is available as a test bed for the evaluation of new segmentation algorithms, through the Cardiac Atlas Project (www.cardiacatlas.org).

Lumen segmentation and stenosis quantification of atherosclerotic carotid arteries in CTA utilizing a centerline intensity prior

PURPOSE

The degree of stenosis is an important biomarker in assessing the severity of cardiovascular disease. The purpose of our work is to develop and evaluate a semiautomatic method for carotid lumen segmentation and subsequent carotid artery stenosis quantification in CTA images.

METHODS

The authors present a semiautomatic stenosis detection and quantification method following lumen segmentation. The lumen of the carotid arteries is segmented in three steps. First, centerlines of the internal and external carotid arteries are extracted with an iterative minimum cost path approach in which the costs are based on a measure of medialness and intensity similarity to lumen. Second, the lumen boundary is delineated using a level set procedure which is steered by gradient information, regional intensity information, and spatial information. Special effort is made in adding terms based on local centerline intensity prior so as to exclude all possible plaque tissues from the segmentation. Third, side branches in the segmented lumen are removed by applying a shape constraint to the envelope of the maximum inscribed spheres of the segmentation. From the segmented lumen, the authors detect and quantify the cross-sectional area-based and cross-sectional diameter-based stenosis degrees according to the North American Symptomatic Carotid En-darterectomy Trial criterion.

RESULTS

The method is trained and tested on a publicly available database from the cls2009 challenge. For the segmentation, the authors obtain a Dice similarity coefficient of 90.2% and a mean absolute surface distance of 0.34 mm. For the stenosis quantification, the authors obtain an average error of 15.7% for cross-sectional diameter-based stenosis and 19.2% for cross-sectional area-based stenosis quantification.

CONCLUSIONS

With these results, the method ranks second in terms of carotid lumen segmentation accuracy, and first in terms of carotid artery stenosis quantification.

Standardized evaluation methodology and reference database for evaluating IVUS image segmentation

This paper describes an evaluation framework that allows a standardized and quantitative comparison of IVUS lumen and media segmentation algorithms. This framework has been introduced at the MICCAI 2011 Computing and Visualization for (Intra)Vascular Imaging (CVII) workshop, comparing the results of eight teams that participated. We describe the available data-base comprising of multi-center, multi-vendor and multi-frequency IVUS datasets, their acquisition, the creation of the reference standard and the evaluation measures. The approaches address segmentation of the lumen, the media, or both borders; semi- or fully-automatic operation; and 2-D vs. 3-D methodology. Three performance measures for quantitative analysis have been proposed. The results of the evaluation indicate that segmentation of the vessel lumen and media is possible with an accuracy that is comparable to manual annotation when semi-automatic methods are used, as well as encouraging results can be obtained also in case of fully-automatic segmentation. The analysis performed in this paper also highlights the challenges in IVUS segmentation that remains to be solved.

Maximal curvature-based segmentation of 3D vessel contours

The segmentation of three-dimensional vascular trees is an important topic in medical image processing. Although it may seem to be an easy task, many different techniques have been proposed in the literature during the last decade and many difficulties remain. One can wonder why the human eye is usually able to understand the connectivity and the topology of the different structures while most algorithms fail to do so. In this paper, we propose an original approach that classifies the different contours by applying a geodesic distance transform on the contours of the vessels, where the evolution speed depends directly on the maximal curvature of the contours. This proposition comes from the observation that the maximal curvature on a standard vessel is usually positive and almost constant while it approaches zero or becomes negative on the contour at the contact with other structures. We describe our method in details and present promising results on synthetic and real images, where the method has been able to detect complex vascular structures without leaking into bones or mixing different vascular networks.

Estimating 3D lumen centerlines of carotid arteries in free-hand acquisition ultrasound

PURPOSE

The purpose of this paper is to present a methodology to estimate the carotid artery lumen centerlines in ultrasound (US) images obtained in a free-hand examination. Challenging aspects here are speckle noise in US images, artifacts, and the lack of contrast in the direction orthogonal to the US beam direction.

METHOD

An algorithm based on a rough lumen segmentation obtained by robust ellipse fitting was developed to deal with these conditions and estimate the lumen center in 2D B-mode scans. In a free-hand sweep examination, continuous image acquisitions are performed through time when the radiologist moves the probe on the patient's neck. The result is a series of images that show 2D cross-sections of the carotid's morphology. A tracking sensor (Flock of Birds) was attached to the probe and both were connected to a PC executing the Stradwin software, which relates spatial information to the acquisition data of the US probe. The spatial information was combined with the 2D lumen center estimates to provide a centerline in 3D. For validation, 19 carotid scans from 15 different patients were scanned, their centerlines calculated by the algorithm and compared with results acquired by manual annotations.

RESULTS

The average Euclidean distance between both among all the examinations was 0.82  mm. For each examination, the percentage of these Euclidean distances below 2  mm was calculated; the average over all examinations was 92%.

CONCLUSION

Automated 3D estimation of carotid artery lumen centerlines in free-hand real-time ultrasound is feasible and can be performed with high accuracy. The algorithm is robust enough to keep the centerlines inside the vessel, even in the absence of contrast in parts of the vessel wall.