Purpose

In conventional magnetic resonance imaging (MRI) system, three gradient coils are commonly used for spatial encoding¹. However, fast switching of the B₀ gradients involves undesired effects such as vibration, acoustic noise, eddy current, and peripheral nerve stimulation (PNS), in addtion to requiring significant hardware and intorducing design constrints on openness. One method to avoid these negative effects is using RF gradients^2,3 for spatiaal encoding, and approach that is especially suitable for low-field applications, where SAR is less problematic. To reconstruct images encoded by RF, the RF magnetic field (B₁) must be known across the entire imaging volume. The B₁ map is often measured using MR images on conventional MR systems⁴ but these methods cannot be used on an MR systems lacking gradient coils. Thus, another way to measure the B₁ map is required for imaging using RF for spatial encoding. The purpose of this work is to measure the B₁ map automatically using a 3D printer and oscilloscope.

Methods

To measure the B_1x, B_1y, B_1z simultaneously, a probe consisting of three pickup loops (square coils, copper wire on the 3D printed frame, 10mm × 10mm × 10mm) was used (Fig. 1). Each of the pickup loops is arranged perpendicular to the others, such that the directions of the magnetic fields created by the three loops are also perpendicular to each other. This configuration also means that, they do not influence the other loops (isolations between the loops were higher than 77 dB at 1 MHz). The probe was fixed at the one end of an acrylic cylinder. If the extruder module of the 3D printer moves in three dimensions, fixing the acrylic cylinder to the extruder module should be sufficient to control the position of the probe. However, many commercial 3D printers are designed such that the extruder moves in two dimensions, and the bed moves in one dimension. The 3D printer used here (Ender 3, Creality, China) also moves the extruder in the x-z direction and the bed in the y direction. Below is a detailed description of how this printer can be used for positioning of the probe. A guide for the acrylic cylinder was designed and printed using the 3D printer and this guide was fixed on the extruder module of the 3D printer. When the extruder module moves in x and/or z direction, the probe moves together but still can slide in y direction. The other end of the acrylic cylinder was fixed in a plastic holder which is also printed using the 3D printer. This plastic holder can slide along the Teflon bar in the x direction. The Teflon bar can slide in the z direction along the two guide bars which are fixed on the bed. Thus, the plastic holder slides in the x-z direction following the movement of extruder, and movement in the y-direction follows the movement of the bed. The movement of the probe was controlled by g-code which was written using Python. Three pickup loops were connected to the digital oscilloscope via co-axial cables. This oscilloscope (Picoscope 4824, Pico technology, UK) was connected to the PC, and saved peak-to-peak values for each second during the B₁ field scanning. A batch file was written for this purpose. B₁ magnitude in x, y, and z direction was calculated using below equation after measurement.$$B_1=\frac{V_{pp}}{2\omega A},$$where $$$V_{pp}$$$, $$$\omega$$$, and A are peak-to-peak voltage, resonance frequency, and cross-sectional area of the loop. Using this set up, the B₁ field generated by circular surface coil was obtained (ROI: 110mm × 110mm × 50mm, resolution: 10mm × 10mm × 10mm, scan time: 14m 29s). Figure 3 shows the scan trajectory. The surface coil (10 mm diameter, 5 turn surface coil) was tuned at 870 kHz and placed on an open magnet. The static magnetic field was also applied during the measurement, with a field strength of 24 mT at the bottom of the magnet⁵.

Results

Figure 4 shows the B₁ map of the surface coil. The magnitudes of B_1x, and B_1y were large near the wire. In particular they were highest where the wire is aligned perpendicular to the magnetic field direction (x and y direction for B_1y and B_1x respectively). The magnitude of B_1z inside of the coil was larger than that outside of the coil.

Discussion

We have automatically measured the amplitude of all three components of the RF magnetic field using a 3D printer and oscilloscope. This approach is useful in developing gradient-free MR system that rely of RF Bloch-Siegert shift for spatial encoding. The design approach demonstrated here can be adapted for higher field MRI since the acrylic cylinder and 3-axis probe are non-magnetic. This system can also be adapted for measuring B₀ field by replacing the 3-axis probe with a Hall effect probe. In addition, B₀ and B₁ field mapping could be performed simultaneously by adding the Hall probe at the center ot the three pickup loops.

Conslusions

We have introduced a method to automatically measure the 3D B₁ field using commercial 3D printer and oscilloscope. We demonstrated that the magnitude of the B_1x, B_1y, and B_1z can be measured simultaneously using three perpendicular loops.

Acknowledgements

No acknowledgement found.

References

1. S. Hidalgo-Tobon, “Theory of gradient coil design methods for magnetic resonance imaging,” concepts Mat. Reson. Part A, vol. 36, no. 4, pp. 223-242, Jul. 2010. 2. J. C. Sharp, S.B. King, “MRI using radiofrequency magnetic field phase gradients,” Magn, Reson, Med., vol. 63, no. 1, pp. 151-161, Jan. 2010. 3. R. Kartäusch, T. Driessel, T. Kampf, T. C. Basse-Lüsebrink, and U. C. Hoelscher, “Spatial phase encoding exploiting the Bloch-Siegert shift effect,“ Magn. Reson. Mat. Phys. Bios. Med., vol. 27, no. 5, pp. 363-371, Oct. 2014. 4. L. I. Sacolick, F. Wiesinger, I. Hancu, and M. W. Vogel, „ B1 mapping by Bloch-Siegert shift,“ Magn. Reson. Med., vol. 63, no. 5, pp. 1315-1322, May 2010. 5. R. T. constable, C. Rogers III, B. Wu, K. Selvaganesan, and G. Galiana, “Design of a novel class of open MRI devices with nonuniform B0, field cycling, and RF spatial encoding,” in Proc. 27th annu. Meet. Int. Soc. Magn. Res. Med., Montreal, Canada, May 10-13, 2019.