Transcript PowerPoint
• How the computer accesses the data is known but: – What kinds of collaborative scenarios do they do within disciplines and between disciplines? (like human fmri to mouse? Or biology to CS) – Or: BIRN may not necessarily collaborate with each other in science- but perhaps in data deployment. Other institutes that are not part of BIRN may want to collaborate with each other taking advantage of the BIRN databases. • How to make your CS person a researcher who can produce “production” work: Make them pick up a hammer and build a piece of hardware. • Hardware “bug” fixes are harder than software. This is the closest CS people will think like bridge builders. Electronic Visualization Laboratory (EVL) University of Illinois at Chicago • Human Brain Project (NIMH/NIH) – Early 90s- with very similar goals. Storing and relating data from genes to behavior. New program in 200205. • What can be learned from HBP? • http://www.nimh.nih.gov/neuroinformatics/index.cfm • “Understanding brain function requires the integration of information from the level of the gene to the level of behavior. At each of these many and diverse levels there has been an explosion of information, with a concomitant specialization of scientists. The price of this progress and specialization is that it is becoming virtually impossible for any individual researcher to maintain an integrated view of the brain and to relate his or her narrow findings to this whole cloth. Although the amount of information to be integrated far exceeds human limitations, solutions to this problem are available from the advanced technologies of computer and information sciences.” Electronic Visualization Laboratory (EVL) University of Illinois at Chicago • Observations on data – Microscopes & data (Mouse) – large data sets • Electron mic – 2Kx2.5K x 512 layers; 4Kx4Kx2K in 2 years; 12Kx12Kx2K in 5-7 years • Confocal laser mic – Data (Morph, Function) – small but many data sets (timeseries function MRI) – 256^3 ~ 1G – Differences in data size will impact the way the tools should be designed to work with it. – Portal-based tools for large data (like Physics) are batch-like – For small data there are opportunities for real time interactivity that will be lost if we assume a 1-interface-fits-all paradigm. • Range of visualization endpoints: – Portal/Web (not sure if collaboration is supported) – High resolution displays – Immersive displays Electronic Visualization Laboratory (EVL) University of Illinois at Chicago Electronic Visualization Laboratory (EVL) University of Illinois at Chicago Large scale brain maps of protein expression in rats. (NCMIR) Electronic Visualization Laboratory (EVL) University of Illinois at Chicago The GeoWall Electronic Visualization Laboratory (EVL) University of Illinois at Chicago Amplified Collaboration Environments Collaborative passive stereo display Collaborative Tiled Display AccessGrid multisite video conferencing Collaborative touch screen whiteboard Wireless laptops & Tablet PCs to steer the displays Electronic Visualization Laboratory (EVL) University of Illinois at Chicago Electronic Visualization Laboratory (EVL) University of Illinois at Chicago