HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Gong, S. Articles by Scott, A. PubMed Articles by Gong, S. Articles by Scott, A.

Investigating the performance of single scatter simulation on Philips Gemini PET/CT

¹ Centre for P.E.T., Austin Hospital, Melbourne, Victoria, Australia

1654

Objectives: The study investigates image artifacts and significant SUV discrepancy observed in the images corrected with single scatter simulation (SSS) comparing to scatter tail fitting (STF). This leads to the systematic validation of the reconstruction protocols for its optimal implementation.

Methods: A series of PET/CT/Cs scans of F-18 lung phantom with hot and cold spheres, F-18 flood and Ge-68 solid uniformity phantoms were acquired on a Philips Gemini PET/CT. Phantom images were reconstructed using various protocols (STF vs. SSS, CT vs. Cs AC, segmentation vs. remapping, 3D vs. 2D RAMLA, 1 vs. 2 iterations, and small vs. normal sinogram). Estimated scatter and images were assessed quantitatively. Images of 20 consecutive patients and 10 large patients (>100kg) were processed in the same way and assessed blindly by 2 physicians.

Results: SUV was 20±3% less in SSS images comparing to STF. The overall quality and contrast show no significant difference with rating scores 3.7±0.48 and 3.9±0.38 for SSS images, 3.8±0.32 and 3.6±0.52 for STF images, respectively (score 4 indicates best and 1 the worst). The implementation of SSS using small sinogram with 1 iteration of 2D RAMLA for scatter estimation achieved similar results to images using big sinogram with 2 iterations of 3D RAMLA but half of its total processing time (3.7±0.6 mins per frame).

Conclusions: SSS offers a feasible alternative to STF within acceptable time for clinical whole-body scans. Scatter scaling and subtraction must be performed prior to 3D RAMLA reconstruction with attenuation correction to obtain accurate compensation. To confirm clinical and research scans assessed within a multi-scanner PET center having a known degree of SUV consistency, calibrations need be performed correspondingly to the correction and reconstruction methods used.

This Article Services Email this article to a friend Similar articles in this journal Alert me to new issues of the journal Download to citation manager Google Scholar Articles by Gong, S. Articles by Scott, A. PubMed Articles by Gong, S. Articles by Scott, A.