Development of magnetic nanoparticles tomography in nuclear and medical electronics division

In this article summary of all accomplishments of Nuclear and Medical Electronics Division in the field of Magnetic Nanoparticles Imaging. Magnetic Nanoparticles Imaging is a new tomographic and molecular imaging method that employs superparamagnetic nanoparticles as the tracer. This article includes the most importuned definition regarding this technique, its most interesting features, as well as report about research conducted in the Division in prospect to advance this imaging method in Poland.


Introduction
Magnetic Particle Imaging (MPI) is novel imaging diagnostic technique. This new molecular imaging method was developed by researchers from Philips Research Laboratories: Bernhard Gleich and Jürgen Weizenecker. In 2005 the first results of their works on imaging using magnetic nanoparticles were published in Nature. Since that publication some research institutes commenced investigations on this imaging technique. Currently the leading scientific centers in this field are University of Lubeck (Thorsten M. Buzug) and University of Berkley, California (Patrick W. Goodwill). Division has ambition to became such a pioneer in Poland.
The increased attention regarding this modality is result of unique features characterizing MPI. First of all, initial research results reviled its high sensitivity, comparable in this regard with the best available molecular imaging techniques such as SPECT or PET, combined with high spatial and time resolution, rivaling Magnetic Resonance Imaging. Secondly, MPI is a method simple to implement. However, not as easily scalable. Although the third and most importuned reason is variety of possible applications of MPI. Some of which are quite unique. Magnetic nanoparticles, which are used as a marker in this technique, can fused with biomolecules to label specific of organic substances or tissues. Which in turn allows study of biological and physiological through observation of time and spatial changes of markers density in the subject. MPI can be applied to function imaging of kidneys or liver, examination of blood flow in the coronary arteries, in diagnostic of pulmonary diseases as well as localization of cancers or implanted stem cells. Potentially, if used along with magnetic hyperthermia, the same nanoparticles or even the same equipment can be applied for both diagnostic and therapy of malignant diseases. MPI scanner can be design as typical tomographic scanner similar in appearance to CT scanner or one-sided scanner of similar uses as mini-gamma camera. At last, this technique can be used to research properties of superparamagnetic nanoparticle itself [1].

Theory of Magnetic Particles Imaging
In MPI superparamagnetic nanoparticles are used as a tracer. The core of such particles is usually manufactured from superparamagnetic iron oxide (SPION) (Fig. 1). It is of radius from 10 to 100 nm and coated with biocompatible polymers. Superparamagnetism is form of magnetism in which ferromagnetic material nanoparticles of sufficiently small size and magnetically isolated from one another can be recognized as separate magnetic domains and randomly flip direction of magnetization vectors under the influence of temperature. In this state nanoparticles behavior can be described by Langevin's theory of paramagnetism and their magnetization curve by equation: where particles concentration, magnetic field strength, , magnetic permeability of vacuum, -Boltzman's constant, temperature, magnetic momentum of saturation, equal to: , where is saturation magnetization of magnetite and is diameter of nanoparticle. The most imported features of nanoparticles that can be deducted from this equation are nonlinear magnetization curve, high magnetic moment compere to normal paramagnets and fact that in high enough magnetic field nanoparticle undergo saturation.
Thanks to this properties detection of nanoparticles is possible if tracers containing them are placed in changing magnetic field (preferably sinusoidal) and signal from their magnetization is simultaneously acquired by appropriate receiving coil. Unfortunately, in this case one would acquire both signal from magnetization changes and excitation signal, which is stronger by about six orders of magnitude.

. Dependence of magnetization signal induced in receiving coils on nonlinear magnetization curve (M-H curve). Excitation signal (bottom left), magnetization curve (top left), superparamagnetic nanoparticels magnetization (top right), voltage signal induced in receiving coil (bottom left)
If nanoparticles magnetization curve was liner, separation of this signal would be impossible (Fig. 2). However, as the magnetization curve is nonlinear signal from nanoparticles contains harmonics of excitation frequency (Fig. 3). It is possible than to separate those high frequencies by filtering out excitation frequency (broadband method) or by measurement of each harmonic separately (narrowband method). Through analysis of this harmonics signals for example particles concentration can be estimated.
Localization of the tracer can be achieved by application of gradient magnetic field, which cause nearly all nanoparticles to undergo saturation thus preventing them from contributing in signal generation, except those in small volume near the zero magnetization field strength, called field free point (FFP). In such circumstances magnetization signal induced in receiving coil originates only from particles in FFP [3]. To scan whole imaged volume FFP can be moved. There are two known procedures of scanning using this method of localizationfrequency mixing method (excitation frequency much higher than movement of FFP) and driving field method (high frequency movement of FFP used as exciting field) [2]. Time resolution of imaging is highly dependent on applied scanning process.

Research conducted in Division
In Nuclear And Medical Electronics Division research on model of tomographic scanner for nanoparticles imaging are being carried out since 2011. In the wake previous studies, conducted under statutory works, one dimensional nanoparticles scanner was created. Presently system for nanoparticles spectroscopy is being developed. Furthermore, numerical calculation MATLAB toolbox is being written that will allow simulation of MPI measurement system, generation of simulated measurement result of such systems and reconstruction of MPI images from both artificial and real data.

MPI scanner developed in Nuclear And Medical Electronics
Division allows one dimensional imaging of nanoparticles concentration [4] (Fig. 4 and 5). It is designed for research of small objects. Current acquisition system employs narrowband detection method using lock-in amplifier and mix frequency method for field free point displacement. Particles are excited with signal of 10 kHz frequency and 8 mT of amplitude. Gradient coils generate field of about 1 T/m strength. Field of view of the scanner has about 30 mm diameter. Using this setup first real measurements were conducted [7] (Fig. 7). The phantoms with the chambers for water solution of nanoparticles with 0.5 mol/L concentration (Fig. 6) were used in the experiments. The phantoms were made using 3D printing technology.

Magnetic Nanoparticles Spectroscopy
Magnetic Particles Spectroscopy (MPS) allows measurement of harmonic spectra of magnetization signal of magnetic nanoparticles. It is used for research of properties of nanoparticles, for example particle diameter or magnetization curve estimation. Magnetic particles spectrometer may can be considered a zero dimensional MPI scanner. To acquire the spectra of magnetization signal harmonics only detection of magnetization signal is necessary. In this case gradient coils are irrelevant. Signal from whole volume is acquired and afterwards amplitude and phase of each harmonic is measured or calculated by Discreet Fourier Transform from signal. Set of amplitudes and phase angles of all harmonics in function of frequency is called magnetic particle spectrum and is unique for specific nanoparticles batch. First trial measurements of amplitude and phase angle spectra of nanoparticles using modified MPI scanner model (Fig. 8) were conducted in Division ( Fig. 9 and 10). Development of new magnetic particles spectroscope is planned in the near future.

Numerical Simulation Toolbox
The last important project connected to MPI developed in Division is work on program for numerical calculations. MPIsim software is written in MATLAB. When completed this toolbox will be ideal scientific help for designing new models of the MPI scanner or coils setups or for verification of results of real measurements. Its first module will allow simulation of magnetic field generated by coils of the scanner [5,6], including gradient field, driving field and excitation field ( Fig. 11 and 12). Second part will be used for simulation of 1D/2D/3D MPI measurements ( Fig. 15), which theoretical with allow numerical estimation of system function of analyzed scanner. Last function of the program will be reconstruction of MPI images and particles concentration distribution. Using this toolbox it will also be possible to numerical calculation of magnetic particles spectra (Fig. 13) and estimation of particle magnetization curve, from both simulated and real data (Fig. 14). This feature will allow not only estimation of magnetic parameters of nanoparticles but also may lead to better theory describing superparamagnets behavior.

Conclusions
Magnetic Particles Imaging is a very promising molecular imaging technique with wide range of potential applications. Its progress will surly contribute to expansion of knowledge in both fields of electronics and medicine. Due to this rezone it became one of the main topics of interest in the Nuclear And Medical Electronics Division. Past results motivate us to continue work in this field. Probable next step will be extension of functionality of current MPI scanner model to 2D or 3D imaging, which in turn will allow further research in this new imaging technique.