Method for imaging based on fitting of magnetization curve to magnetic particle relaxation time

Abstract

The application discloses an imaging method based on magnetization curve fitting of magnetic particle relaxation time, comprising: generating a magnetic field free point (FFP); moving the FFP in a space range where a single magnetic nanoparticle-injected target object is located, generating a pulsed square wave excitation magnetic field at each FFP position, and receiving original signals generated by magnetic nanoparticles in the target object under excitation of the pulsed square wave excitation magnetic field; fitting a target magnetization curve by adjusting relaxation time related parameters in a preset double exponential decay function based on the original signals; taking an adjustment result of the corresponding relaxation time related parameters as a magnetic particle relaxation time detection result of the FFP position; and performing imaging based on at least one same item in the magnetic particle relaxation time detection results obtained based on all FFP positions to obtain a relaxation time imaging graph. The application can separately detect the relaxation time, and combines the FFP scanning to perform multi-color magnetic particle imaging to distinguish plaque, duct and blood vessel regions.

Description

Technical Field

[0001] This invention belongs to the field of magnetic particle biomedical detection and imaging, specifically relating to an imaging method based on fitting the relaxation time of magnetic particles using magnetization curves. Background Technology

[0002] Magnetic Particle Imaging (MPI) is a tracing method based on functional and tomographic imaging techniques to detect the spatial distribution of magnetic nanoparticles (SPIONs). As a functional imaging method, MPI improves imaging resolution without using radioactive materials, making it one of the most promising technologies for clinical translation in the past 20 years. Significant progress has been made in areas such as multimodal in vivo imaging, tumor detection and immunotherapy, cell therapy monitoring, vascular perfusion imaging, inflammation quantification, infectious disease research, precision magnetothermal therapy, targeted drug delivery, and nanomaterials.

[0003] The imaging principle of MPI is based on Langevin's law of magnetization and the nonlinear magnetization curve. The static gradient magnetic field of MPI, or the selective field, causes magnetic particles at different locations to generate distinct signals. The selective field has a corresponding field vector at each spatial location, with the field vector being zero at the center. This point is called the Field Free Region (FFR), where the magnetic field strength is zero; it is also known as the field-free point (FFP) or field-free line (FFL). Rapidly moving the FFR, when it passes through a region containing magnetic particles, particles farther from the FFR reach magnetic saturation and do not react to changes in the overall magnetic field; the receiving coil does not detect a signal. Conversely, particles near the FFP are not magnetically saturated, and their magnetization changes in magnitude and direction, thus detecting a voltage signal in the receiving coil. This voltage signal is distributed to each location of the FFP for reconstruction, yielding the spatial distribution information of the magnetic particles, i.e., the MPI signal, enabling the reconstruction imaging of the concentration distribution of SPIONs.

[0004] Research on magnetic nanoparticles shows that the relaxation time of the magnetization response under an excitation magnetic field is a crucial characteristic. The basic relaxation mechanisms include internal rotation (Nieren relaxation) and external physical rotation (Brown relaxation). The Nieren relaxation time is a function of temperature and the volume of the magnetic core. The Brown relaxation time depends on the fluid viscosity, temperature, and the hydrodynamic volume of the magnetic nanoparticles. Accurate measurement of each relaxation time can be used to describe the in vivo state of magnetic nanoparticles, potentially leading to various clinical applications, such as pathological detection based on viscosity and temperature predictions.

[0005] For example, viscosity within biological tissues can exhibit certain clinical characteristics. The cellular response to the viscoelasticity of the matrix surface is crucial for controlling cellular behavior, such as controlling stem cell proliferation and determining cell fate. In the tumor microenvironment, increased viscosity is caused by the overexpression of indicators such as lymphocytes, lactate, proteins, enzymes, lipids, extracellular secretions, and extracellular matrix components. Increased whole blood viscosity increases cardiovascular mortality and the likelihood of developing Alzheimer's disease. Mice with inflammation and peritonitis exhibit decreased blood viscosity, and so on.

[0006] Temperature is associated with poor prognosis in breast cancer patients. Furthermore, HSP90 is a molecular chaperone in the heat shock response, maintaining the structural and functional integrity of oncogenic proteins in pancreatic cancer cells. The anticancer effects of HSP90 inhibitors can be evaluated by inducing heat shock through magnetothermal therapy combined with temperature monitoring.

[0007] Given the clinical application potential of MPI imaging, the development of multicolor magnetic particle imaging using a specific relaxation time would undoubtedly expand the breadth and depth of MPI applications. However, current methods widely employ sinusoidal excitation to generate dynamically changing magnetic fields to excite magnetic nanoparticles. Both of these relaxation times are functions of the applied magnetic field, decreasing with increasing magnetic field amplitude, and lack a fixed Nilleney or Brownian relaxation time throughout the excitation period. Therefore, under traditional sinusoidal excitation methods, it is difficult to decouple the Nilleney and Brownian relaxation times, thus hindering research development in this area. Summary of the Invention

[0008] To address the aforementioned problems in the prior art, this invention provides an imaging method, apparatus, electronic device, and storage medium based on magnetization curve fitting of magnetic particle relaxation time. The technical problem to be solved by this invention is achieved through the following technical solution:

[0009] In a first aspect, embodiments of the present invention provide an imaging method based on fitting the relaxation time of magnetic particles to a magnetization curve, the method comprising:

[0010] Generates a magnetic field free point (FFP);

[0011] The FFP is moved within the spatial range of the target under test, which has been injected with a single type of magnetic nanoparticle. At each FFP position, a pulsed square wave excitation magnetic field is generated, and the original signal generated by the magnetic nanoparticles in the target under test at that FFP position under the excitation of the pulsed square wave excitation magnetic field is received.

[0012] Based on the original signal, by adjusting the relaxation time-related parameters in the preset double exponential decay function, a target magnetization curve matching the original signal is obtained; the adjustment result of the relaxation time-related parameters corresponding to the target magnetization curve is used as the magnetic particle relaxation time detection result at the FFP position; wherein, the relaxation time-related parameters include the Niehr relaxation time constant, the Brown relaxation time constant, and the percentages of Niehr relaxation and Brown relaxation;

[0013] Imaging is performed based on at least one of the same terms from the magnetic particle relaxation time detection results obtained from all FFP positions to obtain a relaxation time image of the target under test.

[0014] In one embodiment of the present invention, generating the magnetic field free point FFP includes:

[0015] The magnetic field free point (FFP) is generated using a preset static gradient magnetic field.

[0016] In one embodiment of the present invention, generating a pulsed square wave excitation magnetic field at each FFP location and receiving the original signal generated by the magnetic nanoparticles within the target at that FFP location under the excitation of the pulsed square wave excitation magnetic field includes:

[0017] At each FFP location, a pulsed square wave excitation magnetic field is generated using a pre-designed pulsed square wave relaxor, and the original signal generated by the magnetic nanoparticles in the target at that FFP location under the excitation of the pulsed square wave excitation magnetic field is received.

[0018] In one embodiment of the present invention, the pulse square wave relaxor includes:

[0019] A digital acquisition card is used to generate analog signals of pulse square waves and to digitize input sample signals.

[0020] An AC power amplifier is used to amplify the analog signal;

[0021] The transmitting coil is used to transmit amplified analog signals to generate a pulsed square wave excitation magnetic field;

[0022] A current sensor is used to monitor the emission waveform of the AC power amplifier in real time.

[0023] A receiving coil is used to receive the sample signal generated by a single magnetic particle sample under the excitation of the pulsed square wave excitation magnetic field;

[0024] A low-noise preamplifier is used to amplify the sample signal and provide it to the digital acquisition card.

[0025] In one embodiment of the present invention, the step of fitting a target magnetization curve matching the original signal by adjusting the relaxation time correlation parameter in a preset double exponential decay function based on the original signal includes:

[0026] The original signal is integrated to obtain the integrated curve.

[0027] Obtain the relaxation time-related parameters of the current group, and for the preset double exponential decay function, obtain a fitted magnetization curve corresponding to the relaxation time-related parameters of the current group.

[0028] Determine whether the point error between the currently obtained magnetization intensity curve and the integrated curve meets the preset requirements. If yes, use the currently obtained magnetization intensity curve as the target magnetization intensity curve. If not, adjust the parameters using the preset parameter range corresponding to each relaxation time related parameter to obtain a new set of relaxation time related parameters, and return to the process of obtaining the magnetization intensity curve.

[0029] In one embodiment of the present invention, the preset double exponential decay function includes:

[0030]

[0031] Among them, M non-adiabatic M0 represents the non-adiabatic magnetization; t represents time; M0 represents the maximum magnetization under static field amplitude; t0 represents the time of magnetic field reversal; τ N τ represents the Niehr relaxation time constant; B τ represents the Brownian relaxation time constant; N <τ B ; a represents the percentage of Neill relaxation, b represents the percentage of Brownian relaxation, and a+b=1.

[0032] In one embodiment of the present invention, imaging is performed based on at least one term from the magnetic particle relaxation time detection results obtained from all FFP positions to obtain a relaxation time imaging map of the target under test, including:

[0033] Based on the Brownian relaxation time constants obtained from the magnetic particle relaxation time detection results at all FFP locations, an image of the Brownian relaxation time of the target under test is obtained.

[0034] Secondly, embodiments of the present invention provide an imaging device based on magnetization curve fitting of magnetic particle relaxation time, the device comprising:

[0035] The FFP generation module is used to generate magnetic field free points (FFPs).

[0036] The original signal acquisition module is used to move the FFP within the spatial range of the target under test where a single type of magnetic nanoparticle has been injected. At each FFP position, a pulsed square wave excitation magnetic field is generated, and the original signal generated by the magnetic nanoparticles in the target under test at that FFP position under the excitation of the pulsed square wave excitation magnetic field is received.

[0037] The magnetic particle relaxation time detection module is used to fit a target magnetization curve that matches the original signal by adjusting the relaxation time-related parameters in a preset double exponential decay function based on the original signal; and to use the adjustment result of the relaxation time-related parameters corresponding to the target magnetization curve as the magnetic particle relaxation time detection result at the FFP position; wherein, the relaxation time-related parameters include the Niehr relaxation time constant, the Brown relaxation time constant, and the percentages of Niehr relaxation and Brown relaxation;

[0038] The imaging module is used to perform imaging based on at least one of the same terms in the magnetic particle relaxation time detection results obtained from all FFP positions, so as to obtain a relaxation time imaging map of the target under test.

[0039] Thirdly, embodiments of the present invention provide an electronic device, including a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus;

[0040] The memory is used to store computer programs;

[0041] When the processor executes the program stored in the memory, it implements the steps of the imaging method based on magnetization curve fitting of magnetic particle relaxation time provided in the embodiments of the present invention.

[0042] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, it implements the steps of the imaging method based on magnetization curve fitting of magnetic particle relaxation time provided in the embodiments of the present invention.

[0043] The beneficial effects of this invention are:

[0044] In the solution provided by this invention, a magnetic field free point (FFP) is first generated; then, the FFP is moved within the spatial range of the target object where a single type of magnetic nanoparticle has been injected, and at each FFP location, a pulsed square wave excitation magnetic field is generated, and the original signal generated by the magnetic nanoparticles in the target object at that FFP location under the excitation of the pulsed square wave excitation magnetic field is received; then, based on the original signal, a target magnetization intensity curve matching the original signal is obtained by adjusting the relaxation time correlation parameter in a preset double exponential decay function; the adjustment result of the relaxation time correlation parameter corresponding to the target magnetization intensity curve is used as the magnetic particle relaxation time detection result at that FFP location; the relaxation time correlation parameter includes the Niehr relaxation time constant, the Brown relaxation time constant, and the percentages of Niehr relaxation and Brown relaxation; finally, imaging is performed based on at least one identical term in the magnetic particle relaxation time detection results obtained from all FFP locations to obtain a relaxation time imaging map of the target object. This invention utilizes multicolor magnetic particle imaging with combined magnetic field free point scanning. At each FFP location, the Nieer relaxation time and Brownian relaxation time of the magnetic nanoparticles are distinguished and detected using the inversion recovery method. The relaxation time imaging map obtained using the same detection results from all FFP locations can at least be used to distinguish plaque, duct, and vascular regions. Attached Figure Description

[0045] Figure 1 The left and right figures are schematic diagrams of the signal curve and magnetization intensity curve under trapezoidal pulse excitation, respectively;

[0046] Figure 2 The left and right figures show the MH curves of Synomag-D with 30% gelatin and Synomag-D with 1% glycerol under trapezoidal pulse excitation, respectively.

[0047] Figure 3 The graph shows the results of recovering magnetization using the Debye relaxation model during the field rise phase.

[0048] Figure 4 This is a schematic flowchart of an imaging method based on fitting the relaxation time of magnetic particles using a magnetization curve, provided in an embodiment of the present invention.

[0049] Figure 5 This invention provides an embodiment of the use of single-exponential or double-exponential relaxation models to analyze the result curves of magnetization recovery during the flat phase of the magnetic field.

[0050] Figure 6 The left and right figures respectively show the fitted magnetization intensity curves and derived signal curves during the rising and flattening phases of the magnetic field in an embodiment of the present invention;

[0051] Figure 7(a) shows the relaxation time curves of Synomag-D under different magnetic field amplitudes as viscosity increases in the embodiment of the present invention;

[0052] Figure 7(b) is a comparison of two relaxation times at different viscosity levels in the embodiments of the present invention;

[0053] Figure 8 The following are relaxation time characteristic curves of Synomag-D at different temperatures according to embodiments of the present invention;

[0054] Figure 9 The relaxation time characteristics fitted by the magnetization curve in this embodiment of the invention are used to assess viscosity sensitivity.

[0055] Figure 10 This is a schematic diagram of the vascular system model used in the simulation experiment of an embodiment of the present invention;

[0056] Figure 11 This is a schematic diagram of the magnetic field environment of the vascular system model in the simulation experiment of this invention;

[0057] Figure 12 This is a schematic diagram of the signal weighting factor distribution in the simulation experiment of an embodiment of the present invention;

[0058] Figures 13(a) to 13(c) These are, respectively, digital blood vessel phantom images, baseline images, and Brownian relaxation time imaging images from the simulation experiments of this invention embodiment;

[0059] Figure 14 This is a graph showing the variation of Brownian relaxation time of plaque, duct, and vascular regions with the average signal period in a simulation experiment of an embodiment of the present invention.

[0060] Figure 15 This is a schematic diagram of the structure of an imaging device based on fitting the relaxation time of magnetic particles using a magnetization curve, provided in an embodiment of the present invention.

[0061] Figure 16 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] Under traditional sinusoidal excitation methods, it is difficult to decouple the Nieer relaxation time and the Brownian relaxation time. One approximation method for detecting relaxation time under sinusoidal excitation is to use the total relaxation time of the signal broadening and hysteresis. Specifically, the relaxation process of magnetic nanoparticles can be modeled as a first-order Debye process, where the convolution kernel is a single exponential decay function, and the relaxation time constant in the kernel is the so-called Debye relaxation time. This simple model has been experimentally verified to be useful for matching relaxation effects. Based on this, researchers have proposed a relaxation imaging technique called TAURUS (TAU estimation by recovering the underlying mirror symmetry), which estimates the Debye relaxation time constant by recovering the underlying mirror symmetry signal from sinusoidal excitation. Another approximation method for detecting relaxation time is to consider only Brownian relaxation while ignoring Nieer relaxation. This approach is based on the premise that Brownian relaxation dominates when the sinusoidal excitation frequency is high and the magnetic nanoparticle diameter is greater than 25 nm.

[0064] Therefore, there is currently no effective means to distinguish and detect the Nillet relaxation time and Brownian relaxation time of magnetic nanoparticles.

[0065] To facilitate understanding of this scheme, the inventive concept of pulsed square wave excitation, Debye relaxation time in the prior art, and relaxation time detection used in the imaging of this embodiment will be briefly explained first.

[0066] This invention assumes that the period of the pulsed square wave excitation is T. The following explanation uses a trapezoidal pulse as a specific example of a pulsed square wave. For a trapezoidal wave with a period of T, the rising and flat phases of the magnetic field within half a period can be represented as follows:

[0067]

[0068] Where t represents time; H0 represents the magnetic field amplitude during the flat phase; fp represents the ratio of the flat phase time to the total pulse time; and α represents the slope during the rising phase.

[0069]

[0070] In the study of this embodiment of the invention, the half-cycle of the wave is much longer than the relaxation time, which means that within each pulse cycle, the magnetic field is fully magnetized at the end of the flat phase. During the first half-cycle, the non-adiabatic magnetization of the magnetic nanoparticles can be calculated by integrating the received signal:

[0071]

[0072] Among them, M non-adiabatic (t) represents the non-adiabatic magnetization; S received(t') represents the received signal, that is, the signal generated by the received magnetic particles under the excitation of the trapezoidal pulse excitation magnetic field; M0 represents the magnetization intensity at the static excitation field H0.

[0073] See Figure 1 The black curve in the left figure represents the signal curve under trapezoidal pulse excitation, i.e., the curve corresponding to the received signal; the red curve in the right figure represents the magnetization intensity curve under trapezoidal pulse excitation, abbreviated as the M curve; the green curves in both figures represent the corresponding magnetic field intensity curves, abbreviated as the H curve. See also... Figure 1 Understanding this, the magnetization M0 at the static excitation field H0 can be estimated using half of the signal AUC (Area Under Curve, defined as the area under the ROC curve and the coordinate axes) over the entire half-cycle:

[0074]

[0075] from Figure 1 It can be seen that the MH curve under trapezoidal pulse excitation can be divided into a magnetic field rising stage and a magnetic field flat stage. The curve of the magnetic field rising stage has the same trend as the rising excitation magnetic field. However, the MH curve of the rising stage depends on the slope of the rising magnetic field and the relaxation time of the magnetic nanoparticles. Since the excitation field amplitude is constant, the MH curve of the field flat stage is basically a vertical line. The embodiments of the present invention assume that the magnetization is fully recovered at the end of the magnetic field flat stage. The fully recovered magnetization under different magnetic field amplitudes can be fitted with the Langevin function to estimate the particle size. The Langevin function fitting is shown in Equation (5):

[0076]

[0077]

[0078] Where m represents the magnetic moment of the magnetic nanoparticle; M sat The saturation magnetization is 0.551 T / μ0; r represents the particle size in nanometers (nm); k B T represents the Boltzmann constant; μ0 represents the free permeability; T P It represents the absolute temperature in Kelvin (K).

[0079] For the experimental verification results regarding the estimation of magnetic particle size using the Langevin function fitting, please refer to [link to relevant documentation]. Figure 2 understand, Figure 2The left and right plots show the MH curves of Synomag-D magnetic particles and 30% gelatin under pulsed trapezoidal excitation; the right plot shows the MH curves of Synomag-D magnetic particles and 1% glycerol under pulsed trapezoidal excitation. The fully recovered magnetization at the end of each half-cycle under different excitation field amplitudes can be fitted using the Langevin function (black curve) to estimate the particle size of Synomag-D. The particle size estimates obtained in the left and right plots are 9.38 nm and 25.26 nm respectively. These size estimates have been verified to be in good agreement with existing publicly available data.

[0080] During the rising phase of the magnetic field, the relationship between magnetization under non-adiabatic and adiabatic conditions is expressed as follows:

[0081]

[0082] Where u(t) represents the Heaviside step function; * represents convolution in the time domain; non-adiabatic and adiabatic represent non-adiabatic and adiabatic conditions, respectively. In this model, τ D For the Debye relaxation time constant, please refer to [link / reference]. Figure 3 understand, Figure 3 The black dot curve represents the original data, and the blue curve is the magnetization curve recovered using the Debye relaxation model during the field rise phase. Magnetization under adiabatic conditions can be calculated using the Langevin function:

[0083]

[0084] Debye relaxation time can be viewed as a combination of Neill relaxation time and Brown relaxation time:

[0085]

[0086] Where, τ N τ B represents the Niehr relaxation time constant and the Brown relaxation time constant, respectively, with the numerical values ​​representing the corresponding relaxation time.

[0087] Neill and Brownian relaxation times are often difficult to measure separately, and there is currently no suitable formula to calculate them separately.

[0088] This invention, through research, has found that during the flat phase of the magnetic field under pulsed square wave excitation, the relaxation time of magnetic nanoparticles can be considered constant. Pulsed excitation provides a steady-state signal, avoiding signal broadening and hysteresis caused by relaxation effects. Pulsed excitation includes two phases: a magnetic field rise phase and a magnetic field flat phase. In the former, a rapidly changing magnetic field is generated; during this phase, the Nieer and Brownian relaxation times are not constant and cannot be decoupled, allowing only the Debye relaxation process to be evaluated. In the flat phase, the magnetic field does not change with time; the Nieer and Brownian relaxation times are constant and may differ, thus enabling decoupling. Based on this, this invention proposes a double-exponential relaxation theory, a theory of magnetization reversal recovery during pulsed excitation, used to describe the magnetization recovery signal in pulsed square wave excitation. Based on this theory, this invention proposes a method for detecting the relaxation time of magnetic particles. Furthermore, it proposes an imaging method, device, electronic device, and storage medium based on fitting the magnetic particle relaxation time to the magnetization curve.

[0089] Firstly, such as Figure 4 As shown in the embodiment of the present invention, an imaging method based on fitting the relaxation time of magnetic particles to a magnetization curve may include the following steps:

[0090] S1 generates a magnetic field free point FFP;

[0091] Based on MPI imaging technology, it can be understood that the magnetic field free point (hereinafter referred to as FFP) is generated by the selected field.

[0092] This step can generate FFP using a preset static gradient magnetic field. For example, the gradient of this preset static gradient magnetic field can be along the x and y directions, and the field strength can be set as needed. Gradient fields in different directions can be achieved using gradient coil pairs with spacing in the corresponding directions. The coils used can be, for example, Maxwell coils, etc., and there are no restrictions here.

[0093] S2, move the FFP within the spatial range of the target to be tested where a single type of magnetic nanoparticle has been injected, generate a pulsed square wave excitation magnetic field at each FFP position, and receive the original signal generated by the magnetic nanoparticles in the target to be tested at that FFP position under the excitation of the pulsed square wave excitation magnetic field.

[0094] In this embodiment of the invention, the target to be tested can be a human body, animal body, etc. The target to be tested is pre-injected with a single type of magnetic nanoparticle, which means containing only one type of magnetic nanoparticle. This type of magnetic nanoparticle can be superparamagnetic iron oxide nanoparticles (Resovist), specifically various commercially available magnetic nanoparticles, such as iron carboxymethyl glycosides, Perimag, Synomag-D, and the Dongna series magnetic particles, or a self-developed and synthesized magnetic nanoparticle; no specific limitation is made here. However, in one optional embodiment, in order to achieve good relaxation time detection and imaging results, the selected magnetic nanoparticles are those that have been pre-tested and evaluated on different target samples as experimental samples, showing better response and detection effects.

[0095] The magnetic nanoparticles are in a colloidal suspension with a concentration of up to 0.5 mmol Fe / mL. The injection dosage is set according to the weight of the target. The magnetic nanoparticles are administered intravenously, either manually by a doctor or automatically by an instrument. As is well known, magnetic nanoparticles are superparamagnetic. When an external magnetic field is present, the magnetic moments of the magnetic nanoparticles in the liquid will deflect towards the direction of the applied magnetic field, generating a change in magnetic flux in the receiving coil, thereby producing a response voltage signal, which is the original signal in this embodiment of the invention.

[0096] Specifically, the spatial range of the target under test corresponds to the entire field of view (FOV). The field of view device (FFP) is moved within this range to scan the target point by point. At each FFP position, a pulsed square wave excitation magnetic field is generated and the original signal is acquired. Regarding the generation of the pulsed square wave excitation magnetic field and the acquisition of the sample signal, any method or device capable of generating a pulsed square wave excitation magnetic field can be used to generate the required pulsed square wave excitation magnetic field. Any signal receiving method or device can be used to receive the sample signal generated by the single magnetic particle sample under the excitation of the pulsed square wave excitation magnetic field, such as a receiving coil, etc. No specific limitations are imposed here.

[0097] In one optional implementation, generating a pulsed square wave excitation magnetic field at each FFP location and receiving the original signal generated by the magnetic nanoparticles within the target at that FFP location under the excitation of the pulsed square wave excitation magnetic field includes:

[0098] At each FFP location, a pulsed square wave excitation magnetic field is generated using a pre-designed pulsed square wave relaxor, and the original signal generated by the magnetic nanoparticles in the target at that FFP location under the excitation of the pulsed square wave excitation magnetic field is received.

[0099] The pulse square wave relaxor (TWR) of this invention integrates pulse square wave excitation magnetic field generation and sample signal reception functions. The TWR design can eliminate the need for a resonant circuit, allowing trapezoidal waveform excitation. In one optional embodiment, the pulse square wave relaxor may include:

[0100] 1) Digital acquisition card, used to generate analog signals of pulse square waves and to digitize input sample signals;

[0101] 2) An AC power amplifier, used to amplify the analog signal;

[0102] 3) Transmitting coil, used to transmit amplified analog signals to generate pulsed square wave excitation magnetic field;

[0103] 4) A current sensor is used to monitor the emission waveform of the AC power amplifier in real time;

[0104] 5) A receiving coil, used to receive the sample signal generated by the single magnetic particle sample under the excitation of the pulsed square wave excitation magnetic field;

[0105] 6) A low-noise preamplifier is used to amplify the sample signal and provide it to the digital acquisition card.

[0106] The various components of the pulse square wave relaxor can be implemented using existing commercial equipment or self-developed equipment, without specific limitations. For example, in optional embodiments, the data acquisition card can be a DAQ NI6363, the AC power amplifier can be an AE Techron 7548, the low-noise preamplifier can be an SR560, and the transmitting and receiving coils can be implemented using multilayer Litz coils, etc. Of course, the pulse square wave relaxor of this invention is not limited to the structure described above.

[0107] S3, Based on the original signal, by adjusting the relaxation time-related parameters in the preset double exponential decay function, a target magnetization intensity curve matching the original signal is obtained; the adjustment result of the relaxation time-related parameters corresponding to the target magnetization intensity curve is used as the magnetic particle relaxation time detection result at the FFP position.

[0108] The relaxation time-related parameters include the Niehr relaxation time constant, the Brown relaxation time constant, and the percentages of Niehr relaxation and Brown relaxation.

[0109] In this embodiment of the invention, the relaxation process of magnetization during the flat phase of the magnetic field under pulsed excitation is similar to the longitudinal (T1) relaxation process of the reverse recovery pulse in magnetic resonance imaging. Reverse recovery uses an initial pulse of 180° to reverse the longitudinal magnetization intensity, which recovers with a time constant T1. This embodiment of the invention assumes that the magnetic field suddenly switches from a negative peak to a positive peak during pulsed square wave excitation, and vice versa. In this embodiment, after each pulsed square wave excitation with the magnetic field reversed by 180 degrees, the original signal is acquired and curve-fitted. The magnetization intensity of the magnetic nanoparticles is recovered through two relaxation effects. The two-step magnetization response in the Brownian state occurs after the Niehr state. Under pulsed excitation with the magnetic field increasing from zero, the magnetization process is fitted by a preset double exponential decay function. The magnetization curve fitting method is performed in an interactive data language for relaxation time feature extraction; the interactive data language can include IDL, MATLAB, Java, C, Python, etc., and is not limited here. The data file about Synomag-D received in TWR is loaded into the IDL program.

[0110] In one optional implementation, the step of fitting a target magnetization curve matching the original signal by adjusting the relaxation time correlation parameter in a preset double exponential decay function based on the original signal includes:

[0111] (1) Integrate the original signal to obtain the integrated curve;

[0112] This step can be understood by referring to formula (3). The obtained integral curve corresponds to the magnetization intensity under non-adiabatic magnetization conditions, which is the original data curve characterizing the non-adiabatic magnetization intensity.

[0113] (2) Obtain the relaxation time related parameters of the current group, and for the preset double exponential decay function, obtain a fitted magnetization curve corresponding to the relaxation time related parameters of the current group.

[0114] The preset double exponential decay function includes:

[0115]

[0116] Among them, M non-adiabatic M0 represents the non-adiabatic magnetization; t represents time; M0 represents the maximum magnetization under static field amplitude; t0 represents the time of magnetic field reversal; τ N τ represents the Niehr relaxation time constant; B τ represents the Brownian relaxation time constant; N <τ B ; a represents the percentage of Neill relaxation, b represents the percentage of Brownian relaxation, and a+b=1.

[0117] During the preliminary research, this invention found that by assuming a or b is 0, the double exponential recovery process can also be transformed into a single exponential process, but the curve fitting result is worse than that of the double exponential process. For details, please refer to... Figure 5 As shown, Figure 5 The cyan and red curves represent the analysis of magnetization recovery during the flat magnetic field phase using a single-exponential relaxation model and a double-exponential relaxation model, respectively. The black dotted line represents the original data curve. Figure 5 Combination Figure 3 understand. Figure 5 In this study, the double-exponential model exhibits better curve fitting results than the single-exponential model. Therefore, in this embodiment of the invention, a double-exponential decay function is used to fit the recovered magnetization. See [link to relevant documentation]. Figure 6 understand. Figure 6 In the left-hand graph, the black dotted lines represent the fitted magnetization curves during the rising and flattening phases of the magnetic field. In the right-hand graph, the black dotted lines represent the derived signal curves during the rising and flattening phases. In both graphs, the Debye relaxation model is used to fit the cyan magnetization curve during the rising phase, and the double exponential model (represented by Bi-ex in the graph) is used to fit the red magnetization curve during the flattening phase. The signal curve is the reciprocal of the magnetization curve with respect to time, and the magnetic field strength curve is the green curve. When b is very small and close to zero, there is almost no Brownian relaxation effect, only the Niehr relaxation effect.

[0118] The NieR relaxation time under zero magnetic field condition is defined as:

[0119]

[0120] Among them, V C Represents the core volume; K represents the anisotropy constant; τ0 represents the relaxation time under zero field conditions; T P This represents the absolute temperature in Kelvin (K). In trapezoidal excitation, the NieR and Brownian relaxation times extracted from the inversion recovery process during the flattened magnetic field phase may differ from the relaxation times in the zero-field case. The Brownian relaxation time is defined as:

[0121]

[0122] Where η represents the fluid viscosity; V h This represents the hydrodynamic volume of a particle.

[0123] Regarding step (2), the embodiments of the present invention can make (τ) N , τ B ,a) or (τ N , τ B b) As a set of relaxation time-related parameters, and pre-set a preset parameter range for each parameter, for example, τ NThe corresponding preset parameter range can be [0, 10us], τ B The corresponding preset parameter range can be [10µs, 200µs], and the preset parameter range for a or b can be (0, 1), etc. A set of relaxation time-related parameters can be determined as initial values ​​using the lower limit of the preset parameter range for each parameter. Then, by using the preset double exponential decay function to perform curve fitting at different times, a corresponding fitted magnetization curve can be obtained. For an explanation of the concept of curve fitting, please refer to the prior art.

[0124] (3) Determine whether the point error between the currently obtained magnetization intensity curve and the integrated curve meets the preset requirements. If yes, use the currently obtained magnetization intensity curve as the target magnetization intensity curve. If no, adjust the parameters using the preset parameter range corresponding to each relaxation time related parameter to obtain a new set of relaxation time related parameters, and return to the process of obtaining the magnetization intensity curve.

[0125] For the magnetization curve obtained in step (2), determine whether the point error between it and the integrated curve meets the preset requirements. It can be understood that at each time point, there are actual curve points and fitted curve points. The general principle of curve fitting is that for the same time, the error (i.e., the absolute value of the difference) between the points on the actual curve and the points on the fitted curve is the smallest. In this embodiment of the invention, it can be determined whether the error between the two points at the corresponding positions of the target magnetization curve and the integrated curve at the same time is less than a preset threshold, or whether the average value of the two-point errors of all positions is less than a preset average error value. If the above preset requirements are met, the magnetization curve obtained in step (2) is taken as the target magnetization curve, and the curve fitting ends. At this time, the adjustment result of the relaxation time related parameters used when the target magnetization curve is obtained will be obtained, i.e. (τ N , τ B a, b) are used as the magnetic particle relaxation time detection results; if the above preset requirements are not met, the parameters are adjusted to obtain a new set of relaxation time related parameters. For example, at least one of the three parameters can be changed to obtain a new set of relaxation time related parameters. The new set of relaxation time related parameters is used to return to the execution steps (2) to (3) and is used as the current set of relaxation time related parameters in steps (2) to (3) for execution.

[0126] Compared to existing technologies that use sinusoidal excitation, this invention generates a pulsed square wave excitation magnetic field and receives the original signal generated by the magnetic nanoparticles within the target at the FFP location under the excitation of the pulsed square wave excitation magnetic field. Based on the original signal, by adjusting the relaxation time-related parameters in a preset double exponential decay function, a target magnetization curve matching the original signal is fitted. The adjustment result of the relaxation time-related parameters corresponding to the target magnetization curve is used as the magnetic particle relaxation time detection result, including the Nieer relaxation time constant, the Brownian relaxation time constant, and the percentages of Nieer and Brownian relaxation. Therefore, this invention innovatively proposes an effective means to distinguish and detect the Nieer and Brownian relaxation times of magnetic nanoparticles.

[0127] S4. Based on at least one of the same terms in the magnetic particle relaxation time detection results obtained from all FFP positions, an imaging is performed to obtain a relaxation time image of the target under test.

[0128] This step can utilize at least one of the following for imaging: the Niehr relaxation time constant, the Brownian relaxation time constant, the percentage of Niehr relaxation, or the percentage of Brownian relaxation obtained from all FFP locations. Furthermore, depending on the scanning method of the target under test, two-dimensional imaging or three-dimensional imaging can be performed. Specific methods for imaging using point data of different dimensions can be implemented using existing technologies, which is not the focus of this embodiment. It is understood that imaging using any one of the above four items can obtain a relaxation time imaging map corresponding to the target under test. Alternatively, in an optional embodiment, at least two of these items can be combined to obtain a relaxation time imaging map corresponding to the target under test, and so on; all of these are reasonable.

[0129] The following describes the feasibility, imaging purpose, and related applications of relaxation time imaging in the embodiments of the present invention.

[0130] First, in this embodiment of the invention, the relationship between the relaxation time and functional parameters of magnetic nanoparticles was studied experimentally. These functional parameters include viscosity, stiffness, temperature, microenvironment, and the free or bound state of the magnetic particles.

[0131] Taking a single magnetic particle sample composed of Synomag-D magnetic particles as an example, Synomag-D is a multinuclear magnetic nanoparticle coated with dextran with an original iron concentration of 6 mg / ml.

[0132] In the experiment, single magnetic nanoparticles such as Synomag-D can be placed in a buffer solution composed of a mixture of water and glycerol. By changing the composition of the glycerol, the viscosity of the single magnetic particle sample can be altered, thereby simulating different biological viscosity environments. For example, Synomag-D with an original iron concentration of 6 mg / ml has the lowest viscosity, which increases after adding a buffer solution composed of a mixture of water and glycerol. In this embodiment of the invention, to obtain single magnetic particle samples with different viscosities, a series of water-glycerol mixtures with glycerol concentrations ranging from 1% to 70% wt can be prepared as buffer solutions for diluting the magnetic nanoparticles.

[0133] Furthermore, single magnetic nanoparticles such as Synomag-D can be placed in a mixture of water and gelatin gel, using this mixture as an embedding matrix for the single magnetic nanoparticles to further limit their Brownian mobility. In this invention, the hardness of the single magnetic particle sample can be altered by changing the gelatin composition to simulate different biological stiffness environments. For example, this invention can use a water-gelatin gel with a gelatin concentration of 5% to 30% wt as the embedding matrix for the magnetic nanoparticles. The preparation process can, for example, involve adding gelatin powder to ddH2O (double-distilled water) and heating it to a certain temperature, such as 65°C, until the gelatin powder is completely dissolved. The magnetic nanoparticles are then uniformly dispersed in the resulting mixture, and the sample is solidified at a set temperature, such as 4°C. The iron concentration of the single magnetic particle sample can be set to 0.5 mg / ml, etc.

[0134] Furthermore, embodiments of the present invention can also place the obtained single magnetic particle samples at different temperatures to study the relaxation behavior of magnetic nanoparticles at different temperatures. Specifically, for example, a digitally temperature-controlled water bath can be used to heat the single magnetic particle samples to 25°C, 37°C, and 52°C, etc., and relaxation time can be detected and analyzed at each temperature, etc. Different temperatures can simulate different temperature environments in clinical settings.

[0135] This invention's embodiments reveal that the Néel relaxation time is a function of temperature and magnetic core volume. The Brownian relaxation time, on the other hand, depends on the fluid viscosity, temperature, and the hydrodynamic volume of the magnetic nanoparticles. Both Néel and Brownian relaxation times are temperature-dependent, and temperature can serve as a functional parameter of the magnetic particles. Accurate measurement of both relaxation times can provide precise temperature measurement methods. Therefore, the achieved temperature measurement can be further applied in fields such as medicine. For example, real-time temperature imaging is crucial in thermal ablation treatments such as magnetothermal therapy and high-intensity focused ultrasound; temperature is associated with poor prognosis in breast cancer patients; heatstroke, characterized by extreme hyperthermia, also requires accurate temperature measurement; and the anticancer effects of HSP90 inhibitors can be evaluated by using magnetothermal therapy to induce heat shock combined with temperature monitoring, etc.

[0136] Of course, in addition to temperature, the functional parameters of magnetic particles can also include viscosity, stiffness, microenvironment, and the free or bound state of the magnetic particles.

[0137] A significant application of bioviscous environments lies in the medical field. For instance, studying cellular responses to the viscoelasticity of the matrix surface is crucial for controlling cellular behavior, such as stem cell proliferation. Alternatively, in the tumor microenvironment, the overexpression of indicators such as lymphocytes, lactate, proteins, enzymes, lipids, extracellular secretions, and extracellular matrix components can lead to increased viscosity. Increased whole blood viscosity can raise cardiovascular mortality and the risk of Alzheimer's disease, while blood viscosity decreases in mice with inflammation and peritonitis. Biostiffness environments can also be applied in the medical field, such as the changes in stiffness caused by the intervention of different materials like scaffolds and catheters in biological systems. Therefore, predicting viscosity and stiffness holds promise for widespread applications in medicine and other medical fields.

[0138] Among them, the free state of magnetic particles refers to each magnetic particle being independent and unrelated to each other, in a state of free movement; the combined state of magnetic particles refers to multiple magnetic particles being tightly packed together, forming a cluster, unable to move freely, and constrained and restricted by the surrounding magnetic particles.

[0139] The microenvironment refers to the pH value and granzyme environment of the magnetic particles, such as the pH value and granzyme environment of a tumor environment. Granzymes are serine proteases that play an important role in killing virus-infected cells and tumor cells by NK cells and CTLs. Granzymes are granule-related enzymes in cytotoxic T lymphocytes. The pH value and granzymes in the tumor environment can break down bound magnetic particles, causing them to change from a bound state to a free state. This embodiment of the invention can detect the microenvironment by detecting this change.

[0140] This invention employs prior research using known sample data to investigate the relationship between two relaxation times of magnetic nanoparticles and various functional parameters. The following example uses a single magnetic particle sample of Synomag-D. The viscosity change is achieved by altering the glycerol content in a buffer solution composed of a water-glycerol mixture. For instance, a series of water-glycerol mixtures with glycerol concentrations ranging from 1% to 70% wt can be prepared as buffer solutions for diluting Synomag-D. The hardness change is achieved by altering the gelatin concentration in a mixture composed of water and gelatin gel; for example, water-gelatin gels with 5% to 30% wt can be prepared, and so on. Temperature changes can be achieved by heating Synomag-D to multiple temperatures, such as 25°C, 37°C, and 52°C, using a digitally temperature-controlled water bath.

[0141] The following explains the research on some functional parameters:

[0142] (1) Relaxation time characteristics under different viscosities

[0143] As shown in Figure 7(a), the Debye time decreases with decreasing viscosity at low field amplitudes but remains constant at high field amplitudes; the Brownian time increases with increasing viscosity. The Neel time decreases slightly with increasing viscosity. The Brownian percentage first increases and then decreases with increasing viscosity.

[0144] Please refer to Figure 7(b), which shows the two relaxation times at different viscosity levels. Within the range of 0.9–3.2 mPa·s, the Brownian relaxation time exhibits a positive linear relationship with the viscosity level. When the viscosity is greater than 3.2 mPa·s, the Brownian relaxation time reaches saturation and does not change with further viscosity increases. The Néel relaxation time decreases slightly with increasing viscosity. When the viscosity level is greater than 3.2 mPa·s, the Néel relaxation time for all field amplitudes shows a similar saturation effect.

[0145] Therefore, it can be concluded that Brownian relaxation time increases with increasing viscosity.

[0146] (2) Relaxation time characteristics under different stiffnesses

[0147] Analysis shows that as gelatin concentration increases, stiffness increases, the proportion of Brownian relaxation time decreases, and the proportion of Néelian relaxation time widens (details not illustrated here). Therefore, there is a certain correlation between Brownian relaxation time and stiffness, and between Néelian relaxation time and stiffness.

[0148] In clinical practice, stiffness can correspond to conditions such as liver cirrhosis and tumor masses, and its relationship with relaxation time can be used for corresponding detection.

[0149] (3) Relaxation time characteristics at different temperatures

[0150] Based on the magnetization curve fitting method, the Debye relaxation time decreases with increasing temperature. The Brownian relaxation time also decreases with increasing temperature, but this decreasing trend weakens at high viscosities. See details below. Figure 8 As shown. Therefore, it can be concluded that the Brownian relaxation time decreases with increasing temperature.

[0151] The embodiments of the present invention can linearly fit the relaxation time or its characteristics under different magnetic field amplitudes, as the parameters for each functional parameter (i.e., viscosity η, elasticity (hardness) μ, and temperature T). P The function of linear fit. The absolute value of the slope is used to calculate sensitivity:

[0152]

[0153] Paramter sensitivity It can represent the sensitivity of any relaxation time to a certain functional parameter; Parameter initial Indicates the initial value of the function parameter; slope linearfit The absolute value of the slope of the linear fit is represented; Equation (13) describes the percentage change in relaxation time or characteristic caused by each unit change in functional parameter (such as a 1 mPa·s change in viscosity, a 1% change in gelatin concentration, or a 1 °C change in temperature). In the embodiments of the present invention, η initial =0.912 mPa·s; μ initial =5% gelatin;

[0154] Please see Figure 9 This demonstrates the sensitivity of predicting the relaxation time characteristic in viscosity using the magnetization curve fitting method. The Brownian relaxation time is significantly more sensitive to viscosity prediction than the Nieer relaxation time. The sensitivity of the Brownian relaxation time increases with increasing field amplitude, reaching a maximum at approximately 4.5 mT. When the field amplitude is greater than 7.0 mT, the Brownian component disappears, making it impossible to assess the sensitivity of viscosity prediction. At high field amplitudes (9.5 and 10 mT), the Nieer relaxation time does not change with increasing viscosity, exhibiting zero sensitivity.

[0155] Experimental data on the sensitivity of relaxation time features in temperature prediction show that, among magnetization curve fitting methods, Brownian relaxation time exhibits the highest sensitivity in temperature prediction, while Niehr relaxation time features show the lowest sensitivity.

[0156] (4) Relaxation time characteristics under different microenvironments

[0157] Experiments based on embodiments of the present invention lead to the conclusion that as the pH of the microenvironment increases or decreases, and as the amount of granzyme increases, magnetic particles change from a bound state to a free state, resulting in a shorter Brownian relaxation time, an increased peak amplitude, and a higher proportion of the peak. Specific details are not illustrated.

[0158] Therefore, Brownian relaxation time can be used to detect changes in the microenvironment and enable various applications in fields such as biology and medicine. For example, during apoptosis, cytoplasmic viscosity increases. In this invention, magnetic particles can be incorporated into cells, and relaxation analysis can be used to study cytoplasmic / organelle viscosity. For instance, using the aforementioned conclusions, Brownian relaxation time can be used to detect changes in the intracellular microenvironment, thereby determining the number of apoptotic cells. Studies have shown that the method of this invention can detect changes in intracellular viscosity of 1-10 mPa, thus effectively achieving apoptosis monitoring / detection.

[0159] For example, drugs may be released as the microenvironment changes. Based on the relaxation time of magnetic particles and the changing patterns of the microenvironment, combined with pre-determined patterns of drug release with changes in the microenvironment, drug release efficiency can be detected. Of course, the applications of microenvironment detection are not limited to those described above.

[0160] Therefore, based on the above conclusions and sensitivity calculation results, the embodiments of the present invention can predict functional parameters such as viscosity or temperature. For example, a certain correspondence can be pre-constructed based on a known single magnetic particle sample, viscosity value, Brownian relaxation time value at the corresponding viscosity value, and corresponding sensitivity calculation results; or based on a known single magnetic particle sample, temperature value, Brownian relaxation time value at the corresponding temperature value, and corresponding sensitivity calculation results, such as obtaining a corresponding prediction model through numerical fitting or machine learning. Once the corresponding relaxation time is detected, the corresponding functional parameter can be predicted based on the correspondence.

[0161] Secondly, in this embodiment of the invention, a digital blood vessel phantom is constructed through simulation for imaging studies.

[0162] As can be seen from the above research conclusions, functional parameters such as viscosity or stiffness can be predicted by using Brownian relaxation time. If a target to be tested, which has been pre-injected with magnetic nanoparticles, has different viscosities or stiffness in different regions near its blood vessels, it is feasible to distinguish them using Brownian relaxation time. This makes it feasible to distinguish between tumor regions, catheters, stents and normal vascular tissue regions. For easier and more intuitive display, imaging can be used to show the results.

[0163] As an experimental example, this embodiment of the invention constructs a digital blood vessel phantom using simulation methods. Specifically, it is a 5×6cm² vascular system model composed of a container structure with suspended magnetic nanoparticle clusters (MNPs), representing a multicolor MPI digital phantom, such as... Figure 10 As shown, in the diagram, catheter represents a duct; plaque represents a plaque; tumor represents a tumor; and vessel represents a blood vessel. This vascular system model was infused with the same type of magnetic nanoparticles. The diameter of the inserted ducts was ≥2.5 mm around the main branches of the blood vessels, and the diameter of the inserted ducts within the vessels was 0.83 mm. Different concentrations of magnetic nanoparticles were observed at the tumor margin and center. The lower half of the tumor experienced higher temperatures under hyperthermia. Point-by-point scanning of this vascular system model using FFP is illustrated in the diagram. Figure 11 As shown.

[0164] A uniform pulsed square wave excitation magnetic field can be used as the excitation field and combined with a free-floating pressure filter (FFP) for spectral imaging. The relaxation time imaging map can be generated by moving the FFP point-by-point across the entire field of view (FOV).

[0165] Assume the gradient of the chosen field is along the x and y directions, and the excitation field is along the z direction. The formula for calculating the total magnetic field is:

[0166]

[0167] Where t represents time; x, y, and z are three spatial coordinates. It is a unit vector in three coordinate directions; G x G represents the gradient field intensity in the x-direction; y H is the gradient field intensity in the y-direction; H(t) is the magnitude of the magnetic field at time t.

[0168] Near the center of the FFP, the magnetization fully recovered along the z-direction from a single MNP (magnetic nanoparticle) can be calculated as follows:

[0169]

[0170] Where L(H) is the Langevin function; H represents the magnitude of the total magnetic field; m represents the magnetic moment of a single magnetic nanoparticle; r represents the particle size in nanometers (nm); and H0 represents the magnetic field amplitude during the flat phase of the magnetic field. In FFP(G x =G y The magnetization completely recovered from a single MNP along the z-direction at the center of (=0) is expressed as:

[0171] M z (0)=m·L(H0) (16)

[0172] In this embodiment of the invention, the signal weighting factor is assumed to be the ratio of the magnetization in the z-direction at any position to the magnetization at the center of the FFP, as shown in equation (17). See also... Figure 12 When a pixel moves away from the center of the FFP, the signal weighting factor drops rapidly to zero.

[0173]

[0174] The overall signal, or the original signal, is composed of the digital phantom signal S(r,t) (representing the actual magnetic particle distribution) and the signal weighting factor w. s (r) yielded:

[0175] S FFP (r, t) = S(r, t) * w s (r) (18)

[0177] Here, "*" indicates convolution in the FOV spatial domain. The signal weighting factor is a point spread function (PSF), which is closely related to the image resolution.

[0178] Specifically, a selective magnetic field (G) can be used. x =G y =7.0T / m / u0) forms an FFP, which moves point-by-point throughout the FOV. A total of 100 × 100 FFP positions were generated in the simulation. For each FFP position, the signal weighting factor was convolved with the curve of the digital phantom signal to calculate the curve corresponding to the original signal.

[0179] Based on the specific content of S3 above, the magnetic particle relaxation time detection results at each FFP location can be obtained, including (τ N , τ B (a, b).

[0180] For (τ) N , τ B At least one parameter from (a, b) can be used for imaging in embodiment S4 of the present invention, and the values ​​of such parameters at all FFP locations can be used.

[0181] In one alternative implementation, due to the Brownian relaxation time, i.e., τ B The correlation between the observed changes in functional parameters such as viscosity and stiffness, and the imaging based on at least one term from the magnetic particle relaxation time detection results obtained from all FFP locations to obtain a relaxation time imaging map of the target, may include:

[0182] Based on the Brownian relaxation time constants obtained from the magnetic particle relaxation time detection results at all FFP locations, an image of the Brownian relaxation time of the target under test is obtained.

[0183] Specifically, for the digital vascular phantom image, baseline image, and Brownian relaxation time imaging at a field amplitude of 4.5 mT, please refer to [link to relevant documentation]. Figures 13(a) to 13(c) As shown; wherein, the digital blood vessel phantom image in Figure 13(a) is artificially designed; the baseline image in Figure 13(b) is reconstructed by calculating the signal area under the curve during the full excitation period at each FFP location, specifically the image of magnetization obtained by integrating the original signal, the integration is achieved using formula (4), the distribution of blood vessels can be seen on the baseline image, which can be compared with the Brownian relaxation time imaging obtained in the embodiment of the present invention; or in an optional embodiment, a colored Brownian relaxation time imaging can be superimposed on the baseline image to obtain the final imaging image, which is easier to observe, the Brownian relaxation time imaging in Figure 13(c) uses different colors to represent the Brownian relaxation time τ BSince there are no magnetic nanoparticles outside the blood vessel, this part appears black. In the normal area of ​​the blood vessel where no plaque or duct appears, the magnetic nanoparticles are uniformly distributed, so the Brownian relaxation time is the same, and the image color is consistent. However, the plaque has viscous MNPs compared to the normal blood vessel area, so the viscosity is different. The duct area has solidified MNPs compared to the normal blood vessel area, so the stiffness is different. It can be seen that the normal blood vessel area, plaque and duct can be distinguished by Brownian relaxation time imaging.

[0184] Specifically, the Brownian relaxation time was 37.32±1.10 μs in the plaque area, 30.31±0.25 μs in the duct area, and 31.52±1.07 μs in the vascular area. The Nielsen relaxation time was the same for all tissues: 8.14±3.55×8.14±0 μs in the plaque area, 8.14±3.55×10⁻¹⁵ μs in the duct area, and 8.14±1.30×10⁻¹³ μs in the vascular area.

[0185] See Figure 14 , Figure 14 This diagram illustrates the variation of Brownian relaxation times of plaque, duct, and vascular regions with the average signal period in a simulation experiment according to an embodiment of the present invention. At an excitation field amplitude of 4.5 mT, the Brownian relaxation times of the plaque, duct, and vascular regions initially decrease with increasing average signal period. When the average signal period is ≥ 20, the Brownian relaxation times from different tissue regions stabilize. In other words, the Brownian relaxation times of the plaque, duct, and vascular regions decrease and stabilize with increasing average signal period. If 20 excitation periods are used for image acquisition in this embodiment of the present invention, the total scanning time of the digital phantom is approximately 100 seconds (100 × 100 pixels × 20 / 2 kHz).

[0186] Of course, the imaging methods used in the embodiments of the present invention are not limited to Brownian relaxation time. Imaging can also be performed according to the correspondence between the other three items and the changes in functional parameters. For example, the proportion of Brownian relaxation time can be used, or Brownian relaxation time can be used in conjunction with imaging, etc., for applications in fields such as biomedicine. Detailed examples will not be provided here.

[0187] This invention utilizes the inversion recovery method to separate and detect relaxation time, and can combine this method with magnetic field free point scanning for multicolor magnetic particle imaging to distinguish plaques, ducts and vascular regions.

[0188] Secondly, corresponding to the above method embodiments, this invention also provides an imaging device based on fitting the relaxation time of magnetic particles to a magnetization curve, such as... Figure 15 As shown, the device includes:

[0189] FFP generation module 1501 is used to generate magnetic field free points (FFPs).

[0190] The original signal acquisition module 1502 is used to move the FFP within the spatial range of the target under test where a single type of magnetic nanoparticle has been injected, generate a pulsed square wave excitation magnetic field at each FFP position, and receive the original signal generated by the magnetic nanoparticles in the target under test at that FFP position under the excitation of the pulsed square wave excitation magnetic field.

[0191] The magnetic particle relaxation time detection module 1503 is used to, based on the original signal, fit a target magnetization curve that matches the original signal by adjusting the relaxation time-related parameters in a preset double exponential decay function; and use the adjustment result of the relaxation time-related parameters corresponding to the target magnetization curve as the magnetic particle relaxation time detection result at the FFP position; wherein, the relaxation time-related parameters include the Niehr relaxation time constant, the Brown relaxation time constant, and the percentages of Niehr relaxation and Brown relaxation;

[0192] The imaging module 1504 is used to perform imaging based on at least one of the same terms in the magnetic particle relaxation time detection results obtained from all FFP positions, so as to obtain a relaxation time imaging map of the target under test.

[0193] Optionally, the generation of the magnetic field free point FFP includes:

[0194] The magnetic field free point (FFP) is generated using a preset static gradient magnetic field.

[0195] Optionally, when the raw signal acquisition module 1502 generates a pulsed square wave excitation magnetic field at each FFP location it moves to, and receives the raw signal generated by the magnetic nanoparticles within the target under test at that FFP location under the excitation of the pulsed square wave excitation magnetic field, it is specifically used for:

[0196] At each FFP location, a pulsed square wave excitation magnetic field is generated using a pre-designed pulsed square wave relaxor, and the original signal generated by the magnetic nanoparticles in the target at that FFP location under the excitation of the pulsed square wave excitation magnetic field is received.

[0197] Optionally, the pulse square wave relaxor includes:

[0198] A digital acquisition card is used to generate analog signals of pulse square waves and to digitize input sample signals.

[0199] An AC power amplifier is used to amplify the analog signal;

[0200] The transmitting coil is used to transmit amplified analog signals to generate a pulsed square wave excitation magnetic field;

[0201] A current sensor is used to monitor the emission waveform of the AC power amplifier in real time.

[0202] A receiving coil is used to receive the sample signal generated by a single magnetic particle sample under the excitation of the pulsed square wave excitation magnetic field;

[0203] A low-noise preamplifier is used to amplify the sample signal and provide it to the digital acquisition card.

[0204] Optionally, when the magnetic particle relaxation time detection module 1503, based on the original signal, adjusts the relaxation time-related parameters in the preset double exponential decay function to fit a target magnetization curve matching the original signal, it is specifically used for:

[0205] The original signal is integrated to obtain the integrated curve.

[0206] Obtain the relaxation time-related parameters of the current group, and for the preset double exponential decay function, obtain a fitted magnetization curve corresponding to the relaxation time-related parameters of the current group.

[0207] Determine whether the point error between the currently obtained magnetization intensity curve and the integrated curve meets the preset requirements. If yes, use the currently obtained magnetization intensity curve as the target magnetization intensity curve. If not, adjust the parameters using the preset parameter range corresponding to each relaxation time related parameter to obtain a new set of relaxation time related parameters, and return to the process of obtaining the magnetization intensity curve.

[0208] Optionally, the preset double exponential decay function includes:

[0209]

[0210] Among them, M non-adiabatic M0 represents the non-adiabatic magnetization; t represents time; M0 represents the maximum magnetization under static field amplitude; t0 represents the time of magnetic field reversal; τ N τ represents the Niehr relaxation time constant; B τ represents the Brownian relaxation time constant; N <τ B ; a represents the percentage of Neill relaxation, b represents the percentage of Brownian relaxation, and a+b=1.

[0211] Optionally, when the imaging module 1504 performs imaging on at least one term of the magnetic particle relaxation time detection results obtained based on all FFP positions to obtain a relaxation time image of the target under test, it is specifically used for:

[0212] Based on the Brownian relaxation time constants obtained from the magnetic particle relaxation time detection results at all FFP locations, an image of the Brownian relaxation time of the target under test is obtained.

[0213] For details, please refer to the imaging method based on fitting the relaxation time of magnetic particles to the magnetization curve described in the first aspect, which will not be repeated here.

[0214] Thirdly, embodiments of the present invention also provide an electronic device, such as a 16 Figure 16 As shown, it includes a processor 1601, a communication interface 1602, a memory 1603, and a communication bus 1604, wherein the processor 1601, the communication interface 1602, and the memory 1603 communicate with each other through the communication bus 1604.

[0215] The memory is used to store computer programs;

[0216] When the processor executes the program stored in the memory, it implements the steps of any of the imaging methods based on magnetization curve fitting of magnetic particle relaxation time provided in the first aspect of the present invention.

[0217] The communication bus mentioned in the above electronic devices can be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus, etc. This communication bus can be divided into address bus, data bus, control bus, etc. For ease of illustration, only one thick line is used to represent it in the diagram, but this does not mean that there is only one bus or one type of bus.

[0218] The communication interface is used for communication between the aforementioned electronic devices and other devices.

[0219] The memory may include random access memory (RAM) or non-volatile memory (NVM), such as at least one disk storage device. Optionally, the memory may also be at least one storage device located remotely from the aforementioned processor.

[0220] The processors mentioned above can be general-purpose processors, including central processing units (CPUs), network processors (NPs), etc.; they can also be digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0221] The method provided in this invention can be applied to electronic devices. Specifically, the electronic device can be a desktop computer, a portable computer, a smart mobile terminal, a server, etc. No limitation is made herein; any electronic device that can implement this invention falls within the protection scope of this invention.

[0222] Fourthly, corresponding to the imaging method based on magnetization curve fitting of magnetic particle relaxation time provided in the first aspect, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of any of the imaging methods based on magnetization curve fitting of magnetic particle relaxation time provided in the first aspect of the present invention.

[0223] For the embodiments of the device / electronic device / storage medium, since they are basically similar to the method embodiments, the description is relatively simple, and relevant parts can be referred to in the description of the method embodiments.

[0224] It should be noted that the device, electronic device, and storage medium in the embodiments of the present invention are respectively devices, electronic devices, and storage media that apply the above-described imaging method based on fitting the relaxation time of magnetic particles based on magnetization curve. Therefore, all embodiments of the above-described imaging method based on fitting the relaxation time of magnetic particles based on magnetization curve are applicable to the device, electronic device, and storage medium, and can achieve the same or similar beneficial effects.

[0225] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention are included within the scope of protection of the present invention.

Claims

1. An imaging method based on fitting the relaxation time of magnetic particles using magnetization curves, characterized in that, include: Generates a magnetic field free point (FFP); The FFP is moved within the spatial range of the target under test, which has been injected with a single type of magnetic nanoparticle. At each FFP position, a pulsed square wave excitation magnetic field is generated, and the original signal generated by the magnetic nanoparticles in the target under test at that FFP position under the excitation of the pulsed square wave excitation magnetic field is received. Based on the original signal, by adjusting the relaxation time-related parameters in the preset double exponential decay function, a target magnetization curve matching the original signal is obtained; the adjustment result of the relaxation time-related parameters corresponding to the target magnetization curve is used as the magnetic particle relaxation time detection result at the FFP position; wherein, the relaxation time-related parameters include the Niehr relaxation time constant, the Brown relaxation time constant, and the percentages of Niehr relaxation and Brown relaxation; Imaging is performed based on at least one of the same terms from the magnetic particle relaxation time detection results obtained from all FFP positions to obtain a relaxation time image of the target under test.

2. The imaging method based on magnetization curve fitting of magnetic particle relaxation time according to claim 1, characterized in that, The generation of the magnetic field free point FFP includes: The magnetic field free point (FFP) is generated using a preset static gradient magnetic field.

3. The imaging method based on magnetization curve fitting of magnetic particle relaxation time according to claim 1, characterized in that, The process of generating a pulsed square wave excitation magnetic field at each FFP location and receiving the original signal generated by the magnetic nanoparticles within the target at that FFP location under the excitation of the pulsed square wave excitation magnetic field includes: At each FFP location, a pulsed square wave excitation magnetic field is generated using a pre-designed pulsed square wave relaxor, and the original signal generated by the magnetic nanoparticles in the target at that FFP location under the excitation of the pulsed square wave excitation magnetic field is received.

4. The imaging method based on magnetization curve fitting of magnetic particle relaxation time according to claim 3, characterized in that, The pulse square wave relaxor includes: A digital acquisition card is used to generate analog signals of pulse square waves and to digitize input sample signals. An AC power amplifier is used to amplify the analog signal; The transmitting coil is used to transmit amplified analog signals to generate a pulsed square wave excitation magnetic field; A current sensor is used to monitor the emission waveform of the AC power amplifier in real time. A receiving coil is used to receive the sample signal generated by a single magnetic particle sample under the excitation of the pulsed square wave excitation magnetic field; A low-noise preamplifier is used to amplify the sample signal and provide it to the digital acquisition card.

5. The imaging method based on magnetization curve fitting of magnetic particle relaxation time according to claim 4, characterized in that, The step of fitting a target magnetization curve that matches the original signal by adjusting the relaxation time correlation parameter in a preset double exponential decay function based on the original signal includes: The original signal is integrated to obtain the integrated curve. Obtain the relaxation time-related parameters of the current group, and for the preset double exponential decay function, obtain a fitted magnetization curve corresponding to the relaxation time-related parameters of the current group. Determine whether the point error between the currently obtained magnetization intensity curve and the integrated curve meets the preset requirements. If yes, use the currently obtained magnetization intensity curve as the target magnetization intensity curve. If not, adjust the parameters using the preset parameter range corresponding to each relaxation time related parameter to obtain a new set of relaxation time related parameters, and return to the process of obtaining the magnetization intensity curve.

6. The imaging method based on magnetization curve fitting of magnetic particle relaxation time according to claim 1 or 5, characterized in that, The preset double exponential decay function includes: Among them, M non-adiabatic M0 represents the non-adiabatic magnetization; t represents time; M0 represents the maximum magnetization under static field amplitude; t0 represents the time of magnetic field reversal; τ N τ represents the Niehr relaxation time constant; B τ represents the Brownian relaxation time constant; N <τ B ; a represents the percentage of Neill relaxation, b represents the percentage of Brownian relaxation, and a+b=1.

7. The imaging method based on magnetization curve fitting of magnetic particle relaxation time according to claim 1, characterized in that, Imaging is performed on at least one of the magnetic particle relaxation time detection results obtained based on all FFP positions to obtain a relaxation time imaging map of the target under test, including: Based on the Brownian relaxation time constants obtained from the magnetic particle relaxation time detection results at all FFP locations, an image of the Brownian relaxation time of the target under test is obtained.

8. An imaging device based on fitting the relaxation time of magnetic particles to a magnetization curve, characterized in that, include: The FFP generation module is used to generate magnetic field free points (FFPs). The original signal acquisition module is used to move the FFP within the spatial range of the target under test where a single type of magnetic nanoparticle has been injected. At each FFP position, a pulsed square wave excitation magnetic field is generated, and the original signal generated by the magnetic nanoparticles in the target under test at that FFP position under the excitation of the pulsed square wave excitation magnetic field is received. The magnetic particle relaxation time detection module is used to fit a target magnetization curve that matches the original signal by adjusting the relaxation time-related parameters in a preset double exponential decay function based on the original signal; and to use the adjustment result of the relaxation time-related parameters corresponding to the target magnetization curve as the magnetic particle relaxation time detection result at the FFP position; wherein, the relaxation time-related parameters include the Niehr relaxation time constant, the Brown relaxation time constant, and the percentages of Niehr relaxation and Brown relaxation; The imaging module is used to perform imaging based on at least one of the same terms in the magnetic particle relaxation time detection results obtained from all FFP positions, so as to obtain a relaxation time imaging map of the target under test.

9. An electronic device, characterized in that, It includes a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; The memory is used to store computer programs; When the processor executes the program stored in the memory, it implements the steps of the method described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the method described in any one of claims 1-7.

Images