Share Email Print
cover

Proceedings Paper

Optimization of iterative image reconstruction method in PET/SPECT using channelized Hotelling observer
Author(s): Wataru Mukai; Kenya Murase; Shunsuke Fukami; Keiichi Matsumoto; Keiji Shimizu; Xiaomei Yang; Keishi Kitamura; Michio Senda
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

The aim of this study was to apply channelized Hotelling observer (CHO) to optimization of the numbers of subsets and iterations in the ordered-subsets expectation-maximization (OS-EM) algorithm. A positron emission tomography (PET) scanner was used to scan images of a hot-spot phantom and a cylindrical phantom with a diameter of 20 cm filled with 18F-fluorodeoxy glucose. The CHO was applied to 50 slices from the images of each type to compute the detectability index (dCHO). To investigate the optimal numbers of subsets and iterations in the OS-EM algorithm for SPECT, elliptical numerical phantoms containing a hot cylinder of different sizes and contrasts were generated using a simulation tool. In PET, the dCHO value increased with increasing number of iterations when the number of subsets was 1, 2, 4 and 8. When the number of subsets was 32, the dCHO value decreased with increasing number of iterations, i.e., the divergence phenomenon occurred. The optimal numbers of subsets and iterations were 16 and 3. In SPECT, the number of iterations at which the dCHO value became the maximum changed depending on the projection counts and lesion contrast. When the projection counts were 500 k counts/projection and the lesion contrast was 1.5 or 2.0, the dCHO value became the maximum at an iteration number of 5. Our results suggest that CHO is effective for optimizing image reconstruction parameters in the OS-EM algorithm. Although they depend on various factors, they can be quantitatively evaluated using CHO without tedious and time-consuming studies with human observers.

Paper Details

Date Published:
PDF: 10 pages
Proc. SPIE 6515, Medical Imaging 2007: Image Perception, Observer Performance, and Technology Assessment, 65151A; doi: 10.1117/12.709052
Show Author Affiliations
Wataru Mukai, Osaka Univ. (Japan)
Kenya Murase, Osaka Univ. (Japan)
Shunsuke Fukami, Osaka Univ. (Japan)
Keiichi Matsumoto, Osaka Univ. (Japan)
Institute of Biomedical Research and Innovation (Japan)
Keiji Shimizu, Institute of Biomedical Research and Innovation (Japan)
Xiaomei Yang, Osaka Univ. (Japan)
Keishi Kitamura, Shimadzu Corp. (Japan)
Michio Senda, Institute of Biomedical Research and Innovation (Japan)


Published in SPIE Proceedings Vol. 6515:
Medical Imaging 2007: Image Perception, Observer Performance, and Technology Assessment
Yulei Jiang; Berkman Sahiner, Editor(s)

© SPIE. Terms of Use
Back to Top