Gerard van Burken

Automatic active appearance model segmentation of 3D echocardiograms

A fully automated segmentation for 3D echocardiography (3DE) using 3D Active Appearance Models (AAM) was developed and evaluated on 99 patients. The method used ultrasound specific grey value normalization and two matching algorithms were tested. To our knowledge this is the first report on a fully operational 3D AAM employed in 3DE on a large scale. The 3D AAM detected the endocardial contours accurately, even in the presence of large variations in left ventricular appearance and shape. Matching was successful in 91% of patients and resulted in a median point-tosurface error of 2.69 mm (av±sd: 2.91±1.03mm). Results indicate that fully automated AAM analysis of 3DE is practically feasible in datasets of mixed origin and quality.

Side-by-side viewing of anatomically aligned left ventricular segments in three-dimensional stress echocardiography.

Background: Dobutamine stress echocardiography (DSE) suffers from high interobserver and inter‐institution variability in the diagnosis of myocardial ischemia. Therefore, we developed a three‐dimensional (3D) analysis tool that makes it possible to anatomically align 3D rest and stress data systematically, to generate optimal, nonforeshortened standard anatomical cross sections and to analyse the images synchronized and side‐by‐side. Aim of the study: To investigate whether this 3D analysis tool could improve interobserver agreement on myocardial ischemia during 3D DSE. Methods: The study comprised 34 consecutive patients with stable chest pain who underwent both noncontrast and contrast 3D DSE. Two observers scored segmental wall motion using a conventional analysis and the novel analysis with the new 3D tool. Results: The two observers agreed on the presence or absence of myocardial ischemia in 81 of 102 coronary territories (agreement 79%, kappa (κ) 0.28) during noncontrast 3D imaging and 92 of 102 coronary territories (agreement 90%, kappa 0.65) during contrast‐enhanced 3D imaging. With the new 3D analysis software these numbers improved to 98 of 102 coronary territories (agreement 96%, kappa 0.69) during noncontrast 3D imaging and 98 of 102 coronary territories (agreement 96%, kappa 0.82) during contrast‐enhanced 3D imaging. Conclusion: The use of a 3D DSE analysis tool improves interobserver agreement for myocardial ischemia both for noncontrast and contrast images.

Overview of automated quantitation techniques in 2D echocardiography

Many methods for automated quantitation in two-dimensional echocardiography have been published, but few have gained practical importance. This chapter describes the problems and pitfalls of border detection in cardiac ultrasound, gives an overview of methods described in the literature and categorizes the applied techniques into a hierarchy of abstraction levels. Furthermore, a practical system for automated border detection (ECHO-CMS) and its evaluation will be discussed, and the chapter will be concluded with an overview of the general developments anticipated for the near future.

Carotid Intraplaque Neovascularization Quantification Software (CINQS)

Intraplaque neovascularization (IPN) is an important biomarker of atherosclerotic plaque vulnerability. As IPN can be detected by contrast enhanced ultrasound (CEUS), imaging-biomarkers derived from CEUS may allow early prediction of plaque vulnerability. To select the best quantitative imaging-biomarkers for prediction of plaque vulnerability, a systematic analysis of IPN with existing and new analysis algorithms is necessary. Currently available commercial contrast quantification tools are not applicable for quantitative analysis of carotid IPN due to substantial motion of the carotid artery, artifacts, and intermittent perfusion of plaques. We therefore developed a specialized software package called Carotid intraplaque neovascularization quantification software (CINQS). It was designed for effective and systematic comparison of sets of quantitative imaging biomarkers. CINQS includes several analysis algorithms for carotid IPN quantification and overcomes the limitations of current contrast quantification tools and existing carotid IPN quantification approaches. CINQS has a modular design which allows integrating new analysis tools. Wizard-like analysis tools and its graphical-user-interface facilitate its usage. In this paper, we describe the concept, analysis tools, and performance of CINQS and present analysis results of 45 plaques of 23 patients. The results in 45 plaques showed excellent agreement with visual IPN scores for two quantitative imaging-biomarkers (The area under the receiver operating characteristic curve was 0.92 and 0.93).

Sparse appearance model based registration of 3D ultrasound images

In this paper, we propose a sparse appearance model based registration algorithm for segmenting 3D echocardiograms. The end-diastolic model is built in 3D sparsely on 2D planes, representing the anatomical 4-chamber, 2-chamber, and short-axis views. Ultrasound specific intensity normalization and shape-based intensity modeling are employed. The model is matched in an intensity-based registration approach, by perturbing appearance and pose parameters simultaneously. Leave-one-out experiments on 10 patients reveal significant improvement in the segmentation using the normalized cross-correlation metric. The registration method will allow fully automatic extraction of the standard views as used in echocardiography. This will aid in the selection of images for inter- and intra-patient comparison and may provide an alternative for a complete 3D AAM.

Improved Segmentation of Multiple Cavities of the Heart in Wide-View 3-D Transesophageal Echocardiograms.

Minimally invasive interventions in the heart such as in electrophysiology are becoming more and more important in clinical practice. Currently, preoperative computed tomography angiography (CTA) is used to provide anatomic information during electrophysiology interventions, but this does not provide real-time feedback and burdens the patient with additional radiation and side effects of the contrast agent. Three-dimensional transesophageal echocardiography (TEE) is an excellent modality for visualization of anatomic structures and instruments in real time, but some cavities, especially the left atrium, suffer from the limited coverage of the 3-D TEE volumes. This leads to difficulty in segmenting the left atrium. We propose replacing or complementing pre-operative CTA imaging with wide-view TEE. We tested this proposal on 20 patients for which TEE image volumes covering the left atrium and CTA images were acquired. The TEE images were manually registered, and wide-view volumes were generated. Five heart cavities in single-view and wide-view TEE were segmented and compared with atlas based-segmentations derived from the CTA images. We found that the segmentation accuracy (Dice coefficients) improved relative to segmentation of single-view images by 5, 15 and 9 percentage points for the left atrium, right atrium and aorta, respectively. Average anatomic coverage was improved by 2, 29, 62 and 49 percentage points for the right ventricle, left atrium, right atrium and aorta, respectively. This finding confirms that wide-view 3-D TEE can be useful in supporting electrophysiology interventions.

Comparison of spatiotemporal interpolators for 4D image reconstruction from 2D transesophageal ultrasound

°For electrophysiology intervention monitoring, we intend to reconstruct 4D ultrasound (US) of structures in the beating heart from 2D transesophageal US by scanplane rotation. The image acquisition is continuous but unsynchronized to the heart rate, which results in a sparsely and irregularly sampled dataset and a spatiotemporal interpolation method is desired. Previously, we showed the potential of normalized convolution (NC) for interpolating such datasets. We explored 4D interpolation by 3 different methods: NC, nearest neighbor (NN), and temporal binning followed by linear interpolation (LTB). The test datasets were derived by slicing three 4D echocardiography datasets at random rotation angles (θ, range: 0-180) and random normalized cardiac phase (τ, range: 0-1). Four different distributions of rotated 2D images with 600, 900, 1350, and 1800 2D input images were created from all TEE sets. A 2D Gaussian kernel was used for NC and optimal kernel sizes (σθ and στ) were found by performing an exhaustive search. The RMS gray value error (RMSE) of the reconstructed images was computed for all interpolation methods. The estimated optimal kernels were in the range of σθ = 3.24 - 3.69°/ στ = 0.045 - 0.048, σθ = 2.79°/ στ = 0.031 - 0.038, σθ = 2.34°/ στ = 0.023 - 0.026, and σθ = 1.89°/ στ = 0.021 - 0.023 for 600, 900, 1350, and 1800 input images respectively. We showed that NC outperforms NN and LTB. For a small number of input images the advantage of NC is more pronounced.

Simultaneous segmentation of multiple heart cavities in 3D transesophageal echocardiograms

Three-dimensional transesophageal echocardiography (3D TEE) is an excellent modality for real-time visualization of the heart and monitoring of interventions. However, 3D TEE segmentation is still a challenging task due to the complex anatomy, the limited field of view, and typical ultrasound artifacts. To improve the usability of 3D TEE for monitoring interventions, we propose to segment all cavities within the TEE view with a multi-cavity Active Shape Model (ASM) derived from Computed Tomography Angiography (CTA) in conjunction with a tissue/blood classification based on a Gamma Mixture Model (GMM). 3D TEE image data of five patients were acquired with a Philips X7-2t matrix TEE probe. Tissue probability maps were estimated by a two class (blood/tissue) GMM. A statistical shape model containing left and right ventricle, left and right atrium and aorta (LV, LA, RV, LA, Ao) was derived from CTA scans of 151 patients by Principal Component Analysis. Models from individual cavities (ASMpart: ASMLV etc.) and of the whole heart (ASMtot) were generated. First, ASMtot was aligned with the 3D TEE by indicating 3 anatomical landmarks. Second, pose and shape of ASMtot were iteratively updated by a weighted update scheme excluding parts outside of the image sector. Third, shape and pose of each ASMpart were initialized based on shape and pose of ASMtot and iteratively updated in a constrained manner to fit the tissue probability maps. All 3D TEE sets were manually outlined in multiple short and long axis views by two observers. The mean outline of both observers was compared to the ASM segmentations by calculating Dice coefficients. All patients had preoperative CTA scans which were segmented using an atlas approach. The TEE and the CTA segmentation were registered and Dice coefficients were computed. The Dice coefficients of the whole heart between the average observer and ASM segmentations were 0.91, 0.75, 0.87, 0.88, and 0.84 (interobserver variability: 0.95, 0.92, 0.92, 0.88, and 0.90) for TEE set 1 to 5 respectively. The Dice coefficient for the whole hart between CTA and TEE segmentation were 0.85, 0.80, 0.80, 0.81, and 0.71 and showed good agreement. In this work we could successfully show the accuracy and robustness of the proposed multi-cavity segmentation scheme.

Segmentation of 3D transesophageal echocardiograms by multi-cavity active shape model and gamma mixture model

Segmentation of three-dimensional (3D) transesophageal ultrasound (TEE) is highly desired for intervention monitoring and guidance, but it is still a challenging image processing task due to complex local anatomy, limited field of view and typical ultrasound artifacts. We propose to use a multi-cavity active shape model (ASM) derived from Computed Tomography Angiography (CTA) segmentations in conjunction with a blood/tissue classification by Gamma Mixture Models to identify and segment the individual cavities simultaneously. A scheme that utilized successively ASMs of the whole heart and the individual cavities was used to segment the entire heart. We successfully validated our segmentation scheme with manually outlined contours and with CTA segmentations for three patients. The segmentations of the three patients had an average distance of 2.3, 4.9, and 2.1 mm to the manual outlines.

Automatic 3D left ventricular border detection using active appearance models

A fully automated segmentation for 3D echocardiography (3DE) using 3D Active Appearance Models (AAM) was developed and evaluated on end-diastolic (ED) and end-systolic (ES) images of 99 patients. The method used ultrasound specific grey value normalization and employed both regular matching and jacobian tuning. The 3D AAM detected the endocardial contours accurately, even in the presence of large variations in left ventricular appearance and shape. Matching was successful in 87% of patients and resulted in good median point-to-surface errors of 2.65 mm for ED and 3.21 for ES, and good volume regressions (ED: y = −3.2 +1.01×, r=0.95; ES: y = −4.6 +1.01×, r=0.92). Results show that fully automated AAM analysis is practically feasible in 3DE datasets of mixed origin and quality.