Effect of patient arm motion in whole-body PET/CT

Objectives Maintaining the arms in a raised position throughout a whole-body PET/CT study is uncomfortable for many patients and arm motion between CT and PET acquisitions is not uncommon. We investigate the mechanisms that underlie the profound PET artifacts that can result when CT and PET of the arms are misaligned.

Methods A phantom experiment was performed using a central 20 cm diameter Ge-68 cylinder simulating the neck and two peripheral 10 cm diameter F-18 cylinders simulating arms. After motion-free CT and PET on a whole-body PET/CT system, the position of the arms was altered so as to introduce different amounts of misalignment (0-5 cm) within the field-of-view. 20 sequential PET scans were acquired in this position, alternating between 2D and 3D, as the F-18 decayed. Decay of F-18 in the arms, while the activity in the Ge-68 cylinder remained relatively constant, allowed the impact of scatter and attenuation correction errors to be determined.

Results Misalignment artifacts were largely confined to the arms in 2D but extended to the neck in 3D where they manifested as a striking underestimation of intensity that became more significant with increasing misalignment. For 2D, misalignment caused errors in attenuation correction (+6 %) but scatter correction error was negligible. For 3D, scatter correction error (up to -87 %) depended on activity in the arms and was generally more significant than attenuation correction error (+17 %). The 3D scatter correction algorithm scaled the scatter estimate to the tails of the projections (determined incorrectly due to the misaligned CT body outline), resulting in extensive over-correction. 3D image reconstruction without scatter correction substantially eliminated these artifacts in both phantom and patient images.

Conclusions For 3D PET, scatter correction error is a more significant problem than attenuation correction error in cases of patient arm motion and can give rise to severe artifacts. PET image reconstruction without scatter correction is a practical and effective (but non-quantitative) method of suppressing these artifacts