shiun-ying Ju1, Yu-Wei Tang1,
Teng-Yi Huang1, Hsu-Hsia Peng2
1Department of Electrical Engineering,
National Taiwan University of Science and Technology, Taipei, Taiwan; 2Department
of Biomedical Engineering and Environmental Sciences, National Tsing Hua
university, Hsinchu, Taiwan
The
MR flow quantification of the vessels can be used to evaluate the response of
the vessel-related surgeries. However, for the conventional longitudinal studies
of blood flow, manual slice selection in different days can cause measurement
variation and thus degrade the accuracy of evaluation. In our study, in order
to solve this problem, an automatic slice positioning method combined with
previously implemented automatic ROI selection method were proposed to reduce
the inter-scan variation. Furthermore, GPU-accelerated computation was
applied to speed up the image registration. The flow variation of the human
study was reported.