Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Twist-Tuned Bilayer Metasurface for 3T MRI

Ingrid Torres, Raquel Rodriguez, Robert W. Laird, Angela R. Laird, Alex Krasnok

Abstract

Magnetic resonance imaging (MRI) can see deep inside the body without ionizing radiation, but image quality depends strongly on how well the radio-frequency field is controlled. Passive resonant pads and metasurfaces can help, yet they often lose their tuning when they are placed next to water-rich tissue or tissue-like materials. Here we show a simple way to bring such a device back into tune. We built a bilayer metasurface made of two aluminum wire arrays. One layer can rotate relative to the other, and the gap between the two layers can also be adjusted. Bench measurements show that adding a controlled water load shifts the resonance to lower frequency by about \SIrange{4.2}{11.4}{\mega\hertz}. Rotating the layers shifts it back by about \SIrange{13.2}{14.9}{\mega\hertz}, which is much stronger than changing the gap alone. One loaded setting lands essentially at the proton frequency used in \SI{3}{\tesla} MRI. In a proof-of-concept scan on a clinical \SI{3}{\tesla} system, the metasurface made internal features in a structured pineapple phantom easier to see than in a substrate-only control. These results show that a passive MRI metasurface can be tuned after fabrication and retuned under load using geometry alone, opening a practical route to simple adjustable RF accessories for MRI.

Twist-Tuned Bilayer Metasurface for 3T MRI

Abstract

Magnetic resonance imaging (MRI) can see deep inside the body without ionizing radiation, but image quality depends strongly on how well the radio-frequency field is controlled. Passive resonant pads and metasurfaces can help, yet they often lose their tuning when they are placed next to water-rich tissue or tissue-like materials. Here we show a simple way to bring such a device back into tune. We built a bilayer metasurface made of two aluminum wire arrays. One layer can rotate relative to the other, and the gap between the two layers can also be adjusted. Bench measurements show that adding a controlled water load shifts the resonance to lower frequency by about \SIrange{4.2}{11.4}{\mega\hertz}. Rotating the layers shifts it back by about \SIrange{13.2}{14.9}{\mega\hertz}, which is much stronger than changing the gap alone. One loaded setting lands essentially at the proton frequency used in \SI{3}{\tesla} MRI. In a proof-of-concept scan on a clinical \SI{3}{\tesla} system, the metasurface made internal features in a structured pineapple phantom easier to see than in a substrate-only control. These results show that a passive MRI metasurface can be tuned after fabrication and retuned under load using geometry alone, opening a practical route to simple adjustable RF accessories for MRI.
Paper Structure (7 sections, 4 equations, 4 figures, 2 tables)

This paper contains 7 sections, 4 equations, 4 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Results and Discussion
  3. Bench measurements identify the loading-induced shift
  4. Gap trims the resonance, whereas rotation recovers the band
  5. A coupled-resonator picture explains the tuning hierarchy
  6. A proof-of-concept 3T scan shows an imaging consequence
  7. Conclusions

Figures (4)

  • Figure 1: Bilayer wire metasurface used in this study. (a) Conceptual drawing of the device. The two wire layers are separated by an adjustable gap $G$ and rotated by a relative in-plane angle $\phi$. The inset shows the intended placement of the pad beneath the sample inside the MRI bore. (b) Photograph of the prototype used for the bench and MRI measurements.
  • Figure 2: Representative calibrated reflection spectra at different interlayer gaps. Normalized versions of calibrated $S_{11}$ traces are shown for Family C, the gap-sweep family that remains nearest the proton band after loading, at gaps of (a) 0mm, (b) 10mm, (c) 20mm, and (d) 30mm. In every panel, the water-loaded state shifts the tracked resonance to lower frequency than the unloaded state.
  • Figure 3: Rotation-driven coarse tuning at the minimal-gap setting. Normalized versions of calibrated $S_{11}$ traces are shown for the two indexed rotation states in (a) the unloaded case and (b) the water-loaded case. The tracked resonance shifts by 14.92MHz without the added water load and by 13.24MHz with the added water load, showing that rotation is the dominant tuning variable.
  • Figure 4: MRI proof-of-concept demonstration on a structured, water-rich fruit phantom. Representative images from the matched scan pair acquired with the posterior/spine receive array disabled are shown for (a) the substrate-only control and (b) the metasurface-supported scan at the nominal 10° indexed state. The metasurface condition shows clearer internal radial structure and a sharper boundary between the bright outer region and the darker core.