The present application is the US national stage of International Application PCT/CN2014/075302 filed on Apr. 14, 2014 which, in turn, claims priority to Chinese Patent Application CN2014101286595 filed on Apr. 1, 2014.
The invention relates to the field of nano measurement technology, and more particularly to a method for one-dimensional in-vivo temperature imaging based on paramagnetic property of magnetic nanoparticles.
In-vivo temperature imaging refers to temperature imaging of tissues of a complete and survival individual. In the biomedical field such as hyperthermia cancer therapy, as it is so difficult to obtain temperature field distribution information in vivo accurately that many medical treatments cannot be used effectively. At present, in-vivo temperature measurement is categorized as invasive measurement and non-invasive measurement. Invasive measurement is simple and is able to monitor temperature of a lesion directly with high accuracy in real time. However, it is highly traumatic, the probe insertion tends to cause transfer of infected cells, the radiation of a heating source have affect on the probe directly may leading to decrease of measuring accuracy, and the measured temperature data is a point temperature which cannot construct temperature field distribution of the whole lesion. Meanwhile, non-invasive measurement can avoid wound infection or proliferation of cancer cells effectively and realize real-time imaging of an in-vivo temperature field with comparatively high accuracy, which enables it to have widely potential applications in the biomedical field.
Non-invasive measurement mainly includes infrared temperature measurement, ultrasonic temperature measurement, NMR temperature measurement and magnetic nanoparticle remote temperature measurement. Infrared temperature measurement measures temperature of a measured object according to infrared radiation intensity thereof, which is applicable for surface temperature measurement of objects instead of temperature field measurement deeply in tissues, and is vulnerable to emissivity of an object and aerosol particles. The key of ultrasonic temperature measurement is to measure propagation time of an ultrasonic accurately, which requires to measure acoustic characteristics and temperature characteristics of tissues in advance. However, temperature characteristics of tissues are instable and differ greatly therebetween, which affects temperature measurement significantly. NMR temperature measurement features high price and limited space resolution and temperature resolution, and is unfavorable for widespread applications. Non-invasive temperature field imaging by magnetic nanoparticles may overcome the above shortcomings, by which in-vivo temperature imaging may be realized thereby monitoring a hyperthermia cancer therapy process in real-time so as to make adjustments timely and effectively.
However, current non-invasive in-vivo temperature measurement method based on magnetic nanoparticles can only realize single-point temperature measurement instead of obtaining temperature field distribution deeply in tissues. Besides, accuracy of temperature measurement is affected by concentration distribution of magnetic nanoparticles deeply in tissues. Therefore, it is an urgent problem to be resolved in the field of magnetic nanoparticle hyperthermia cancer therapy that developing a method capable of realizing in-vivo temperature field imaging without knowing concentration distribution of magnetic nanoparticles.
In view of the above-mentioned problems, it is an objective of the invention to provide a method for in-vivo temperature imaging based on paramagnetic property of magnetic nanoparticles, so as to detect a one-dimensional in-vivo temperature field accurately without knowing concentration distribution of magnetic nanoparticles.
To achieve the above objective, in accordance with one embodiment of the present invention, there is provided a method for in-vivo temperature imaging by using magnetic nanoparticle, comprising steps of:
for the DC magnetic field in step (2) while keeping the AC excited magnetic field in step (2) unchanged, where x represents position, x1 is the starting position of a DC gradient magnetic field f(x) in the one-dimensional space, and Δx is the width of the DC gradient magnetic field f(x), and collecting the AC magnetization signal of the magnetic tracer agents again thereby obtaining amplitudes B1, B3 . . . B2j-1 of odd harmonics thereof;
thereby obtaining in-vivo temperature
of the area of [x1, x1+Δx], where c=NMS, N is concentration of the magnetic nanoparticles, Ms is effective atomic magnetic moment of the magnetic tarcer agent, k0 is the Boltzmann's constant, the magnetization of the magnetic nanoparticles is described by the Langevin's function, c·f2j-1(y, z(rk)) is the amplitude of a (2j−1)th harmonic obtained by finite terms of Taylor series expansion of the Langevin's function, z(rk) is the intensity of the DC magnetic field at a kth discrete point rk in the area of [x1, x1+Δx], and m is the number of discrete points in the area of [x1, x1+Δx]; and
In a class of this embodiment, the starting position of the DC gradient magnetic field in the measured one-dimensional space is changed by moving a DC magnetic field generating device or by changing current of the exciting coil.
In a class of this embodiment, in steps (2) and (3), amplitudes of odd harmonics of the AC magnetization signal of the magnetic nanoparticles are detected by a digital phase-sensitive detection method or by a least squares system parameter identification method.
In accordance with another embodiment, there is provided a magnetic nanoparticle temperature imaging system, comprising:
to the one-dimensional space where the magnetic tracer agents are positioned, where x represents position, x1 is the starting position of a DC gradient magnetic field f(x) in the one-dimensional space, and Δx is the width of the DC gradient magnetic field f(x);
thereby obtaining in-vivo temperature
of the area of [x1, x1+Δx], where c=NMS, N is concentration of the magnetic nanoparticles, Ms is effective atomic magnetic moment of the magnetic nanoparticle, k0 is the Boltzmann's constant, the magnetization of the magnetic nanoparticles is described by the Langevin's function, c·f2j-1(y, z(rk)) is the amplitude of a (2j−1)th harmonic obtained by finite terms of Taylor series expansion of the Langevin's function, z(rk) is the intensity of the DC magnetic field at a kth discrete point rk in the area of [x1, x1+Δx], m is the number of discrete points in the area of [x1, x1+Δx], and the number of the harmonics j≥2.
Advantages of the invention comprise:
According to the present invention, firstly, the DC magnetic field and the AC magnetic field are applied simultaneously to the area where magnetic nanoparticles are positioned, and the AC magnetization signal of the magnetic nanoparticles is collected thereby obtaining amplitudes of harmonics thereof. Secondly, the combined DC magnetic field and the AC magnetic field are applied simultaneously to the area where the magnetic nanoparticles are positioned, and the AC magnetization signal of the magnetic nanoparticles is collected thereby obtaining amplitudes of harmonics thereof. Thirdly, amplitude variations of odd harmonics are obtained by amplitudes of odd harmonics in the second round minus amplitudes of odd harmonics in the first round. Under the above two magnetic excitations, the amplitude variations of odd harmonics in the whole measured area contain only signals of magnetic nanoparticles in an area with a width of Δx, namely only being relevant to temperature and concentration of the magnetic nanoparticles in the area with the width of Δx. Therefore, it is the concentration distribution of magnetic nanoparticles in an area with a width of Δx instead of in the whole one-dimensional area which affects temperature measurement. As Δx is very small, the concentration of magnetic nanoparticles in an area with a width of Δx may be considered as a constant, and therefore a whole one-dimensional space can be divided into multiple small areas with a width of Δx by applying different excited magnetic fields to magnetic nanoparticles, so that temperature imaging of the one-dimensional space is transformed to single-point temperature measurement of each of the small areas. The present invention is able to obtain temperature field of a one-dimensional space accurately and rapidly without knowing concentration of magnetic nanoparticles, and is especially applicable for temperature imaging of thermal motion at bio-molecular level. Simulation experiments show that it has a measurement error less than 0.79 K under a noise environment with an SNR of 80 dB.
For clear understanding of the objectives, features and advantages of the invention, detailed description of the invention will be given below in conjunction with accompanying drawings and specific embodiments. It should be noted that the embodiments are only meant to explain the invention, and not to limit the scope of the invention.
Referring to
(1) positioning a magnetic tracer agents at a measured object;
(2) collecting amplitudes of odd harmonics for the first time;
(21) applying a DC magnetic field Hdc=b and an AC excited magnetic field H(t)=H0 sin(2πft) simultaneously to an area where the magnetic tracer agent is positioned, where a total magnetic field H(t)=H0 sin(2πft)+Hdc, H0 is the amplitude of the AC excited magnetic field, f is frequency thereof, and t represents time, as shown in
As only finite terms of Taylor series expansion of the Langevin's function is used for in-vivo temperature calculation in following steps, considering truncation errors of the model, both amplitude of the AC excited magnetic field H0 and that of the DC magnetic field b should be small. However, if an applied magnetic field were too weak, the background noise would become greater comparatively, which decreases the SNR and is unfavorable for signal extraction. Therefore, it is of great importance to choose the intensity of the AC excited magnetic field and that of the DC magnetic field reasonably, and amplitudes thereof may be adjusted according to experimental results.
(22) collecting the AC magnetization signal of the magnetic tracer agents in the measured area;
(23) detecting amplitudes A1, A3 . . . A2j-1 of odd harmonics of the AC magnetization signal, where the number of the harmonics j≥2;
(3) collecting amplitudes of odd harmonics for the second time;
(31) removing the constant DC magnetic field, and applying a combined DC magnetic field
to the area where the magnetic tracer agents are positioned while keeping the AC excited magnetic field H(t)=H0 sin(2πft) unchanged, where x represents position, x1 is the starting position of a DC gradient magnetic field f(x) in the one-dimensional space, and Δx is the width of the DC gradient magnetic field f(x), a DC magnetic field z is a linear or nonlinear gradient magnetic field in an area of Δx, z=b and z=−b respectively on two sides of the gradient magnetic field, and a total magnetic field H(t)=H0 sin(2πft)+Hdc′, as shown in
It should be noted that the intensity of the magnetic field on two sides of the DC gradient magnetic field should equal that of the constant DC magnetic field in step (2), and the amplitude and frequency of the AC excited magnetic field should keep unchanged. Besides, too great magnetic field intensity b may lead to changes of positive-negative signs of odd harmonics of the magnetic nanoparticles under the gradient magnetic field, which brings errors to in-vivo temperature measurement.
(32) collecting an AC magnetization signal of the magnetic tracer agents in the measured area;
(33) detecting amplitudes B1, B3 . . . B2j-1 of odd harmonics of the AC magnetization signal, where the number of the harmonics j≥2;
(4) calculating variations S1, S3 . . . S2j-1 between amplitudes B1, B3 . . . B2j-1 of the odd harmonics in step (3) and those A1, A3 . . . A2j-1 of the odd harmonics in step (2);
(5) calculating in-vivo temperature according to equations between the amplitude variations of odd harmonics and in-vivo temperature;
Magnetization of magnetic nanoparticles may be described by the Langevin's function as follows:
where
H is a magnetic field applied to the magnetic nanoparticles, N is concentration of the magnetic nanoparticles, Ms is effective atomic magnetic moment of the magnetic nanoparticles, k0 is the Boltzmann's constant, and T is absolute temperature of a measured object.
When H(t)=H0 sin(2πft)+z, namely an AC magnetic field and a DC magnetic field are applied simultaneously to the magnetic nanoparticles, the amplitudes of harmonics may be derived approximately by finite terms of Taylor series expansion.
Amplitude of the first harmonic is as follows:
which is simplified as Amp1=c·f(y,z);
amplitude of the second harmonic is as follows:
which is simplified as Amp2=c·f(y,z);
amplitude of the third harmonic is as follows:
which is simplified as Amp3=c·f3(y,z); and
amplitude of the nth harmonic is simplified as Ampn=c·fn(y,z),
where c=NMs,
N is concentration of the magnetic nanoparticles, Ms is effective atomic magnetic moment of the sample, k0 is the Boltzmann's constant, T is absolute temperature of the measured object, z is intensity of the DC magnetic field, and H0 is an amplitude of the AC magnetic field.
It can be derived by mathematical induction that odd harmonics are even functions with respect to the DC magnetic field z, and even harmonics are odd functions with respect to the DC magnetic field z, namely f2j-1(z)=f2j-1(−z) and f2j(−z)=−f2j(z), where j≥1. As shown in
In order to further illustrate the present invention, amplitudes of harmonics changing with an applied excited magnetic field is analyzed in combination with
Regarding the measured area as temperature points T1, T2 . . . Ti . . . Tn, each of which takes an area with a width equal to that of the gradient magnetic field of the combined DC magnetic field Δx. As Δx is very small, concentration of magnetic nanoparticles in an area of Δx may be considered as a constant.
When a magnetic field H(t)=H0 sin(2πft)+Hdc is applied to the measured area, according to the Langevin's function, amplitude of a first harmonic of magnetic nanoparticles in an area of Δx with a same temperature Ti is constant, as shown in
When a magnetic field H(t)=H0 sin(2πft)+Hdc′ is applied to the measured area, and the gradient magnetic field with a width of Δx is located at a temperature point Ti, as a first harmonic is an even function with respect to a DC magnetic field z, namely f1(z)=f1(−z), absolute values and signs of first harmonics of magnetic nanoparticles under the DC magnetic field where z=±b keep unchanged, and only the amplitude of a first harmonic of magnetic nanoparticles in an area where the DC magnetic field z is a linear or nonlinear gradient magnetic field change. Amplitudes of first harmonics of magnetic nanoparticles at the temperature points in this case are shown in
Under the above two kinds of magnetic fields, amplitude variation between first harmonics of the whole measured area may be represented by S1,i=B1−A1. In combination with
Based on the above analysis, amplitude variation between first harmonics in an area of [x1, x1+Δx] may be represented by:
where C=NMs, N is concentration of the magnetic nanoparticles, Ms is effective atomic magnetic moment of the magnetic nano sample, and c·f1(y,z(r)) represents amplitude of a first harmonic at rϵΩ, where Ω=[x1, x1+Δx].
Similarly, amplitude variation between third harmonics in an area of [x1, x1+Δx] may be represented by:
where c·f3(y,z(r)) represents amplitude of a third harmonic at rϵΩ, where Ω=[x1, x1+Δx].
Similarly, amplitude variation between (2j−1)th harmonics of the whole measured area may be represented by:
where c·f2j-1 (y,z(r)) represents amplitude of a (2j−1)th harmonic at rϵΩ, where Ω=[x1, x1+Δx] and j≥2.
Overall, relationship between amplitude variations of odd harmonics of the magnetic nanoparticles and in-vivo temperature is established.
By discretizing the area of Ω where the ith temperature point is located into m segments, equations between amplitude variations of odd harmonics S1,i . . . S2j-1,i and in-vivo temperature are as follows:
ci and yi at the ith temperature point can be obtained by resolving the discrete equations thereby obtaining in-vivo temperature at the ith temperature point
and its corresponding concentration of magnetic nanoparticles
When the number of detected odd harmonics j=2, ci and yi may be obtained by resolving the equations directly thereby obtaining in-vivo temperature at the ith temperature point
When the number of detected odd harmonics j≥3, ci and yi may be obtained by resolving overdetermined equations using algorithms such as the least square method thereby obtaining in-vivo temperature at the ith temperature point
(6) changing the starting position x1 of the DC gradient magnetic field in the one-dimensional space by mechanical scanning or excited magnetic field scanning, so that the DC gradient magnetic field with the width of Δx moves to a next area, and recording a corresponding coordinate; and returning to step (2) if detection of the whole measured area is not completed, and ending the process if detection of the whole measured area is completed;
In mechanical scanning, the gradient magnetic field is enabled to scan the whole one-dimensional space by moving a magnetic field generating device via a motor or by moving the magnetic nanoparticles directly, and corresponding coordinates are recorded.
In excited magnetic field scanning, the gradient magnetic field is enabled to scan the whole one-dimensional space by methods such as changing current of an exciting coil, and corresponding coordinates are recorded.
When the scanning speed is high enough or the temperature changes slowly, A1 detected in the first round is almost constant, and scanning of the whole imaging area may be completed by returning to step (3) directly instead of returning to step (2), where A1 detected in the first round is subtracted from a signal obtained in each scanning, which may decrease temperature imaging time without affecting the accuracy of temperature measurement.
When distribution of magnetic nanoparticles in the one-dimensional space is comparatively uniform, amplitude differences of second harmonics may be used to locate coordinates of the temperature points. By drawing trend of amplitude differences of second harmonics, it may be inferred that the projection of the amplitude differences of the second harmonics of each temperature point onto the coordinate axis corresponds to its coordinate.
1. Simulation Model and Test Explanation
In order to study the effectiveness and feasibility of the method for in-vivo temperature imaging, simulation data containing noise are used in the simulation to test the method. Effective magnetic moment Ms of particles of a agent applied in the simulation is measured as 8.5×10−19 (the effective magnetic moment is determined by the type of the tracer agents), and the width of the DC gradient magnetic field Δx=5 mm, which moves 1 mm each time by simulating mechanical movement. Considering truncation errors of an approximate model obtained by the first eight terms of Taylor series expansion of the Langevin's function, in the simulation, amplitude of the AC magnetic field H0=50 Oe and frequency thereof is 160 Hz, the amplitude of the DC magnetic field is 30 Oe, and noise is added to AC magnetization signals by a awgn function in MATLAB. Following tests are conducted for different purposes:
1. Simulation Model and Test Explanation
In order to study effectiveness of the method for magnetic nanoparticle temperature imaging when the DC gradient magnetic field is nonlinear, effective magnetic moment Ms of particles of a agent applied in the simulation is measured as 8.5×10−19 (the effective magnetic moment is determined by the type of the agents), the imaging range is 40 mm, the width of the DC gradient magnetic field Δx=20 mm, which moves 1 mm each time by simulating mechanical movement. A nonlinear DC gradient magnetic field is applied in the simulation. Amplitude of the AC magnetic field H0=60 Oe and frequency thereof is 1.6 kHz, magnetization of the DC magnetic field is 30 Oe, and noise is added to AC magnetization signals by a awgn function in MATLAB. Concentration distribution of the magnetic nanoparticles is uniform, noise with an SNR of 90 dB or 80 dB is added in temperature imaging, test temperature is in a range of 300K˜310K, and simulation results are shown in
2. Simulation Results
As a result, accuracy, stability and reproducibility of the method for magnetic nanoparticle temperature imaging are guaranteed, which provides a reliable method for accurate, rapid and noninvasive temperature field imaging of a living body under complicated circumstances.
While preferred embodiments of the invention have been described above, the invention is not limited to disclosure in the embodiments and the accompanying drawings. Any changes or modifications without departing from the spirit of the invention fall within the scope of the invention.
Number | Date | Country | Kind |
---|---|---|---|
2014 1 0128659 | Apr 2014 | CN | national |
Filing Document | Filing Date | Country | Kind |
---|---|---|---|
PCT/CN2014/075302 | 4/14/2014 | WO | 00 |
Publishing Document | Publishing Date | Country | Kind |
---|---|---|---|
WO2015/149388 | 10/8/2015 | WO | A |
Number | Name | Date | Kind |
---|---|---|---|
8498837 | Liu | Jul 2013 | B2 |
9301693 | Liu | Apr 2016 | B2 |
9534963 | Liu | Jan 2017 | B2 |
Number | Date | Country | |
---|---|---|---|
20160296175 A1 | Oct 2016 | US |