Experimental investigation of ghosting artefacts in in-beam MRI during proton pencil beam scanning


Experimental investigation of ghosting artefacts in in-beam MRI during proton pencil beam scanning

Gantz, S.; Hietschold, V.; Hoffmann, A. L.

Introduction
The integration of real-time MRI is expected to improve the targeting precision of proton therapy. We have developed a first prototype setup of an in-beam MRI scanner at a proton pencil beam scanning (PBS) beam line. The aim of this study was to investigate the effects of the dynamic magnetic fringe fields of the PBS beam steering magnets on the MR image quality during simultaneous irradiation and MR image acquisition.
Materials and methods
A 0.22 T open MR scanner was positioned in front of the horizontal PBS research beam line. 2D planar dose spot application was achieved by magnetic beam steering in horizontal and vertical direction through a pair of fast scanning magnets.
A proton pencil beam of 220 MeV was subsequently scanned along a horizontal and vertical central line in the MR imaging field. The irradiation time matched the acquisition time of a single-slice gradient echo sequence, while imaging a homogeneous transversal slice of the ACR Small Phantom. The image quality was evaluated qualitatively and compared to reference images acquired without beam scanning.
Results and conclusions
MR images acquired during vertical beam scanning showed no visual differences to reference images, whereas images acquired during horizontal beam scanning showed coherent ghosting artefacts in phase encoding direction. The artefacts exhibit a systematic behavior in which the number of ghosts is inversely proportional to the number of dose spots scanned. The phase maps of the k-space data prove that the artefacts are caused by phase offsets between adjacent lines, which result from changes in the MR resonance frequency due to the dynamic fringe fields of the beam scanning magnets in the PBS nozzle.
Now the origin of the ghosting artefacts is well understood, appropriate means for magnetic shielding or k-space data post-processing have to be implemented and studied to eliminate these artefacts.

Keywords: MRI; proton therapy

  • Contribution to proceedings
    8th MR in RT Symposium 2021, 19.-21.04.2021, Heidelberg, Germany

Permalink: https://www.hzdr.de/publications/Publ-30017