SNM Annual Meeting Abstracts
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     



J Nucl Med. 2008; 49 (Supplement 1):62P
This Article
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Kadrmas, D.
Right arrow Articles by Panin, V.
PubMed
Right arrow Articles by Kadrmas, D.
Right arrow Articles by Panin, V.

Instrumentation & Data Analysis: Image Generation

PET - Optimization and Evaluation

Impact of TOF on PET tumor detection: An LROC study

Dan Kadrmas1, Michael Casey2 and Vladimir Panin2

1 Department of Radiology, University of Utah, Salt Lake City, Utah; 2 Siemens Medical Solutions, Knoxville, Tennessee

249

Objectives: Time-of-flight PET provides a degree of depth-localization along coincidence lines between detector elements, leading to improved noise characteristics in the reconstructed image. In this work, the effect of TOF PET upon focal lesion detection was assessed by localization receiver operating characteristics (LROC) analysis with both human and numerical observers.

Methods: A multi-compartment anthropomorphic phantom was setup to mimic whole-body FDG cancer imaging and scanned 12 times on a TruePoint Biograph with TrueV and prototype TOF capability (<600 ps) (J Nucl Med 2006(47):184P). Eight of the scans had 26 Ge-68 "shell-less" lesions (6 to 16 mm diam.) placed throughout the phantom. This provided lesion-present and lesion-absent datasets with known truth appropriate for LROC analysis. Baseline reconstructions were performed by fully-3D line-of-response LOR-OSEM with and without point spread function (HD.PET), reconstructions were then repeated with TOF information. A channelized non-prewhitened (CNPW) observer, previously developed for use with such phantom data, was used to study the effect of iteration number and post-filter on lesion detection. Reconstructions with the optimal parameters were scored by human observers, computing the tumor localization performance and area under the LROC curve (ALROC) for all reconstructions.

Results: The use of TOF information significantly improved lesion detectability: ALROC = 0.68 with TOF, vs. 0.54 without TOF; offering improvement in addition to that gained by the use of an accurate PSF model over the LOR reconstruction without such a model.

Conclusions: Inclusion of TOF information in PET reconstruction yielded a significant improvement in tumor detection for the medium-sized phantom used in this work, and likely has even greater impact for larger subjects.





This Article
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Google Scholar
Right arrow Articles by Kadrmas, D.
Right arrow Articles by Panin, V.
PubMed
Right arrow Articles by Kadrmas, D.
Right arrow Articles by Panin, V.