Two-color magnetic nanoparticle imaging method based on system matrix real-imaginary part separation

Abstract

The invention relates to a two-color magnetic nanoparticle imaging method based on system matrix real-imaginary part separation, which comprises the following steps: respectively moving two different types of magnetic nanoparticle samples to a central zero magnetic field point, applying an excitation magnetic field and a scanning magnetic field, and obtaining magnetization signals; constructing system matrixes A1 and A2 based on the magnetization signal of the magnetic nanoparticle sample; measuring a to-be-measured object formed by the two kinds of particles to obtain a measurement magnetization signal U; splicing the system matrixes A1 and A2 of the two different types of magnetic nanoparticles and the measurement magnetization signal U to obtain a new system matrix A'and a new magnetization signal U '; the new system matrix A'is combined with the new magnetization signal U ', particle concentration distribution of two different kinds of magnetic nanoparticles is obtained, the concentration distribution of the two kinds of magnetic nanoparticles is sent into different color channels respectively, and the concentration distribution is displayed in different colors, and the method has important significance on application of the magnetic nanoparticle imaging technology in clinical medicine.

Description

Technical Field

[0001] The invention relates to the technical field of magnetic nanoparticle imaging, in particular to a dual-color magnetic nanoparticle imaging method based on real-imaginary part separation of a system matrix. Background Art

[0002] Magnetic Particle Imaging (MPI) is a new type of medical imaging technology. It measures the nonlinear magnetization response signal of magnetic nanoparticles and inverts the multi-parameter information such as the concentration of magnetic nanoparticles in the imaging space, the temperature and viscosity of the surrounding medium. This technology was proposed by German scientists in the early 21st century and has been in a rapid development stage in the past decade. Compared with existing medical imaging technologies such as CT, ultrasound, nuclear magnetic resonance (MRI) and PET, MPI has the characteristics of high sensitivity, high spatiotemporal resolution, no tissue depth limitation and no radioactivity. It has great application potential in medical imaging fields such as cell tracking, tumor detection, blood pool angiography and precision magnetic hyperthermia.

[0003] Two-color magnetic nanoparticle imaging refers to a single scan of the detection area to achieve two-color quantitative visualization of different particles in the same field of view. The combination of magnetic nanoparticles with different characteristics and specific biological molecules is expected to achieve quantitative detection of multiple molecular markers in vivo, providing new ideas for biomedical imaging. However, due to the crosstalk between the two particle signals during the reconstruction process, the quantitative imaging accuracy of the two ions is low; on the other hand, the elements in the system matrix and the measurement signal are all complex numbers, and the complex values ​​need to be projected into the real space in each iteration, resulting in a time-consuming and slow reconstruction process.

[0004] Therefore, the present invention urgently needs a dual-color imaging method that can effectively reduce signal crosstalk between particles and speed up image reconstruction, improve the detection accuracy and imaging speed of dual-color magnetic nanoparticles, and promote the application of magnetic nanoparticles in clinical medicine. Summary of the invention

[0005] In order to solve the problems existing in the above-mentioned prior art, the purpose of the present invention is to provide a two-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix, by separating the real part and the imaginary part in the particle system matrix and the measurement signal, making full use of the real and imaginary information of the system matrix and the measurement signal, converting complex operations into real operations, minimizing signal crosstalk between particles, reducing the number of iterations, and effectively improving the two-color magnetic nanoparticle imaging accuracy and imaging speed, which is of great significance to the application of magnetic nanoparticle imaging technology in clinical medicine.

[0006] To achieve the above object, the present invention provides the following solutions:

[0007] A two-color magnetic nanoparticle imaging method based on real-imaginary separation of a system matrix, comprising:

[0008] Based on two different types of magnetic nanoparticles, two different types of magnetic nanoparticle samples are moved to the central zero magnetic field point respectively, and an excitation magnetic field and a scanning magnetic field are applied to obtain magnetization signals; wherein the magnetization signals are generated by different types of magnetic nanoparticles at different positions relative to the zero magnetic field point; based on the magnetization signals of the magnetic nanoparticle samples, a system matrix A is constructed. 1 and A 2 ;

[0009] The object to be measured composed of two kinds of particles is measured to obtain a measured magnetization signal U;

[0010] The system matrix A of two different types of magnetic nanoparticles 1 , A 2 And the measured magnetization signal U is spliced ​​to obtain the new system matrix A ’ and the new magnetization signal U ’ ;

[0011] Through the new system matrix A ’ Combined with the new magnetization signal U ’ , obtain the particle concentration distribution of the two different types of magnetic nanoparticles, send the concentration distribution of the two magnetic particles to different color channels respectively, and display them in different colors.

[0012] Optionally, acquiring the new system matrix includes:

[0013]

[0014] in, , System matrix for the first magnetic nanoparticle sample The real and imaginary parts of , System matrix for the second magnetic nanoparticle sample The real and imaginary parts of .

[0015] Optionally, acquiring the new magnetization signal includes:

[0016]

[0017] in, , To measure the magnetization signal The real and imaginary parts of .

[0018] Optionally, obtaining the particle concentrations of the two different types of magnetic nanoparticles includes:

[0019] According to the new system matrix, a weight factor is obtained, a weight matrix is ​​constructed by the weight factor, and a magnetic nanoparticle concentration distribution model is constructed based on the weight matrix, the new system matrix and the new measured magnetization signal;

[0020] The magnetic nanoparticle concentration distribution model is iteratively reconstructed to solve the particle concentrations of the two different types of magnetic nanoparticles.

[0021] Optionally, obtaining the weight factor includes:

[0022]

[0023] in, is the weight factor, is the system matrix No. A vector of row elements.

[0024] Optionally, constructing the weight matrix includes:

[0025]

[0026] in, is a diagonal matrix, The diagonal line Weight factor for the row.

[0027] Optionally, constructing the magnetic nanoparticle concentration distribution model includes:

[0028]

[0029] in, is the relaxation factor, is the residual.

[0030] Optionally, solving the particle concentrations of the two different types of magnetic nanoparticles includes:

[0031] Will , , the magnetic nanoparticle concentration distribution model is iteratively reconstructed to solve the particle concentrations of the two different types of magnetic nanoparticles:

[0032]

[0033]

[0034] in, is the number of iterations, , is the system matrix the number of rows, is the system matrix No. A vector of row elements, The magnetization signal No. Elements of a row.

[0035] The beneficial effects of the present invention are:

[0036] The present invention separates the complex elements in the two particle system matrices and the measurement signal into real and imaginary parts and recombines them, and weights them according to the inverse of the second norm of the row vector of the new matrix. This makes full use of the real and imaginary information of the system matrix and the measurement signal, effectively reduces the signal crosstalk between particles, and improves the image reconstruction accuracy.

[0037] The present invention separates and recombine the real and imaginary parts of the two particle system matrices and the measurement signal, converts the operations in the complex space into operations in the real space, avoids the projection of the complex concentration value into the real space during the reconstruction process, effectively reduces the number of iterations while ensuring the image quality, speeds up the convergence of the reconstruction results, and greatly improves the reconstruction speed of the two-color magnetic nanoparticle image. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0039] Figure 1 This is a flow chart of a two-color magnetic nanoparticle imaging method based on real-imaginary separation of a system matrix according to an embodiment of the present invention;

[0040] Figure 2 The figure is a flow chart of a weighted algebraic reconstruction algorithm according to an embodiment of the present invention. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0043] like Figure 1 As shown, this embodiment discloses a two-color magnetic nanoparticle imaging method based on real-imaginary separation of a system matrix, comprising: based on two different types of magnetic nanoparticles, two different types of magnetic nanoparticle samples are moved to the central zero magnetic field point respectively, an excitation magnetic field and a scanning magnetic field are applied, and a magnetization signal is obtained; wherein the magnetization signal is generated by different types of magnetic nanoparticles at different positions relative to the zero magnetic field point; based on the magnetization signal of the magnetic nanoparticle sample, a system matrix A is constructed. 1 and A 2 ; Measure the object to be tested composed of two kinds of particles to obtain the measured magnetization signal U; The system matrix A of two different types of magnetic nanoparticles 1 , A 2 And the measured magnetization signal U is spliced ​​to obtain the new system matrix A ’ and the new magnetization signal U ’ ; Through the new system matrix A ’ Combined with the new magnetization signal U ’ , obtain the particle concentration distribution of two different types of magnetic nanoparticles, send the concentration distribution of these two magnetic particles to different color channels respectively, and display them in different colors.

[0044] Specifically, this embodiment discloses a dual-color magnetic nanoparticle imaging method based on real-imaginary separation of a system matrix, comprising:

[0045] Step S10, using the magnetic nanoparticle imaging system, respectively move the unit volume samples of two different magnetic nanoparticles to the central zero magnetic field point, apply an excitation magnetic field to the imaging field of view to magnetize the samples, and simultaneously apply a scanning magnetic field to move the zero magnetic field point for scanning, obtain the magnetization signals of the two particles at different positions relative to the zero magnetic field point, and construct a system matrix of the two particles and ;

[0046] Step S20, placing the object under test in the imaging field of view, and the concentration distribution of the two magnetic nanoparticles contained in the object under test and , similarly apply the excitation magnetic field and the scanning magnetic field to collect the magnetization signal generated by the object under test ;

[0047] Step S30, the system matrix of the two particles , and magnetization signal The real and imaginary parts of the complex elements in are split and reassembled to obtain the new system matrix and magnetization signal , the concentration distribution of the two magnetic nanoparticles and Splice to get the particle concentration .

[0048] Step S40, based on the new system matrix obtained in S30 , magnetization signal and particle concentration Construct a new equation to be solved and solve for the particle concentration .

[0049] Further, step S30 includes:

[0050] Step S31, the system matrix of the two particles , The real and imaginary parts of the complex elements are separated and concatenated in the direction of the matrix columns to obtain a new system matrix ; Correspondingly, the magnetization signal The real part and the imaginary part of the same splicing are performed to obtain the magnetization signal :

[0051] ,

[0052] in, , is a system matrix of a particle The real and imaginary parts of , is the system matrix of another particle The real and imaginary parts of , The magnetization signal The real and imaginary parts of .

[0053] Step S32, the concentration distribution of the two magnetic nanoparticles and Splicing to get particle concentration :

[0054]

[0055] in, and is the concentration distribution of the two magnetic nanoparticles.

[0056] Furthermore, obtaining the particle concentrations of two different types of magnetic nanoparticles includes: obtaining weight factors according to a new system matrix, constructing a weight matrix through the weight factors, and constructing a magnetic nanoparticle concentration distribution model based on the weight matrix, the new system matrix and the new measured magnetization signal; iteratively reconstructing the magnetic nanoparticle concentration distribution model to solve the particle concentrations of two different types of magnetic nanoparticles.

[0057] Specifically, step S40 includes:

[0058] Step S41, based on the system matrix obtained in step S30 , calculate the weight factor :

[0059]

[0060] in, is the system matrix No. A vector of row elements.

[0061] Step S42, based on the weight factor obtained in step S41 , calculate the weight matrix :

[0062]

[0063] in, is a diagonal matrix, On its diagonal Weight factor for the row.

[0064] Step S43, based on the system matrix obtained in steps S30, S41, and S42 , magnetization signal and the weight matrix , construct the concentration distribution of magnetic nanoparticles The equation is:

[0065]

[0066] in, is the relaxation factor and , is the residual.

[0067] Step S44, using an algebraic reconstruction algorithm to reconstruct the concentration of magnetic nanoparticles To solve, remember , , the iterative formula is:

[0068]

[0069]

[0070] in, is the number of iterations, , is the system matrix the number of rows, is the system matrix No. A vector of row elements, The magnetization signal No. Elements of a row.

[0071] The actual applicable scenarios of the present invention include single harmonic narrowband magnetic nanoparticle imaging and multi-harmonic broadband magnetic nanoparticle imaging. Here, single harmonic narrowband magnetic nanoparticle imaging is taken as an example for description.

[0072] First, the magnetic nanoparticle imaging system is used to set a certain imaging field of view and excitation intensity. Two different unit volume magnetic nanoparticle samples are placed at the zero magnetic field point, and the excitation magnetic field and scanning magnetic field are applied to the imaging field of view. The single harmonic (such as the third harmonic) point spread function of the particle magnetization signal is extracted, and the system matrix of different particles is constructed based on this. and .

[0073] The object to be measured is placed in a magnetic nanoparticle imaging system. The concentration distributions of the two magnetic nanoparticles contained in the object to be measured are and , use the same excitation conditions to excite the object under test and extract the single harmonic part of its magnetization signal (such as the 3rd harmonic) , the particle system matrix , , particle concentration , and magnetization signal The relationship between them is as follows:

[0074]

[0075] After collecting the system matrix of the particles and the magnetization signal of the object under test, each element in the system matrix and the magnetization signal is a complex value, that is, it consists of a real number and an imaginary number. Therefore, the complex system matrix and magnetization signal can be split into real and imaginary parts and spliced ​​to form a new equation:

[0076]

[0077] in, , is a system matrix of a particle The real and imaginary parts of , is the system matrix of another particle The real and imaginary parts of , The magnetization signal The real and imaginary parts of the re-split and recombined system matrix and magnetization signal can be expressed as and , particle concentration , Combination :

[0078] , ,

[0079] In the actual measurement process, the amplitude of the real part and the amplitude of the imaginary part are often inconsistent. Considering the influence of noise, a weight factor is introduced. and the residual vector , construct a new particle concentration solution equation:

[0080]

[0081] in, is the relaxation factor and , is the weight matrix.

[0082] Based on the system matrix , calculate the weight factor :

[0083]

[0084] in, is the system matrix No. A vector of row elements.

[0085] Based on weighting factor , calculate the weight matrix :

[0086]

[0087] in, is a diagonal matrix, On its diagonal Weight factor for the row.

[0088] The above equations can be solved using an algebraic reconstruction algorithm. The specific algorithm flow chart is as follows: Figure 2 As shown:

[0089] First, enter the system matrix of the two particles , measure the magnetization signal , Iterations and relaxation factor ;

[0090] According to the system matrix Calculate weight factor and the weight matrix , The value of the system matrix No. The inverse of the row vector two-norm of the row, is a diagonal matrix, The elements on its diagonal.

[0091] Initialize the residual vector , according to the set number of iterations and the number of system equations, traverse the projection and calculate the concentration of the two particles .remember , , the specific iterative solution formula is as follows:

[0092]

[0093]

[0094] in, is the number of iterations, , is the system matrix the number of rows, is the system matrix No. A vector of row elements, The magnetization signal No. Elements of a row.

[0095] The present invention proposes a two-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix. By separating the real part and the imaginary part of the system matrix and the magnetization signal, the real and imaginary information of the system matrix and the measurement signal are fully utilized, and complex operations are converted into real operations, thereby avoiding the projection of complex numbers into the real number space, effectively improving the two-color magnetic nanoparticle imaging accuracy and image reconstruction speed, which is of great significance for promoting the clinical application of magnetic nanoparticles.

[0096] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A two-color magnetic nanoparticle imaging method based on real-imaginary separation of a system matrix, characterized in that: include: Based on two different types of magnetic nanoparticles, two different types of magnetic nanoparticle samples are moved to the central zero magnetic field point respectively, and an excitation magnetic field and a scanning magnetic field are applied to obtain magnetization signals; wherein the magnetization signals are generated by different types of magnetic nanoparticles at different positions relative to the zero magnetic field point; based on the magnetization signals of the magnetic nanoparticle samples, system matrices A1 and A2 are constructed; The object to be measured composed of two kinds of particles is measured to obtain a measured magnetization signal U; The system matrices A1 and A2 of two different types of magnetic nanoparticles and the measured magnetization signal U are spliced ​​to obtain a new system matrix A ’ and the new magnetization signal U ’ ; Through the new system matrix A ’ Combined with the new magnetization signal U ’ , obtain the particle concentration distribution of the two different types of magnetic nanoparticles, send the concentration distribution of the two magnetic particles to different color channels respectively, and display them in different colors.

2. The dual-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix according to claim 1, characterized in that: Acquiring the new system matrix includes: in, , System matrix for the first magnetic nanoparticle sample The real and imaginary parts of , System matrix for the second magnetic nanoparticle sample The real and imaginary parts of .

3. The dual-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix according to claim 1, characterized in that: Acquiring the new magnetization signal comprises: in, , To measure the magnetization signal The real and imaginary parts of .

4. The dual-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix according to claim 1, characterized in that: Obtaining the particle concentrations of the two different types of magnetic nanoparticles comprises: According to the new system matrix, a weight factor is obtained, a weight matrix is ​​constructed by the weight factor, and a magnetic nanoparticle concentration distribution model is constructed based on the weight matrix, the new system matrix and the new measured magnetization signal; The magnetic nanoparticle concentration distribution model is iteratively reconstructed to solve the particle concentrations of the two different types of magnetic nanoparticles.

5. The dual-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix according to claim 4, characterized in that: Obtaining the weight factor includes: in, is the weight factor, is the system matrix No. A vector of row elements.

6. The dual-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix according to claim 4, characterized in that: Constructing the weight matrix includes: in, is a diagonal matrix, is the weight factor of the i-th row on the diagonal.

7. The dual-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix according to claim 4, characterized in that: Constructing the magnetic nanoparticle concentration distribution model includes: in, is the relaxation factor, is the residual.

8. The dual-color magnetic nanoparticle imaging method based on real-imaginary separation of the system matrix according to claim 7, characterized in that: Solving the particle concentrations of the two different types of magnetic nanoparticles includes: Will , , the magnetic nanoparticle concentration distribution model is iteratively reconstructed to solve the particle concentrations of the two different types of magnetic nanoparticles: in, is the number of iterations, , is the system matrix the number of rows, is the system matrix No. A vector of row elements, The magnetization signal No. Elements of a row.

Images