RT Journal Article
SR Electronic
T1 Image quality assessment for a routine use of amplitude thresholding in FDG PET
JF Journal of Nuclear Medicine
JO J Nucl Med
FD Society of Nuclear Medicine
SP 207
OP 207
VO 52
IS supplement 1
A1 Frederic Schoenahl
A1 Dagmar Steiner
A1 Susanne Schmidt
A1 Richard Bauer
YR 2011
UL http://jnm.snmjournals.org/content/52/supplement_1/207.abstract
AB 207 Objectives Numerous studies confirm the image quality improvement of gating procedures in the thorax region, however phase triggering (4D) requires long acquisitions and regular breathing motions and is hardly used in routine as a consequence. We investigate an alternative amplitude based thresholding technique 1) on routine patient gated data 2) and using data of moving lesions in a phantom. Methods Several patients with difficulties to breath were acquired and their complete respiratory signal correlated with the acquisition data signal. The signal baseline was automatically identified and corrected. PET events were filtered to provide a reduced listmode using a user chosen amplitude threshold, and its reconstruction compared to original reconstructions. A phantom study was performed for objective assessment using 2 moving hot lesions sizes and 3 contrasts to a hot background. Comparison of both 4D and amplitude gating versus no motion scheme was based on SUVs. All data were reconstructed with the clinical defaults on a Siemens Biograph TruePoint® scanner. Results Subjective analysis on images shows sharper images and lesion detectability is dramatically increased on modified data. Quantitative evaluation on phantom static lesions versus amplitude gated lesions shows SUV indices maintaining within 5% for the three contrasts (10:1, 5:1, 3:1) and multiple lesion sizes (1, 2, 5, 10 ml), whereas 4D gating results drop by up to 25% despite the simple movement simulated. We determined empirically a thresholding value of 40% to provide best results and fastest protocols (4 vs. 10 min). Conclusions These results confirm the relevance of amplitude based gating approaches for the purpose of diagnostic with organ motion when a trigger signal is available. The implementation is partially automatic and a fast preprocessing makes it clinically viable. Further automatization of the method is foreseen (automatic amplitude band determination) and towards other applications (better image quality in therapy planning)