Premium
TU‐E‐141‐09: Impact of Attenuation Correction Mode On 4D PET/CT for Target Definition in Lung Cancer Patients
Author(s) -
Bowen SR,
Nyflot MJ,
Zeng J,
Sandison GA,
Patel SA,
Kinahan PE
Publication year - 2013
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4815437
Subject(s) - nuclear medicine , correction for attenuation , voxel , attenuation , lung cancer , medicine , positron emission tomography , pet ct , standardized uptake value , mathematics , physics , radiology , pathology , optics
Purpose: The impact of attenuation correction on respiratory‐gated (4D) PET/CT images to define lung cancer targets, relative to 4D maximum intensity projection (MIP) or 3D static images, is poorly understood. Changes from 3DPET to 4DPET SUV metrics and target volumes were characterized in patients as a function of PET acquisition and CT attenuation correction (CTAC). Methods: Lung cancer patients underwent PET/CT list‐mode examinations spanning 3 bed positions. Static and respiratory phase‐gated sinograms were unlisted to generate 3DPET and 4DPET images. Emission data were attenuation‐corrected with phase‐averaged (4DAVG) or phase‐matched (4DMATCH) 4DCT and reconstructed with OSEM (2 iterations, 28 subsets, 5 mm post‐filter) on 2.0×2.0×3.3 mm3 voxel grids. PET target volumes were defined in MIM™5.6.3 using a gradient search method. Differences in SUVmax, SUVmean, SUVpeak, and defined volumes were compared between 3D, 4D MIP, and individual 4D phases by pairwise sign‐rank tests. 4DAVG and 4DMATCH CTAC‐PET image metrics were compared between respiratory phases using multi‐group non‐parametric ANOVA. Results: For 13 FDG‐PET avid lesions, SUVmax, SUVmean and SUVpeak were significantly higher in 4DMIP and 4D phase‐gated images relative to 3D images (9–11%, p<0.03). 4DMIP volumes were larger than 3D and 4D phase volumes (6–14%, p<0.01). 4DMATCH SUVmax was higher than 4DAVG SUVmax near end‐expiration phase and lower around peak‐inspiration phase (p=0.001). 4DMATCH SUVmean exceeded 4DAVG SUVmean at end‐expiration (p=0.02), while SUVpeak trended similarly but was insufficiently powered (p=0.06). 4D phase volumes were not significantly different between attenuation correction methods (p>0.37). Conclusion: 4DPET provides improved image quantification over 3DPET. 4DPET quantification near end‐expiration appears superior with phase‐matched CTAC, but may be less reliable during phase‐gates with residual motion and phase‐sorting CT artifacts. Optimization of CTAC from patient‐specific respiratory patterns may improve robustness of 4D PET/CT for lung cancer target definition and dose painting.