Investigation of large field-of-view transmission imaging for SPECT attenuation compensation with Gd-153, Tc-99m and Ce-139 sources

Citation
Mp. Tornai et al., Investigation of large field-of-view transmission imaging for SPECT attenuation compensation with Gd-153, Tc-99m and Ce-139 sources, IEEE NUCL S, 47(3), 2000, pp. 1182-1191
Citations number
28
Categorie Soggetti
Apllied Physucs/Condensed Matter/Materiales Science","Nuclear Emgineering
Journal title
IEEE TRANSACTIONS ON NUCLEAR SCIENCE
ISSN journal
00189499 → ACNP
Volume
47
Issue
3
Year of publication
2000
Part
3
Pages
1182 - 1191
Database
ISI
SICI code
0018-9499(200006)47:3<1182:IOLFTI>2.0.ZU;2-D
Abstract
A recently developed, large field-of-view (FOV), dual head SPECT system was investigated for fast sequential fan-beam (FB) transmission computed tomog raphy (TCT) with various transmission sources. For TCT, the system utilizes a fixed line source at the 77 cm focal line of a symmetric FB collimator. The TCT maps are used for non-uniform attenuation compensation (NUAC) of Tc -99m myocardial SPECT. The heads each have three degrees of freedom (transa xial tilt, radial linear and transaxial linear travel). When the heads are tilted and height adjusted to accommodate different size FOVs (34 similar t o 53 cm diameter), the symmetric FB geometry only samples part of the FOV. Thus, with full 360 degree acquisitions, the pseudo-asymmetric FB TCT detec tor utilizes measured conjugate views to avoid truncation artifacts without sacrificing spatial resolution, and minimizes emission contamination compa red with other TCT geometries. Transmission line source energies of 100 to 166 keV (from Gd-153, Tc-99m, and Ce-139 line sources) were utilized with r od and sphere, and anthropomorphic phantoms. The TCT images were reconstruc ted with iterative ordered subsets estimation maximization (OSEM). Comparis ons were made between the emission reconstructions utilizing filtered backp rojection (without and with iterative and multiplicative Chang NUAC) and OS EM (without and with NUAC). Based on a derived noise equivalent count rate determination of the three sources, the highest energy yet weakest intensit y TCT source outperforms the other sources in the primary myocardial region of interest by approximately a factor of two. The emission contamination f or the highest energy transmission source was an order of magnitude smaller than for the other sources with this acquisition geometry. These line sour ces each demonstrate the potential for suitably compensated images and quan titative myocardial activity profiles with this geometry. Higher energy tra nsmission sources in particular result in more physically accurate attenuat ion maps useful for SPECT quantitation.