CEST lipid artifact elimination method, medium, device, and magnetic resonance imaging apparatus

By incorporating fat signals into the Bloch-McConnell equation model for fitting and normalization in magnetic resonance imaging, the problem of fat artifact elimination is solved, the accuracy and robustness of CEST imaging is improved, and the scope of application is wider and the reduction of Collection time.

WO2024235016A1PCT designated stage expired Publication Date: 2024-11-21ZHEJIANG UNIV

Patent Information

Application Number
PCT/CN2024/090878
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2023-05-15
Filing Date
2024-04-30
Publication Date
2024-11-21

AI Technical Summary

Technical Problem

Existing magnetic resonance imaging techniques are difficult to effectively remove fat artifacts when applying chemical exchange saturation transfer imaging, especially in fat-rich tissues, resulting in inaccuracy of results and complexity of analysis.

Method used

Using a method based on magnetization transfer and fat signal fitting differential analysis, the fat signal is included in the Bloch-McConnell equation model, and fit and normalize it to eliminate fat artifacts and improve the accuracy of CEST imaging.

Benefits of technology

Without adjusting the acquisition sequence, fat artifact elimination in the CEST signal is achieved, which improves imaging accuracy and robustness, has a wider range of application and reduces acquisition time.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2024090878_21112024_PF_FP_ABST
    Figure CN2024090878_21112024_PF_FP_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of magnetic resonance. Disclosed are a CEST lipid artifact elimination method, a medium, a device, and a magnetic resonance imaging apparatus. According to the present invention, a lipid multi-peak model is incorporated into a simplified water-magnetization transfer (MT) two-pool Bloch-McConnell equation model for iterative solving, and fitting is performed on water, MT, and lipid multi-peak signals to obtain a reference signal for a CEST effect without direct lipid artifacts, and finally, the indirect influence of a lipid signal on a CEST signal is eliminated by means of normalization. A Bloch-McConnell equation involving lipids used in the present invention can more truly reflect the composition of a collected signal, so as to obtain a more accurate CEST reference signal. The present invention can eliminate lipid artifacts in CEST signals without adjusting the collection sequence, thereby improving the accuracy of CEST imaging.
Need to check novelty before this filing date? Find Prior Art

Description

CEST fat artifact elimination method, medium, equipment and magnetic resonance imaging device Technical Field

[0001] The present invention belongs to the field of magnetic resonance imaging, and in particular relates to the removal of fat artifacts and the extraction of contrast in chemical exchange saturation transfer imaging. Background Art

[0002] Chemical Exchange Saturation Transfer (CEST) imaging in magnetic resonance imaging can provide information at the molecular level and reflect different pathological changes. At present, CEST has been used to a considerable extent in the brain, for example, the protein and metabolite information provided by CEST is used to detect and grade brain gliomas. However, when CEST is applied to the body, especially fat-rich tissues such as the breast, it will be affected by strong fat artifacts. The analysis method of CEST is usually to subtract the reference signal from the labeled signal, and use the difference to reflect the CEST effect of the target group. The asymmetric magnetization transfer (MTR) commonly used in CEST analysis asym ) analysis uses the signal at the symmetric frequency of the target group with respect to the water resonance frequency as the reference signal. This method assumes that the signal at the symmetric frequency does not contain the CEST effect and that other effects are the same as the marker signal. The difference between the two is the CEST signal. However, in tissues containing fat, the resonant frequency range of fat usually overlaps with the frequency range of the reference signal, and the fat signal is not symmetric with respect to the water resonance frequency, which means that MTR asym The influence of fat cannot be excluded during analysis, resulting in fat artifacts in the results; in addition, magnetization transfer (MT) will also affect MTR to a certain extent. asym Analysis

[0003] Current methods for eliminating fat artifacts can be roughly divided into (1) methods based on acquisition sequence design and (2) methods based on post-processing analysis. Acquisition sequence design methods include fat suppression sequence, water excitation sequence, and CEST-Dixon water-fat separation method. These methods require additional sequence design, which is not conducive to the promotion of clinical application. The fat suppression sequence method has limited effect in suppressing fat and may have residual fat signals; the water excitation sequence increases the energy absorbed by the human body, limiting its acquisition speed and application range; the CEST-Dixon water-fat separation method requires multiple data for analysis, resulting in a significant increase in acquisition time. The post-processing analysis method is mainly based on the fitting analysis method of CEST signal normalization. The specific fitting method used is mainly multi-pool Lorentz fitting, and the single-pool Lorentz difference method, extrapolated magnetization transfer signal fitting method, etc. can also be used for analysis. The single-cell Lorentz difference method and the extrapolated magnetization transfer signal fitting method do not incorporate fat signals into the model, resulting in artifacts in the processing results; the multi-cell Lorentz fitting method assumes that low-power saturation reaches a steady state, which may not be met in actual applications. In addition, this method has many fitting parameters and requires a large amount of data for fitting, resulting in longer acquisition time and lower robustness.

[0004] Summary of the Invention

[0005] The primary objective of this invention is to overcome the shortcomings of existing methods by providing a more comprehensive and robust post-processing, widely applicable, and improved CEST fat artifact removal method based on differential analysis with fitted magnetization transfer and lipid signals (DIGITAL). This method incorporates fat signals into a model based on the Bloch-McConnell equation, obtains a reference signal through fitting, and then performs normalization to eliminate fat artifacts.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] In a first aspect, the present invention provides a CEST fat artifact removal method based on magnetization transfer (MT) and fat signal fitting and differentiation. The method performs fat artifact removal and signal extraction processing on each voxel in the raw data collected by CEST (Chemical Exchange Saturation Transfer, CEST). The processing process for each voxel includes:

[0008] Step 1: Perform linear interpolation correction of the main magnetic field frequency B0 field offset on the raw data collected by CEST to obtain the corrected Z spectrum corresponding to the target voxel;

[0009] Step 2: Using the data containing only MT in the corrected Z spectrum, as well as the data containing both direct water saturation effects and direct fat saturation effects, as fitting samples, the model parameters of the three-pool Bloch-McConnell equation for water, MT, and fat were fitted within their respective given upper and lower bounds; the three-pool Bloch-McConnell equation for water, MT, and fat was obtained by incorporating the multimodal model of fat into the two-pool Bloch-McConnell equation for water and MT;

[0010] Step 3: Substitute the fitted model parameters into the water-MT-fat three-pool Bloch-McConnell equation model, and generate the CEST reference signal for each frequency point in the corrected Z spectrum. Subtract the actual collected signal in the corrected Z spectrum from the reference signal corresponding to each frequency point to obtain the CEST effect value of the corresponding frequency point in the CEST effect curve;

[0011] Step 4: Normalize the CEST effect curve by dividing it by (1-FF), where FF is the fat fraction obtained by fitting. In the normalized CEST effect curve, the value at the frequency of interest is the CEST effect value after eliminating fat artifact interference.

[0012] Preferably, the water-MT-fat three-pool Bloch-McConnell equation model is as follows:

[0013] Where exp represents the exponential function with the natural constant e as the base; Δt represents the increment at time t; is the total magnetization vector of water and fat as the reference signal at time t+Δt, They represent the magnetization vector of the water part at time t The magnetization vector of the fat part, in Matrix M w =[M xw , M yw , M zw , M zm ] T , M xw 、M yw 、M zw 、M zm They are the x-direction magnetization intensity component of the water group, the y-direction magnetization intensity component of the water group, the z-direction magnetization intensity component of the water group, and the z-direction magnetization intensity component of the MT group, respectively. The matrix M f =[Mxf1 , M yf1 , M zf1 , M xf2 , M yf2 , M zf2 ,...,M xf7 , M yf7 , M zf7 ] T , M xf1 、M xf2、 ...、M xf7 are the x-direction magnetization intensity components of the 1st to 7th groups in fat, M yf1 、M yf2 ,...,M yf7 are the y-direction magnetization intensity components of the 1st to 7th groups in fat, M zf1 、M zf2 ,...,M zf7 are the z-direction magnetization intensity components of the 1st to 7th groups in fat respectively; fat fraction FF = M 0f / (M 0f +M 0w ) represents the ratio of fat signal to total water and fat signal, where M 0f is the magnetization intensity component M of the 1st to 7th groups in the fat under steady state 0f1 、M 0f2 ,...,M 0f7 The sum, M 0w is the magnetization intensity component of the water group in the steady state; the coefficient matrix where the matrix Vector C w =[0, 0, R 1w M 0w , R 1m M 0m ] T ; Δω w =ω RF -ω w -ΔB0,ω RF represents the applied RF pulse frequency, ω w represents the resonance frequency of the water cluster, ΔB0 represents the residual main magnetic field inhomogeneity after correction, ω1 represents the intensity of the applied RF pulse, and k wm 、k mw They represent the exchange rates from water group to MT group and MT group to water group, R rfm R represents the absorption rate of MT group to RF pulse, 1w =1 / T 1w 、R 2w =1 / T 2w、 R 1m =1 / T 1m 、R2m =1 / T 2m They represent the longitudinal relaxation rate of water group, the transverse relaxation rate of water group, the longitudinal relaxation rate of MT group, the transverse relaxation rate of MT group, and M 0m is the magnetization intensity of the MT group in steady state; matrix Vector C f =[0, 0, R 1f1 M 0f1 ,0,0,R 1f2 M 0f2 ,...,0,0,R 1f7 M 0f7 ] T , where the matrix A f1 ~A f7 The nth matrix in The relative amplitude M of the 7 fat groups 0f1 、M 0f2 ,...,M 0f7 The relationship between Δω and Δω is estimated by three parameters: CL, ndb, and nmidb. Parameter CL represents the fatty acid chain length, parameter ndb represents the number of double bonds per molecule, and parameter nmidb represents the number of methylene double bonds. fn =ω RF -ω fn -ΔB0,ω fn represents the resonance frequency of the nth group in fat, R 1fn =1 / T 1fn 、R 2fn =1 / T 2fn They represent the longitudinal relaxation rate of the nth group in fat and the transverse relaxation rate of the nth group in fat respectively; the longitudinal relaxation time of the 7 fat groups is set to be the same: T 1f =T 1f1 =T 1f2 =…=T 1f7 , the transverse relaxation time of the seven fat groups is also set to the same: T 2f =T 2f1 =T 2f2 =…=T 2f7 ;

[0014] The parameters to be fitted in the model are: T 1w , T 2w , T 2m , M 0m , k wm , FF, T 1f , T 2f ,ΔB0,CL,ndb,nmidb.

[0015] Preferably, the longitudinal relaxation rate R of the MT is1m As well as the resonance frequency of the fat peak, the preset values ​​were input into the model as prior conditions.

[0016] Preferably, the intensity ω1 of the applied radio frequency pulse is calculated from an additionally acquired radio frequency field intensity map and then input into the model as a priori condition.

[0017] Preferably, in step 2, the data containing only MT are data points whose frequencies in the corrected Z spectrum are within the range of 80 to 6 ppm; the data containing direct water saturation effects are data points whose frequencies in the corrected Z spectrum are within the range of 1.5 to -1.5 ppm; and the data containing direct fat saturation effects are data points whose frequencies in the corrected Z spectrum are within the range of -1 to -6 ppm.

[0018] Preferably, in step 2, the data used as fitting samples are data with frequencies in the range of 80 to 10 ppm and 0.75 to -6 ppm in the corrected Z spectrum.

[0019] As an example, in step 2, the parameter to be fitted in the model T 1w ,T 2w ,T 2m ,m 0m ,k wm ,FF,T 1f ,T 2f ,ΔB0,CL,ndb,nmidb are respectively [1.50, 0.100, 5e -6 , 0.4500, 40, 1, 0.400, 0.150, 64, 18, 5.5, 2.5], and the lower bounds are [1.35, 0.015, 1e -6 , 0.0001, 20, 0, 0.250, 0.025, -64, 17, 1, 0], and the initial values ​​of the fitting are [1.35, 0.015, 1e -6 , 0.0001,20,1,0.400,0.025,0,17,1,0].

[0020] Preferably, in step 2, the absorption rate of the MT group to the radio frequency pulse is Where g(·) is a linear function, including Gaussian, Lorentzian, and super-Lorentzian types. In breast data, the linear function is preferably super-Lorentzian.

[0021] Preferably, in step 2, a higher weight is set for the data in the range of 80 to 10 ppm during fitting, which is twice the weight of the data in the range of 0.75 to -6 ppm.

[0022] Preferably, in step 4, the frequency of interest is 3.5 ppm, and the CEST effect value at 3.5 ppm in the normalized CEST effect curve is APT # value.

[0023] In a second aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as described in any one of the above methods is implemented.

[0024] In a third aspect, the present invention provides a computer electronic device comprising a processor and a storage medium, wherein a computer program is stored on the storage medium. When the computer program is executed by the processor, the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as described in any one of the above methods is implemented.

[0025] In a fourth aspect, the present invention provides a magnetic resonance imaging device for removing CEST fat artifacts, which includes a magnetic resonance scanner and a control unit, wherein the magnetic resonance scanner is used to obtain CEST images through magnetic resonance CEST imaging; the control unit can obtain the CEST images and a computer program is stored in the control unit. When the computer program is executed, it is used to implement the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as described in any one of the above methods, and output the CEST effect value of each voxel for eliminating fat artifacts.

[0026] The present invention has the following beneficial effects:

[0027] This method incorporates a multimodal model of fat into the Bloch-McConnell equation for an iterative solution, simultaneously fitting the multimodal signals of water, magnetization transfer, and fat. The Bloch-McConnell equation, which includes fat, more accurately reflects the composition of the acquired signal, thereby obtaining a more accurate CEST reference signal. This eliminates fat artifacts in the CEST signal without adjusting the acquisition sequence, improving the accuracy of CEST imaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] FIG1 is a schematic diagram of a specific implementation process of the present invention;

[0029] FIG2 is a data curve collected and a fitted data curve in the embodiment;

[0030] Figure 3 shows the dynamic enhancement image collected in the embodiment and the APT obtained by this method. # picture;

[0031] FIG4 is a fat fraction diagram obtained by fitting using the method proposed in the present invention in an embodiment. DETAILED DESCRIPTION

[0032] The present invention will be further described below in conjunction with specific embodiments and accompanying drawings.

[0033] The present invention provides a post-processing, widely applicable, and more robust CEST fat artifact removal method based on differential analysis with fitted magnetization transfer and lipid signals (DIGITAL). This method incorporates the seven resonant peaks of fat into a simplified two-pool Bloch-McConnell equation model based on water-MT. By fitting the water, MT, and fat signals, a reference signal of the CEST effect, free of direct fat artifacts, is obtained. Finally, normalization is performed to eliminate the indirect influence of the fat signal on the CEST signal.

[0034] The core of this method lies in rationally incorporating fat into the simplified two-tank Bloch-McConnell equation for water and MT, thereby establishing a three-tank Bloch-McConnell equation for water, MT, and fat. Simultaneously, this model is rationally simplified, reducing the computational resource requirements and thus expanding the scope of this method's application. The following further explains the three-tank Bloch-McConnell equation for water, MT, and fat in this method:

[0035] The simplified Bloch-McConnell equation for the water-MT two-tank (water tank w, MT tank w) commonly used in the field can be expressed as:

[0036] Among them, M xw 、M yw 、M zw 、M zm They are the x-direction magnetization intensity component of the water group, the y-direction magnetization intensity component of the water group, the z-direction magnetization intensity component of the water group, and the z-direction magnetization intensity component of the MT group, Δω w =ω RF -ω w ,ω RF represents the applied RF pulse frequency, ω w represents the resonance frequency of the water cluster, ω1(t) represents the intensity of the RF pulse applied at time t, k wm 、kmw They represent the exchange rates from water group to MT group and MT group to water group, R rfm R represents the absorption rate of MT group to RF pulse, 1w =1 / T 1w、 R 2w =1 / T 2w 、R 1m =1 / T 1m 、R 2m =1 / T 2m They represent the longitudinal relaxation rate of water group, the transverse relaxation rate of water group, the longitudinal relaxation rate of MT group, the transverse relaxation rate of MT group, and M 0w 、M 0m are the magnetization intensity of the water group in steady state and the magnetization intensity of the MT group in steady state, respectively.

[0037] The absorption rate R of MT group to radio frequency pulse rfb It can be expressed as Where g(·) is a linear function, including Gaussian, Lorentzian, and super-Lorentzian types. It is generally believed in the field that super-Lorentzian linear functions are closer to the MT signal in the human body.

[0038] For super-Lorentzian linear functions:

[0039] where Δω m =ω RF -ω m ,ω m Indicates the resonant frequency of the water cluster.

[0040] The simplified water-MT two-pool Bloch-McConnell equation described above can only describe the exchange process between water and the magnetization transfer signal. When fat signals are present, the CEST data curve will contain fat resonance peaks. During two-pool fitting, fat signals may be present in the fitted sample data, resulting in deviations in the two-pool fitting results. In addition, the preprocessing of CEST data is also affected by fat, and the CEST signal is distorted by the modulation of fat content.

[0041] For a resonance peak signal of fat, it can be described by the Bloch equation:

[0042] Among them, M xf 、M yf 、M zf are the x, y, and z direction magnetization intensity components, Δω f =ω RF -ω f ,ωRF represents the applied RF pulse frequency, ω f represents the resonant frequency of the resonance peak, ω1(t) represents the intensity of the RF pulse applied at time t, and R 1f =1 / T 1f、 R 2f =1 / T 2f Represent the longitudinal relaxation rate and the transverse relaxation rate, M 0f is the magnetization intensity in the steady state.

[0043] For the water-MT two-pool Bloch-McConnell equation, it is written in matrix form as follows:

[0044] Among them, M w =[M xw ,M yw ,M zw ,M zm ] T , C w =[0,0,R 1w M 0w ,R 1m M 0m ] T

[0045] Further simplification yields:

[0046] in,

[0047] The solution of the above formula is:

[0048] in is the magnetization vector of the Bloch-McConnell equation model of water and MT pools at time (t+Δt).

[0049] Similarly, for a single peak of fat, the Bloch equation is simplified and solved according to the above method, and its solution is:

[0050] in is the magnetization vector of a single resonance peak of fat at time (t+Δt), The superscript “1” in the formula represents a fat signal with one resonance peak.

[0051] Since there is no chemical exchange or other interactions between the various resonance peaks of fat, the fat multi-peak signal can be obtained by summing the signals of multiple fat peaks according to the amplitude ratio of the fat peaks. The summation process can be completed through matrix operations, let

[0052] in Vector C f =[0, 0, R 1f1 M 0f1 ,0,0,R 1f2 M 0f2 ,...,0,0,R 1f7 M 0f7 ] T , where the matrix A f1 ~A f7 The nth matrix in The relative amplitude M of the 7 fat groups 0f1 、M of2 ,...,M 0f7 The relationship between them can be estimated by three parameters: fatty acid chain length (CL), number of double bonds per molecule (ndb), and number of methylene double bonds (nmidb). Specifically, the number of hydrogen protons contained in each of the seven groups in each fat molecule can be expressed using the three parameters:

[0053] The number of hydrogen protons in group #1 is M1 = 9

[0054] The number of hydrogen protons in group #2, M2 = ((CL-4)×6)-(ndb×8)+(nmidb×2)

[0055] The number of hydrogen protons contained in group #3 is M3 = 6

[0056] The number of hydrogen protons contained in group #4 is M4 = 6 + (ndb-nmidb) × 4

[0057] The number of hydrogen protons contained in group #5 M5 = nmidb × 2

[0058] The number of hydrogen protons contained in group #6 is M6 = 4

[0059] The number of hydrogen protons contained in group #7 M7=ndb×2+1

[0060] The relationship between the relative amplitudes of the seven fat groups is determined by the chemical composition of the fat. The number of hydrogen protons M in any nth group is n Dividing by the total number of hydrogen protons contained in a fat molecule, we can get the relative amplitude between the groups, n = 1, 2, ..., 7.

[0061] Δω fn =ω RF -ω fn -ΔB0,ω fn represents the resonance frequency of the nth group in fat, R1fn =1 / T 1fn 、R 2fn =1 / T 2fn They represent the longitudinal relaxation rate of the nth group in fat and the transverse relaxation rate of the nth group in fat, respectively.

[0062] Among them, M f =[M xf1 , M yf1 , M zf1 , M xf2 , M yf2 , M zf2 ,...,M xf7 , M yf7 , M zf7 ] T , M xf1、 M xf2 ,...,M xf7 are the x-direction magnetization intensity components of the 1st to 7th groups in fat, M yf1 、M yf2 ,...,M yf7 are the y-direction magnetization intensity components of the 1st to 7th groups in fat, M zf1 、M zf2 ,...,M zf7 They are the z-direction magnetization intensity components of the 1st to 7th groups in fat, respectively.

[0063] Through the above changes, the sum of the fat multi-peak signals can be expressed as:

[0064] By using formula (1) and formula (3), the water and fat at a specific saturation pulse frequency ω can be iteratively calculated respectively. RF The signal M after saturation w (ω RF ), M f (ω RF There is no chemical exchange between water and fat, and the final collected signal can be simply expressed as the sum of water and fat signals, that is, M acq (ω RF )=M w (ω RF )+M f (ω RF ).

[0065] In CEST data processing, the collected data needs to be divided by the unsaturated image data to calculate the Z spectrum, and the reference Z spectrum M can be obtained by dividing the fitted signal by the unsaturated data. ref :

[0066] Let the fat fraction FF be the ratio of unsaturated fat to unsaturated water-fat signal, FF = M 0f / (M 0f +M 0w ), the above formula can be simplified as:

[0067] M in the above formula 0w With M 0f This can be eliminated by setting the steady-state value during fitting to 1. Note that the steady-state value during fitting does not have to be equal to the steady-state value in fat fraction. Thus, the above formula can be further simplified to: ref (ω RF )=M w (ω RF )·(1-FF)+M f (ω RF )·FF#(4)

[0068] In this way, it is equivalent to directly fitting the percentage of water and fat signals after saturation as unsaturated signals, that is, directly fitting the Z spectrum, and then adding them according to the ratio of water and fat unsaturated signals.

[0069] From formulas (1), (3), and (4), we can get the formula for the reference signal changing with time:

[0070] Consider a specific frequency of interest ω i The CEST effect at the reference signal is calculated directly using the CEST effect value CEST(ω i ) can be expressed as: CEST(ω i )=M ref (ω i )-M acq (ω i )

[0071] Among them, M acq (ω i ) is the saturation frequency ω i The CEST effect is caused by the exchange of solute molecules with water, so the fat part of the reference signal and the collected signal are the same. So: CEST(ω i )=(M wref -M wacq )·(1-FF)

[0072] Among them, M wref 、M wacqare the water part signals in the reference and collected data, respectively. According to the above formula, the CEST effect directly calculated is affected by (1-FF). Dividing both sides of the above formula by (1-FF) can eliminate the scaling of the CEST effect by fat fraction. The CEST effect after fat fraction correction is recorded as CEST′(ω i ), there may be:

[0073] Formula (5) is the model used in the present invention, and formula (6) is the algorithm for calculating the CEST effect using this model.

[0074] Since the relaxation rates of fat are relatively close, we assume that T 1f =T 1f1 =T 1f2 =…=T 1f7 , T 2f =T 2f1 =T 2f2 =…=T 2f7 To reduce the fitting operation. The longitudinal relaxation rate R of MT 1m The resonance frequency of the fat peak is input into the model as a priori conditions according to the data in the literature. Finally, the model of the present invention contains 12 unknowns, which are T 1w , T 2w , T 2m , M 0m , k wm , FF, T 1f , T 2f , ΔB0, CL, ndb, nmidb, and all other parameters are known. Model fitting yields a reference signal that incorporates both fat and MT effects, eliminating the direct influence of fat on the CEST effect. The CEST effect is then renormalized according to the CEST effect calculation formula to eliminate the scaling effect of fat on the CEST effect, resulting in a CEST effect value that excludes fat artifacts.

[0075] Therefore, based on the above theoretical discussion, the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference in the present invention can be specifically implemented as follows:

[0076] Each voxel in the raw data collected by CEST is processed for fat artifact removal and signal extraction. The processing of each voxel includes:

[0077] Step 1: Perform linear interpolation correction of the main magnetic field frequency B0 field offset on the raw data collected by CEST to obtain the corrected Z spectrum corresponding to the target voxel.

[0078] Step 2: Using the data containing only MT in the corrected Z spectrum, as well as the data containing both direct water saturation effects and direct fat saturation effects, as fitting samples, the model parameters of the water-MT-fat three-pool Bloch-McConnell equation model were fitted within the given upper and lower bounds. The above three-pool water-MT-fat Bloch-McConnell equation model is described in Equation (5) and will not be repeated here.

[0079] It should be noted that when extracting data points from the corrected Z spectrum as fitting samples, the specific frequency ranges selected are determined based on the frequency ranges of the three effects. Generally speaking, data containing only MT are those within the 80-6 ppm range in the corrected Z spectrum; data containing direct water saturation effects are those within the 1.5-1.5 ppm range; and data containing direct fat saturation effects are those within the -1-6 ppm range. Theoretically, data points within these three frequency ranges in the corrected Z spectrum can be used as fitting samples. However, to ensure the best fitting results, it is best not to use these data points near the final frequency of interest. For example, if the frequency of interest is 3.5 ppm, data points near 3.5 ppm should not be used as fitting samples. Therefore, it is preferable to use data within the 80-10 ppm and 0.75-6 ppm ranges in the corrected Z spectrum as fitting samples.

[0080] Step 3: Substitute the fitted model parameters into the water-MT-fat three-pool Bloch-McConnell equation model, and generate the CEST reference signal for each frequency point in the corrected Z spectrum. Subtract the actual collected signal in the corrected Z spectrum from the reference signal corresponding to each frequency point to obtain the CEST effect value of the corresponding frequency point in the CEST effect curve.

[0081] Step 4: Normalize the CEST effect curve by dividing it by (1-FF), where FF is the fat fraction obtained by fitting. In the normalized CEST effect curve, the value at the frequency of interest is the CEST effect value after eliminating fat artifact interference.

[0082] It should be noted that the selection of the frequency of interest is arbitrary, but in the research of this field, the frequency of interest is usually selected as the resonance frequency of the group containing the CEST effect to reflect the signal intensity of the group at the frequency of interest. For example, when the frequency of interest is selected as 3.5ppm, the CEST effect value calculated by this method is APT #The amide proton transfer signal from digital analysis (CEST) is a fat artifact elimination method based on magnetization transfer and fat signal fitting and differentiation. Other frequencies that reflect the CEST effect can also be selected as frequencies of interest; this method does not impose any restrictions on the selection of frequencies of interest.

[0083] Below, this application will further illustrate the method proposed in the present invention through the following non-limiting examples:

[0084] Example

[0085] In this embodiment, the specific implementation process of the DIGITAL method is shown in Figure 1. The method proposed by the present invention requires first collecting CEST image data, and then performing the DIGITAL method on each voxel in the collected CEST image data, that is, performing steps 1 to 4 in sequence, and calculating the APT of each voxel after eliminating fat artifacts. # The specific implementation process of steps 1 to 4 will be described in detail below.

[0086] Step 1: Perform linear interpolation correction of the main magnetic field frequency B0 field offset on the raw data collected by CEST to obtain the corrected Z spectrum.

[0087] Step 2: The Z-spectrum data containing only MT, as well as the data containing both direct water saturation effects and direct fat saturation effects, are used as fitting samples. The pre-defined three-pool Bloch-McConnell equation model for water, MT, and fat is fitted within the given upper and lower bounds. This fitting model is used to calculate the Z-spectrum reference signal. As described above, the three-pool Bloch-McConnell equation model for water, MT, and fat used in this method is as follows:

[0088] Where exp represents the exponential function with the natural constant e as the base; Δt represents the increment at time t; is the total magnetization vector of water and fat as the reference signal at time t+Δt, They represent the magnetization vector of the water part at time t Magnetization vector of fat part in Matrix M w =[M xw , M yw , M zw , M zm ] T , M xw 、M yw、M zw 、M zm They are the x-direction magnetization intensity component of the water group, the y-direction magnetization intensity component of the water group, the z-direction magnetization intensity component of the water group, and the z-direction magnetization intensity component of the MT group. f =[M xf1 , M yf1 , M zf1 , M xf2 , M yf2 , M zf2 ,...,M xf7 , M yf7 , M zf7 ] T , M xf1 、M xf2、 ...、M xf7 are the x-direction magnetization intensity components of the 1st to 7th groups in fat, M yf1 、M yf2 ,...,M yf7 are the y-direction magnetization intensity components of the 1st to 7th groups in fat, M zf1 、M zf2 ,...,M zf7 are the z-direction magnetization intensity components of the 1st to 7th groups in fat respectively; fat fraction FF = M 0f / (M 0f +M 0w ) represents the ratio of fat signal to total water and fat signal, where M 0f is the magnetization intensity component M of the 1st to 7th groups in the fat under steady state 0f1 、M 0f2 ,...,M 0f7 The sum, M 0w is the magnetization intensity component of the water group in the steady state; the coefficient matrix where the matrix Vector C w =[0, 0, R 1w M 0w , R 1m M 0m ] T Δω w =ω RF -ω w -ΔB0,ω RF represents the applied RF pulse frequency, ω w represents the resonance frequency of the water cluster, ΔB0 represents the residual main magnetic field inhomogeneity after correction, ω1 represents the intensity of the applied RF pulse, and k wm 、k mw represents the exchange rate from water group to MT group and MT group to water group, where g(·) is a super-Lorentz type, which represents the absorption rate of the MT group to the RF pulse, R 1w =1 / T 1w 、R 2w =1 / T 2w 、R 1m =1 / T 1m 、R 2m =1 / T 2m They represent the longitudinal relaxation rate of water group, the transverse relaxation rate of water group, the longitudinal relaxation rate of MT group, the transverse relaxation rate of MT group, and M 0m is the magnetization intensity of the MT group in steady state, the matrix Vector C f =[0, 0, R 1f1 M 0f1 ,0,0,R 1f2 M 0f2 ,...,0,0,R 1f7 M 0f7 ] T , where the matrix A f1 ~A f7 The nth matrix in The relative amplitude M of the 7 fat groups 0f1 、M of2 ,...,M 0f7 The relationship between them can be estimated by three parameters: fatty acid chain length (CL), number of double bonds per molecule (ndb), and number of methylene double bonds (nmidb). Specifically, the number of hydrogen protons contained in each of the seven groups in each fat molecule can be expressed using the three parameters:

[0089] The number of hydrogen protons in group #1 is M1 = 9

[0090] The number of hydrogen protons in group #2, M2 = ((CL-4)×6)-(ndb×8)+(nmidb×2)

[0091] The number of hydrogen protons contained in group #3 is M3 = 6

[0092] The number of hydrogen protons contained in group #4 is M4 = 6 + (ndb-nmidb) × 4

[0093] The number of hydrogen protons contained in group #5 M5 = nmidb × 2

[0094] The number of hydrogen protons contained in group #6 is M6 = 4

[0095] The number of hydrogen protons contained in group #7 M7=ndb×2+1

[0096] The relationship between the relative amplitudes of the seven fat groups is determined by the chemical composition of the fat. The number of hydrogen protons M in any nth group is n Dividing by the total number of hydrogen protons in a fat molecule gives the relative magnitude of each group, n = 1, 2, ..., 7. Δω fn =ω RF -ω fn -ΔB0,ω fn represents the resonance frequency of the nth group in fat, R 1fn =1 / T 1fn 、R 2fn =1 / T 2fn They represent the longitudinal relaxation rate of the nth group in fat and the transverse relaxation rate of the nth group in fat, respectively.

[0097] In the above model, since fat contains 7 resonance peaks, there are too many unknown parameters in the model, so we simplify the 7-peak model of fat. The main differences between the 7 peaks of fat are the differences in their resonance frequencies and relative amplitudes. Their relaxation rates do not differ much. Therefore, we assume that in the model, the relaxation rates of different resonance peaks of fat are the same, that is, T 1f =T 1f1 =T 1f2 =…=T 1f7 , T 2f =T 2f1 =T 2f2 =…=T 2f7 In addition, this method requires additional acquisition of the RF field intensity map to calculate the RF pulse intensity ω1 actually received by each voxel.

[0098] After the above simplification, the model used in this method contains 12 undetermined parameters, namely T 1w , T 2w , T 2m , M 0m , k wm , FF, T 1f , T 2f , ΔB0, CL, ndb, nmidb. These undetermined parameters are fitted by selecting data from the Z spectrum containing only MT, as well as data containing direct water saturation effects and direct fat saturation effects. In this example, the data selected are within the frequency ranges of 80-10 ppm and 0.75--6 ppm. During fitting, the weight of the data in the 80-10 ppm range is preferably set to twice that of the data in the 0.75--6 ppm range to balance the difference in the two data sets.

[0099] In this embodiment, the 12 undetermined parameters T 1w, T 2w , T 2m , M 0m , k wm , FF, T 1f , T 2f ΔB0, CL, ndb, and nmidb can be estimated directly by fitting samples, obtaining fitting values ​​that minimize the total fitting error in the model used in this method within the upper and lower bounds. The total fitting error can be expressed as the minimum mean square error, and the weight setting can be achieved by changing the error coefficient when calculating the minimum root mean square error. The specific fitting method can be implemented using software such as MATLAB and Python, or other existing technologies.

[0100] During fitting, the undetermined parameter T 1w , T 2w , T 2m , M 0m , k wm , FF, T 1f , T 2f , the upper and lower bounds of ΔB0, CL, ndb, and nmidb can be adjusted based on the fitted data. When fitting different human tissues or water models, the undetermined parameters need to be adjusted based on the normal parameter range of the tissue or water model. In this example, breast data is fitted, and the parameter T 1w , T 2w , T 2m , M om , k wm , FF, T 1f , T 2f , the upper bounds of ΔB0, CL, ndb, and nmidb are set to [1.50, 0.100, 5e -6 , 0.4500, 40, 1, 0.400, 0.150, 64, 18, 5.5, 2.5], and the lower bounds are set to [1.35, 0.015, 1e -6 , 0.0001, 20, 0, 0.250, 0.025, -64, 17, 1, 0], and the initial values ​​of the fitting are [1.35, 0.015, 1e -6 ,0.0001,20,1,0.400,0.025,0,17,1,0]

[0101] Step 3: Substitute the above fitting parameters into the model proposed by this method to calculate the reference curve including MT effect, water direct saturation effect, and fat direct saturation effect. The horizontal axis of the curve is the input fitting frequency ω, and the vertical axis is the reference signal The CEST effect curve is obtained by subtracting the Z spectrum curve after the main magnetic field B0 correction from the reference curve point by point. It should be noted that since the corrected Z spectrum curve is actually composed of discrete data points, when calculating the CEST effect curve, the corresponding reference signal should be calculated in the model with the fitting parameters substituted based on the frequency of each data point in the Z spectrum curve. The original value of the corresponding frequency on the Z spectrum curve is then subtracted from the reference signal to obtain the CEST effect value of the corresponding evaluation rate in the CEST effect curve.

[0102] Step 4: Renormalize the CEST effect curve calculated in step 3 by dividing it by (1-FF) obtained by fitting the voxel to eliminate the scaling of the CEST effect by the fat signal, and obtain the corrected CEST effect curve. The data at the frequency of 3.5ppm in this curve is the APT. # Effect value.

[0103] Of course, when calculating the reference curve in the above steps, it is not necessary to calculate the complete reference curve. It is only necessary to calculate the value at 3.5ppm. The CEST effect value at 3.5ppm can be obtained by dividing it by (1-FF) to obtain the corrected CEST effect value. # Effect value.

[0104] After performing steps 1 to 4 above on all voxels in the CEST image, each voxel corresponds to an APT # Value, these APT # The value corresponds to each voxel in the CEST image one by one, and an APT # picture.

[0105] The above-mentioned DIGITAL method was applied to a breast tumor patient for testing to further demonstrate the technical effect of the method. In this embodiment, the imaging sequence for acquiring CEST data includes a CEST saturation module and a fast spin echo acquisition module. The CEST saturation module contains 10 Gaussian waveform saturation pulses with a duration of 100ms, an effective amplitude of 1uT, and an interval between saturation pulses of 10ms. It should be noted that this method does not require fat suppression during CEST imaging, so there is no need to add a fat suppression module during CEST acquisition. In addition, additional main magnetic field B0 images and radio frequency field B1 images are also collected for B0 correction and parameter fitting.

[0106] As shown in Figure 1, the Z spectrum data collected in CEST is corrected for the main magnetic field B0 offset based on the collected main magnetic field B0 map. The collected Z spectrum is shown in the experimental data in Figure 2. Then, digital fitting is performed on all voxels. The fitting samples are data with frequencies between 80 and 10 ppm and 0.75 and -6 ppm. The collected radio frequency field B1 map is also input into the fitting process to calculate the radio frequency ω1. The upper and lower bounds and initial values ​​of the parameters are the same as those in step 2 above. The samples used for fitting and the reference curve obtained by fitting are shown in Figure 2.

[0107] The experimental data obtained in this example are shown in Figure 3. # The image has a corresponding high signal area in the high signal area of ​​the dynamic enhanced image (indicated by the black arrow in Figure 3). The color bar on the right only indicates the APT # Figure. The fat fraction figure obtained by fitting in this experiment is shown in Figure 4, where obvious changes in fat fraction can be observed. # Figure 3 and fat fraction, APT of glandular area with low fat fraction and glandular area with high fat fraction # There was no significant difference in effect size, suggesting that fat artifact had been eliminated.

[0108] Similarly, based on the same inventive concept, another preferred embodiment of the present invention further provides a computer electronic device corresponding to the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference provided in the above embodiment, which includes a memory and a processor;

[0109] The memory is used to store computer programs;

[0110] The processor is configured to implement the aforementioned CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference when executing the computer program.

[0111] Similarly, based on the same inventive concept, another preferred embodiment of the present invention also provides a computer-readable storage medium corresponding to the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference provided in the above embodiment, and the storage medium stores a computer program. When the computer program is executed by the processor, the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as described above is implemented.

[0112] It is understood that the above-mentioned storage medium and memory may include random access memory (RAM) and non-volatile memory (NVM), such as at least one disk storage. At the same time, the storage medium may also be various media that can store program code, such as a USB flash drive, a mobile hard disk, a magnetic disk, or an optical disk. Of course, with the widespread use of cloud servers, the above-mentioned software programs can also be installed on the cloud platform to provide corresponding services. Therefore, computer-readable storage media are not limited to the form of local hardware.

[0113] It is understandable that the above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0114] It should also be noted that those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working process of the device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the various embodiments provided in this application, the division of steps or modules in the device and method is only a logical function division. In actual implementation, there may be other division methods, for example, multiple modules or steps can be combined or integrated together, and a module or step can also be split.

[0115] Furthermore, the logic instructions in the aforementioned memory can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention.

[0116] Similarly, based on the same inventive concept, another preferred embodiment of the present invention further provides a magnetic resonance imaging device corresponding to the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference provided in the above embodiment, which includes a magnetic resonance scanner and a control unit;

[0117] The magnetic resonance scanner is used to obtain CEST images through magnetic resonance CEST imaging;

[0118] The control unit can acquire the CEST image and a computer program is stored in the control unit. When the computer program is executed, it is used to implement the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as described above, and output the CEST effect value of each voxel for eliminating fat artifacts.

[0119] It should be noted that the MRI device may be any MRI scanner capable of implementing parallel imaging methods. Its structure is state-of-the-art, and a mature commercial product may be employed, with no specific limitation on the specific model. Furthermore, in addition to storing the aforementioned computer program, the control unit of the MRI device should also contain the imaging sequences and other software programs necessary to implement CEST imaging.

[0120] The modules and functions involved in the above-mentioned embodiments of the present invention can be implemented by circuits, other hardware or executable program codes, as long as the corresponding functions can be realized. If code is used, the code can be stored in a storage device and executed by corresponding components in a computing device. The implementation of the present invention is not limited to any specific combination of hardware and software. All hardware models in the present invention can be commercially available products and can be selected according to actual user needs. Of course, the magnetic resonance CEST imaging sequence and device also need to be coordinated with other necessary hardware or software, which will not be described here.

[0121] The above-described embodiments are merely embodiments of the present application and are not intended to limit the present invention. Any changes and improvements to the present invention, or direct or indirect applications in other related technical fields, without departing from the spirit and scope of the present invention, are included in the present invention and are protected by the appended claims.

Claims

1. A CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference, It is characterized in that Each voxel in the raw data collected by CEST is processed for fat artifact removal and signal extraction. The processing of each voxel includes: Step 1: Perform linear interpolation correction of the main magnetic field frequency B0 field offset on the raw data collected by CEST to obtain the corrected Z spectrum corresponding to the target voxel; Step 2: Using the data containing only MT in the corrected Z spectrum and the data containing the direct saturation effect of water and the direct saturation effect of fat as fitting samples, the model parameters in the water-MT-fat three-pool Bloch-McConnell equation model are fitted within the respective given upper and lower bounds; the water-MT-fat three-pool Bloch-McConnell equation model is obtained by incorporating the multi-peak model of fat into the water-MT two-pool Bloch-McConnell equation; Step 3: Substitute the fitted model parameters into the water-MT-fat three-pool Bloch-McConnell equation model, and generate the CEST reference signal for each frequency point in the corrected Z spectrum. Subtract the actual collected signal in the corrected Z spectrum from the reference signal corresponding to each frequency point to obtain the CEST effect value of the corresponding frequency point in the CEST effect curve; Step 4: Divide the CEST effect curve by (1-FF) for normalization, where FF is the fat fraction obtained by fitting; in the normalized CEST effect curve, the value at the frequency of interest is the CEST effect value after eliminating fat artifact interference.

2. The CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as claimed in claim 1, It is characterized in that The water-MT-fat three-pool Bloch-McConnell equation model is as follows: Where exp represents an exponential function with the natural constant e as the base; Δt represents the increment at time t; is the total magnetization vector of water and fat as the reference signal at time t+Δt, They represent the magnetization vector of the water part at time t. Magnetization vector of fat part in Matrix M w =[M xw , M yw , M zw , M zm ] T , M xw 、M yw 、M zw 、M zm They are The x-direction magnetization intensity component of the water group, the y-direction magnetization intensity component of the water group, the z-direction magnetization intensity component of the water group, the z-direction magnetization intensity component of the MT group, the matrix M f =[M xf1 , M yf1 , M zf1 , M xf2 , M yf2 , M zf2 , ..., M xf7 , M yf7 , M zf7 ] T , M xf1 、M xf2 , …, M xf7 are the x-direction magnetization intensity components of the 1st to 7th groups in fat, M yf1 、M yf2 , ..., M yf7 are the y-direction magnetization intensity components of the 1st to 7th groups in fat, M zf1 、M zf2 , …, M zf7 They are the z-direction magnetization intensity components of the 1st to 7th groups in fat, respectively; Fat fraction FF = M 0f / (M 0f +M 0w ) represents the ratio of fat signal to total water and fat signal, where M 0f is the magnetization intensity component M of the 1st to 7th groups of fat in steady state 0f1 、M 0f2 , ..., M 0f7 The sum, M 0w is the magnetization intensity component of the water group in the steady state; the coefficient matrix The matrix Vector C w =[0, 0, R 1w M 0w , R 1m M 0m ] T ; Δω w =ω RF -ω w -ΔB 0 ,ω RF represents the applied RF pulse frequency, ω w represents the resonant frequency of the water cluster, ΔB 0 represents the residual main magnetic field inhomogeneity after correction, ω 1 represents the intensity of the applied RF pulse, k wm , k mw They represent the exchange rate from water group to MT group and from MT group to water group, R rfm R is the absorption rate of MT group to RF pulse, 1w =1 / T 1w , R 2w =1 / T 2w , R 1m =1 / T 1m , R 2m =1 / T 2m They represent the longitudinal relaxation rate of the water group, the transverse relaxation rate of the water group, the longitudinal relaxation rate of the MT group, and the transverse relaxation rate of the MT group, respectively. 0m is the magnetization intensity of the MT group in steady state; matrix Vector C f =[0, 0, R 1f1 M 0f1 ,0,0,R 1f2 M 0f2 , ..., 0, 0, R 1f7 M 0f7 ] T , where the matrix A f1 ~A f7 The nth matrix in n=1,2,…,7;relative amplitude of 7 fat groups Value M 0f1 、M 0f2 , …, M 0f7 The relationship between Δω and Δω is estimated by three parameters: CL, ndb, and nmidb. CL represents the fatty acid chain length, ndb represents the number of double bonds per molecule, and nmidb represents the number of methylene double bonds. fn =ω RF -ω fn -ΔB 0 ,ω fn represents the resonance frequency of the nth group in fat, R 1fn =1 / T 1fn , R 2fn =1 / T 2fn They represent the longitudinal relaxation rate of the nth group in fat and the transverse relaxation rate of the nth group in fat respectively; the longitudinal relaxation time of the seven groups in fat is set to be the same: T 1f =T 1f1 =T 1f2 =…=T 1f7 , the transverse relaxation time of the seven fat groups is also set to the same: T 2f =T 2f1 =T 2f2 =…=T 2f7 ; The parameters to be fitted in the model are: T 1w , T 2w , T 2m , M 0m , k wm , FF, T 1f , T 2f , ΔB 0 , CL, ndb, nmidb.

3. The CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as claimed in claim 2, It is characterized in that Preferably, the longitudinal relaxation rate R of the MT is 1m and the resonance frequency of the fat peak, all of which are input into the model as a priori conditions; preferably, the intensity of the applied radio frequency pulse ω 1 The RF field intensity map obtained by the additional sampling is calculated and input into the model as a priori condition.

4. The CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as claimed in claim 1, It is characterized in that In step 2, the data containing only MT are data points whose frequencies in the corrected Z spectrum are preferably within the range of 80 to 6 ppm; the data containing direct water saturation effects are data points whose frequencies in the corrected Z spectrum are preferably within the range of 1.5 to -1.5 ppm; and the data containing direct fat saturation effects are data points whose frequencies in the corrected Z spectrum are preferably within the range of -1 to -6 ppm. More preferably, the data used as fitting samples are data with frequencies in the range of 80 to 10 ppm and 0.75 to -6 ppm in the corrected Z spectrum.

5. The CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as claimed in claim 2, It is characterized in that Preferably, in step 2, the parameter to be fitted in the model is 1w , T 2w , T 2m , M 0m , k wm , FF, T 1f , T 2f , ΔB 0 , the upper bounds of CL, ndb, and nmidb during fitting are [1.50, 0.100, 5e -6 , 0.4500, 40, 1, 0.400, 0.150, 64, 18, 5.5, 2.5], and the lower bounds are [1.35, 0.015, 1e -6 , 0.0001, 20, 0, 0.250, 0.025, -64, 17, 1, 0], and the initial values ​​of the fitting are [1.35, 0.015, 1e -6 , 0.0001, 20, 1, 0.400, 0.025, 0, 17, 1, 0]; Preferably, in step 2, a higher weight is set for the data in the range of 80 to 10 ppm during fitting, and the weight is twice the weight of the data in the range of 0.75 to -6 ppm.

6. The CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as claimed in claim 2, It is characterized in that In step 2, the absorption rate of the RF pulse by the MT group Where g(·) is a linear function, including Gaussian, Lorentzian, and super-Lorentzian types. In breast data, the linear function is preferably super-Lorentzian.

7. The CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as claimed in claim 1, It is characterized in that In step 4, the frequency of interest is 3.5 ppm, and the CEST effect value at 3.5 ppm in the normalized CEST effect curve is APT # value.

8. A computer-readable storage medium, It is characterized in that The storage medium stores a computer program, and when the computer program is executed by the processor, the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as described in any one of claims 1 to 7 is implemented.

9. A computer electronic device, It is characterized in that The method comprises a processor and a storage medium, wherein a computer program is stored on the storage medium. When the computer program is executed by the processor, the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as claimed in any one of claims 1 to 7 is implemented.

10. A magnetic resonance imaging device for removing CEST fat artifacts, It is characterized in that The invention comprises a magnetic resonance scanner and a control unit, wherein the magnetic resonance scanner is used to obtain a CEST image through magnetic resonance CEST imaging; the control unit can obtain the CEST image and a computer program is stored in the control unit, and when the computer program is executed, it is used to implement the CEST fat artifact elimination method based on magnetization transfer and fat signal fitting difference as described in any one of claims 1 to 7, and output the CEST effect value of each voxel for eliminating fat artifacts.

Citation Information

Patent Citations

  • CESTR measuring method and system for magnetic resonance chemical exchange saturation transfer imaging

    CN104997511A

  • Magnetic resonance spectrum full-automatic post-processing method for quantifying fat content

    CN108872901A

  • Chemical exchange saturation transfer imaging method for eliminating fat artifact

    CN110068783A

  • Z-spectrum fitting method based on pseudo-Friedel-Crafts offline type and CEST MRI quantification method based on pseudo-Friedel-Crafts offline type

    CN110706199A

  • CEST data fitting method and device based on extrapolation magnetization transfer signal and medium

    CN113171076A

Cited By

  • Rapid magnetic resonance CEST quantitative method based on deep learning and imaging equipment

    CN121385757A