This article has Open Peer Review reports available.
Theory, simulation and experimental results of the acoustic detection of magnetization changes in superparamagnetic iron oxide
© Gleich et al; licensee BioMed Central Ltd. 2011
Received: 26 April 2010
Accepted: 29 June 2011
Published: 29 June 2011
Magnetic Particle Imaging is a novel method for medical imaging. It can be used to measure the local concentration of a tracer material based on iron oxide nanoparticles. While the resulting images show the distribution of the tracer material in phantoms or anatomic structures of subjects under examination, no information about the tissue is being acquired. To expand Magnetic Particle Imaging into the detection of soft tissue properties, a new method is proposed, which detects acoustic emissions caused by magnetization changes in superparamagnetic iron oxide.
Starting from an introduction to the theory of acoustically detected Magnetic Particle Imaging, a comparison to magnetically detected Magnetic Particle Imaging is presented. Furthermore, an experimental setup for the detection of acoustic emissions is described, which consists of the necessary field generating components, i.e. coils and permanent magnets, as well as a calibrated microphone to perform the detection.
The estimated detection limit of acoustic Magnetic Particle Imaging is comparable to the detection limit of magnetic resonance imaging for iron oxide nanoparticles, whereas both are inferior to the theoretical detection limit for magnetically detected Magnetic Particle Imaging. Sufficient data was acquired to perform a comparison to the simulated data. The experimental results are in agreement with the simulations. The remaining differences can be well explained.
It was possible to demonstrate the detection of acoustic emissions of magnetic tracer materials in Magnetic Particle Imaging. The processing of acoustic emission in addition to the tracer distribution acquired by magnetic detection might allow for the extraction of mechanical tissue parameters. Such parameters, like for example the velocity of sound and the attenuation caused by the tissue, might also be used to support and improve ultrasound imaging. However, the method can also be used to perform imaging on its own.
Magnetic Particle imaging (MPI) is a novel method for medical imaging. It can be used to measure the local concentration of a tracer material based on iron oxide nanoparticles. While the resulting images show the distribution of the tracer material in phantoms or anatomic structures of subjects under examination, no information about the circumjacent tissue is being acquired. To expand MPI into the detection of soft tissue properties, a new method is proposed, which detects acoustic emissions caused by magnetization changes in superparamagnetic iron oxide. These signals may allow for the in vivo determination of soft tissue parameters, such as attenuation and velocity of sound. This article presents an introduction to acoustically detected MPI, a comparison to magnetically detected MPI, as well as first simulations and experimental results.
Magnetic Particle Imaging (MPI) determines the presence of iron oxide nanoparticles from their nonlinear magnetization behavior . Such nanoparticles are well known as superparamagnetic iron oxide (SPIO). They are available as a clinically approved contrast agent for liver examinations in magnetic resonance imaging (MRI)  and are usually administered into the bloodstream via intravenous injection. In the context of MPI, such nanoparticles are used as a tracer material. To examine the magnetization properties of the tracer, MPI deploys a field free point (FFP) realized in a strong gradient field, called the selection field. This field free point is moved across the sample or phantom to gather information about the distribution and concentration of the tracer material. Using additional homogeneous magnetic fields, called drive fields, this movement can be realized in a very fast manner, leading to dynamic 2D  or 3D  imaging. The very signal generation takes place when the FFP passes a point in space containing tracer material and the magnetization of the nanoparticles is changed. This change in magnetization results in an induced signal that can be picked up in a recording coil. Depending on the steepness of the magnetization curve and the strength of the selection field, information about the distribution and concentration of the tracer material can be obtained with high spatial resolution. To reconstruct an image from this information, the properties of the instrumentation, e.g. coil sensitivities, and the characteristics of the tracer have to be determined. This information is usually encapsulated into the so-called system function. Practically, the system function can be determined by measuring the MPI signal generated by a small sample at a sufficient number of points within the field of view. In a recent paper  it was shown, that for the one-dimensional case the system function can be composed of Chebyshev polynomials of the second kind, assuming idealized particles.
with α being an abbreviation for μ/kBT and B being the magnitude of the field in x-direction, i.e. B = |Bx|, The value of μ depends on the particle volume and thus on the particle diameter. Gx is dBx/dx, the gradient of the Bx component, which can assumed to be constant for the region of interest.
with A being the detector area and Δf being the bandwidth. is the real part of the sound impedance, which is given by the product of density and speed of sound, as long as the damping is negligible. Neglecting the damping is justified for all tissues that allow a penetration of the sound wave of more than one wavelength. For a square millimeter sized detector and material properties of water at 310 K, the noise pressure is . The signal generation by magnetic particles can be estimated by assuming a cube of 1 mm3 filled with 0.5 mol(Fe)/l present as magnetite, leading to a saturation magnetization of approximately 4.4 mT μ0-1. In a gradient field of 5 T μ0-1/m this translates into a force of 18 μN or a pressure of 18 Pa and results in a signal-to-noise ratio of 1.1 × 105 for a measuring time of one second. A reasonable detection limit would be 22 μmol(Fe)/l, which corresponds to five standard deviations and should be detectable within one second. For comparison, a typical dosage of the applied material in MRI is 8 to 40 μmol(Fe)/l  with a detectable local concentration of about 50 μmol(Fe)/l . In MPI, an ultimate detection limit of about 20 nmol(Fe)/l has been estimated .
There are some remarkable differences between simulation and experiment. Most notably, the symmetry is broken in the experimental data. This may be attributed to a small tilt of the sample with respect to the vertical direction. Additionally, the membrane can exhibit an asymmetric stiffness, caused by the gluing process. Together with off-axis components of the force acting on the membrane, this can break the symmetry. A background signal can be the cause of symmetry breaking for the second harmonic. This background signal can change the relative height of the peaks in figure 5, 6, and 7, depending on its relative phase. The presence of such a background in the experiment is evident, as the measured amplitude never drops to zero.
Since the microphone used for the measurements is calibrated, the recorded signal can be transformed to sound pressure values. The fundamental frequency shows a peak amplitude of approximately 7 mPa. From the amount of magnetic material and the diameter of the tube, a pressure of 2 Pa is expected, i.e. a factor 300 higher. This discrepancy might be attributed to the fact that the particles in Resovist do not reach saturation at 2.5 mT μ0-1 , as most of the particles seem to be very small, i.e. below ten nm . As a consequence, only a small fraction of the particles contribute to signal generation. Additionally, the membrane itself might be a cause for the discrepancy. During a change in pressure, the tension of the membrane counteracts the force on the membrane.
Using an experimental magnetic particle imaging setup, it was possible to detect acoustic emissions of a magnetic tracer material. A theoretical model was applied to simulate and predict the spatial distribution of the signal. The experimentally detected sound amplitudes were in good agreement with the simulations.
For medical applications, the information gained from acoustic detection might be used to generate an additional contrast to magnetically detected MPI images, which only show the distribution and concentration of the tracer. A well-defined sound wave is emitted from the position of the field free point (FFP) and travels through the tissue to the detector. During this propagation, the wave is attenuated, scattered, and delayed. As a consequence, it should be possible to image mechanical tissue properties, such as sound velocity and attenuation, which are not directly accessible to ultrasound imaging.
As an additional benefit, these tissue properties would be useful to improve image quality in ultrasound imaging by eliminating multi-scattering artefacts. For these types of applications to be feasible, it would be necessary to realize an efficient mechanism of sound generation from the FFP. If the experiment described herein would be performed in tissue, the signal would probably be very low, as the pressure rapidly decreases with distance from the FFP. The main reason for this is destructive interference between positive and negative pressure waves, as observed in box-less loudspeaker designs. This effect may be circumvented by moving the FFP faster than the local sound velocity, analogous to the Cerenkov Effect  for charged particles which are moving faster than light in the given medium.
This work was financially supported by the German Federal Ministry of Education and Research (BMBF) under the grant number FKZ 13N9079.
- Gleich B, Weizenecker J: Tomographic imaging using the nonlinear response of magnetic particles. Nature. 2005, 435: 1214-1217. 10.1038/nature03808.View ArticlePubMedGoogle Scholar
- Chikazumi S: Physics of Magnetism. 1964, John-Wiley, New YorkGoogle Scholar
- Gleich B, Weizenecker J, Borgert J: Experimental results on fast 2D-encoded magnetic particle imaging. Phys Med Biol. 2008, 53: N81-N84. 10.1088/0031-9155/53/6/N01.View ArticlePubMedGoogle Scholar
- Weizenecker J, Gleich B, Rahmer J, Dahnke H, Borgert J: Three-dimensional real-time in vivo magnetic particle imaging. Phys Med Biol. 2009, 54: L1-L10. 10.1088/0031-9155/54/5/L01.View ArticlePubMedGoogle Scholar
- Rahmer J, Weizenecker J, Gleich B, Borgert J: Signal encoding in magnetic particle imaging: properties of the system function. BMC Medical Imaging. 2009, 9: 4-10.1186/1471-2342-9-4.View ArticlePubMedPubMed CentralGoogle Scholar
- Weizenecker J, Borgert J, Gleich B: A simulation study on the resolution and sensitivity of Magnetic Particle Imaging. Phys Med Biol. 2007, 52 (21): 6363-6374. 10.1088/0031-9155/52/21/001.View ArticlePubMedGoogle Scholar
- Farlow R, Hayward G: The minimum signal force detectable in air with a piezoelectric plate transducer. Proc R Soc Lond A. 2001, 467: 2741-2755.View ArticleGoogle Scholar
- Patient information leaflet for Resovist. 2004, Schering AGGoogle Scholar
- Dahnke H, Schäffter T: Limits of detection of SPIO at 3.0 T Using T2* Relaxometry. Magn, Reson Med. 2005, 53: 1202-1206. 10.1002/mrm.20435.View ArticleGoogle Scholar
- Lawaczek R, Bauer H, Frenzel T, Hasegawa M, Ito Y, Kito K, Miwa N, Tsutsui H, Vogler H, Weinmann HJ: Magnetic iron oxide particles coated with carboxydextran for parenteral administration and liver contrasting. Pre-clinical profile of SH U555A. Acta Radiol. 1997, 38 (4 Pt 1): 584-597.Google Scholar
- Cerenkov PA, P A: Visible Radiation Produced by Electrons Moving in a Medium with Velocities Exceeding that of Light. Phys Rev. 1937, 52: 378-381. 10.1103/PhysRev.52.378.View ArticleGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2342/11/16/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.