Integrating a low-field open MR scanner with a static proton therapy research beamline: Characterisation of in-beam imaging performance

Purpose
For the successful integration of MRI and proton therapy (PT), the mutual electromagnetic interaction between both systems needs to be investigated. So far, no combined system existed to investigate the MR imaging performance in the presence of a proton beam. Here, the aim is to characterize the imaging performance of a first in-beam MRI scanner during simultaneous proton irradiation and imaging.

Materials & methods
A 0.22 T open MRI scanner (MrJ2200, Paramed) has been installed at the horizontal static research beamline of our PT facility. The imaging performance characterization included both magnetic field homogeneity (MFH) measurements and ACR phantom tests for image quality. The MFH was mapped over a 22 cm diameter spherical volume by a magnetic field camera (MFC3045/48, Metrolab) being placed in the centre of the scanner’s field-of-view (FOV). To assess the effect of magnetic fringe fields of the nearby beamline magnets, the MFH was measured without and with magnets energized for beam energies between 70220 MeV. Phantom imaging tests were performed with the ACR small phantom being centrally positioned in the scanner FOV inside a knee coil (Fig. 1). Images were acquired by performing T1- and T2-weighted spin echo (SE) sequences with parameter settings according to the ACR phantom test protocol. Additionally, T1 and T2*- weighted gradient echo (GES and GEL, respectively) scans were performed. The phantom was irradiated by a 125 MeV pencil beam (Ø=10 mm) at dose rates of 1 and 80 Gy/min. Images were acquired for six different scenarios, starting from a reference scan with beamline magnets and beam off, followed by sequentially switching on first the beamline magnets and then the beam during both radiofrequency calibration and image acquisition or during image acquisition alone. A validated software tool (MATLAB) was used to extract the ACR imaging parameters and to estimate geometric transformations from image pairs acquired for the different scenarios.

Results
The peak-to-peak MFH was 88 ppm, which is within the scanner’s operating specifications. The MFH measurements with and without energized beamline magnets showed no significant differences (<3 ppm), but the mean baseline resonance frequency was increased by 70110 Hz depending on beam energy. The SE and GEL image quality was sufficient for analysis. GES images showed a low SNR (<15) and suffered from banding artefacts, which prevented automated evaluation of ACR parameters. For all six scenarios, differences in ACR parameters were within measurement uncertainties. A sequence-dependent uniform image translation of 0.53.0 mm in frequency encoding direction was observed due to operating the beamline. These image translations were in accordance with the baseline resonance frequency shift and showed to be inversely proportional to the gradient strengths of the sequences used (0.75.7 mT/m).

Conclusion
The imaging performance of a first low-field in-beam MRI scanner integrated with a static PT research beamline meets vendor and ACR image quality specifications. No degradation of the imaging performance was observed during simultaneous operation of the MRI and PT systems. Beamline induced off-resonance image translations need to be compensated for in the different imaging sequences.