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


     




J Nucl Med. 2013; 54 (Supplement 2):375
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 Armstrong, I.
Right arrow Articles by Matthews, J.
PubMed
Right arrow Articles by Armstrong, I.
Right arrow Articles by Matthews, J.

Instrumentation & Data Analysis

Data Analysis & Management V: Quantification

Impact of PSF modelling and time-of-flight reconstruction on SUVmax, SUVpeak and total lesion glycolysis in lung lesions

Ian Armstrong1, Matthew Kelly2, Heather Williams1 and Julian Matthews3

1 Nuclear Medicine, Central Manchester University Hospitals, Manchester, United Kingdom 2 Molecular Imaging, Siemens Healthcare, Oxford, United Kingdom 3 MAHSC, University of Manchester, Manchester, United Kingdom

Abstract No. 375

Objectives: To assess the impact of point spread function (PSF) modelling and time-of-flight (TOF) reconstruction on SUVmax, SUVpeak and total lesion glycolysis (TLG) in lung lesions.

Methods: Routine FDG images from 74 patients acquired on a Siemens Biograph mCT were reconstructed with OSEM, PSF, TOF and PSF+TOF with smoothing filters chosen to match image noise in the liver. SUVmax, SUVpeak and TLG were measured in 79 lung tumours. TLG was computed as the product of volume and mean SUV for a 40% threshold of SUVmax.

Results: Significant differences (p<0.05) for SUVmax and SUVpeak were seen for all advanced reconstructions compared with OSEM and only for PSF and TOF for TLG. Mean [range] ratios for SUVmax compared with OSEM were 1.31 [1.00-1.58], 1.09 [0.93-1.33] and 1.48 [1.01-2.06] for PSF, TOF and PSF+TOF, respectively; for SUVpeak: 1.17 [1.05-1.31], 1.07 [0.96-1.31] and 1.26 [1.00-1.56], respectively; for TLG: 0.92 [0.59-1.12], 1.02 [0.76-1.22] and 0.93 [0.55-1.21] respectively. An inverse non-linear relationship was seen between SUV increase and lesion volume, which was strongest for PSF+TOF. No relationship was seen between TLG change and lesion volume. SUVmax in 2 lesions increased from <2.5 with OSEM to >4.0 with PSF+TOF. SUVmax in 7 lesions increased from <4.0 with OSEM to >4.0 with PSF+TOF. These cut-off values are used locally as guidance for discrimination of lesions (2.5 to 4.0: likely malignancy; >4.0 definite malignancy).

Conclusions: Evidence suggests that implementing advanced reconstructions with PSF and/or TOF is likely to result in improved resolution, signal to noise ratio, and lesion detectability. However, as shown, there is a significant impact on SUV quantitation that may change threshold based lesion classification. Using TLG, instead of SUVmax and SUVpeak, observed changes are notably smaller enabling use of thresholds which are more robust to reconstruction algorithms.





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 Armstrong, I.
Right arrow Articles by Matthews, J.
PubMed
Right arrow Articles by Armstrong, I.
Right arrow Articles by Matthews, J.