Image Feature Extraction on a Grid Testbed

For medical imagery (confocal laser-scanning microscopes, CT, MRI and PET), NCHC does image processing, analysis and 3D reconstruction. For biotechnology imagery (such as microarray biochips), NCHC uses a data clustering procedure for feature extraction that provides insight into an image, such as identifying diseases caused by some protein. Grid techniques enable the use of distributed computing resources and shared data. High-speed networks enable fast processing. For these technologies to be useful in daily medical activities, doctors need responses to procedures in typically five seconds.

Contact
Sang-Liang Chu
National Center for High Performance Computing (NCHC), Taiwan
c00chu00@nchc.gov.tw

Fang-Pang Lin
NCHC, Taiwan
fplin@nchc.gov.tw

Collaborators
Sang-Liang Chu, Chang Huain Hsieh, Fang-Pang Lin, Whey-Fone Tsai, Hsiu-Mei Chou, Yu-Chung Chen, Sun-In Lin, Grace Hong, NCHC, Taiwan
Chun-Ho Chen, Institute of Statistical Science Academia Sinica, Taiwan
Matthias Mueller, High Performance Computing Center, Rechenzentrum Universität Stuttgart, Germany

http://motif.nchc.gov.tw/DataGrid/