D. Ayyagari, N. Ramesh, D. Yatsenko, T. Tasdizen, C. Atria. Image reconstruction using priors from deep learning, In Medical Imaging 2018: Image Processing, SPIE, March, 2018.
Tomosynthesis, i.e. reconstruction of 3D volumes using projections from a limited perspective is a classical inverse, ill-posed or under constrained problem. Data insufficiency leads to reconstruction artifacts that vary in severity depending on the particular problem, the reconstruction method and also on the object being imaged. Machine learning has been used successfully in tomographic problems where data is insufficient, but the challenge with machine learning is that it introduces bias from the learning dataset. A novel framework to improve the quality of the tomosynthesis reconstruction that limits the learning dataset bias by maintaining consistency with the observed data is proposed. Convolutional Neural Networks (CNN) are embedded as regularizers in the reconstruction process to introduce the expected features and characterstics of the likely imaged object. The minimization of the objective function keeps the solution consistent with the observations and limits the bias introduced by the machine learning regularizers, improving the quality of the reconstruction. The proposed method has been developed and studied in the specific problem of Cone Beam Tomosynthesis Flouroscopy (CBT-fluoroscopy)1 but it is a general framework that can be applied to any image reconstruction problem that is limited by data insufficiency.
Nonlinear stability and time step selection for the MPM method, In Computational Particle Mechanics, Jan, 2018.
The Material Point Method (MPM) has been developed from the Particle in Cell (PIC) method over the last 25 years and has proved its worth in solving many challenging problems involving large deformations. Nevertheless there are many open questions regarding the theoretical properties of MPM. For example in while Fourier methods, as applied to PIC may provide useful insight, the non-linear nature of MPM makes it necessary to use a full non-linear stability analysis to determine a stable time step for MPM. In order to begin to address this the stability analysis of Spigler and Vianello is adapted to MPM and used to derive a stable time step bound for a model problem. This bound is contrasted against traditional Speed of sound and CFL bounds and shown to be a realistic stability bound for a model problem.
H. Bhatia, A.G. Gyulassy, V. Lordi, J.E. Pask, V. Pascucci, P.T. Bremer.
TopoMS: Comprehensive topological exploration for molecular and condensed‐matter systems, In Journal of Computational Chemistry, Vol. 39, No. 16, Wiley, pp. 936--952. March, 2018.
We introduce TopoMS, a computational tool enabling detailed topological analysis of molecular and condensed‐matter systems, including the computation of atomic volumes and charges through the quantum theory of atoms in molecules, as well as the complete molecular graph. With roots in techniques from computational topology, and using a shared‐memory parallel approach, TopoMS provides scalable, numerically robust, and topologically consistent analysis. TopoMS can be used as a command‐line tool or with a GUI (graphical user interface), where the latter also enables an interactive exploration of the molecular graph. This paper presents algorithmic details of TopoMS and compares it with state‐of‐the‐art tools: Bader charge analysis v1.0 (Arnaldsson et al., 01/11/17) and molecular graph extraction using Critic2 (Otero‐de‐la‐Roza et al., Comput. Phys. Commun. 2014, 185, 1007). TopoMS not only combines the functionality of these individual codes but also demonstrates up to 4× performance gain on a standard laptop, faster convergence to fine‐grid solution, robustness against lattice bias, and topological consistency. TopoMS is released publicly under BSD License. © 2018 Wiley Periodicals, Inc.
We present the development of an open-source software called OpenSpace that bridges the gap between scientific discoveries and public dissemination and thus paves the way for the next generation of science communication and data exploration. We describe how the platform enables interactive presentations of dynamic and time-varying processes by domain experts to the general public. The concepts are demonstrated through four cases: Image acquisitions of the New Horizons and Rosetta spacecraft, the dissemination of space weather phenomena, and the display of high-resolution planetary images. Each case has been presented at public events with great success. These cases highlight the details of data acquisition, rather than presenting the final results, showing the audience the value of supporting the efforts of the scientific discovery.
The biophysical basis for electrocardiographic evaluation of myocardial ischemia stems from the notion that ischemic tissues develop, with relative uniformity, along the endocardial aspects of the heart. These injured regions of subendocardial tissue give rise to intramural currents that lead to ST segment deflections within electrocardiogram (ECG) recordings. The concept of subendocardial ischemic regions is often used in clinical practice, providing a simple and intuitive description of ischemic injury; however, such a model grossly oversimplifies the presentation of ischemic disease—inadvertently leading to errors in ECG-based diagnoses. Furthermore, recent experimental studies have brought into question the subendocardial ischemia paradigm suggesting instead a more distributed pattern of tissue injury. These findings come from experiments and so have both the impact and the limitations of measurements from living organisms. Computer models have often been employed to overcome the constraints of experimental approaches and have a robust history in cardiac simulation. To this end, we have developed a computational simulation framework aimed at elucidating the effects of ischemia on measurable cardiac potentials. To validate our framework, we simulated, visualized, and analyzed 226 experimentally derived acute myocardial ischemic events. Simulation outcomes agreed both qualitatively (feature comparison) and quantitatively (correlation, average error, and significance) with experimentally obtained epicardial measurements, particularly under conditions of elevated ischemic stress. Our simulation framework introduces a novel approach to incorporating subject-specific, geometric models and experimental results that are highly resolved in space and time into computational models. We propose this framework as a means to advance the understanding of the underlying mechanisms of ischemic disease while simultaneously putting in place the computational infrastructure necessary to study and improve ischemia models aimed at reducing diagnostic errors in the clinic.
Computational models of myocardial ischemia often use oversimplified ischemic source representations to simulate epicardial potentials. The purpose of this study was to explore the influence of biophysically justified, subject-specific ischemic zone representations on epicardial potentials.
We developed and implemented an image-based simulation pipeline, using intramural recordings from a canine experimental model to define subject-specific ischemic regions within the heart. Static epicardial potential distributions, reflective of ST segment deviations, were simulated and validated against measured epicardial recordings.
Simulated epicardial potential distributions showed strong statistical correlation and visual agreement with measured epicardial potentials. Additionally, we identified and described in what way border zone parameters influence epicardial potential distributions during the ST segment.
From image-based simulations of myocardial ischemia, we generated subject-specific ischemic sources that accurately replicated epicardial potential distributions. Such models are essential in understanding the underlying mechanisms of the bioelectric fields that arise during ischemia and are the basis for more sophisticated simulations of body surface ECGs.
Background: Noninvasive localization of premature ventricular complexes (PVCs) to guide ablation therapy is one of the emerging applications of electrocardiographic imaging (ECGI). Because of its increasing clinical use, it is essential to compare the many implementations of ECGI that exist to understand the specific characteristics of each approach.
Objective: Our consortium is a community of researchers aiming to collaborate in the field of ECGI, and to objectively compare and improve methods. Here, we will compare methods to localize the origin of PVCs with ECGI.
Methods: Our consortium hosts a repository of ECGI data on its website. For the current study, participants analysed simulated electrocardiograms from premature beats, freely available on that website. These PVCs were simulated to originate from eight ventricular locations and the resulting body-surface potentials were computed. These body-surface electrocardiograms (and the torso-heart geometry) were then provided to the study participants to apply their ECGI algorithms to determine the origin of the PVCs. Participants could choose freely among four different source models, i.e., representations of the bioelectric fields reconstructed from ECGI: 1) epicardial potentials (POTepi), 2) epicardial & endocardial potentials (POTepi&endo), 3) transmembrane potentials on the endocardium and epicardium (TMPepi&endo) and 4) transmembrame potentials throughout the myocardium (TMPmyo). Participants were free to employ any software implementation of ECGI and were blinded to the ground truth data.
Results: Four research groups submitted 11 entries for this study. The figure shows the localization error between the known and reconstructed origin of each PVC for each submission, categorized per source model. Each colour represents one research group and some groups submitted results using different approaches. These results demonstrate that the variation of accuracy was larger among research groups than among the source models. Most submissions achieved an error below 2 cm, but none performed with a consistent sub-centimetre accuracy.
Conclusion: This study demonstrates a successful community-based approach to study different ECGI methods for PVC localization. The goal was not to rank research groups but to compare both source models and numerical implementations. PVC localization with these methods was not as dependent on the source representation as it was on the implementation of ECGI. Consequently, ECGI validation should not be performed on generic methods, but should be specifically performed for each lab's implementation. The novelty of this study is that it achieves this in the first open, international comparison of approaches using a common set of gold standards. Continued collaborative validation is essential to understand the effect of implementation differences, in order to reach significant improvements and arrive at clinically-relevant sub-centimetre accuracy of PVC localization.
M. Cluitmans, D. H. Brooks, R. MacLeod, O. Dössel, M. S. Guillem, P. M. van Dam, J. Svehlikova, B. He, J. Sapp, L. Wang, L. Bear.
Validation and Opportunities of Electrocardiographic Imaging: From Technical Achievements to Clinical Applications, In Frontiers in Physiology, Vol. 9, Frontiers Media SA, pp. 1305. 2018.
Electrocardiographic imaging (ECGI) reconstructs the electrical activity of the heart from a dense array of body-surface electrocardiograms and a patient-specific heart-torso geometry. Depending on how it is formulated, ECGI allows the reconstruction of the activation and recovery sequence of the heart, the origin of premature beats or tachycardia, the anchors/hotspots of re-entrant arrhythmias and other electrophysiological quantities of interest. Importantly, these quantities are directly and noninvasively reconstructed in a digitized model of the patient’s three-dimensional heart, which has led to clinical interest in ECGI’s ability to personalize diagnosis and guide therapy.
Despite considerable development over the last decades, validation of ECGI is challenging. Firstly, results depend considerably on implementation choices, which are necessary to deal with ECGI’s ill-posed character. Secondly, it is challenging to obtain (invasive) ground truth data of high quality. In this review, we discuss the current status of ECGI validation as well as the major challenges remaining for complete adoption of ECGI in clinical practice.
Specifically, showing clinical benefit is essential for the adoption of ECGI. Such benefit may lie in patient outcome improvement, workflow improvement, or cost reduction. Future studies should focus on these aspects to achieve broad adoption of ECGI, but only after the technical challenges have been solved for that specific application/pathology. We propose ‘best’ practices for technical validation and highlight collaborative efforts recently organized in this field. Continued interaction between engineers, basic scientists and physicians remains essential to find a hybrid between technical achievements, pathological mechanisms insights, and clinical benefit, to evolve this powerful technique towards a useful role in clinical practice.
A. Erdemir, P.J. Hunter, G.A. Holzapfel, L.M. Loew, J. Middleton, C.R. Jacobs, P. Nithiarasu, R. Löhner, G. Wei, B.A. Winkelstein, V.H. Barocas, F. Guilak, J.P. Ku, J.L. Hicks, S.L. Delp, M.S. Sacks, J.A. Weiss, G.A. Ateshian, S.A. Maas, A.D. McCulloch, G.C.Y. Peng.
Perspectives on Sharing Models and Related Resources in Computational Biomechanics Research, In Journal of Biomechanical Engineering, Vol. 140, No. 2, ASME International, pp. 024701. Jan, 2018.
The role of computational modeling for biomechanics research and related clinical care will be increasingly prominent. The biomechanics community has been developing computational models routinely for exploration of the mechanics and mechanobiology of diverse biological structures. As a result, a large array of models, data, and discipline-specific simulation software has emerged to support endeavors in computational biomechanics. Sharing computational models and related data and simulation software has first become a utilitarian interest, and now, it is a necessity. Exchange of models, in support of knowledge exchange provided by scholarly publishing, has important implications. Specifically, model sharing can facilitate assessment of reproducibility in computational biomechanics and can provide an opportunity for repurposing and reuse, and a venue for medical training. The community's desire to investigate biological and biomechanical phenomena crossing multiple systems, scales, and physical domains, also motivates sharing of modeling resources as blending of models developed by domain experts will be a required step for comprehensive simulation studies as well as the enhancement of their rigor and reproducibility. The goal of this paper is to understand current perspectives in the biomechanics community for the sharing of computational models and related resources. Opinions on opportunities, challenges, and pathways to model sharing, particularly as part of the scholarly publishing workflow, were sought. A group of journal editors and a handful of investigators active in computational biomechanics were approached to collect short opinion pieces as a part of a larger effort of the IEEE EMBS Computational Biology and the Physiome Technical Committee to address model reproducibility through publications. A synthesis of these opinion pieces indicates that the community recognizes the necessity and usefulness of model sharing. There is a strong will to facilitate model sharing, and there are corresponding initiatives by the scientific journals. Outside the publishing enterprise, infrastructure to facilitate model sharing in biomechanics exists, and simulation software developers are interested in accommodating the community's needs for sharing of modeling resources. Encouragement for the use of standardized markups, concerns related to quality assurance, acknowledgement of increased burden, and importance of stewardship of resources are noted. In the short-term, it is advisable that the community builds upon recent strategies and experiments with new pathways for continued demonstration of model sharing, its promotion, and its utility. Nonetheless, the need for a long-term strategy to unify approaches in sharing computational models and related resources is acknowledged. Development of a sustainable platform supported by a culture of open model sharing will likely evolve through continued and inclusive discussions bringing all stakeholders at the table, e.g., by possibly establishing a consortium.
M.D. Foote, B. Zimmerman, A. Sawant, S. Joshi. Real-Time Patient-Specific Lung Radiotherapy Targeting using Deep Learning, In 1st Conference on Medical Imaging with Deep Learning (MIDL 2018), Amsterdam, The Netherlands, 2018.
Radiation therapy has presented a need for dynamic tracking of a target tumor volume. Fiducial markers such as implanted gold seeds have been used to gate radiation delivery but the markers are invasive and gating significantly increases treatment time. Pretreatment acquisition of a 4DCT allows for the development of accurate motion estimation for treatment planning. A deep convolutional neural network and subspace motion tracking is used to recover anatomical positions from a single radiograph projection in real-time. We approximate the nonlinear inverse of a diffeomorphic transformation composed with radiographic projection as a deep network that produces subspace coordinates to define the patient-specific deformation of the lungs from a baseline anatomic position. The geometric accuracy of the subspace projections on real patient data is similar to accuracy attained by original image registration between individual respiratory-phase image volumes.
Direct stimulation of the cortical surface is used clinically for cortical mapping and modulation of local activity. Future applications of cortical modulation and brain-computer interfaces may also use cortical stimulation methods. One common method to deliver current is through electrocorticography (ECoG) stimulation in which a dense array of electrodes are placed subdurally or epidurally to stimulate the cortex. However, proximity to cortical tissue limits the amount of current that can be delivered safely. It may be desirable to deliver higher current to a specific local region of interest (ROI) while limiting current to other local areas more stringently than is guaranteed by global safety limits. Two commonly used global safety constraints bound the total injected current and individual electrode currents. However, these two sets of constraints may not be sufficient to prevent high current density locally (hot-spots). In this work, we propose an efficient approach that prevents current density hot-spots in the entire brain while optimizing ECoG stimulus patterns for targeted stimulation. Specifically, we maximize the current along a particular desired directional field in the ROI while respecting three safety constraints: one on the total injected current, one on individual electrode currents, and the third on the local current density magnitude in the brain. This third set of constraints creates a computational barrier due to the huge number of constraints needed to bound the current density at every point in the entire brain. We overcome this barrier by adopting an efficient two-step approach. In the first step, the proposed method identifies the safe brain region, which cannot contain any hot-spots solely based on the global bounds on total injected current and individual electrode currents. In the second step, the proposed algorithm iteratively adjusts the stimulus pattern to arrive at a solution that exhibits no hot-spots in the remaining brain. We report on simulations on a realistic finite element (FE) head model with five anatomical ROIs and two desired directional fields. We also report on the effect of ROI depth and desired directional field on the focality of the stimulation. Finally, we provide an analysis of optimization runtime as a function of different safety and modeling parameters. Our results suggest that optimized stimulus patterns tend to differ from those used in clinical practice.
L. Guo, A. Narayan, T. Zhou. A gradient enhanced ℓ1-minimization for sparse approximation of polynomial chaos expansions, In Journal of Computational Physics, Vol. 367, Elsevier BV, pp. 49--64. Aug, 2018.
We investigate a gradient enhanced ℓ 1-minimization for constructing sparse polynomial chaos expansions. In addition to function evaluations, measurements of the function gradient is also included to accelerate the identification of expansion coefficients. By designing appropriate preconditioners to the measurement matrix, we show gradient enhanced ℓ 1 minimization leads to stable and accurate coefficient recovery. The framework for designing preconditioners is quite general and it applies to recover of functions whose domain is bounded or unbounded. Comparisons between the gradient enhanced approach and the standard ℓ 1-minimization are also presented and numerical examples suggest that the inclusion of derivative information can guarantee sparse recovery at a reduced computational cost.
L. Guo, A. Narayan, L. Yan, T. Zhou.
Weighted approximate fekete points: sampling for least-squares polynomial approximation, In SIAM Journal on Scientific Computing, Vol. 40, No. 1, SIAM, pp. A366--A387. Jan, 2018.
We propose and analyze a weighted greedy scheme for computing deterministic sample configurations in multidimensional space for performing least-squares polynomial approximations on $L^2$ spaces weighted by a probability density function. Our procedure is a particular weighted version of the approximate Fekete points method, with the weight function chosen as the (inverse) Christoffel function. Our procedure has theoretical advantages: when linear systems with optimal condition number exist, the procedure finds them. In the one-dimensional setting with any density function, our greedy procedure almost always generates optimally conditioned linear systems. Our method also has practical advantages: our procedure is impartial to the compactness of the domain of approximation and uses only pivoted linear algebraic routines. We show through numerous examples that our sampling design outperforms competing randomized and deterministic designs when the domain is both low and high dimensional.
Topological data analysis is an emerging area in exploratory data analysis and data mining. Its main tool, persistent homology, has become a popular technique to study the structure of complex, high-dimensional data. In this paper, we propose a novel method using persistent homology to quantify structural changes in time-varying graphs. Specifically, we transform each instance of the time-varying graph into a metric space, extract topological features using persistent homology, and compare those features over time. We provide a visualization that assists in time-varying graph exploration and helps to identify patterns of behavior within the data. To validate our approach, we conduct several case studies on real-world datasets and show how our method can find cyclic patterns, deviations from those patterns, and one-time events in time-varying graphs. We also examine whether a persistence-based similarity measure satisfies a set of well-established, desirable properties for graph metrics.
M. Hajij, B. Wang, P. Rosen. MOG: Mapper on Graphs for Relationship Preserving Clustering, In CoRR, 2018.
The interconnected nature of graphs often results in difficult to interpret clutter. Typically techniques focus on either decluttering by clustering nodes with similar properties or grouping edges with similar relationship. We propose using mapper, a powerful topological data analysis tool, to summarize the structure of a graph in a way that both clusters data with similar properties and preserves relationships. Typically, mapper operates on a given data by utilizing a scalar function defined on every point in the data and a cover for scalar function codomain. The output of mapper is a graph that summarize the shape of the space. In this paper, we outline how to use this mapper construction on an input graphs, outline three filter functions that capture important structures of the input graph, and provide an interface for interactively modifying the cover. To validate our approach, we conduct several case studies on synthetic and real world data sets and demonstrate how our method can give meaningful summaries for graphs with various complexities
For practical model-based demands, such as design space exploration and uncertainty quantification (UQ), a high-fidelity model that produces accurate outputs often has high computational cost, while a low-fidelity model with less accurate outputs has low computational cost. It is often possible to construct a bi-fidelity model having accuracy comparable with the high-fidelity model and computational cost comparable with the low-fidelity model. This work presents the construction and analysis of a non-intrusive (i.e., sample-based) bi-fidelity model that relies on the low-rank structure of the map between model parameters/uncertain inputs and the solution of interest, if exists. Specifically, we derive a novel, pragmatic estimate for the error committed by this bi-fidelity model. We show that this error bound can be used to determine if a given pair of low- and high-fidelity models will lead to an accurate bi-fidelity approximation. The cost of this error bound is relatively small and depends on the solution rank. The value of this error estimate is demonstrated using two example problems in the context of UQ, involving linear and non-linear partial differential equations.
The automatic inverse design of three-dimensional plasmonic nanoparticles enables scientists and engineers to explore a wide design space and to maximize a device's performance. However, due to the large uncertainty in the nanofabrication process, we may not be able to obtain a deterministic value of the objective, and the objective may vary dramatically with respect to a small variation in uncertain parameters. Therefore, we take into account the uncertainty in simulations and adopt a classical robust design model for a robust design. In addition, we propose an efficient numerical procedure for the robust design to reduce the computational cost of the process caused by the consideration of the uncertainty. Specifically, we use a global sensitivity analysis method to identify the important random variables and consider the non-important ones as deterministic, and consequently reduce the dimension of the stochastic space. In addition, we apply the generalized polynomial chaos expansion method for constructing computationally cheaper surrogate models to approximate and replace the full simulations. This efficient robust design procedure is performed by varying the particles' material among the most commonly used plasmonic materials such as gold, silver, and aluminum, to obtain different robust optimal shapes for the best enhancement of electric fields.
As the finite element method (FEM) and the finite volume method (FVM), both traditional and high-order variants, continue their proliferation into various applied engineering disciplines, it is important that the visualization techniques and corresponding data analysis tools that act on the results produced by these methods faithfully represent the underlying data. To state this in another way: the interpretation of data generated by simulation needs to be consistent with the numerical schemes that underpin the specific solver technology. As the verifiable visualization literature has demonstrated: visual artifacts produced by the introduction of either explicit or implicit data transformations, such as data resampling, can sometimes distort or even obfuscate key scientific features in the data. In this paper, we focus on the handling of elemental continuity, which is often only C0 continuous or piecewise discontinuous, when visualizing primary or derived fields from FEM or FVM simulations. We demonstrate that traditional data handling and visualization of these fields introduce visual errors. In addition, we show how the use of the recently proposed line-SIAC filter provides a way of handling elemental continuity issues in an accuracy-conserving manner with the added benefit of casting the data in a smooth context even if the representation is element discontinuous.
Targeting Neuronal Fiber Tracts for Deep Brain Stimulation Therapy Using Interactive, Patient-Specific Models, In Journal of Visualized Experiments, No. 138, MyJove Corporation, Aug, 2018.
Deep brain stimulation (DBS), which involves insertion of an electrode to deliver stimulation to a localized brain region, is an established therapy for movement disorders and is being applied to a growing number of disorders. Computational modeling has been successfully used to predict the clinical effects of DBS; however, there is a need for novel modeling techniques to keep pace with the growing complexity of DBS devices. These models also need to generate predictions quickly and accurately. The goal of this project is to develop an image processing pipeline to incorporate structural magnetic resonance imaging (MRI) and diffusion weighted imaging (DWI) into an interactive, patient specific model to simulate the effects of DBS. A virtual DBS lead can be placed inside of the patient model, along with active contacts and stimulation settings, where changes in lead position or orientation generate a new finite element mesh and solution of the bioelectric field problem in near real-time, a timespan of approximately 10 seconds. This system also enables the simulation of multiple leads in close proximity to allow for current steering by varying anodes and cathodes on different leads. The techniques presented in this paper reduce the burden of generating and using computational models while providing meaningful feedback about the effects of electrode position, electrode design, and stimulation configurations to researchers or clinicians who may not be modeling experts.
Many biomedical image analysis applications require segmentation. Convolutional neural networks (CNN) have become a promising approach to segment biomedical images; however, the accuracy of these methods is highly dependent on the training data. We focus on biomedical image segmentation in the context where there is variation between source and target datasets and ground truth for the target dataset is very limited or non-existent. We use an adversarial based training approach to train CNNs to achieve good accuracy on the target domain. We use the DRIVE and STARE eye vasculture segmentation datasets and show that our approach can significantly improve results where we only use labels of one domain in training and test on the other domain. We also show improvements on membrane detection between MIC-CAI 2016 CREMI challenge and ISBI2013 EM segmentation challenge datasets.