The NIH/NIGMS
Center for Integrative Biomedical Computing

Background - Radiofrequency ablation is routinely used to treat cardiac arrhythmias, but gaps remain in ablation lesion sets because there is no direct visualization of ablation-related changes. In this study, we acutely identify and target gaps using a real-time magnetic resonance imaging (RT-MRI) system, leading to a complete and transmural ablation in the atrium.

Methods and Results - A swine model was used for these studies (n=12). Ablation lesions with a gap were created in the atrium using fluoroscopy and an electroanatomic system in the first group (n=5). The animal was then moved to a 3-tesla MRI system where high-resolution late gadolinium enhancement MRI was used to identify the gap. Using an RT-MRI catheter navigation and visualization system, the gap area was ablated in the MR scanner. In a second group (n=7), ablation lesions with varying gaps in between were created under RT-MRI guidance, and gap lengths determined using late gadolinium enhancement MR images were correlated with gap length measured from gross pathology. Gaps up to 1.0 mm were identified using gross pathology, and gaps up to 1.4 mm were identified using late gadolinium enhancement MRI. Using an RT-MRI system with active catheter navigation gaps can be targeted acutely, leading to lesion sets with no gaps. The correlation coefficient (R2) between the gap length was identified using MRI, and the gross pathology was 0.95.

Conclusions - RT-MRI system can be used to identify and acutely target gaps in atrial ablation lesion sets. Acute targeting of gaps in ablation lesion sets can potentially lead to significant improvement in clinical outcomes.

We propose to aid the interactive visualization of time-varying spatial datasets by simplifying node position data over the entire simulation as opposed to over individual states. Our approach is based on two observations. The first observation is that the trajectory of some nodes can be approximated well without recording the position of the node for every state. The second observation is that there are groups of nodes whose motion from one state to the next can be approximated well with a single transformation. We present dataset simplification techniques that take advantage of this node data redundancy. Our techniques are general, supporting many types of simulations, they achieve good compression factors, and they allow rigorous control of the maximum node position approximation error. We demonstrate our approach in the context of finite element analysis data, of liquid flow simulation data, and of fusion simulation data.

Conventional shadow mapping relies on uniform sampling for producing hard shadow in an efficient manner. This approach trades image quality in favor of efficiency. A number of approaches improve upon shadow mapping by combining multiple shadow maps or using complex data structures to produce shadow maps with multiple resolutions. By sacrificing some performance, these adaptive methods produce shadows that closely match ground truth.

This paper introduces Rectilinear Texture Warping (RTW) for efficiently generating adaptive shadow maps. RTW images combine the advantages of conventional shadow mapping - a single shadow map, quick construction, and constant time pixel shadow tests, with the primary advantage of adaptive techniques - shadow map resolutions which more closely match those requested by output images. RTW images consist of a conventional texture paired with two 1-D warping maps that form a rectilinear grid defining the variation in sampling rate. The quality of shadows produced with RTW shadow maps of standard resolutions, i.e. 2,048×2,048 texture for 1080p output images, approaches that of raytraced results while low overhead permits rendering at hundreds of frames per second.

Keywords: Rendering, Shadow Algorithms, Adaptive Sampling

We propose an automatic approach for segmenting the left atrium from MRI images. In particular, the thoracic aorta is detected and used as a salient feature to find a seed region that lies inside the left atrium. A hybrid energy that combines robust statistics and localized region intensity information is employed to evolve active contours from the seed region to capture the whole left atrium. The experimental results demonstrate the accuracy and robustness of our approach.

Atrial fibrillation (AF) is the most common sustained arrhythmia encountered in adult cardiology.1,2 Several studies have demonstrated that AF is associated with electrical, contractile, and structural remodeling (SRM) in the left atrium (LA) that contributes to the persistence and sustainability of the arrhythmia.3-7 It has also been shown that the end result of this remodeling process is loss of atrial myocytes and increased collagen content and hence fibrosis of the LA wall.5 Delayed enhancement MRI (DE-MRI) using gadolinium contrast has been demonstrated to localize and quantify the degree of SRM or fibrosis associated with AF in the LA.8

DE-MRI has also been shown to be useful in localizing and quantifying scar formation in the LA following radiofrequency ablation (RFA).9,10 The pulmonary vein (PV) antral region can be visualized to assess circumferential PV scarring that results from RFA lesions/ablation. In addition, the amount of scarring to the LA after catheter ablation can be quantified as a proportion of the total left atrial volume.

Rhythm control of AF using catheter ablation has yielded varying results in different patient populations.11 Identifying the ideal candidate for catheter ablation remains a significant challenge. In addition, a number of different approaches to catheter ablation have been reported and most experts agree that 1 ablation strategy does not fit allAF patients.11-15 Therefore, selecting the proper strategy for a particular patient is also an important determinant of procedure success.

We used DE-MRI to quantify both the degree of SRM/fibrosis pre-ablation and scar formation post ablation. Our aim was to identify predictors of successful ablation in a group of patients stratified according to pre-ablation fibrosis. This would help select the most appropriate ablation strategy for the individual AF ablation candidate.

Most visualizations produced in academia or industry have a specific niche audience that is well versed in either the often complicated visualization methods or the scientific domain of the data. Sometimes it is useful to produce visualizations that can communicate results to a broad audience that will not have the domain specific knowledge often needed to understand the results. In this work, we present our experiences in disseminating the results of two studies to national audience. The resulting visualizations and press releases allowed the studies’ researchers to educate a national, if not global, audience.

Computational modeling in electrocardiography often requires the examination of cardiac forward and inverse problems in order to non-invasively analyze physiological events that are otherwise inaccessible or unethical to explore. The study of these models can be performed in the open-source SCIRun problem solving environment developed at the Center for Integrative Biomedical Computing (CIBC). A new toolkit within SCIRun provides researchers with essential frameworks for constructing and manipulating electrocardiographic forward and inverse models in a highly efficient and interactive way. The toolkit contains sample networks, tutorials and documentation which direct users through SCIRun-specific approaches in the assembly and execution of these specific problems.

We present a system for visualizing memory reference traces, the records of the memory transactions performed by a program at runtime. The visualization consists of a structured layout representing the levels of a cache and a set of data glyphs representing the pieces of data in memory being operated on during application runtime. The data glyphs move in response to events generated by a cache simulator, indicating their changing residency in the various levels of the memory hierarchy. Within the levels, the glyphs arrange themselves into higher-order shapes representing the structure of the cache levels, including the composition of their associative cache sets and eviction ordering. We make careful use of different visual channels, including structure, motion, color, and size, to convey salient events as they occur. Our abstract visualization provides a high-level, global view of memory behavior, while giving insight about important events that may help students or software engineers to better understand their software’s performance and behavior.

An ensemble of biological shapes can be represented and analyzed with a dense set of point correspondences. In previous work, optimal point placement was determined by optimizing an information theoretic criterion that depends on relative spatial locations on different shapes combined with pairwise Euclidean distances between nearby points on the same shape. These choices have prevented such methods from effectively characterizing shapes with complex geometry such as thin or highly curved features. This paper extends previous methods for automatic shape correspondence by taking into account the underlying geometry of individual shapes. This is done by replacing the Euclidean distance for intrashape pairwise particle interactions by the geodesic distance. A novel set of numerical techniques for fast distance computations on curved surfaces is used to extract these distances. In addition, we introduce an intershape penalty term that incorporates surface normal information to achieve better particle correspondences near sharp features. Finally, we demonstrate this new method on synthetic and biological datasets.

Keywords: namic

Identification of electrical activation or depolarization times on sparsely-sampled complex heart surfaces is of importance to clinicians and researchers in cardiac electrophysiology. We introduce a spatiotemporal approach for activation time estimation which combines prior results using spatial and temporal methods with our own progress on gradient estimation on triangulated surfaces. Results of the method applied to simulated and canine heart data suggest that improvements are possible using this novel combined approach.

In inverse electrocardiography (ECG), the problem of finding activation times on the heart noninvasively from body surface potentials is typically formulated as a nonlinear least squares optimization problem. Current solutions rely on iterative algorithms which are sensitive to the presence of local minima. As a result, improved initialization approaches for this problem have been of considerable interest. However, in experiments conducted on a subject with Wolff-Parkinson-White syndrome, we have observed that there may be a mismatch between favorable solutions of the optimization problem and solutions with the desired physiological characteristics. In this work, we use a method based on a convex optimization framework to explore the solution space and analyze whether the optimization criteria target their intended objective.

In a number of medical imaging modalities, including measurements or estimates of electrical activity on cortical or cardiac surfaces, it is often useful to estimate spatial derivatives of data on curved anatomical surfaces represented by triangulated meshes. Assuming the triangle vertices are points on a smooth manifold, we derive a method for estimating gradients and Hessians on locally 2D surfaces embedded in 3D directly in the global coordinate system. Accuracy of the method is validated through simulations on both smooth and corrugated surfaces.

This paper presents an efficient, fine-grained parallel algorithm for solving the Eikonal equation on triangular meshes. The Eikonal equation, and the broader class of Hamilton–Jacobi equations to which it belongs, have a wide range of applications from geometric optics and seismology to biological modeling and analysis of geometry and images. The ability to solve such equations accurately and efficiently provides new capabilities for exploring and visualizing parameter spaces and for solving inverse problems that rely on such equations in the forward model. Efficient solvers on state-of-the-art, parallel architectures require new algorithms that are not, in many cases, optimal, but are better suited to synchronous updates of the solution. In previous work [W. K. Jeong and R. T. Whitaker, SIAM J. Sci. Comput., 30 (2008), pp. 2512–2534], the authors proposed the fast iterative method (FIM) to efficiently solve the Eikonal equation on regular grids. In this paper we extend the fast iterative method to solve Eikonal equations efficiently on triangulated domains on the CPU and on parallel architectures, including graphics processors. We propose a new local update scheme that provides solutions of first-order accuracy for both architectures. We also propose a novel triangle-based update scheme and its corresponding data structure for efficient irregular data mapping to parallel single-instruction multiple-data (SIMD) processors. We provide detailed descriptions of the implementations on a single CPU, a multicore CPU with shared memory, and SIMD architectures with comparative results against state-of-the-art Eikonal solvers.

Although the survival rates of warfighters in recent conflicts are among the highest in military history, those who have sustained proximal limb amputations may present additional rehabilitation challenges. In some of these cases, traditional prosthetic limbs may not provide adequate function for service members returning to an active lifestyle. Osseointegration has emerged as an acknowledged treatment for those with limited residual limb length and those with skin issues associated with a socket together. Using this technology, direct skeletal attachment occurs between a transcutaneous osseointegrated implant (TOI) and the host bone, thereby eliminating the need for a socket. While reports from the first 100 patients with a TOI have been promising, some rehabilitation regimens require 12-18 months of restricted weight bearing to prevent overloading at the bone-implant interface. Electrically induced osseointegration has been proposed as an option for expediting periprosthetic fixation and preliminary studies have demonstrated the feasibility of adapting the TOI into a functional cathode. To assure safe and effective electric fields that are conducive for osseoinduction and osseointegration, we have developed multiscale modeling approaches to simulate the expected electric metrics at the bone--implant interface. We have used computed tomography scans and volume segmentation tools to create anatomically accurate models that clearly distinguish tissue parameters and serve as the basis for finite element analysis. This translational computational biological process has supported biomedical electrode design, implant placement, and experiments to date have demonstrated the clinical feasibility of electrically induced osseointegration.

Fundamental to high angular resolution diffusion imaging (HARDI), is the estimation of a positive-semidefinite orientation distribution function (ODF) and extracting the diffusion properties (e.g., fiber directions). In this work we show that these two goals can be achieved efficiently by using homogeneous polynomials to represent the ODF in the spherical deconvolution approach, as was proposed in the Cartesian Tensor-ODF (CT-ODF) formulation. Based on this formulation we first suggest an estimation method for positive-semidefinite ODF by solving a linear programming problem that does not require special parametrization of the ODF. We also propose a rank-k tensor decomposition, known as CP decomposition, to extract the fibers information from the estimated ODF. We show that this decomposition is superior to the fiber direction estimation via ODF maxima detection as it enables one to reach the full fiber separation resolution of the estimation technique. We assess the accuracy of this new framework by applying it to synthetic and experimentally obtained HARDI data.

For radiotherapy planning, contouring of target volume and healthy structures at risk in CT volumes is essential. To automate this process, one of the available segmentation techniques can be used for many thoracic organs except the esophagus, which is very hard to segment due to low contrast. In this work we propose to initialize our previously introduced model based 3D level set esophagus segmentation method with a principal curve tracing (PCT) algorithm, which we adapted to solve the esophagus centerline detection problem. To address challenges due to low intensity contrast, we enhanced the PCT algorithm by learning spatial and intensity priors from a small set of annotated CT volumes. To locate the esophageal wall, the model based 3D level set algorithm including a shape model that represents the variance of esophagus wall around the estimated centerline is utilized. Our results show improvement in esophagus segmentation when initialized by PCT compared to our previous work, where an ad hoc centerline initialization was performed. Unlike previous approaches, this work does not need a very large set of annotated training images and has similar performance.

The examination of the dermis/epidermis junction (DEJ) is clinically important for skin cancer diagnosis. Reflectance confocal microscopy (RCM) is an emerging tool for detection of skin cancers in vivo. However, visual localization of the DEJ in RCM images, with high accuracy and repeatability, is challenging, especially in fair skin, due to low contrast, heterogeneous structure and high inter- and intra-subject variability. We recently proposed a semi-automated algorithm to localize the DEJ in z-stacks of RCM images of fair skin, based on feature segmentation and classification. Here we extend the algorithm to dark skin. The extended algorithm first decides the skin type and then applies the appropriate DEJ localization method. In dark skin, strong backscatter from the pigment melanin causes the basal cells above the DEJ to appear with high contrast. To locate those high contrast regions, the algorithm operates on small tiles (regions) and finds the peaks of the smoothed average intensity depth profile of each tile. However, for some tiles, due to heterogeneity, multiple peaks in the depth profile exist and the strongest peak might not be the basal layer peak. To select the correct peak, basal cells are represented with a vector of texture features. The peak with most similar features to this feature vector is selected. The results show that the algorithm detected the skin types correctly for all 17 stacks tested (8 fair, 9 dark). The DEJ detection algorithm achieved an average distance from the ground truth DEJ surface of around 4.7μm for dark skin and around 7-14μm for fair skin.

Keywords: confocal reflectance microscopy, image analysis, skin, classification