Abstract #5037

MR-Assisted PET Respiratory Motion Correction Using Deep-Learning Based Short-Scan Motion Fields

Sihao Chen¹, Tyler J. Fraum¹, Cihat Eldeniz¹, Joyce Mhlanga¹, Weijie Gan¹, Thomas Vahle², Uday B. Krishnamurthy³, David Faul⁴, H. Michael Gach¹, Michael M. Binkley¹, Ulugbek S. Kamilov¹, Richard Laforest¹, and Hongyu An¹

¹Washington University in St. Louis, Saint Louis, MO, United States, ²Siemens Healthcare GmbH, Erlangen, Germany, ³Siemens Medical Solutions USA, Saint Louis, MO, United States, ⁴Siemens Medical Solutions USA, Malvern, PA, United States

Synopsis

Respiratory motion causes signal blurring and image artifacts. Simultaneous PET/MRI allows for MR-assisted motion correction (MoCo) in PET imaging, leading to improved PET images for detection and evaluation of lesions. In this study, we proposed and examined a PET MoCo approach using motion vector fields (MVFs) from a deep-learning reconstructed MRI scan. MRI-based MVFs were derived from either 2000 spokes (MoCo₂₀₀₀, 5-6 minutes acquisition time) using a Fourier transform reconstruction or 200 spokes (MoCo_P2P200, 30-40 seconds acquisition time) using a deep-learning Phase2Phase (P2P) reconstruction and then incorporated into PET MoCo reconstruction.

This abstract and the presentation materials are available to members only; a login is required.

Join Here