Speaker:Prof. Albert Lawrence, University of California 
Time:2:30pm--4:00pm, Mar. 26, 2009 (Thursday) 
Place:Room 440, 4th Floor, ICT

Abstract:  
The electron microscope (EM) has long been used to probe the structure of materials beyond the resolution of the light microscope, down to the molecular level. Immense volumes of data have been collected to build a picture coherent with light microscope data.  A crude estimate of the scale of the volume covered by comparable 3D reconstructions may up to 10^15 longer than the scales probed by EM!  Bridging this gap is an enormous undertaking.
We report on recent developments at the National Center for Microscopy and Imaging Research (NCMIR), which is dedicated to the large scale acquisition of biological data. Also we report on a software package, Transform Based Tracking, Bundle Adjustment and Reconstruction, (TxBR), which is designed to compensate for the geometric nonlinearities of EM which arise for large detector arrays. In order to achieve the high throughput and rapid turnaround needed to assemble large data bases, and search through large numbers of reconstructions for relatively rare events we must utilize parallel processing. One relatively inexpensive and effective approach for tomography is to use the graphics processor (GPU). We review our efforts in this direction.
Finally, we describe, very briefly, some long-term projects in the imaging sciences aimed at the improvement of reconstruction technology, and the automation of data processing tasks which are bottlenecks in the data processing and analysis.

Bio:
Albert Lawrence obtained his PhD degree in mathematics at the University of Chicago in 1969.  Understanding life processes in general and the workings of the brain in particular has held a particular fascination for him throughout most of his scientific career. Accordingly, he has been studying the applications of mathematics in biology for the past forty years. His present research is funded by the National Institutes of Health through the National Center for Microscopy and Imaging Research and the National Biomedical Computational Resource at the University of California, San Diego.