Published February 2006 | Version v1
Journal article

Accuracy of attenuation coefficient obtained by 137Cs single-transmission scanning in PET. Comparison with conventional germanium line source

  • 1. Inst. of Biomedical Research and Innovation, Kobe, Hyogo (Japan)
  • 2. Shimadzu Corp., Medical Systems Div., Kyoto, Kyoto (Japan)
  • 3. Osaka Univ., Graduate School of Medicine, Suita, Osaka (Japan)

Description

Transmission scanning can be successfully performed with a Cs-137 single-photon-emitting point source for three-dimensional PET imaging. This method was effective for postinjection transmission scanning because of differences in physical energy. However, scatter contamination in the transmission data lowers measured attenuation coefficients. The purpose of this study was to investigate the accuracy of the influence of object scattering by measuring the attenuation coefficients on the transmission images. We also compared the results with the conventional germanium line source method. Two different types of PET scanner, the SET-3000 G/X (Shimadzu Corp.) and ECAT EXACT HR+ (Siemens/CTI), were used. For the transmission scanning, the SET-3000 G/X and ECAT HR+ were the Cs-137 point source and Ge-68/Ga-68 line source, respectively. With the SET-3000 G/X, we performed transmission measurement at two energy gate settings, the standard 600-800 keV as well as 500-800 keV. The energy gate setting of the ECAT HR2+ was 350-650 keV. The effects of scattering in a uniform phantom with different cross-sectional areas ranging from 201 cm2 to 314 cm2 to 628 cm2 (apposition of the two 20 cm diameter phantoms) and 943 cm2 (stacking of the three 20 cm diameter phantoms) were acquired without emission activity. First, we evaluated the attenuation coefficients of the two different types of transmission scanning using region of interest (ROI) analysis. In addition, we evaluated the attenuation coefficients with and without segmentation for Cs-137 transmission images using the same analysis. The segmentation method was a histogram-based soft-tissue segmentation process that can also be applied to reconstructed transmission images. In the Cs-137 experiment, the maximum underestimation was 3% without segmentation, which was reduced to less than 1% with segmentation at the center of the largest phantom. In the Ge-68/Ga-68 experiment, the difference in mean attenuation coefficients was stable with all phantoms. We evaluated the accuracy of attenuation coefficients of Cs-137 single-transmission scans. The results for Cs-137 suggest that scattered photons depend on object size. Although Cs-137 single-transmission scans contained scattered photons, attenuation coefficient error could be reduced using by the segmentation method. (author)

Additional details

Publishing Information

Journal Title
Nippon Hoshasen Gijutsu Gakkai Zasshi
Journal Volume
62
Journal Issue
2
Journal Page Range
p. 289-296
ISSN
0369-4305