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


     



J Nucl Med. 2008; 49 (Supplement 1):121P
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 Necib, H.
Right arrow Articles by Buvat, I.
PubMed
Right arrow Articles by Necib, H.
Right arrow Articles by Buvat, I.

Instrumentation & Data Analysis: Data Analysis & Management

Oncology

Detection of the tumor changes between two FDG PET scans using parametric imaging

H. Necib1, M. Dusart2, P. Tylski1, B. Vanderlinden2 and I. Buvat1

1 UMR 8165 CNRS, Orsay, France; 2 Institut Bordet, Bruxelles, Belgium

480

Objectives: Patient follow-up based on PET scans is a promising approach for the assessment of tumor response to therapy. We introduce a parametric imaging method to detect and analyze the tumor changes between 2 consecutive PET scans.

Methods: 14 pairs of consecutive PET/CT tumor images acquired between 10 and 12 weeks apart when monitoring lung cancer patients undergoing chemotherapy were considered. For each pair, after CT-based registration of the PET images, the two PET volumes were converted in SUV units and subtracted. A biparametric graph of subtracted voxel values versus voxel values in the first scan was obtained and analyzed using a Gaussian mixture model. Parametric images of significant SUV changes in tumors were deduced. The mean SUV changes in the clusters of voxels seen on the parametric images were calculated and compared to the clinician’s assessment of the tumor response based on visual analysis.

Results: The CT-based registration of the PET images was accurate enough to subtract the registered PET images without obvious misalignement. The parametric images correctly showed 9 tumors with complete response (mean change in SUV in the 9 corresponding clusters was -4.4±3.0), 12 tumors with partial response (mean change in SUV was -3.9±2.0) and 4 tumor progressions (mean change in SUV was +4.4±1.5). 2 tumors seen as stable by the clinician were detected in the parametric images, with SUV changes of +1.3 and +1.6. Additional PET/CT scans showed that these tumors later progressed. The parametric images did not show any cluster that did not correspond to a tumor.

Conclusions: The proposed parametric imaging approach detects tumor changes occurring between two successive scans, without the need to first identify the tumors. Unlike an ROI–based analysis, it retains information at the voxel level.





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 Necib, H.
Right arrow Articles by Buvat, I.
PubMed
Right arrow Articles by Necib, H.
Right arrow Articles by Buvat, I.