4432
Simulation of external magnetic field changes due to head motion during 7 Tesla MRI scan1SPMIC, University of Nottingham, Nottingham, United Kingdom
Synopsis
One potential method for monitoring the effects of head movement in the scanner consists of using a fixed array of field probes to measure field changes produced outside the head by small changes in head position and angulation. This method has the advantage of requiring neither attachment of markers or probes to the head, nor modification of the imaging sequence. Here, we use realistic head models to simulate the external field changes produced by typical head movements in a 7T scanner and use the results to explore the relationship between the magnetic field perturbation and changes in head position.
Introduction
Head movement is a major issue in MRI. A variety of methods for monitoring head position inside the MRI scanner have therefore been developed and used for retrospective and prospective motion correction1. One potential method for monitoring head movement in the scanner consists of using a fixed array of field probes to measure field changes produced outside the head by small changes in head position and angulation (Figure 1). This method of monitoring head position has the advantage of requiring neither attachment of markers or probes to the head, nor modification of the imaging sequence2. Here, we use realistic head models to simulate the external field changes produced by typical head movements in a 7T scanner and use the results to explore the relationship between the magnetic field perturbation and changes in head position.Methods
Simulations mimicked 7T experimental measurements in which the field variation was measured using 16 NMR probes3 fixed in four co-axial rings around the head, and head movement was simultaneously monitored by optically tracking a Moiré Phase Tracking (MPT) marker rigidly coupled to the head via a dental mould (Figure1). Voxelated models of two subjects were formed from MRI data acquired at 3T (voxels classified as fat, water or air), and positioned relative to the field probes using 7T images acquired during the same session as the field recordings. Repeated field simulations were carried out using a Fourier-based method4 with the model translated and rotated based on the optical measurements, or moved by varying a single co-ordinate to evaluate the linearity of field variation with position. Movement parameters were also predicted from the simulated field values using a Partial Least Square (PLS) regression. The modelled space consisted of a cube of 7003 voxels, each of 0.5 x 0.5 x 0.5 mm3 size. A region at the base of the model was classified as water in order to avoid field changes being produced by movements of the artificial boundary at the inferior extreme of the head/neck model (Figure 1).Results
Figure 2 shows the field perturbations measured using the field camera when Subject 1 executed nodding movements in the 7T scanner5. The field changes at the field probe positions produced by applying the recorded movements to the simulations are also shown. To determine the relationship between the change in magnetic field (ΔB) and the change in position (ΔT) and orientation (ΔR), the head pose of Subject 2’s head model was changed either as a pure translation or as a pure rotation. We considered movements in the range ±10 mm or ±10° (Figure 3) and ±2 mm or ±2° (Figure 4). The results5 indicate that the field variation is approximately linear with position co-ordinate and angulation for the smaller range of movements, but for the larger movements (which are greater in extent than would normally occur during scanning) the field variation is non-linear. This is particularly the case for the probe sited near the nose. Figure 5 shows that the PLS method can accurately predict movements based on the field measurements (R2 >0.9 for both models) for the smaller range of movements (± 2mm /±2o).Discussion
The advantages of using simulations is that it allows the factors that most strongly influence the experimental results to be identified. These are the relative position of the head and the probes, with probes that come into very close proximity with the nose or ears being most likely to show non-linear field variations with head position. Initial experiments using the HUGO6 head model also allowed us to study the influence of the tissue segmentation on the field perturbations, with the results indicating that the susceptibility values allocated to superficial tissues have the greatest influence on the measured external fields.Conclusion
Simulation allows us to explore the effect of a wide range of well-determined head movements, and to probe the relationship between head movement and extra-cranial field changes. The results show that the fields change linearly with head position for small translations (± 2mm) and rotations (±2°) and that small changes in head position can be calculated from measurements of the external field change by partial least squares regression. Simulations of the field changes produced outside the head by realistic head movements are in reasonable agreement with measurements made when subjects execute the same movements.Acknowledgements
No acknowledgement found.References
1. Luesebrink, Falk; Sciarra, Alessandro; Mattern, Hendrik; et al. T-1-weighted in vivo human whole brain MRI dataset with an ultrahigh isotropic resolution of 250 mm .Scientific Data, May 2017; 4, 170062
2. Aranovitch A., Haeberlin M., Gross S. et al. Prospective Motion Correction With NMR Markers Using Only Native Sequence Elements. Magn Reson Med. 2018; 79:2046–2056.
3. Bischof L., Smith J., Mougin O. et al. Relating external magnetic field changes to head movement using motion and field cameras. Hawaii,USA ISMRM; 0303. 2017
4. J.P. MARQUES, R. BOWTELL .Application of a Fourier-Based Method for Rapid Calculation of Field Inhomogeneity Due to Spatial Variation of Magnetic Susceptibility. Concepts in Magnetic Resonance Part B (Magnetic Resonance Engineering) 2005; Vol. 25B(1) 65–78
5. Bortolotti L., Smith J., Gowland P., Bowtell R. . Simulating changes in external magnetic field due to head motion in a 7T scanner. Oxford, UK BCISMRM: PP13. 2018
6. Christopher M. Collins, Bei Yang, Qing X. Yang, Michael B. Smith. Numerical calculations of the static magnetic field in three-dimensional multi-tissue models of the human head. Magnetic Resonance Imaging 2002; 20,413–424
Figures
Left: Head models were produced by segmenting a 3T MR image into three different tissue compartments (fat χ=-7.779ppm, water χ=-9ppm and air χ= 0ppm)
Right: 16 NMR probes (Skope, Zurich.) were fixed in three co-axial rings around the head. Head movement was monitored optically, using a Moiré Phase Tracking (MPT) marker coupled to the head via a dental mould. A simulation of the field perturbation produced in a central sagittal slice when the head is exposed to a 7T field is shown in the bottom right sub-figure (which also indicates the positions of the field probes projections onto this slice)
Left: field simulation and probe positions for subject 1;
Right: (top) changes in head position measured using the MPT system; (middle) measured and (bottom) simulated changes in field produced at the probe positions by these head movements.
