Automatic model-based segmentation of the heart in CT images

A Learning-Based CT Prostate Segmentation Method via Joint Transductive Feature Selection and Regression

In1 recent years, there has been a great interest in prostate segmentation, which is a important and challenging task for CT image guided radiotherapy. In this paper, a learning-based segmentation method via joint transductive feature selection and transductive regression is presented, which incorporates the physician's simple manual specification (only taking a few seconds), to aid accurate segmentation, especially for the case with large irregular prostate motion. More specifically, for the current treatment image, experienced physician is first allowed to manually assign the labels for a small subset of prostate and non-prostate voxels, especially in the first and last slices of the prostate regions. Then, the proposed method follows the two step: in prostate-likelihood estimation step, two novel algorithms: tLasso and wLapRLS, will be sequentially employed for transductive feature selection and transductive regression, respectively, aiming to generate the prostate-likelihood map. In multi-atlases based label fusion step, the final segmentation result will be obtained according to the corresponding prostate-likelihood map and the previous images of the same patient. The proposed method has been substantially evaluated on a real prostate CT dataset including 24 patients with 330 CT images, and compared with several state-of-the-art methods. Experimental results show that the proposed method outperforms the state-of-the-arts in terms of higher Dice ratio, higher true positive fraction, and lower centroid distances. Also, the results demonstrate that simple manual specification can help improve the segmentation performance, which is clinically feasible in real practice.

Segmentation and optical flow estimation in cardiac CT sequences based on a spatiotemporal PDM with a correction scheme and the Hermite transform

PURPOSE

The left ventricle and the myocardium are two of the most important parts of the heart used for cardiac evaluation. In this work a novel framework that combines two methods to isolate and display functional characteristics of the heart using sequences of cardiac computed tomography (CT) is proposed. A shape extraction method, which includes a new segmentation correction scheme, is performed jointly with a motion estimation approach.

METHODS

For the segmentation task we built a Spatiotemporal Point Distribution Model (STPDM) that encodes spatial and temporal variability of the heart structures. Intensity and gradient information guide the STPDM. We present a novel method to correct segmentation errors obtained with the STPDM. It consists of a deformable scheme that combines three types of image features: local histograms, gradients and binary patterns. A bio-inspired image representation model based on the Hermite transform is used for motion estimation. The segmentation allows isolating the structure of interest while the motion estimation can be used to characterize the movement of the complete heart muscle.

RESULTS

The work is evaluated with several sequences of cardiac CT. The left ventricle was used for evaluation. Several metrics were used to validate the proposed framework. The efficiency of our method is also demonstrated by comparing with other techniques.

CONCLUSION

The implemented tool can enable physicians to better identify mechanical problems. The new correction scheme substantially improves the segmentation performance. Reported results demonstrate that this work is a promising technique for heart mechanical assessment.

Landmark constellation models for medical image content identification and localization

PURPOSE

Many medical imaging tasks require the detection and localization of anatomical landmarks, for example for the initialization of model-based segmentation or to detect anatomical regions present in an image. A large number of landmark and object localization methods have been described in the literature. The detection of single landmarks may be insufficient to achieve robust localization across a variety of imaging settings and subjects. Furthermore, methods like the generalized Hough transform yield the most likely location of an object, but not an indication whether or not the landmark was actually present in the image.

METHODS

For these reasons, we developed a simple and computationally efficient method combining localization results from multiple landmarks to achieve robust localization and to compute a localization confidence measure. For each anatomical region, we train a constellation model indicating the mean relative locations and location variability of a set of landmarks. This model is registered to the landmarks detected in a test image via point-based registration, using closed-form solutions. Three different outlier suppression schemes are compared, two using iterative re-weighting based on the residual landmark registration errors and the third being a variant of RANSAC. The mean weighted residual registration error serves as a confidence measure to distinguish true from false localization results. The method is optimized and evaluated on synthetic data, evaluating both the localization accuracy and the ability to classify good from bad registration results based on the residual registration error.

RESULTS

Two application examples are presented: the identification of the imaged anatomical region in trauma CT scans and the initialization of model-based segmentation for C-arm CT scans with different target regions. The identification of the target region with the presented method was in 96 % of the cases correct.

CONCLUSION

The presented method is a simple solution for combining multiple landmark localization results. With appropriate parameters, outlier suppression clearly improves the localization performance over model registration without outlier suppression. The optimum choice of method and parameters depends on the expected level of noise and outliers in the application at hand, as well as on the focus on localization, classification, or both. The method allows detecting and localizing anatomical fields of view in medical images and is well suited to support a wide range of applications comprising image content identification, anatomical navigation and visualization, or initializing the pose of organ shape models.

Automatic Liver Segmentation Based on Shape Constraints and Deformable Graph Cut in CT Images

Liver segmentation is still a challenging task in medical image processing area due to the complexity of the liver's anatomy, low contrast with adjacent organs, and presence of pathologies. This investigation was used to develop and validate an automated method to segment livers in CT images. The proposed framework consists of three steps: 1) preprocessing; 2) initialization; and 3) segmentation. In the first step, a statistical shape model is constructed based on the principal component analysis and the input image is smoothed using curvature anisotropic diffusion filtering. In the second step, the mean shape model is moved using thresholding and Euclidean distance transformation to obtain a coarse position in a test image, and then the initial mesh is locally and iteratively deformed to the coarse boundary, which is constrained to stay close to a subspace of shapes describing the anatomical variability. Finally, in order to accurately detect the liver surface, deformable graph cut was proposed, which effectively integrates the properties and inter-relationship of the input images and initialized surface. The proposed method was evaluated on 50 CT scan images, which are publicly available in two databases Sliver07 and 3Dircadb. The experimental results showed that the proposed method was effective and accurate for detection of the liver surface.

A bi-ventricular cardiac atlas built from 1000+ high resolution MR images of healthy subjects and an analysis of shape and motion

Atlases encode valuable anatomical and functional information from a population. In this work, a bi-ventricular cardiac atlas was built from a unique data set, which consists of high resolution cardiac MR images of 1000+ normal subjects. Based on the atlas, statistical methods were used to study the variation of cardiac shapes and the distribution of cardiac motion across the spatio-temporal domain. We have shown how statistical parametric mapping (SPM) can be combined with a general linear model to study the impact of gender and age on regional myocardial wall thickness. Finally, we have also investigated the influence of the population size on atlas construction and atlas-based analysis. The high resolution atlas, the statistical models and the SPM method will benefit more studies on cardiac anatomy and function analysis in the future.

Cardiac Multi-detector CT Segmentation Based on Multiscale Directional Edge Detector and 3D Level Set

Extraction of the cardiac surfaces of interest from multi-detector computed tomographic (MDCT) data is a pre-requisite step for cardiac analysis, as well as for image guidance procedures. Most of the existing methods need manual corrections, which is time-consuming. We present a fully automatic segmentation technique for the extraction of the right ventricle, left ventricular endocardium and epicardium from MDCT images. The method consists in a 3D level set surface evolution approach coupled to a new stopping function based on a multiscale directional second derivative Gaussian filter, which is able to stop propagation precisely on the real boundary of the structures of interest. We validated the segmentation method on 18 MDCT volumes from healthy and pathologic subjects using manual segmentation performed by a team of expert radiologists as gold standard. Segmentation errors were assessed for each structure resulting in a surface-to-surface mean error below 0.5 mm and a percentage of surface distance with errors less than 1 mm above 80%. Moreover, in comparison to other segmentation approaches, already proposed in previous work, our method presented an improved accuracy (with surface distance errors less than 1 mm increased of 8-20% for all structures). The obtained results suggest that our approach is accurate and effective for the segmentation of ventricular cavities and myocardium from MDCT images.

Differences in Regional Brain Volumes Two Months and One Year after Mild Traumatic Brain Injury

Conventional structural imaging is often normal after mild traumatic brain injury (mTBI). There is a need for structural neuroimaging biomarkers that facilitate detection of milder injuries, allow recovery trajectory monitoring, and identify those at risk for poor functional outcome and disability. We present a novel approach to quantifying volumes of candidate brain regions at risk for injury. Compared to controls, patients with mTBI had significantly smaller volumes in several regions including the caudate, putamen, and thalamus when assessed 2 months after injury. These differences persisted but were reduced in magnitude 1 year after injury, suggesting the possibility of normalization over time in the affected regions. More pronounced differences, however, were found in the amygdala and hippocampus, suggesting the possibility of regionally specific responses to injury.

Assessment of left sided filling dynamics in diastolic dysfunction using cardiac computed tomography

BACKGROUND

Left ventricular (LV) diastolic dysfunction (DD) often accompanies coronary artery disease but is difficult to assess since it involves a complex interaction between LV filling and left atrial (LA) emptying.

OBJECTIVE

To characterize simultaneous changes in LA and LV volumes using cardiac computed tomography (CT) in a group of patients with various grades of DD based on echocardiography.

METHODS

We identified 35 patients with DD by echocardiography, who had also undergone cardiac CT, and 35 age-matched normal controls. LV and LA volumes were measured every 10% of the RR interval, using semi-automatic software. From these, - systolic, early-diastolic and late-diastolic volume changes were calculated, and additional parameters of diastolic filling derived. Conduit volume was defined as the difference between the LV and LA early-diastolic volume change.

RESULTS

Patients with DD had significantly larger LV mass, and LA volumes, reduced early emptying volumes and increased conduit volume as percent of early LV filling (All p<0.001). LA function, manifesting as total emptying fraction (LATEF), decreased proportionately with worsening grades of DD (p<0.001). LA contractile function was maintained until advanced grade-3 DD. By receiver operating characteristic analysis, LATEF had an AUC of 0.88 to separate between normals and DD. At a threshold of <42.5%, LATEF has 97% sensitivity and 69% specificity to detect DD.

CONCLUSIONS

DD is characterized by reduced LA function and an alteration in the relative contributions of the atrial emptying and conduit volume components of early LV filling. In patients undergoing cardiac CT, it is possible to identify the presence and severity of DD.

A Pipeline for the Generation of Realistic 3D Synthetic Echocardiographic Sequences: Methodology and Open-Access Database

Quantification of cardiac deformation and strain with 3D ultrasound takes considerable research efforts. Nevertheless, a widespread use of these techniques in clinical practice is still held back due to the lack of a solid verification process to quantify and compare performance. In this context, the use of fully synthetic sequences has become an established tool for initial in silico evaluation. Nevertheless, the realism of existing simulation techniques is still too limited to represent reliable benchmarking data. Moreover, the fact that different centers typically make use of in-house developed simulation pipelines makes a fair comparison difficult. In this context, this paper introduces a novel pipeline for the generation of synthetic 3D cardiac ultrasound image sequences. State-of-the art solutions in the fields of electromechanical modeling and ultrasound simulation are combined within an original framework that exploits a real ultrasound recording to learn and simulate realistic speckle textures. The simulated images show typical artifacts that make motion tracking in ultrasound challenging. The ground-truth displacement field is available voxelwise and is fully controlled by the electromechanical model. By progressively modifying mechanical and ultrasound parameters, the sensitivity of 3D strain algorithms to pathology and image properties can be evaluated. The proposed pipeline is used to generate an initial library of 8 sequences including healthy and pathological cases, which is made freely accessible to the research community via our project web-page.

Computer Vision Techniques for Transcatheter Intervention

Minimally invasive transcatheter technologies have demonstrated substantial promise for the diagnosis and the treatment of cardiovascular diseases. For example, transcatheter aortic valve implantation is an alternative to aortic valve replacement for the treatment of severe aortic stenosis, and transcatheter atrial fibrillation ablation is widely used for the treatment and the cure of atrial fibrillation. In addition, catheter-based intravascular ultrasound and optical coherence tomography imaging of coronary arteries provides important information about the coronary lumen, wall, and plaque characteristics. Qualitative and quantitative analysis of these cross-sectional image data will be beneficial to the evaluation and the treatment of coronary artery diseases such as atherosclerosis. In all the phases (preoperative, intraoperative, and postoperative) during the transcatheter intervention procedure, computer vision techniques (e.g., image segmentation and motion tracking) have been largely applied in the field to accomplish tasks like annulus measurement, valve selection, catheter placement control, and vessel centerline extraction. This provides beneficial guidance for the clinicians in surgical planning, disease diagnosis, and treatment assessment. In this paper, we present a systematical review on these state-of-the-art methods. We aim to give a comprehensive overview for researchers in the area of computer vision on the subject of transcatheter intervention. Research in medical computing is multi-disciplinary due to its nature, and hence, it is important to understand the application domain, clinical background, and imaging modality, so that methods and quantitative measurements derived from analyzing the imaging data are appropriate and meaningful. We thus provide an overview on the background information of the transcatheter intervention procedures, as well as a review of the computer vision techniques and methodologies applied in this area.

Automatic determination of differential coronary artery motion minima for cardiac computed tomography optimal phase selection

RATIONALE AND OBJECTIVES

Selecting the optimal phase for coronary artery evaluation can be challenging, especially at higher heart rates, given that the optimal phase may differ for each of the coronary arteries. This study aimed to evaluate a novel vessel-specific algorithm which automatically outputs the minimum motion phase per coronary artery.

MATERIALS AND METHODS

The study included 44 patients who underwent 256-slice cardiac computed tomography for evaluation of chest pain. End-systolic and mid-diastolic minimal motion phases were automatically calculated by a previously validated global motion algorithm and by a new vessel-specific algorithm which calculates the minimum motion for each of the three main coronary arteries, separately. Two readers blindly evaluated all coronary segments for image quality. Median scores per coronary artery were compared by the Wilcoxon signed rank test.

RESULTS

The variation, per patient, between the optimal phases of the three coronary arteries was 5.0 ± 4.5% (1%-22%) for end systole and 4.8 ± 4.1% (0%-19%) for mid diastole. The mean image quality scores per coronary artery were 4.0 ± 0.61 for the vessel-specific approach and 3.80 ± 0.69 for the global phase selection (P < .001). Overall, 46 of 122 arteries had a better score with the vessel-specific approach and five with the standard global approach. Interreader agreement was substantial (k = 0.72).

CONCLUSIONS

This study has shown that multiple phases are required to ensure optimal image quality for all three coronary arteries and that a vessel-specific phase selection algorithm achieves superior results to the standard global approach.

Myocardial perfusion analysis in cardiac computed tomography angiographic images at rest

Cardiac computed tomography angiography (CTA) is a non-invasive method for anatomic evaluation of coronary artery stenoses. However, CTA is prone to artifacts that reduce the diagnostic accuracy to identify stenoses. Further, CTA does not allow for determination of the physiologic significance of the visualized stenoses. In this paper, we propose a new system to determine the physiologic manifestation of coronary stenoses by assessment of myocardial perfusion from typically acquired CTA images at rest. As a first step, we develop an automated segmentation method to delineate the left ventricle. Both endocardium and epicardium are compactly modeled with subdivision surfaces and coupled by explicit thickness representation. After initialization with five anatomical landmarks, the model is adapted to a target image by deformation increments including control vertex displacements and thickness variations guided by trained AdaBoost classifiers, and regularized by a prior of deformation increments from principal component analysis (PCA). The evaluation using a 5-fold cross-validation demonstrates the overall segmentation error to be 1.00 ± 0.39 mm for endocardium and 1.06 ± 0.43 mm for epicardium, with a boundary contour alignment error of 2.79 ± 0.52. Based on our LV model, two types of myocardial perfusion analyzes have been performed. One is a perfusion network analysis, which explores the correlation (as network edges) pattern of perfusion between all pairs of myocardial segments (as network nodes) defined in AHA 17-segment model. We find perfusion network display different patterns in the normal and disease groups, as divided by whether significant coronary stenosis is present in quantitative coronary angiography (QCA). The other analysis is a clinical validation assessment of the ability of the developed algorithm to predict whether a patient has significant coronary stenosis when referenced to an invasive QCA ground truth standard. By training three machine learning techniques using three features of normalized perfusion intensity, transmural perfusion ratio, and myocardial wall thickness, we demonstrate AdaBoost to be slightly better than Naive Bayes and Random Forest by the area under receiver operating characteristics (ROC) curve. For the AdaBoost algorithm, an optimal cut-point reveals an accuracy of 0.70, with sensitivity and specificity of 0.79 and 0.64, respectively. Our study shows perfusion analysis from CTA images acquired at rest is useful for providing physiologic information in diagnosis of obstructive coronary artery stenoses.

Pulmonary Hypertension: A Nomogram Based on CT Pulmonary Angiographic Data for Prediction in Patients without Pulmonary Embolism

PURPOSE

To use cardiovascular data from computerized tomographic (CT) pulmonary angiography for facilitating the identification of pulmonary hypertension (PH) in patients without acute pulmonary embolism.

MATERIALS AND METHODS

The institutional human research committee approved this retrospective study; informed consent was waived. Patients without pulmonary embolism who underwent CT pulmonary angiography and echocardiography within 24 hours of each other between December 2008 and October 2012 were retrospectively identified. The diameters of the pulmonary artery, aorta, and right and left ventricles and the severity of reflux of contrast material were assessed. The volumes of each cardiac compartment were calculated. Doppler echocardiography served as a reference standard for PH. A prediction model for PH was built by using backward logistic regression and was presented on a nomogram. The prediction model was evaluated with 10-fold cross-validation, and a test group of patients was studied between November 2012 and June 2014.

RESULTS

The final study group included 182 patients, of whom 98 (54%) were given a diagnosis of PH on the basis echocardiographic results. Age of 67 years or older (odds ratio [OR] = 4.46), reflux grade of 3 or higher (OR = 2.63), right atrial volume of greater than or equal to 106 cm(3) (OR = 3.59), pulmonary artery diameter greater than or equal to 28 mm (OR = 2.52) and pulmonary artery diameter to aorta diameter ratio of greater than or equal to 0.86 (OR = 2.17) were independently associated with PH. The logistic model showed good discrimination ability (area under the curve = 0.844, discrimination slope = 0.359). Tenfold cross-validation showed 85.7% sensitivity, 60.7% specificity, 71.3% positive predictive value, and 76.1% negative predictive value for identification of PH, while the test group showed similar results (84.1%, 60.5%, 71.2%, and 76.7%, respectively).

CONCLUSION

Cardiovascular data derived from CT pulmonary angiography are associated with PH, and a nomogram can be created that may facilitate identification of PH after exclusion of acute pulmonary embolism.

Three-dimensional cardiac computational modelling: methods, features and applications

The combination of computational models and biophysical simulations can help to interpret an array of experimental data and contribute to the understanding, diagnosis and treatment of complex diseases such as cardiac arrhythmias. For this reason, three-dimensional (3D) cardiac computational modelling is currently a rising field of research. The advance of medical imaging technology over the last decades has allowed the evolution from generic to patient-specific 3D cardiac models that faithfully represent the anatomy and different cardiac features of a given alive subject. Here we analyse sixty representative 3D cardiac computational models developed and published during the last fifty years, describing their information sources, features, development methods and online availability. This paper also reviews the necessary components to build a 3D computational model of the heart aimed at biophysical simulation, paying especial attention to cardiac electrophysiology (EP), and the existing approaches to incorporate those components. We assess the challenges associated to the different steps of the building process, from the processing of raw clinical or biological data to the final application, including image segmentation, inclusion of substructures and meshing among others. We briefly outline the personalisation approaches that are currently available in 3D cardiac computational modelling. Finally, we present examples of several specific applications, mainly related to cardiac EP simulation and model-based image analysis, showing the potential usefulness of 3D cardiac computational modelling into clinical environments as a tool to aid in the prevention, diagnosis and treatment of cardiac diseases.

MRI segmentation of the human brain: challenges, methods, and applications

Image segmentation is one of the most important tasks in medical image analysis and is often the first and the most critical step in many clinical applications. In brain MRI analysis, image segmentation is commonly used for measuring and visualizing the brain's anatomical structures, for analyzing brain changes, for delineating pathological regions, and for surgical planning and image-guided interventions. In the last few decades, various segmentation techniques of different accuracy and degree of complexity have been developed and reported in the literature. In this paper we review the most popular methods commonly used for brain MRI segmentation. We highlight differences between them and discuss their capabilities, advantages, and limitations. To address the complexity and challenges of the brain MRI segmentation problem, we first introduce the basic concepts of image segmentation. Then, we explain different MRI preprocessing steps including image registration, bias field correction, and removal of nonbrain tissue. Finally, after reviewing different brain MRI segmentation methods, we discuss the validation problem in brain MRI segmentation.

Available transcatheter aortic valve replacement technology

Transcatheter aortic valve replacement (TAVR) is an alternative and now recommended therapy for patients who meet indications for surgical valve replacement and are high or prohibitive surgical risk. Available TAVR technologies are rapidly emerging to treat this complex patient population. This review discusses the specific valve designs of the transcatheter heart valves currently used, newer generation modifications to overcome limitations of earlier valve designs, and novel imaging modalities, such as computed tomographic angiography-fluoroscopy and echocardiography-fluoroscopy fusion imaging, available for pre-procedural planning and intra-procedural guidance.

Real-time automated thickness measurement of the in vivo human tympanic membrane using optical coherence tomography

BACKGROUND

Otitis media (OM), an infection in the middle ear, is extremely common in the pediatric population. Current gold-standard methods for diagnosis include otoscopy for visualizing the surface features of the tympanic membrane (TM) and making qualitative assessments to determine middle ear content. OM typically presents as an acute infection, but can progress to chronic OM, and after numerous infections and antibiotic treatments over the course of many months, this disease is often treated by surgically inserting small tubes in the TM to relieve pressure, enable drainage, and provide aeration to the middle ear. Diagnosis and monitoring of OM is critical for successful management, but remains largely qualitative.

METHODS

We have developed an optical coherence tomography (OCT) system for high-resolution, depth-resolved, cross-sectional imaging of the TM and middle ear content, and for the quantitative assessment of in vivo TM thickness including the presence or absence of a middle ear biofilm. A novel algorithm was developed and demonstrated for automatic, real-time, and accurate measurement of TM thickness to aid in the diagnosis and monitoring of OM and other middle ear conditions. The segmentation algorithm applies a Hough transform to the OCT image data to determine the boundaries of the TM to calculate thickness.

RESULTS

The use of OCT and this segmentation algorithm is demonstrated first on layered phantoms and then during real-time acquisition of in vivo OCT from humans. For the layered phantoms, measured thicknesses varied by approximately 5 µm over time in the presence of large axial and rotational motion. In vivo data also demonstrated differences in thicknesses both spatially on a single TM, and across normal, acute, and chronic OM cases.

CONCLUSIONS

Real-time segmentation and thickness measurements of image data from both healthy subjects and those with acute and chronic OM demonstrate the use of OCT and this algorithm as a robust, quantitative, and accurate method for use during real-time in vivo human imaging.

Computed tomographic image analysis based on FEM performance comparison of segmentation on knee joint reconstruction

The demand for an accurate and accessible image segmentation to generate 3D models from CT scan data has been increasing as such models are required in many areas of orthopedics. In this paper, to find the optimal image segmentation to create a 3D model of the knee CT data, we compared and validated segmentation algorithms based on both objective comparisons and finite element (FE) analysis. For comparison purposes, we used 1 model reconstructed in accordance with the instructions of a clinical professional and 3 models reconstructed using image processing algorithms (Sobel operator, Laplacian of Gaussian operator, and Canny edge detection). Comparison was performed by inspecting intermodel morphological deviations with the iterative closest point (ICP) algorithm, and FE analysis was performed to examine the effects of the segmentation algorithm on the results of the knee joint movement analysis.

Multi-atlas segmentation with augmented features for cardiac MR images

Multi-atlas segmentation infers the target image segmentation by combining prior anatomical knowledge encoded in multiple atlases. It has been quite successfully applied to medical image segmentation in the recent years, resulting in highly accurate and robust segmentation for many anatomical structures. However, to guide the label fusion process, most existing multi-atlas segmentation methods only utilise the intensity information within a small patch during the label fusion process and may neglect other useful information such as gradient and contextual information (the appearance of surrounding regions). This paper proposes to combine the intensity, gradient and contextual information into an augmented feature vector and incorporate it into multi-atlas segmentation. Also, it explores the alternative to the K nearest neighbour (KNN) classifier in performing multi-atlas label fusion, by using the support vector machine (SVM) for label fusion instead. Experimental results on a short-axis cardiac MR data set of 83 subjects have demonstrated that the accuracy of multi-atlas segmentation can be significantly improved by using the augmented feature vector. The mean Dice metric of the proposed segmentation framework is 0.81 for the left ventricular myocardium on this data set, compared to 0.79 given by the conventional multi-atlas patch-based segmentation (Coupé et al., 2011; Rousseau et al., 2011). A major contribution of this paper is that it demonstrates that the performance of non-local patch-based segmentation can be improved by using augmented features.

Tracking using motion estimation with physically motivated inter-region constraints

We propose a method for tracking structures (e.g., ventricles and myocardium) in cardiac images (e.g., magnetic resonance) by propagating forward in time a previous estimate of the structures using a new physically motivated motion estimation scheme. Our method estimates motion by regularizing only within structures so that differing motions among different structures are not mixed. It simultaneously satisfies the physical constraints at the interface between a fluid and a medium that the normal component of the fluid's motion must match the normal component of the medium's motion and the No-Slip condition, which states that the tangential velocity approaches zero near the interface. We show that these conditions lead to partial differential equations with Robin boundary conditions at the interface, which couple the motion between structures. We show that propagating a segmentation across frames using our motion estimation scheme leads to more accurate segmentation than traditional motion estimation that does not use physical constraints. Our method is suited to interactive segmentation, prominently used in commercial applications for cardiac analysis, where segmentation propagation is used to predict a segmentation in the next frame. We show that our method leads to more accurate predictions than a popular and recent interactive method used in cardiac segmentation.

Compounding local invariant features and global deformable geometry for medical image registration

Using deformable models to register medical images can result in problems of initialization of deformable models and robustness and accuracy of matching of inter-subject anatomical variability. To tackle these problems, a novel model is proposed in this paper by compounding local invariant features and global deformable geometry. This model has four steps. First, a set of highly-repeatable and highly-robust local invariant features, called Key Features Model (KFM), are extracted by an effective matching strategy. Second, local features can be matched more accurately through the KFM for the purpose of initializing a global deformable model. Third, the positional relationship between the KFM and the global deformable model can be used to precisely pinpoint all landmarks after initialization. And fourth, the final pose of the global deformable model is determined by an iterative process with a lower time cost. Through the practical experiments, the paper finds three important conclusions. First, it proves that the KFM can detect the matching feature points well. Second, the precision of landmark locations adjusted by the modeled relationship between KFM and global deformable model is greatly improved. Third, regarding the fitting accuracy and efficiency, by observation from the practical experiments, it is found that the proposed method can improve % of the fitting accuracy and reduce around 50% of the computational time compared with state-of-the-art methods.

Images as drivers of progress in cardiac computational modelling

Computational models have become a fundamental tool in cardiac research. Models are evolving to cover multiple scales and physical mechanisms. They are moving towards mechanistic descriptions of personalised structure and function, including effects of natural variability. These developments are underpinned to a large extent by advances in imaging technologies. This article reviews how novel imaging technologies, or the innovative use and extension of established ones, integrate with computational models and drive novel insights into cardiac biophysics. In terms of structural characterization, we discuss how imaging is allowing a wide range of scales to be considered, from cellular levels to whole organs. We analyse how the evolution from structural to functional imaging is opening new avenues for computational models, and in this respect we review methods for measurement of electrical activity, mechanics and flow. Finally, we consider ways in which combined imaging and modelling research is likely to continue advancing cardiac research, and identify some of the main challenges that remain to be solved.

Three-dimensional reconstruction of a cardiac outline by magnetocardiography

A 3-D cardiac visualization is significantly helpful toward clinical applications of magnetocardiography (MCG), but the cardiac reconstruction requires a segmentation process using additional image modalities. This paper proposes a 3-D cardiac outline reconstruction method using only MCG measurement data without further imaging techniques. The cardiac outline was reconstructed by a combination of both spatial filtering and coherence mapping method. The strength of cardiac activities was first estimated by the array-gain constraint minimum-norm spatial filter with recursively updated gram matrix (AGMN-RUG). Then, waveforms were reconstructed at whole source grids, and the maximum source points of an atrium and ventricle were selected as a reference, respectively. Next, the coherence between each maximum source point and whole source points was compared by the coherence mapping method. A reconstructed cardiac outline was validated by comparing with an overlapped volume ratio when the reconstructed volume was identically matched with the original volume. The results obtained by the AGMN-RUG were compared to the results by other spatial filters. The accuracy of numerical simulation and phantom experiment by the AGMN-RUG was superior 10% and 8%, respectively, than the accuracy by the standardized low-resolution electromagnetic tomography. This accuracy demonstrated the efficacy of the proposed 3-D cardiac reconstruction method.

Improvement of automated right ventricular segmentation using dual-bolus contrast media injection with 256-slice coronary CT angiography

RATIONALE AND OBJECTIVES

To investigate the effect of dual-bolus contrast media injection (dual-CM) on the accuracy of automated right ventricular (RV) segmentation on coronary computed tomography angiography (CCTA).

MATERIALS AND METHODS

A total of 104 patients with suspected and known coronary artery disease underwent 256-slice CCTA with retrospective electrocardiographic (ECG) gating. The patients were divided into 51 patients who underwent single-bolus CM injection (single-CM) and 53 patients who underwent dual-CM. The dual-CM method consisted of an initial bolus of CM followed by an injection of dilute CM with saline (CM:saline, 1:9). Three-dimensional CCTA images were automatically segmented into the RV, left ventricle (LV), and myocardium using commercially available software (Comprehensive Cardiac Analysis; Philips Medical Systems, Cleveland, OH). Prevalence and locations of segmentation errors were compared between single-CM and dual-CM. Segmentation errors were defined as segment deviation of >1 cm from the actual ventricular margin.

RESULTS

Prevalence of segmentation errors was significantly lower with dual-CM than with single-CM in the diastolic phase (4/41 vs. 20/41, respectively; P < .01), and there was no difference between the two methods in the systolic phase (2/12 vs. 2/10, respectively). With dual-CM and single-CM, the locations of segmentation errors were mostly the RV wall (4/53 and 18/51, respectively) and secondly the LV wall (2/53 and 9/51, respectively).

CONCLUSIONS

Dual-CM improved the accuracy of automated ventricular segmentation using diastolic data from 256-slice CCTA.

A framework to measure myocardial extracellular volume fraction using dual-phase low dose CT images

PURPOSE

Myocardial extracellular volume fraction (ECVF) is a surrogate imaging biomarker of diffuse myocardial fibrosis, a hallmark of pathologic ventricular remodeling. Low dose cardiac CT is emerging as a promising modality to detect diffuse interstitial myocardial fibrosis due to its fast acquisition and low radiation; however, the insufficient contrast in the low dose CT images poses great challenge to measure ECVF from the image.

METHODS

To deal with this difficulty, the authors present a complete ECVF measurement framework including a point-guided myocardial modeling, a deformable model-based myocardium segmentation, nonrigid registration of pre- and post-CT, and ECVF calculation.

RESULTS

The proposed method was evaluated on 20 patients by two observers. Compared to the manually delineated reference segmentations, the accuracy of our segmentation in terms of true positive volume fraction (TPVF), false positive volume fraction (FPVF), and average surface distance (ASD), were 92.18% ± 3.52%, 0.31% ± 0.10%, 0.69 ± 0.14 mm, respectively. The interobserver variability measured by concordance correlation coefficient regarding TPVF, FPVF, and ASD were 0.95, 0.90, 0.94, respectively, demonstrating excellent agreement. Bland-Altman method showed 95% limits of agreement between ECVF at CT and ECVF at MR.

CONCLUSIONS

The proposed framework demonstrates its efficiency, accuracy, and noninvasiveness in ECVF measurement and dramatically advances the ECVF at cardiac CT toward its clinical use.

Automatic 4D reconstruction of patient-specific cardiac mesh with 1-to-1 vertex correspondence from segmented contours lines

We propose an automatic algorithm for the reconstruction of patient-specific cardiac mesh models with 1-to-1 vertex correspondence. In this framework, a series of 3D meshes depicting the endocardial surface of the heart at each time step is constructed, based on a set of border delineated magnetic resonance imaging (MRI) data of the whole cardiac cycle. The key contribution in this work involves a novel reconstruction technique to generate a 4D (i.e., spatial-temporal) model of the heart with 1-to-1 vertex mapping throughout the time frames. The reconstructed 3D model from the first time step is used as a base template model and then deformed to fit the segmented contours from the subsequent time steps. A method to determine a tree-based connectivity relationship is proposed to ensure robust mapping during mesh deformation. The novel feature is the ability to handle intra- and inter-frame 2D topology changes of the contours, which manifests as a series of merging and splitting of contours when the images are viewed either in a spatial or temporal sequence. Our algorithm has been tested on five acquisitions of cardiac MRI and can successfully reconstruct the full 4D heart model in around 30 minutes per subject. The generated 4D heart model conforms very well with the input segmented contours and the mesh element shape is of reasonably good quality. The work is important in the support of downstream computational simulation activities.

A complete system for automatic extraction of left ventricular myocardium from CT images using shape segmentation and contour evolution

The left ventricular myocardium plays a key role in the entire circulation system and an automatic delineation of the myocardium is a prerequisite for most of the subsequent functional analysis. In this paper, we present a complete system for an automatic segmentation of the left ventricular myocardium from cardiac computed tomography (CT) images using the shape information from images to be segmented. The system follows a coarse-to-fine strategy by first localizing the left ventricle and then deforming the myocardial surfaces of the left ventricle to refine the segmentation. In particular, the blood pool of a CT image is extracted and represented as a triangulated surface. Then, the left ventricle is localized as a salient component on this surface using geometric and anatomical characteristics. After that, the myocardial surfaces are initialized from the localization result and evolved by applying forces from the image intensities with a constraint based on the initial myocardial surface locations. The proposed framework has been validated on 34-human and 12-pig CT images, and the robustness and accuracy are demonstrated.

Automatic detection of left and right ventricles from CTA enables efficient alignment of anatomy with myocardial perfusion data

BACKGROUND

Accurate alignment between cardiac CT angiographic studies (CTA) and nuclear perfusion images is crucial for improved diagnosis of coronary artery disease. This study evaluated in an animal model the accuracy of a CTA fully automated biventricular segmentation algorithm, a necessary step for automatic and thus efficient PET/CT alignment.

METHODS AND RESULTS

Twelve pigs with acute infarcts were imaged using Rb-82 PET and 64-slice CTA. Post-mortem myocardium mass measurements were obtained. Endocardial and epicardial myocardial boundaries were manually and automatically detected on the CTA and both segmentations used to perform PET/CT alignment. To assess the segmentation performance, image-based myocardial masses were compared to experimental data; the hand-traced profiles were used as a reference standard to assess the global and slice-by-slice robustness of the automated algorithm in extracting myocardium, LV, and RV. Mean distances between the automated and the manual 3D segmented surfaces were computed. Finally, differences in rotations and translations between the manual and automatic surfaces were estimated post-PET/CT alignment. The largest, smallest, and median distances between interactive and automatic surfaces averaged 1.2 ± 2.1, 0.2 ± 1.6, and 0.7 ± 1.9 mm. The average angular and translational differences in CT/PET alignments were 0.4°, -0.6°, and -2.3° about x, y, and z axes, and 1.8, -2.1, and 2.0 mm in x, y, and z directions.

CONCLUSIONS

Our automatic myocardial boundary detection algorithm creates surfaces from CTA that are similar in accuracy and provide similar alignments with PET as those obtained from interactive tracing. Specific difficulties in a reliable segmentation of the apex and base regions will require further improvements in the automated technique.

An automatic service for the personalization of ventricular cardiac meshes

Computational cardiac physiology has great potential to improve the management of cardiovascular diseases. One of the main bottlenecks in this field is the customization of the computational model to the anatomical and physiological status of the patient. We present a fully automatic service for the geometrical personalization of cardiac ventricular meshes with high-order interpolation from segmented images. The method is versatile (able to work with different species and disease conditions) and robust (fully automatic results fulfilling accuracy and quality requirements in 87% of 255 cases). Results also illustrate the capability to minimize the impact of segmentation errors, to overcome the sparse resolution of dynamic studies and to remove the sometimes unnecessary anatomical detail of papillary and trabecular structures. The smooth meshes produced can be used to simulate cardiac function, and in particular mechanics, or can be used as diagnostic descriptors of anatomical shape by cardiologists. This fully automatic service is deployed in a cloud infrastructure, and has been made available and accessible to the scientific community.

Automatic segmentation of the left atrium from MR images via variational region growing with a moments-based shape prior

The planning and evaluation of left atrial ablation procedures are commonly based on the segmentation of the left atrium, which is a challenging task due to large anatomical variations. In this paper, we propose an automatic approach for segmenting the left atrium from magnetic resonance imagery. The segmentation problem is formulated as a problem in variational region growing. In particular, the method starts locally by searching for a seed region of the left atrium from an MR slice. A global constraint is imposed by applying a shape prior to the left atrium represented by Zernike moments. The overall growing process is guided by the robust statistics of intensities from the seed region along with the shape prior to capture the entire atrial region. The robustness and accuracy of our approach are demonstrated by experimental results from 64 human MR images.

Quantitative magnetic resonance imaging analysis of the relationship between contact force and left atrial scar formation after catheter ablation of atrial fibrillation

BACKGROUND

Catheter contact force (CF) is an important determinant of radiofrequency (RF) lesion quality during pulmonary vein isolation (PVI). Late gadolinium enhancement (LGE) magnetic resonance imaging (MRI) allows good visualization of ablation lesions.

OBJECTIVE

This study describes a new technique to examine the relationship between CF during RF delivery and LGE signal intensity (SI) following PVI.

METHODS

Six patients underwent PVI for paroxysmal AF using a CF-sensing catheter and following preprocedural MRI. During ablation, CF-time integral (FTI) and position was documented for each RF application. All patients underwent repeat LGE MRI 3 months later. The LGE SIs were projected onto a MRI-derived 3-dimensional left atrial (LA) shell and a CF map was generated on the same shell. The entire LA surface was divided into 5 mm(2) segments. Force and LGE maps were fused and compared for each 5 mm(2) zone. An effective lesion was defined when MRI-defined scar occupied >90% of a 5 mm(2) analysis zone.

RESULTS

Acute PVI was achieved in 100%. Two hundred sixty-eight RF lesions were tagged on the LA shells and given a lesion-specific FTI. Increasing FTI correlated with increased LGE SI, which was greater when the FTI was > 1,200 gs. Below an FTI of 1,200 gs, an increment in the FTI resulted in only a small increment in scar, whereas above 1,200 gs an increment in the FTI resulted in a large change of scar.

CONCLUSION

There is a correlation between FTI and LGE SI in MRI following AF ablation. Real-time FTI maps are feasible and may prevent inadequate lesion formation.

Automatic delineation of the myocardial wall from CT images via shape segmentation and variational region growing

Prognosis and diagnosis of cardiac diseases frequently require quantitative evaluation of the ventricle volume, mass, and ejection fraction. The delineation of the myocardial wall is involved in all of these evaluations, which is a challenging task due to large variations in myocardial shapes and image quality. In this paper, we present an automatic method for extracting the myocardial wall of the left and right ventricles from cardiac CT images. In the method, the left and right ventricles are located sequentially, in which each ventricle is detected by first identifying the endocardium and then segmenting the epicardium. To this end, the endocardium is localized by utilizing its geometric features obtained on-line from a CT image. After that, a variational region-growing model is employed to extract the epicardium of the ventricles. In particular, the location of the endocardium of the left ventricle is determined via using an active contour model on the blood-pool surface. To localize the right ventricle, the active contour model is applied on a heart surface extracted based on the left ventricle segmentation result. The robustness and accuracy of the proposed approach is demonstrated by experimental results from 33 human and 12 pig CT images.

3D strain assessment in ultrasound (Straus): a synthetic comparison of five tracking methodologies

This paper evaluates five 3D ultrasound tracking algorithms regarding their ability to quantify abnormal deformation in timing or amplitude. A synthetic database of B-mode image sequences modeling healthy, ischemic and dyssynchrony cases was generated for that purpose. This database is made publicly available to the community. It combines recent advances in electromechanical and ultrasound modeling. For modeling heart mechanics, the Bestel-Clement-Sorine electromechanical model was applied to a realistic geometry. For ultrasound modeling, we applied a fast simulation technique to produce realistic images on a set of scatterers moving according to the electromechanical simulation result. Tracking and strain accuracies were computed and compared for all evaluated algorithms. For tracking, all methods were estimating myocardial displacements with an error below 1 mm on the ischemic sequences. The introduction of a dilated geometry was found to have a significant impact on accuracy. Regarding strain, all methods were able to recover timing differences between segments, as well as low strain values. On all cases, radial strain was found to have a low accuracy in comparison to longitudinal and circumferential components.

Point-guided modeling and segmentation of myocardium for low dose cardiac CT images

Cardiac CT is emerging as a preferable modality to detect myocardial stress/rest perfusion; however the insufficient contrast of myocardium on CT image makes its segmentation difficult. In this paper, we present a point-guided modeling and deformable model-based segmentation method. This method first builds a triangular surface model of myocardium through Bézier contour fitting based on a few points selected by clinicians. Then, a deformable model-based segmentation method is developed to refine the segmentation result. The experiments on 8 cases show the accuracy of the segmentation in terms of true positive volume fraction, false positive volume fractions, and average surface distance can reach 91.0%, 0.3%, and 0.6mm, respectively. The comparison between the proposed method and a graph cut-based method is performed. The results demonstrate that this method is effective in improving the accuracy further.

Objective quantification of aortic valvular structures by cardiac computed tomography angiography in patients considered for transcatheter aortic valve implantation

PURPOSE

To test the ability of a model-based segmentation of the aortic root for consistent assessment of aortic valve structures in patients considered for transcatheter aortic valve implantation (TAVI) who underwent 256-slice cardiac computed tomography (CT).

METHODS

Consecutive patients (n = 49) with symptomatic severe aortic stenosis considered for TAVI and patients without aortic stenosis (n = 17) underwent cardiac CT. Images were evaluated by two independent observers who measured the diameter of the aortic annulus and its distance to both coronary ostia (1) manually and (2) software-assisted. All acquired measures were compared with each other and to (3) fully automatic quantification.

RESULTS

High correlations were observed for 3D measures of the aortic annulus conducted on multiple oblique planes (r = 0.87 and 0.84 between observers and model-based measures, and r = 0.81 between observers). Reproducibility was further improved by software-assisted versus manual assessment for all the acquired variables (r = 0.98 versus 0.81 for annulus diameter, r = 0.94 versus 0.85 for distance to the left coronary ostium, P < 0.01 for both). Thus, using software-assisted measurements very low limits of agreement were observed for the annulus diameter (95%CI of -1.2 to 0.6 mm) and within very low time-spent (0.6 ± 0.1 min for software-assisted versus 1.6 ± 0.3 min per patient for manual assessment, P < 0.001). Assessment of the aortic annulus using the 3D model-based instead of manual 2D-coronal measurements would have modified the implantation strategy in 12 of 49 patients (25%) with aortic stenosis. Four of 12 patients with potentially modified implantation strategy yielded postprocedural moderate paravalvular regurgitation, which may have been avoided by implantation of a larger prosthesis, as suggested by automatic 3D measures.

CONCLUSION

Our study highlights the usefulness of software-assisted preprocedural assessment of the aortic annulus in patients considered for TAVI.

Spine segmentation in medical images using manifold embeddings and higher-order MRFs

We introduce a novel approach for segmenting articulated spine shape models from medical images. A nonlinear low-dimensional manifold is created from a training set of mesh models to establish the patterns of global shape variations. Local appearance is captured from neighborhoods in the manifold once the overall representation converges. Inference with respect to the manifold and shape parameters is performed using a higher-order Markov random field (HOMRF). Singleton and pairwise potentials measure the support from the global data and shape coherence in manifold space respectively, while higher-order cliques encode geometrical modes of variation to segment each localized vertebra models. Generic feature functions learned from ground-truth data assigns costs to the higher-order terms. Optimization of the model parameters is achieved using efficient linear programming and duality. The resulting model is geometrically intuitive, captures the statistical distribution of the underlying manifold and respects image support. Clinical experiments demonstrated promising results in terms of spine segmentation. Quantitative comparison to expert identification yields an accuracy of 1.6 ± 0.6 mm for CT imaging and of 2.0 ± 0.8 mm for MR imaging, based on the localization of anatomical landmarks.

On mixed reality environments for minimally invasive therapy guidance: systems architecture, successes and challenges in their implementation from laboratory to clinic

Mixed reality environments for medical applications have been explored and developed over the past three decades in an effort to enhance the clinician's view of anatomy and facilitate the performance of minimally invasive procedures. These environments must faithfully represent the real surgical field and require seamless integration of pre- and intra-operative imaging, surgical instrument tracking, and display technology into a common framework centered around and registered to the patient. However, in spite of their reported benefits, few mixed reality environments have been successfully translated into clinical use. Several challenges that contribute to the difficulty in integrating such environments into clinical practice are presented here and discussed in terms of both technical and clinical limitations. This article should raise awareness among both developers and end-users toward facilitating a greater application of such environments in the surgical practice of the future.

Share and enjoy: anatomical models database--generating and sharing cardiovascular model data using web services

Sharing data between scientists and with clinicians in cardiac research has been facilitated significantly by the use of web technologies. The potential of this technology has meant that information sharing has been routinely promoted through databases that have encouraged stakeholder participation in communities around these services. In this paper we discuss the Anatomical Model Database (AMDB) (Gianni et al. Functional imaging and modeling of the heart. Springer, Heidelberg, 2009; Gianni et al. Phil Trans Ser A Math Phys Eng Sci 368:3039-3056, 2010) which both facilitate a database-centric approach to collaboration, and also extends this framework with new capabilities for creating new mesh data. AMDB currently stores cardiac geometric models described in Gianni et al. (Functional imaging and modelling of the heart. Springer, Heidelberg, 2009), a number of additional cardiac models describing geometry and functional properties, and most recently models generated using a web service. The functional models represent data from simulations in geometric form, such as electrophysiology or mechanics, many of which are present in AMDB as part of a benchmark study. Finally, the heartgen service has been added for producing left or bi-ventricle models derived from binary image data using the methods described in Lamata et al. (Med Image Anal 15:801-813, 2011). The results can optionally be hosted on AMDB alongside other community-provided anatomical models. AMDB is, therefore, a unique database storing geometric data (rather than abstract models or image data) combined with a powerful web service for generating new geometric models.

Generating anatomical models of the heart and the aorta from medical images for personalized physiological simulations

The anatomy and motion of the heart and the aorta are essential for patient-specific simulations of cardiac electrophysiology, wall mechanics and hemodynamics. Within the European integrated project euHeart, algorithms have been developed that allow to efficiently generate patient-specific anatomical models from medical images from multiple imaging modalities. These models, for instance, account for myocardial deformation, cardiac wall motion, and patient-specific tissue information like myocardial scar location. Furthermore, integration of algorithms for anatomy extraction and physiological simulations has been brought forward. Physiological simulations are linked closer to anatomical models by encoding tissue properties, like the muscle fibers, into segmentation meshes. Biophysical constraints are also utilized in combination with image analysis to assess tissue properties. Both examples show directions of how physiological simulations could provide new challenges and stimuli for image analysis research in the future.