TY - JOUR T1 - Lung cancer imaging with respiratory motion-compensated simultaneous PET/MR JF - Journal of Nuclear Medicine JO - J Nucl Med SP - 1417 LP - 1417 VL - 57 IS - supplement 2 AU - Joyita Dutta AU - Georges El Fakhri AU - Quanzheng Li Y1 - 2016/05/01 UR - http://jnm.snmjournals.org/content/57/supplement_2/1417.abstract N2 - 1417Objectives Motion artifacts introduced by respiration degrade both the quality and quantitative accuracy of thoracic PET images. The goal of this study is to evaluate the efficacy of motion-compensated simultaneous PET/MR in lung cancer imaging.Methods Simultaneous PET/MR imaging was performed on subjects with cancerous lesions in one or both lungs. [18F]FDG PET list-mode data and MR k-space data were simultaneously acquired using a Golden-angle RAdial Navigated Gradient Echo (GRANGE) pulse sequence. The MR acquisition parameters in the studies were as follows: TR 3.3 ms/slice, bandwidth 1 kHz/pixel, readout oversampling factor of 2,256 samples per radial line, and 26 coronal slices with a slice thickness of 8 mm and with 4000 radial lines per slice. The PET list-mode events and MR k-lines were grouped into six amplitude-based gates per respiratory cycle. Gated MR images were reconstructed using the k-t FOCUSS reconstruction technique. Non-rigid registration based on diffeomorphic demons was used to compute deformation fields from the reconstructed MR images. PET list-mode data were acquired for 7 min and binned into gated sinograms. Motion compensated (MC) PET images corresponding to the end-exhalation phase were reconstructed using the sinograms and deformation fields by employing a maximum a posteriori approach. As reference, one-gate (OG) and ungated (UG) PET images were computed.Results Overestimation of the volume and underestimation of the intensity of high-intensity features (such as lesions) are well-known consequences of motion-induced blurring. We therefore quantitatively compared the OG, UG, and MC cases by computing the lesion volume and lesion to background contrast-to-noise ratio (CNR). Our results indicate that, while, UG leads to an overestimation of lesion volume by 11%-38% relative to the OG case, MC exhibits a 2%-19% range for the same metric. Furthermore, MC leads to a CNR improvement of 19%-196% relative to OG and 6%-51% relative to UG.Conclusions Compensation of respiratory motion led to substantial enhancement of image contrast and reduction of variability in PET images of patients with lung lesions. Motion compensated simultaneous PET/MR is therefore promising for quantitative imaging of lung lesions. ER -