Biomedical Sciences and Visualization

An Image-Based Approach to Scientific Visualization

Sponsor: Department of Energy (DOE)

Effective analysis requires continuous thought and rapid feedback to queries (visual queries for scientific visualization). This method requires a substantial paradigm shift from the current method of storing all of the data for later analysis. This work offers such a shift in paradigms. It is built upon the premise that many precomputed "animations" can be constructed to satisfy more than 90% of a user's queries about the simulation. With training, experience, and the right tools, we can approach 100%. Compare this to current visualization tools, which can be applied to answer most (but not all) queries on a single data set, but offer little support for multiple data sets. The latency involved in using these tools with very large data sets, tries even the most patient scientist. As a result, many questions are left unanswered for even a single data set, because of this formidable bottleneck.

The long term objective of this research is to develop a more synchronous compute and investigate procedure that will not require the user to manipulate cumbersome and prodigious computational data.

Funding for this project is made possible through a grant from the U.S. Department of Energy

Further Reading
Interactive Medical Data on Demand: A High-Performance Imaged-Based Approach Across Heterogeneous Environments. Stredney D, Agrawal A, Barber D, Crawfis R, fend WC, Hou, J, Panda DK, Sadayappan P, Powell K, Schmalbrock P, Sessanna D, Wiet GJ, Shareef N and J Bryan, Proceedings of MMVR8, 2000, Westwood, (Eds). IOS Press Amsterdam: 327-333

Interactive Volume Visualization for Synchronous and Asynchronous Remote Collaboration, Stredney D, Crawfis R, Wiet GJ, Sessanna D, Shareef N and J Bryan, Proceedings of MMVR7 1999, Westwood et. al., (Eds). IOS Press, Amsterdam: 344-350