The Visualization Friday Forum

Fall 2007 * Noon-1pm * D106 LSRC

August 31 - Redgraph: Interacting with Networks in Three-Dimensions
Harry Halpin
Visualization Technology Group

Redgraph, a software package developed by the Visualization Technology Group at Duke, lets users explore complex networks through interactive extrusion of nodes and links. The extrusion method lets a graph well-known graph lay-out algorithm display the network in two dimensions in the DIVE, and the so allows the user to "pick" a node or group of nodes a and pull them into three dimensional space. The software is based on the generic RDF (Resource Description Framework), a W3C Web standard for network-based data, and so should be easily usable on wide ranges of data sets.
The talk will also give a general introduction to network theory and the importance of visualization in understanding "real-world" networks, the Semantic Web data format RDF, empirical data on extrusion, and preliminary results from interdisciplinary work with Tim Lenoir in using the network visualizer to chart the history of biotechnology innovaton in North Carolina through patent visualization.

September 7 - Visualization Capabilities of Matlab
Rachael Brady
Visualization Technology Group

see a video of the talk
The ppt slides and matlab .m files are here.

Matlab is an excellent tool for prototyping simulations or doing quick computations. Most people are familiar with the 2D and 3D plot capabilities, and probably the image capabilities of Matlab. This talk will review good uses of those tools, point out some interesting features, and then illustrate the 3D visualization aspects of Matlab.
Note: The content is very similar to the lecture I gave at the CSEM Matlab workshop on April 10, 2007.

September 14 - Medical Imaging Simulation of Mice and Men
Paul Segars
Radiology

see a video of the talk

Computer simulation methods are finding an increasingly important role in medical imaging research for characterizing, evaluating, and optimizing medical imaging systems. Computer simulation involves computer generated phantoms, models of the imaging process, and fast computational methods. Computer phantoms provide a model of the subject's anatomy and physiology. Given a model of the physics of the imaging process, acquired data of a computer phantom can be generated using the computational methods. A major advantage to using computer-generated phantoms in simulation studies is that the exact anatomy and physiological functions of the phantom are known, thus providing a gold standard from which to evaluate and improve medical imaging devices and image processing and reconstruction techniques. Other advantages are that computer phantoms are always willing participants and can be altered easily to model different anatomies and medical situations providing a large population of subjects from which to perform research. A vital aspect of simulation is to have a realistic phantom or model of the subject's anatomy. Without this, the results of the simulation may not be indicative of what would occur in actual patients or animal subjects and would, therefore, have limited practical value. This talk concerns the development and application of two realistic digital phantoms for use in medical imaging research, the 4D NCAT and 4D MOBY phantoms. The 4D NCAT phantom was developed to model the human anatomy and physiology for imaging studies while the 4D MOBY phantom was developed to model the mouse anatomy and physiology for small animal studies. Both phantoms are based on state-of-the-art computer graphics techniques and when combined with accurate models for the imaging process are capable of providing a wealth of realistic imaging data from subjects of various anatomies. With this ability, the phantoms have enormous potential to study the effects of anatomical, physiological, physical, and instrumentational factors on medical and small animal imaging and to research new instrumentation, image acquisition strategies, image processing and reconstruction methods and image visualization and interpretation techniques.

September 21 - Predictive models that infer geometry and dependence
Sayan Mukherjee
IGSP

see a video of the talk

A framework for predictive modeling that simultaneously infers the geometry and statistical dependencies of the input space is developed. The central idea in this approach is the estimation of gradients of regression or classification functions. Given the gradient estimate one can perform simultaneous dimension reduction and regression, inference of graphical models, and nonlinear dimension reduction. An extension of this to the multi-task setting will be developed. The main applications will be in genomics: modeling tumor progression. However, illustrations such as classifying digits will also be described. Lots of pictures will be shown.
Joint work with: Qiang Wu, Justin Guinney, Mauro Maggioni, and Elena Edelman

September 28 - Building Very Small Mobile Micro-Robots
Bruce Donald
Computer Science

I will survey the challenges of building and controlling mobile micro-robots based on microelectromechanical systems (MEMS). I will present my laboratory's work on steerable, untethered micro-robots, each with dimensions up to 100 times smaller than prior systems. I will finally discuss possible applications of such MEMS and micro-robotics, and speculate on the future outlook for their science and engineering.

October 5 - Fall Break

October 12 - Visualization in the Sciences
Russell Taylor II
Computer Science, UNC

see a video of the talk

The "Visualization in the Sciences" course taught at UNC since 2001 describes the current state of the art in computer-based visualization as applied in the natural and medical sciences. It includes teaching the available visualization techniques and their characteristics (including enough description of perception to enable understanding why some techniques work and others do not). It is aimed both at computer science students (who would implement such visualizations) and at natural science students (who would use them). It includes teaching how to use an available visualization toolkit to enable rapid development of visualizations. A final project teams a computer science student as a tool smith with a natural science student as a client liaison to develop a visualization tuned for a particular problem.
The lecture will include a fast-forward preview of the topics covered in the course (1 slide per lecture), a description of several of the final projects resulting from the course, and Q&A on how the course has gone.

October 19 - Scientific Graphics with Adobe Illustrator: Don't be scared -- really, it's fun!
Eric Monson
Visualization Technology Group

see a video of the talk

Do you ever have to draw simple diagrams for proposals or publications? Have you ever struggled to get a graphing program to make your plots look just right? Adobe Illustrator is a vector graphics editing program which can be very useful for scientists and engineers in just these situations, but many people avoid it because of the seemingly steep learning curve. In this brief tutorial, I will run through some simple examples of Illustrator's capabilities, showing you how to start using its tools to modify your graphs and create scientific figures.

October 26 - Social Systems, Spatialization & Cognitive Masking/Demasking
Pedro Lasch
Art, Art History, and Visual Studies

see a video of the talk

This workshop will focus on my 'Naturalizations Series,' an ongoing interdisciplinary experiment based on the use of a set of mirror-masks designed in 2002. The initial perception created by these masks is one of spatial and psychological confusion. Subjects are reversed if only one person is wearing the mask. If several people wear them, their faces disappear and transform into an endless set of reflections of other mirrors, other faces, environments, and objects. Landscape and subject are one and many. Subjects are inseparable from each other, their bodies dismembered by rectangular planes departing and arriving through reflected gazes. Light breaks and travels on these masks with unpredictable speed and variety. Space and movement become counter-intuitive. We will together enter this perceptual, psychological, and social labyrinth to reconsider the value of such low-tech cognitive tools -- formerly called philosophical toys, as useful complements to the expensive technology and encoded language most prevalent in (scientific) visualization. Network theory, systems theory, and current ideas around spatialization in art and science will also be discussed before and after our experience with the masks.

November 2 - Visualizing root networks, inside out
Philip Benfey
Biology

Our goal is to characterize the networks of genes that regulate root development. We have developed a technology platform, the RootArray, to visualize the activity of genes at high spatiotemporal resolution in response to external stimuli. We have also developed means of imaging the growth of roots in real time and mathematically describing the resulting physical networks. We are applying insights gained from these technologies toward improving agriculture, biofuels and carbon sequestration.

November 9 - Spatiotemporal characterization of fluctuating oxygenation in tumors
Isabel Cardenas-Navia
Radiation Oncology (Mark Dewhirst's Lab)

see a video of the talk

Tumor hypoxia is one of the most studied physiological phenomena in cancer research and has been shown to be prognostically significant in many clinical studies, independent of treatment type. Although many investigators have measured the presence of so-called 'intermittent' or 'acute' hypoxia in preclinical and clinical tumors, instabilities in tumor oxygenation were presumed to be temporally rare, spatially isolated events attributed to transient episodes of vascular stasis. In this talk we show data directly measuring temporal fluctuations in vascular pO2 in three rat tumors, showing that O2 delivery to tumors is constantly fluctuating. We will discuss visualization problems in our initial approach of the data, the application of spatial statistics to characterizing the spatial as well as temporal differences between tumor types, and the application of spatial statistics to garner insight into the biological mechanisms behind fluctuations in tumor oxygenation.

November 16 - Quantifying the Benefits of Immersive Visualization
Doug Bowman
Computer Science, Virginia Tech

see a video of the talk

Immersive virtual reality (VR), typified by technologies such as stereoscopic projected displays and 3D tracking systems, has been available for many years, but there are only a few examples of real-world applications of this technology, most of which are focused on providing a realistic "experience" to the user so that he experiences a sense of "presence." We believe that immersive technologies can provide other benefits, such as increased spatial understanding or reduced information clutter. We are running a series of empirical studies to demonstrate these benefits and to determine "how much immersion is enough." Our results so far indicate that higher levels of immersion do lead to improved spatial understanding in complex visualizations, and that greater immersion can produce more efficient interaction in difficult task situations.

November 23 - Thanksgiving Break

November 30 - Medical Simulation for Training and Beyond
Xunlei Wu
RENCI

see a video of the talk

Medical simulation is an inter-disciplinary science using computer-based and/or mannequin-based learning method in which users practice tasks and procedures in lifelike circumstances to improve skills without putting patients at risk. I will describe three still active Medical Simulation research projects which I have been involved in before joining RENCI. These projects are: 1. EVE, an interactive endovascular procedure simulator; 2. COMETS, an Autonomous Casualty Simulator; 3. SOFA, a simulation open framework architecture. The purpose of my talk is to present the overall picture of each project, their capabilities, and potential applications and, more importantly, foster future research collaborations.

December 7 - Serious Challenges and Serious Issues Developing Serious Games
Randy Brown
CTO, Virtual Heroes

see a video of the talk
The ppt slides are here.

Serious Games are game technologies used for purposes other than pure entertainment, such as training and education. There are many challenges to overcome in applying current real-time, interactive, immersive game technologies to issues relating to "real-world" problems. In this presentation I will discuss some of the technical hurdles, development issues, and challenges of mapping various types of learning to specific methodologies of immersive 3D gameplay. All in 30 minutes. Seriously!