Senior Scholars Program

UMMS Affiliation

Department of Radiology

Faculty Mentor

Michael A. King

Publication Date


Document Type



Life Sciences | Medicine and Health Sciences | Radiology


Background: Patient motion during emission imaging can create artifacts in the reconstructed emission distributions, which may mislead the diagnosis. For example, in myocardial-perfusion imaging, these artifacts can be mistaken for defects. Various software and hardware approaches have been developed to detect and compensate for motion. There are various ways of testing the effectiveness of motion correction methods applied in emission tomography, including the use of realistic digital anthropomorphic phantoms.

Purpose: The purpose of this study was to create 3D digital anthropomorphic phantoms based on MRI data of volunteers undergoing a series of clinically relevant motions. These phantoms with combined position tracking were used to investigate both imaging-data-driven and motion tracking strategies to estimate and correct for patient motion.

Methods: MRI scans were obtained of volunteers undergoing a series of clinically relevant movements. During the MRI, the motions were recorded by near-infra-red cameras tracking using external markers on the chest and abdomen. Individual-specific extended cardiac-torso (XCAT) phantoms were created fit to our volunteer MRI imaging data representing pre- and post-motion states. These XCAT phantoms were then used to generate activity and attenuation distributions. Monte Carlo methods will then be performed to simulate SPECT acquisitions, which will be used to evaluate various motion estimation and correction strategies.

Results: Three volunteers were scanned in the MRI with concurrent external motion tracking. Each volunteer performed five separate motions including an axial slide, roll, shoulder twist, spine bend, and arm motion. These MRI scans were then manually digitalized into 3D anthropomorphic XCAT phantoms. Activity and attenuation distributions were created for each XCAT phantom, representing fifteen individual-specific motions.

Conclusions: Our results will be combined with the external motion tracking data to determine if external motion tracking accurately reflects heart position in patients undergoing cardiac SPECT imaging. This data will also be used to evaluate other motion correction methods in the future.


Motion, Anthropometry, Computer Simulation, Models, Biological, Imaging, Three-Dimensional, Magnetic Resonance Imaging

Rights and Permissions

Copyright is held by the author(s), with all rights reserved.

DOI of Published Version


Journal/Book/Conference Title

Senior Scholars Program


Medical student Caitlin Connelly participated in this study as part of the Senior Scholars research program at the University of Massachusetts Medical School.