AWM at ICIAM 2011 Abstracts
Monday, July 18, 2011, 10:00 am – 12:00 pm, Room: 214
AWM Meeting: Women at the Forefront of Applied Mathematics
Models for Cell Polarization and Crawling
Leah Edelstein-Keshet, University of British Columbia, Canada, keshet@math.ubc.ca
Abstract. The motility of many types of living cells involves crawling and requires the assembly and degradation of the internal structural proteins (cytoskeleton) of the cell. Regulating this structure are signaling proteins that first undergo symmetry breaking and polarization in response to localized or graded stimuli received by the cell. Here I report on work in my group on understanding the biological mechanisms and the interesting mathematical and computational issues that arose as a result.
Imaging Biomechanical Parameters for Disease Characterization
Joyce R. McLaughlin, Rensselaer Polytechnic Institute, USA, mclauj@rpi.edu
Abstract. Shear stiffness imaging is an outgrowth of the palpation exam where doctors press against the skin to feel abnormalities in tissue. We discuss the technologies that enable shear stiffness imaging, the viscoelastic models that yield simulated data that best match experimental data, and the mathematical issues that occur when developing imaging algorithms with those models.
How Do Insects Fly and Turn?
Jane Wang, Cornell University, USA, jane.wang@cornell.edu
Abstract. Insects’ aerial acrobatics result from the concerted efforts of their brains, flight muscles, and flapping wings. To understand insect flight, we started from the outer scale, analyzing the unsteady aerodynamics of flapping flight, and are gradually working toward the inner scale, deducing control algorithms. In this approach, the dynamics of flight informs us about the internal control scheme for a specific behavior.
I will first describe the aerodynamic tricks that dragonflies employ to hover and fly efficiently. I will then discuss how fruit flies recover from aerial stumbles, and how they make subtle wing movements to induce sharp turns in tens of wing beats, or 40-80ms. The observed yaw maneuver can be explained by a quantitative mechanical model that connects a single control variable to the body dynamics.
Selected Developments in Non-derivative Optimization
Margaret H. Wright, Courant Institute of Mathematical Sciences, New York University, USA, mhw@cs.nyu.edu
Abstract. For at least 50 years, practitioners in science, engineering, and medicine have happily solved important real-world optimization problems with non-derivative methods, despite the until-recent shortage (in some instances, a complete lack) of rigorous convergence results. Research on both theoretical and computational properties of non-derivative methods has flourished in recent years, including lively disagreements about (i) which methods are “most effective’, (ii) why, and (iii) what “effective’ means in this context. This talk will briefly survey selected significant developments in non-derivative optimization and speculate about the future role of these methods.
Monday July 18, 2011, 10:00 am – 12:00 pm, Room: 214
AWM Meeting – Workshop: Research Talks by Recent Ph.D.s I
Modeling the Association of Epstein-Barr Virus Infection with the Development of Nasopharyngeal Carcinoma
Giao Huynh, Oakland University, USA, huynh@oakland.edu
Abstract. Nasopharyngeal carcinoma (NPC) is a cancer of epithelial cells in the nasopharynx that has distinct patterns of incidence in different populations. Latent Epstein-Barr virus (EBV) infection within epithelial cells in the pharynx, genetic factors, and diet play key roles in the pathogenesis of NPC. We develop a mathematical model to study critical breakdown in regulation of the infection that can induce elevated level of latently infected epithelial cells, and hence an increased risk of NPC.
A New Microscale Model for Electrical Activity of Myocardial Cells Incorporating the Effects of Ephaptic Coupling
Joyce Lin, University of Utah, USA, joyce.lin@utah.edu
Abstract. Hydration and dehydration of myocardial tissue and the resultant effect on extracellular volume have implications on medical pathologies associated with edema, such as heart failure, hypertension, and arrhythmia. This talk will present a new model for the electrical activity in myocardial tissue that is able to capture the irregular, complex cellular geometry and is numerically more efficient than previously studied models. Results and analysis from simulations and characterizations of parameter trends will be presented.
A Mathematical Model of Platelet Deposition and Blood Coagulation under Flow
Karin Leiderman, , kleiderman@ucmerced.edu
Abstract. To explore how blood flow affects the growth of thrombi (blood clots) and how the growing masses, in turn, feed back and affect the blood flow, we have developed the first spatial-temporal mathematical model of blood platelet aggregation and coagulation under flow that includes detailed descriptions of coagulation biochemistry, chemical activation and deposition of blood platelets, as well as the two-way interaction between the fluid dynamics and the growing platelet mass. We present this model and use it to explain what underlies the threshold behavior of the coagulation system’s production of important enzymes and to show how wall shear rate and near-wall enhanced platelet concentrations affect the development of growing thrombi. By accounting for the porous nature of the thrombus, we also demonstrate how advective and diffusive transport to and within the thrombus affects its growth at different stages and spatial locations.
Monday, July 18, 2011, 3:00 – 4:00 pm, Room: 214
AWM Meeting – Workshop: Opportunities Beyond Academia
Helping Others to Solve Their Important Problems
Kristyn Maschhoff, Cray, Inc., USA, kristyn@cray.com
Abstract. Helping or enabling others to do their research probably doesn’t sound like the most glamorous of job descriptions, but fairly accurately describes my position at Cray. In this presentation I will talk about my role at Cray as an applications specialist working in the future architectures group and highlight various aspects of the work which I find the most rewarding; working with current customers to help customize their algorithms in order to run more efficiently on existing systems; and working closely with the Cray system architects to add in features or provide new capabilities to our future systems to enable more productive use of these systems. The reward comes from helping customers to use our machines to solve problems previously not possible.
One Career in Discrete Mathematics and Combinatorial Optimization at Sandia National Laboratories
Cynthia Phillips, Sandia National Laboratories, USA, caphill@sandia.gov
Abstract. I will discuss the research environment at a national laboratory which has some of the best aspects of academia and industry. I will describe some of the varied projects I have worked on over my 20-year (so far) career; the balance between theory and impact; and general tips on how to climb the technical ladder. While many tips will be specific to the labs, others apply to research careers in general.
An Academic Perspective on Non-academic Collaborations
Marsha Berger, Courant Institute of Mathematical Sciences, New York University, USA, berger@cims.nyu.edu
Abstract. I have always enjoyed the more practical side of scientific computing research. In this talk I will describe how this came about, how lucky I was to find the collaborations I have, particularly a very long standing one at NASA Ames, and some of the challenges of trying to do “non-academic” work in an academic environment.
Tuesday, July 19, 2011, 3:00 PM – 5:00 PM, Room: 214
AWM Meeting – Workshop: Research Talks by Recent Ph.D.s II
Gamma-Convergence Approximation to Piecewise Smooth Medical Image Segmentation
Jung-Ha An, California State University, Stanislaus, U.S., jan@csustan.edu;
Mikael Rousson, Apple, Inc., U.S., mikael.rousson@gmail.com;
Chenyang Xu, Siemens Corporation Research, Germany, chenyang.xu@siemens.com
Abstract. Accurate extraction of structures of interest still remains difficult in many medical imaging applications. A variational region based algorithm that is able to deal with spatial perturbations of the image intensity directly is proposed. Image segmentation is obtained by using a gamma-Convergence approximation for a multi-scale piecewise smooth model. The proposed model is implemented efficiently using recursive Gaussian convolutions. Numerical experiments on 2-dimensional human liver MR images show the effectiveness of the suggested model. Do not include references or citations separately at the end of the abstract. Instead, all citations must be in text in the general form [Authorname, Title, etc].
Numerical Optimization for Constrained Image Registration with Application to Biomedical Imaging
Raya Horesh, University of Minnesota, USA, rhoresh@ima.umn.edu;
Eldad Haber, University of British Columbia, Canada, haber@eoas.ubc.ca;
Jan Modersitzki, University of Lübeck, Germany, jan.modersitzki@mic.uni-luebeck.de
Abstract. Image registration is a technique aimed at aligning two or more images of the same scene taken at different times, from different viewpoints and/or by different sensors. It is an essential tool for various applications in medicine, geosciences, and other disciplines. However, obtaining plausible deformations is a complex task as often transformations are required to be locally invertible or even more challenging, are required to bound volume changes within a reasonable range. In this work, solutions to the registration problem were obtained by directly imposition of a volume constraint upon each voxel in a discretized domain. This study focuses on the development of an efficient and robust numerical algorithm and in particular, the application of an augmented Lagrangian method with a multigrid as preconditioner. We demonstrate that this combination yields an almost optimal solver for the problem. The numerical algorithm was tested on 3D SPECT images of a swine heart.
Optimal Management Controls for Maximizing the Recovery of an Endangered Fish Species
Rachael Miller Neilan, Louisiana State University, USA, rmill48@lsu.edu;
Kenneth Rose, University of Louisiana, USA, karose@lsu.edu;
Wim Kimmer, San Francisco State University, USA, kimmerer@sfsu.edu;
William Bennett, University of California, Davis, USA, wabennett@ucdavis.edu;
Karen Edwards, San Francisco State University, USA, kpe@sfsu.edu
Abstract. A large-scale, computationally expensive individual-based model (IBM) was used to simulate the population decline of delta smelt during 1995 to 2005 in the Upper San Francisco Estuary. We approximated the IBM’s output with a spatially-explicit matrix projection model. By applying optimal control theory to the matrix model, we determined cost-effective management actions involving redirected movement and improved habitat (affecting mortality and growth) that would maximize long-term population growth during this period of decline.
Analysis of Fully Coupled NavierStokes/DarcyTransport Problem
Aycil Cesmelioglu, Institute for Mathematics and its Applications, USA, aycil.cesmelioglu@ima.umn.edu
Abstract. We study a multiphysics problem arising from the problem of groundwater contamination through rivers. Specifically, a convectiondiffusion type transport is fully coupled with the NavierStokes/ Darcy flow via velocity field and concentration. We prove the existence of a weak solution and provide numerical results for the special case where the the coupling is one way, i.e., transport is coupled to flow through velocity.