Abstract
3289
Introduction: Due to comparatively long measurement times in PET/MRI imaging, patient motion are likely during the examination. These lead to artefacts, which have a negative effect on the visual assessability and quantitative validity of the image data and, in the worst case, lead to misjudgements. Simultaneous PET/MRI systems allow registration of the motion and enable correction of the PET data. To assess the effectiveness of motion correction (MoCo) methods, it is necessary to carry out measurements on phantoms that are moved in a reproducible way. This study explores the possibility of using such a phantom-based setup to evaluate MoCo strategies in PET/MR of the human head.
Methods: An MR-compatible robotic system was used to generate rigid motion of a head-like phantom (1). Different tools, either from the manufacturer or open-source software, were used to estimate and correct for motion based on the PET data itself (SIRF (2) with SPM (3) and NiftyReg (4)) and MR data acquired simultaneously, FSL-MCLFIRT (5) and BrainCompass (BC, Siemens Healthineers AG). Different motion estimates were compared using data acquired during robot-induced motion. The effectiveness of MoCo of PET data was evaluated by determining the segmented volume of an activity-filled flask inside the phantom. In addition, the segmented volume was used to determine the centre-of-mass (COM) and the change in maximum activity concentration (Amax).
Results: The results showed a volume increase between 2.7 and 36.3% could be induced by the experimental setup depending on the motion pattern. Both, BC and MCFLIRT, produced corrected PET images, by reducing the volume increase to 0.7% - 4.7% (BC) and to -2.8% - 0.4% (MCFLIRT). The same was observed for example for the COM, where the results show that MCFLIRT (0.2 to 0.6 mm after MoCo) had a smaller deviation from the reference position than BC (0.5 to 1.8 mm) for all displacements. The consideration of Amax provided a deviation of MCFLIRT from the reference value (Amax before motion) is less than 1.2% for all translation amplitudes while the BrainCompass provided deviations between 1.3% and 2.2%.
Conclusions: The experimental setup is suitable for the reproducible generation of motion patterns. Using open-source software for MoCo is a viable alternative to the vendor-provided MoCo software. The evaluation showed that MoCo methods lead to a minimization of volume increase through motion, which may result in a better localization of PET data.
(1) Harries, J. et al. (2020). EJNMMI physics, 7(1), 1-12.
(2) Ovtchinnikov, E. et al. (2020). Computer Physics Communications, 249, 107087.
(3) Friston, K. J. et al. (1995). Human Brain Mapping, 3(3), 165-189.
(4) Modat, M. et al. (2014). Journal of Medical Imaging, 1(2), 024003.
(5) Jenkinson, M. et al. (2002). NeuroImage, 17(2), 825-841.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.