WO2022095142A1 - 一种adc-t2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振*** - Google Patents

一种adc-t2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振*** Download PDF

Info

Publication number
WO2022095142A1
WO2022095142A1 PCT/CN2020/130626 CN2020130626W WO2022095142A1 WO 2022095142 A1 WO2022095142 A1 WO 2022095142A1 CN 2020130626 W CN2020130626 W CN 2020130626W WO 2022095142 A1 WO2022095142 A1 WO 2022095142A1
Authority
WO
WIPO (PCT)
Prior art keywords
echo signals
magnetic resonance
echo
formula
adc
Prior art date
Application number
PCT/CN2020/130626
Other languages
English (en)
French (fr)
Inventor
张洁莹
潘子异
王伟谦
吴子岳
叶洋
Original Assignee
无锡鸣石峻致医疗科技有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 无锡鸣石峻致医疗科技有限公司 filed Critical 无锡鸣石峻致医疗科技有限公司
Publication of WO2022095142A1 publication Critical patent/WO2022095142A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/50—NMR imaging systems based on the determination of relaxation times, e.g. T1 measurement by IR sequences; T2 measurement by multiple-echo sequences
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
    • G01R33/00—Arrangements or instruments for measuring magnetic variables
    • G01R33/20—Arrangements or instruments for measuring magnetic variables involving magnetic resonance
    • G01R33/44—Arrangements or instruments for measuring magnetic variables involving magnetic resonance using nuclear magnetic resonance [NMR]
    • G01R33/48—NMR imaging systems
    • G01R33/54—Signal processing systems, e.g. using pulse sequences ; Generation or control of pulse sequences; Operator console
    • G01R33/543—Control of the operation of the MR system, e.g. setting of acquisition parameters prior to or during MR data acquisition, dynamic shimming, use of one or more scout images for scan plane prescription

Definitions

  • the invention belongs to the technical field of nuclear magnetic resonance imaging, in particular to the technology of magnetic resonance spectrometer, and in particular to a method, device, computer equipment and non-uniform field magnetic resonance system for measuring the decay time constant of transverse magnetization vector.
  • Nuclear magnetic resonance technology is a technology that uses the nuclear magnetic resonance phenomenon of hydrogen protons to image or detect the composition and structure of substances.
  • Nuclei in the human body that contain a single number of protons, such as the nucleus of hydrogen, have protons that have spin motion.
  • the spin motions of charged nuclei are physically analogous to individual small magnets whose directional distributions are random in the absence of external conditions.
  • these small magnets will rearrange according to the magnetic field lines of the external magnetic field.
  • the nuclei are excited with radio frequency pulses of a specific frequency, so that the spins of these nuclei (small magnets) are deflected and resonance occurs. This is the phenomenon of nuclear magnetic resonance.
  • the excited atomic nuclei (resonant small magnets) will gradually return to the state before excitation.
  • electromagnetic wave signals will be released, and the magnetic resonance images will be obtained after receiving and processing the nuclear magnetic resonance signals through special equipment. Or the composition and structure information of a substance.
  • T2 also known as the spin-spin relaxation time T2
  • spin echo sequence Spin Echo, SE
  • T2 CPMG nuclear magnetic resonance sequence
  • the CPMG sequence is short and can avoid the accumulation of errors due to inaccurate 180-degree RF pulses (due to limited RF field uniformity).
  • signals are also typically acquired using CPMG sequences.
  • T2 in solids is much shorter than T2 in liquids.
  • the T2 in typical biological tissues is in the range of 20-150ms, and the free water T2 is much longer than the bound water T2. Clinically, the observed T2 prolongation at the lesion is interpreted as an increased free water ratio.
  • Molecules in matter all have a certain degree of diffusion motion, and its direction is random, which is called thermal motion or Brownian motion of molecules. If the diffusion motion of water molecules is not subject to any constraints, it can be called free diffusion. In the human body, water molecules such as cerebrospinal fluid and urine have relatively little restrictions on their diffusive movement and are considered free diffusion. In fact, the diffusion movement of water molecules in biological tissues will be restricted to different degrees due to the constraints of the surrounding medium, which is called restricted diffusion. The diffusion movement of water molecules in general tissues is restricted diffusion.
  • the apparent diffusion coefficient is a physical quantity that describes the ability of water molecules to diffuse in tissues. After the magnetic resonance signal is excited, the diffusion motion of water molecules in the direction of the gradient magnetic field will cause the attenuation of the magnetic resonance signal. The greater the change in the magnetic field, the greater the attenuation of the tissue signal. Therefore, the apparent diffusion coefficient of an object can be measured by nuclear magnetic resonance technology, thereby indirectly reflecting the microstructure characteristics and changes of the object.
  • the apparent diffusion coefficient (ADC, Apparent Diffusion Coefficient) is widely used as an important clinical diagnostic index. It is generally measured by Diffusion Weighted Imaging (DWI), such as spin echo-planar imaging (SE-EPI), that is, spin echo sequence (SE) for diffusion gradient coding , echo plane sequence (EPI) for signal readout.
  • DWI Diffusion Weighted Imaging
  • SE-EPI spin echo-planar imaging
  • SE-EPI spin echo sequence
  • EPI echo plane sequence
  • a non-uniform magnetic field magnetic resonance system also referred to as a non-uniform field magnetic resonance system
  • a similar diffusion-weighted imaging technique is introduced to measure the apparent diffusion coefficient of substances.
  • FIG. 1 is a spin echo-CPMG sequence (SE-CPMG), that is, diffusion gradient encoding is performed based on spin echo, and then an ultra-fast CPMG sequence is used for signal readout.
  • SE-CPMG spin echo-CPMG sequence
  • DSE-CPMG dual spin echo-CPMG sequence
  • STE-CPMG stimulated echo-CPMG sequence
  • the ADC measurement pulse sequence is composed of a diffusion gradient encoding module and a signal readout module.
  • the gradient magnetic field is very large, usually 2 to 3 orders of magnitude higher than that of a conventional magnetic resonance imaging system (Magnetic Resonance Imaging, MRI), and it is impossible to control the gradient magnetic field in the signal readout stage.
  • MRI Magnetic Resonance Imaging
  • DWI technology in the MRI system can control the gradient field reduction in the signal readout stage, while in a non-uniform magnet, the static gradient magnetic field G is often very large and is a constant uncontrollable gradient field
  • the gradient magnetic field In the signal readout stage it will still play the role of diffusion encoding, which will affect the accuracy of the apparent diffusion coefficient ADC measurement.
  • the purpose of the present invention is to provide a method, device, computer equipment, non-uniform field magnetic resonance system and computer-readable storage medium for measuring the decay time constant of the transverse magnetization vector.
  • the present invention provides a method for measuring the two-dimensional spectrum of ADC-T2, comprising:
  • the dimension of the ADC-T2 two-dimensional spectrum to be solved is set as p ⁇ q, and then the integral formula is converted into the following matrix formula:
  • s' vect(S'), S' represents S'(i,j), vect(S') represents the vector form of S'(i,j), and K represents the form of M ⁇ N ⁇ p ⁇ q
  • K represents the form of M ⁇ N ⁇ p ⁇ q
  • repmat() represents the function for copying and tiling the matrix
  • vect(repmat()) represents the vector form of repmat()
  • vect(S) represents the vector form of S(D, T 2 );
  • F(s) represents the function about the variable s
  • 2 represents the two-norm of the vector
  • ⁇ represents the constraint coefficient
  • the solution result of the ADC-T2 two-dimensional atlas is obtained by reduction.
  • the measurement method in a nuclear magnetic resonance system with a very uneven magnetic field or a very short echo time, it is possible to obtain echo signals from multiple CPMG sequences with different echo intervals.
  • the ADC coefficient and T2 value are fitted from the echo signal of the group, so that the ADC-T2 spectrum can be measured, so that the complex diffusion weighting sequence is no longer required, and the algorithm is simple and has low system requirements, which can reduce the need for spectrometers. Cost of hardware systems such as equipment, RF power amplifiers, and RF coils.
  • the measurement method also has the characteristics of stable algorithm, is not easily affected by flowing liquid, and is also applicable to substances with small T1/T2.
  • the measurement sequence is simple and fast, and no diffusion gradient coding module is required, which is easier to implement than the existing technology; for extremely inhomogeneous magnetic fields, or for nuclear magnetic resonance systems that cannot achieve extremely short echo times, this method can still measure ADCs -T2 spectrum; based on the aforementioned two advantages, the system hardware requirements can be reduced when designing a non-uniform field NMR system for ADC-T2 measurement, thereby reducing hardware costs; the ADC-T2 measurement method is not easily affected by flowing liquids , and it is also applicable to substances with smaller T1/T2; it can more accurately distinguish the water and fat components in the tested object.
  • the value of the diffusion sensitivity coefficient b is proportional to (t EE ) 3 , where t EE represents the time from the center point of the first 90° excitation pulse to the first time during the acquisition of a single set of echo signals The duration of the center point time of the echo signal.
  • the smallest fitting vector s greater than 0 is obtained by solving the optimization problem, including:
  • the matrix K is subjected to the following dimension reduction processing based on singular value decomposition:
  • U M ⁇ N ⁇ M ⁇ N represents the M ⁇ N ⁇ M ⁇ N order unitary matrix
  • the elements on the diagonal of ⁇ M ⁇ N ⁇ p ⁇ q are singular values
  • k′ represents dimensionality reduction
  • U M ⁇ N ⁇ k′ represents the first k′column matrix of U M ⁇ N ⁇ M ⁇ N ;
  • the vector s' is reduced from the M ⁇ N dimension to the k' dimension according to the following formula:
  • s' svd represents the vector when the vector s' is reduced from the M ⁇ N dimension to the k' dimension, is the transpose matrix of the matrix U M ⁇ N ⁇ k′ ;
  • the fitting vector s is obtained by solving:
  • the method before performing integral formula fitting on the multiple sets of echo signals, the method further includes performing the following preprocessing on the multiple sets of echo signals:
  • fast Fourier transform is performed on the multiple sets of echo signals to obtain frequency domain data, and then the low frequency part lower than the preset frequency threshold is retained and averaged to obtain the Multiple sets of echo signals with no difference in the dimension of the number of sampling points.
  • the method before performing integral formula fitting on the multiple sets of echo signals, the method further includes performing the following preprocessing on the multiple sets of echo signals:
  • averaging processing is performed on the multiple groups of echo signals to obtain multiple groups of echo signals with no difference in the dimension of average number of times.
  • the method before performing integral formula fitting on the multiple sets of echo signals, the method further includes performing the following preprocessing on the multiple sets of echo signals:
  • filtering processing based on singular value decomposition is performed on the plurality of sets of echo signals, so as to obtain a plurality of sets of echo signals without the influence of noise in the length dimension of the echo chain.
  • the present invention provides an ADC-T2 two-dimensional spectrum measurement device, comprising an echo signal acquisition module, an integral formula fitting module, a matrix formula conversion module, a problem solving conversion module, and a fitting vector that are sequentially connected in communication. Solving module and solving result restoration module;
  • the echo signal acquisition module is used to acquire multiple sets of echo signals acquired based on the CPMG nuclear magnetic resonance sequence, wherein the CPMG nuclear magnetic resonance sequence is used to perform diffusion gradient coding and read out each set of echo signals, and Different echo intervals are used for each group of echo signals;
  • the integral formula fitting module is used to fit the following integral formulas on the multiple sets of echo signals:
  • the matrix formula conversion module is used to set the dimension of the ADC-T2 two-dimensional atlas to be solved as p ⁇ q, and then convert the integral formula into the following matrix formula:
  • s' vect(S'), S' represents S'(i,j), vect(S') represents the vector form of S'(i,j), and K represents the form of M ⁇ N ⁇ p ⁇ q
  • K represents the form of M ⁇ N ⁇ p ⁇ q
  • repmat() represents the function for copying and tiling the matrix
  • vect(repmat()) represents the vector form of repmat()
  • vect(S) represents the vector form of S(D, T 2 );
  • the problem-solving conversion module is used to convert the problem of solving S(D, T 2 ) into a problem-solving optimization problem of the following formula according to the matrix formula:
  • F(s) represents the function about the variable s
  • 2 represents the two-norm of the vector
  • ⁇ represents the constraint coefficient
  • the fitting vector solving module is used to obtain the smallest fitting vector s greater than 0 by solving the optimization problem according to the solution;
  • the solution result restoration module is configured to restore the solution result of the ADC-T2 two-dimensional atlas according to the obtained fitting vector s.
  • the present invention provides a computer device, comprising a communicatively connected memory and a processor, wherein the memory is used to store a computer program, and the processor is used to read the computer program, and execute the first aspect Or any one of the first aspect may design the measurement method.
  • the present invention provides a non-uniform field magnetic resonance system, including a console, a nuclear magnetic resonance spectrometer, a magnet and a radio frequency subsystem;
  • the console connected to the nuclear magnetic resonance spectrometer in communication, is used to send instructions to the nuclear magnetic resonance spectrometer to control the parameter selection of the measurement sequence and the location of the region of interest, and to receive the data collected by the nuclear magnetic resonance spectrometer.
  • the magnetic resonance signal is obtained, and data processing is completed, wherein the data processing includes performing the measurement method according to the first aspect or any one of the possible designs of the first aspect;
  • the nuclear magnetic resonance spectrometer is connected to the radio frequency subsystem in communication, and is used for executing the instructions from the console, and transmits the radio frequency excitation signal of the measurement sequence and receives the magnetic resonance signal through the radio frequency subsystem;
  • the magnet is used to be arranged just above the object to be inspected and the signal transceiving component in the radio frequency subsystem;
  • the radio frequency subsystem is used for transmitting radio frequency excitation signals of the measurement sequence and receiving the magnetic resonance signals under the control of the nuclear magnetic resonance spectrometer.
  • the radio frequency subsystem includes a radio frequency power amplifier, a preamplifier, a transceiver switch and a radio frequency coil;
  • the signal input end of the radio frequency power amplifier is electrically connected to the signal output end of the nuclear magnetic resonance spectrometer, and the signal output end of the radio frequency power amplifier is electrically connected to the first switch end of the transceiving switch;
  • the signal input end of the preamplifier is electrically connected to the second switching end of the transceiving switch, and the signal output end of the preamplifier is electrically connected to the signal input end of the nuclear magnetic resonance spectrometer;
  • the controlled end of the transceiving switch is communicatively connected to the output end of the nuclear magnetic resonance spectrometer, and the switching common terminal of the transceiving switch is electrically connected to the radio frequency coil;
  • the radio frequency coil is used as a signal transceiver component of the radio frequency subsystem to transmit the radio frequency excitation signal to the object under inspection and receive the magnetic resonance signal from the object under inspection.
  • the present invention provides a computer-readable storage medium, where instructions are stored on the computer-readable storage medium, and when the instructions are executed on a computer, any one of the first aspect or the first aspect is executed. a possible design of the measurement method.
  • the present invention provides a computer program product comprising instructions that, when executed on a computer, cause the computer to perform the measurement as described in the first aspect or any one of the possible designs of the first aspect method.
  • FIG. 1 is an example diagram of the ADC measurement pulse sequence used in the non-uniform field magnetic resonance system in the prior art, wherein (a) SE-CPMG sequence is used; (b) DSE-CPMG sequence is used; (c) STE is used -CPMG sequence.
  • FIG. 2 is a schematic flowchart of the measurement method provided by the present invention.
  • FIG. 3 is an example diagram of the ADC-T2 measurement pulse sequence adopted in the non-uniform field magnetic resonance system provided by the present invention.
  • FIG. 5 is an example diagram of the ADC-T2 two-dimensional spectrum measured when the tested substances are 0.5mmol/L MnCl2 solution and peanut oil provided by the present invention.
  • FIG. 6 is a schematic structural diagram of the measuring device provided by the present invention.
  • FIG. 7 is a schematic structural diagram of a computer device provided by the present invention.
  • FIG. 8 is a schematic structural diagram of a non-uniform field magnetic resonance system provided by the present invention.
  • first, second, etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one unit from another. For example, a first element could be referred to as a second element, and similarly a second element could be referred to as a first element, without departing from the scope of example embodiments of this invention.
  • the measurement method of the ADC-T2 two-dimensional spectrum provided in the first aspect of this embodiment can be, but is not limited to, a method with an extremely uneven magnetic field or an extremely short echo time that cannot be achieved.
  • the console in the nuclear magnetic resonance system (which is connected to the nuclear magnetic resonance spectrometer in order to control the parameter selection of the measurement sequence and the positioning of the region of interest, and receives the magnetic resonance signals collected by the nuclear magnetic resonance spectrometer to complete the data processing) executes , in order to obtain an accurate ADC-T2 two-dimensional map.
  • the measurement method of the ADC-T2 two-dimensional spectrum includes but is not limited to the following steps S101-S106.
  • a typical ⁇ -2 ⁇ -2 ⁇ -2 ⁇ ... RF pulse sequence is used: the flip angle of the first excitation pulse is ⁇ , followed by several refocusing pulses, flipping The angle is 2 ⁇ ; the phase difference between the first excitation pulse and the first refocusing pulse is 90 degrees, the time interval between the first excitation pulse and the first refocusing pulse is ⁇ /2, and the first refocusing pulse to the first The time interval between the sampling windows is ⁇ /2; the time interval between the returning pulses is ⁇ , which is called the echo interval.
  • the constant gradient field is the natural gradient field of the magnet and does not require control.
  • N echo signals can be collected in one excitation (where N is a positive integer), it needs to be repeated many times in order to improve the signal-to-noise ratio by averaging the signals, and to obtain ADC-T2 two-dimensional It is necessary to collect multiple sets of echo signals by changing the echo interval ⁇ .
  • the multiple sets of echo signals acquired by collection can be represented by a 4-dimensional array S(m,n,a,q), wherein the first-dimensional array S(m) is based on different echoes interval, the length is a positive integer M, that is, the echo interval ⁇ corresponding to M different lengths; the second-dimensional array S(n) is based on different echo chain lengths, and the length is a positive integer N; the third-dimensional array S(a) is based on Different average times during repeated collection, the length is a positive integer A; the fourth dimension array S(q) is based on the different sampling points of the single read data, and the length is a positive integer Q.
  • the multiple sets of echo signals are a 4-dimensional array S(m,n,a,q), with too many dimensions and noise influence, which will increase the unnecessary integral calculation amount and bring about Therefore, it is necessary to perform dimensionality reduction and/or denoising processing first, that is, before performing integration formula fitting on the multiple sets of echo signals, it also includes but is not limited to performing the following methods on the multiple sets of echo signals ( A) to any one or any combination of preprocessing in (C): (A) in the dimension of the number of sampling points of the single readout data, fast Fourier transform is performed on the multiple groups of echo signals to obtain frequency domain data, Then keep the low-frequency part below the preset frequency threshold and perform averaging processing to obtain multiple sets of echo signals with no difference in the dimension of the number of sampling points; (B) in the dimension of the average number of times, for the multiple sets of echo signals The signals are averaged to obtain multiple sets of echo signals that are indistinguishable in the dimension of the average number of times; (C) in
  • the SVD algorithm is mainly used in feature decomposition, recommendation systems, and natural language processing in dimensionality reduction algorithms.
  • a linear transformation is decomposed into two linear transformations, one linear transformation represents rotation, and the other linear transformation represents stretching) filtering processing, so as to obtain no noise in the length dimension of the echo chain. of multiple sets of echo signals. Therefore, through the processing of the aforementioned methods (A) and (B), the dimension of the four-dimensional array S(m,n,a,q) can be reduced into a two-dimensional array S(m,n), which is beneficial to reduce subsequent calculations.
  • s' vect(S'), S' represents S'(i,j), vect(S') represents the vector form of S'(i,j), and K represents the form of M ⁇ N ⁇ p ⁇ q
  • K represents the form of M ⁇ N ⁇ p ⁇ q
  • repmat() represents the function for copying and tiling the matrix
  • vect(repmat()) represents the vector form of repmat()
  • vect(S) represents the vector form of S(D, T 2 ).
  • F(s) represents the function about the variable s
  • 2 represents the two-norm of the vector
  • ⁇ represents the constraint term coefficient
  • the specific process of solving the fitting vector s according to the solving optimization problem is an existing conventional solving method.
  • the dimensionality reduction solution may be performed but not limited to the following steps S501-S503.
  • U M ⁇ N ⁇ M ⁇ N represents the M ⁇ N ⁇ M ⁇ N order unitary matrix
  • the elements on the diagonal of ⁇ M ⁇ N ⁇ p ⁇ q are singular values
  • k′ represents dimensionality reduction
  • U M ⁇ N ⁇ k′ represents the first k′column matrix of U M ⁇ N ⁇ M ⁇ N ;
  • s' svd represents the vector when the vector s' is reduced from the M ⁇ N dimension to the k' dimension, is the transpose matrix of the matrix U M ⁇ N ⁇ k′ .
  • the solution result of the ADC-T2 two-dimensional map can be obtained by reduction in a conventional manner. That is to say, considering the strong gradient field existing in the non-uniform field, in the CPMG sequence, it can not only play the role of frequency coding, but also always have the role of diffusion coding. That is to say, if the echo interval of the CPMG sequence is larger, the signal will be lower.
  • the influence of the diffusion effect on the CPMG signal can be described by the following formula:
  • ⁇ represents the magnetic spin ratio of the nucleus
  • D' represents the ADC coefficient of the substance
  • G represents the magnitude of the gradient magnetic field
  • ⁇ represents the echo interval of the CPMG sequence.
  • the experimental results of this embodiment are specifically described below by taking the detection of 0.5 mmol/L MnCl 2 solution and peanut oil on a non-uniform field nuclear magnetic resonance system as an example.
  • Fig. 4 it can be seen that the signal attenuation of CPMG sequences with different echo intervals is different, which is due to the influence of molecular diffusion;
  • Fig. 5 it can be seen that the 0.5mmol/L MnCl 2 solution is a uniform peak on the ADC-T2 two-dimensional spectrum , while peanut oil showed double peaks in ADC-T2 two-dimensional spectrum, ADC coefficients and T2 values of the two substances were close to the theoretical values.
  • the peaks of MnCl 2 solution and peanut oil are clearly distinguished on the spectrum: the ADC of MnCl 2 solution is larger, while that of peanut oil has a smaller ADC.
  • a plurality of acquired echo-intervals with different echo intervals can be obtained.
  • ADC coefficients and T2 values are fitted from multiple sets of echo signals, so that the ADC-T2 spectrum can be measured, so that the complex diffusion weighting sequence is no longer needed, and the algorithm is simple and has system requirements.
  • the low cost can reduce the cost of hardware systems such as spectrometer equipment, RF power amplifiers and RF coils.
  • the measurement method also has the characteristics of stable algorithm, is not easily affected by flowing liquid, and is also applicable to substances with small T1/T2. That is, the measurement sequence is simple and fast, and no diffusion gradient coding module is required, which is easier to implement than the existing technology; for extremely inhomogeneous magnetic fields, or for nuclear magnetic resonance systems that cannot achieve extremely short echo times, this method can still measure ADCs -T2 spectrum; based on the aforementioned two advantages, the system hardware requirements can be reduced when designing a non-uniform field NMR system for ADC-T2 measurement, thereby reducing hardware costs; the ADC-T2 measurement method is not easily affected by flowing liquids , and it is also applicable to substances with smaller T1/T2; it can more accurately distinguish the water and fat components in the tested object.
  • a second aspect of this embodiment provides a virtual device for implementing the measurement method in the first aspect or any one of the possible designs of the first aspect, including an echo signal acquisition module that is sequentially communicatively connected, Integral formula fitting module, matrix formula conversion module, solving problem conversion module, fitting vector solving module and solving result restoration module;
  • the echo signal acquisition module is used to acquire multiple sets of echo signals acquired based on the CPMG nuclear magnetic resonance sequence, wherein the CPMG nuclear magnetic resonance sequence is used to perform diffusion gradient coding and read out each set of echo signals, and Different echo intervals are used for each group of echo signals;
  • the integral formula fitting module is used to fit the following integral formulas on the multiple sets of echo signals:
  • the matrix formula conversion module is used to set the dimension of the ADC-T2 two-dimensional atlas to be solved as p ⁇ q, and then convert the integral formula into the following matrix formula:
  • s' vect(S'), S' represents S'(i,j), vect(S') represents the vector form of S'(i,j), and K represents M ⁇ N ⁇ p ⁇ q
  • K represents M ⁇ N ⁇ p ⁇ q
  • repmat() represents the function for copying and tiling the matrix
  • vect(repmat()) represents the vector form of repmat()
  • vect(S) represents the vector form of S(D, T 2 );
  • the problem-solving conversion module is used to convert the problem of solving S(D, T 2 ) into a problem-solving optimization problem of the following formula according to the matrix formula:
  • F(s) represents the function about the variable s
  • 2 represents the two-norm of the vector
  • ⁇ represents the constraint coefficient
  • the fitting vector solving module is used to obtain the smallest fitting vector s greater than 0 by solving the optimization problem according to the solution;
  • the solution result restoration module is configured to restore the solution result of the ADC-T2 two-dimensional atlas according to the obtained fitting vector s.
  • a third aspect of this embodiment provides a computer device for executing the measurement method in the first aspect or any possible design of the first aspect, including a communicatively connected memory and a processor, wherein, The memory is used to store a computer program, and the processor is used to read the computer program, and execute the measurement method according to the first aspect or any one of the possible designs of the first aspect.
  • the memory may include, but is not limited to, random access memory (Random-Access Memory, RAM), read-only memory (Read-Only Memory, ROM), flash memory (Flash Memory), first-in first-out memory (First Input Memory) First Output, FIFO) and/or first-in-last-out memory (First Input Last Output, FILO), etc.; the processor may not be limited to using a microprocessor of the STM32F105 series.
  • the computer equipment may also include, but is not limited to, a power module, a display screen and other necessary components.
  • a fourth aspect of this embodiment provides a non-uniform field magnetic resonance system for performing the measurement method in the first aspect or any possible design of the first aspect, including a console, a nuclear magnetic resonance Spectrometers, magnets and RF subsystems;
  • the console connected to the nuclear magnetic resonance spectrometer in communication, is used to send instructions to the nuclear magnetic resonance spectrometer to control the parameter selection of the measurement sequence and the location of the region of interest, and to receive the data collected by the nuclear magnetic resonance spectrometer.
  • the magnetic resonance signal is obtained, and data processing is completed, wherein the data processing includes performing the measurement method according to the first aspect or any one of the possible designs of the first aspect;
  • the nuclear magnetic resonance spectrometer is connected to the radio frequency subsystem in communication, and is used for executing the instructions from the console, and transmits the radio frequency excitation signal of the measurement sequence and receives the magnetic resonance signal through the radio frequency subsystem;
  • the magnet is used to be arranged just above the object to be inspected and the signal transceiving component in the radio frequency subsystem;
  • the radio frequency subsystem is used for transmitting radio frequency excitation signals of the measurement sequence and receiving the magnetic resonance signals under the control of the nuclear magnetic resonance spectrometer.
  • the magnet is generally designed as a permanent magnet, such as a unilateral permanent magnet, so that there is still a highly non-uniform magnetic field in the region of interest.
  • the radio frequency subsystem includes a radio frequency power amplifier, a preamplifier, a transceiver switch and a radio frequency coil;
  • the signal input end of the radio frequency power amplifier is electrically connected to the signal output end of the nuclear magnetic resonance spectrometer, and the signal output end of the radio frequency power amplifier is electrically connected to the first switch end of the transceiving switch;
  • the signal input end of the preamplifier is electrically connected to the second switching end of the transceiving switch, and the signal output end of the preamplifier is electrically connected to the signal input end of the nuclear magnetic resonance spectrometer;
  • the controlled terminal of the transceiving switch is communicatively connected to the output terminal of the nuclear magnetic resonance spectrometer, and the switching common terminal of the transceiving switch is electrically connected to the radio frequency coil;
  • the radio frequency coil is used as a signal transceiver component of the radio frequency subsystem to transmit the radio frequency excitation signal to the object under inspection and receive the magnetic resonance signal from the object under inspection.
  • the radio frequency power amplifier is used to amplify the radio frequency excitation signal to be transmitted; the preamplifier is used to amplify the received magnetic resonance signal.
  • the transceiving switch is used for switching control, so that the radio frequency coil can transmit the radio frequency excitation signal and receive the magnetic resonance signal.
  • a fifth aspect of this embodiment provides a computer-readable storage medium that stores an instruction containing the measurement method in the first aspect or any possible design of the first aspect, that is, the computer-readable storage medium stores a memory on the computer-readable storage medium.
  • the instructions when executed on a computer, perform the measurement method as described in the first aspect or any one of the possible designs of the first aspect.
  • the computer-readable storage medium refers to a carrier for storing data, which may include, but is not limited to, a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive, and/or a memory stick (Memory Stick), etc.
  • the computer may be a general-purpose computer, a special-purpose A computer, computer network, or other programmable device.
  • a sixth aspect of this embodiment provides a computer program product containing instructions, when the instructions are run on a computer, the computer is caused to perform the measurement described in the first aspect or any possible design of the first aspect method.
  • the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

Landscapes

  • Physics & Mathematics (AREA)
  • High Energy & Nuclear Physics (AREA)
  • Condensed Matter Physics & Semiconductors (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)

Abstract

一种ADC-T2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振***，可以在具有极不均匀的磁场或者是无法实现极短的回波时间的核磁共振***中，通过获取的多个采用不同回波间隔的CPMG序列采集的回波信号，从多组回波信号中拟合出ADC系数和T2值，从而能测量出ADC-T2图谱，不再需要复杂的扩散加权序列，具有算法简单且对***要求低的优点，可以降低谱仪设备、射频功放和射频线圈等硬件***的成本。同时该测量方法还具有算法稳定的特点，不易受流动液体影响，对于T1/T2较小的物质同样适用。

Description

一种ADC-T2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振*** 技术领域
本发明属于核磁共振成像技术领域，具体而言，涉及磁共振频谱仪技术，特别地涉及一种横向磁化矢量衰减时间常数的测量方法、装置、计算机设备及非均匀场磁共振***。
背景技术
核磁共振技术是利用氢质子的核磁共振现象进行成像或者检测物质成分和结构的一种技术。人体内包含单数质子的原子核，例如氢原子核，其质子具有自旋运动。带电原子核的自旋运动，在物理上类似于单独的小磁体，在没有外部条件影响下这些小磁体的方向性分布是随机的。当人体置于外部磁场中时，这些小磁体将按照外部磁场的磁力线重新排列。这时，用特定频率的射频脉冲激发原子核，使这些原子核的自旋(小磁铁)发生偏转，产生共振，这就是核磁共振现象。停止发射射频脉冲后，被激发的原子核(共振的小磁体)会逐渐恢复到激发前的状态，在恢复的过程中会释放电磁波信号，通过专用设备接收并处理核磁共振信号后即获得磁共振图像或者物质的成分和结构信息。
原子核横向磁化矢量趋于零的速度取决于自旋-自旋相互作用的强度，用横向磁化矢量衰减时间常数T2(也称为自旋-自旋弛豫时间T2)来描述。在磁共振技术中，一般使用自旋回波序列(Spin Echo，SE)来测量T2，如CPMG核磁共振序列(aNMR pulse sequencenamed by several scientists Carr,Purcell,Meiboom and Gill，即由Carr,Purcell,Meiboom和Gill等人命名的核磁共振序列)，即在x轴施加90度射频脉冲，之后在t＝τ,3τ,5τ,…,(2n-1)τ时在y轴施加180度射频脉冲，于是在t＝2τ,4τ,6τ,…,2nτ时可得到回波信号。CPMG序列时间短，且可以避免由于180度射频脉冲不准确(由于射频场均匀性有限)而造成的误差累积。在非均匀场磁共振***中，通常也使用CPMG序列采集信号。一般的，固体中T2比液体中T2短得多。典型的生物组织中的T2位于20-150ms范围，自由水T2比束缚水T2要长得多。临床上，在病灶处观察到T2延长被解释为自由水比例增高。
物质中的分子都存在一定程度的扩散运动，其方向是随机的，称为分子的热运动或布朗运动。如果水分子的扩散运动不受任何约束，可称为自由扩散。在人体中，脑脊液和尿液等水分子扩散运动所受到的限制相对小，被视为自由扩散。事实上，生物组织内的水分子因受周围介质的约束，其扩散运动将受到不同程度的限制，称之为限制性扩散，一般组织中水分子的扩散运动则属于限制性扩散。表观扩散系数就是描述水分子在组织中的扩散能力的一种物理量。磁共振信号被激发后，水分子在梯度磁场方向上的扩散运动将造成磁共振信号的衰减，如果水分子在梯度磁场方向上扩散越自由，则在梯度磁场施加期间扩散距离越大，经历的磁场变化也越大，组织信号衰减越明显。因此，可以通过核磁共振技术测量物体的表观扩 散系数，从而间接的反映物体微观结构特点及其变化。
在磁共振成像技术中，表观扩散系数(ADC，Apparent Diffusion Coefficient)作为一个重要的临床诊断指标，被广泛的应用。一般通过扩散加权成像技术(Diffusion Weighted Imaging，DWI)进行测量，如自旋回波-平面回波序列(Spin echo-echo planar imaging，SE-EPI)，即自旋回波序列(SE)进行扩散梯度编码，平面回波序列(EPI)进行信号读出。在非均匀磁场磁共振***(也简称非均匀场磁共振***)中，类似的扩散加权成像技术被引入，用于测量物质的表观扩散系数。几种典型的测量表观扩散系数的脉冲序列如图1所示。图1中的(a)为自旋回波-CPMG序列(Spin echo-CPMG sequence，SE-CPMG)，即基于自旋回波进行扩散梯度编码，然后用超快速的CPMG序列进行信号读出。图1中的(b)为双自旋回波-CPMG序列(Dual spin echo-CPMG sequence，DSE-CPMG)，即基于双回波序列进行扩散梯度编码，同样用超快速的CPMG序列进行信号读出，该方法能降低低速液体流动带来的影响。图1中的(c)为受激回波-CPMG序列(Stimulated echo-CPMG sequence，STE-CPMG)，即基于受激回波序列进行扩散梯度编码，该方法能降低T1恢复的影响，当被检测物体的T1/T2比较小时，用该序列测量表观扩散系数(Apparent diffusion coefficient，ADC)可以提升测量准确度。
现有技术中，ADC测量脉冲序列都是由扩散梯度编码模块和信号读出模块组成。由于在非均匀场磁共振***中，梯度磁场非常大，通常比常规磁共振成像***(Magnetic Resonance Imaging，MRI)的梯度场高2～3个数量级，且无法控制该梯度磁场在信号读出阶段的改变(MRI***中的DWI技术，在信号读出阶段可以控制梯度场降低，而在非均匀磁体中，由于静态梯度磁场G往往非常大，且为恒定不可控梯度场)，而该梯度磁场在信号读出阶段仍然会起到扩散编码的作用，使得会影响到表观扩散系数ADC测量的准确性。因此需要采用超快速信号读出模块，才能降低读出信号过程中扩散效应的影响。即现有技术都是采用超短回波间隔的CPMG序列进行信号读出，例如使用的回波间隔为40us。这样对核磁共振***的谱仪设备、射频功放和射频线圈等硬件要求都非常高。
发明内容
为了解决在具有极不均匀磁场的核磁共振***中需采用超快速信号读出模块才能准确测量ADC和T2的二维图谱分布，导致对谱仪设备、射频功放和射频线圈等硬件要求都非常高的问题，本发明目的在于提供一种横向磁化矢量衰减时间常数的测量方法、装置、计算机设备、非均匀场磁共振***及计算机可读存储介质。
第一方面，本发明提供了一种ADC-T2二维图谱的测量方法，包括：
获取基于CPMG核磁共振序列采集得到的多组回波信号，其中，所述CPMG核磁共振序列用于进行扩散梯度编码和各组回波信号的读出，并针对各组回波信号采用有不同的回波间隔；
对所述多组回波信号进行如下积分公式的拟合：
Figure PCTCN2020130626-appb-000001
式中，S′(i,j)表示在所述多组回波信号中与变量i和j对应的回波信号，i＝1,2,…,M，j＝1,2,…,N，τ i表示在第i组回波信号采集过程中所采用的回波间隔，M表示所述多组回波信号的总组数，n j表示在单组回波信号中第j个回波信号，N表示单组回波信号的回波链长度，D max表示表观扩散系数ADC的最大值边界，D min表示表观扩散系数ADC的最小值边界，T 2-max表示横向磁化矢量衰减时间常数T2的最大值边界，T 2-min表示横向磁化矢量衰减时间常数T2的最小值边界，S(D,T 2)表示待求解的ADC-T2二维图谱，D表示表观扩散系数ADC，T 2表示横向磁化矢量衰减时间常数T2，e表示自然对数的底数；
将所述待求解的ADC-T2二维图谱的维度设定为p×q，然后将所述积分公式转换为如下的矩阵公式：
s′＝K·s
式中，s′＝vect(S′)，S′表示S′(i,j)，vect(S′)表示S′(i,j)的向量形式，K表示由M·N×p·q个元素K a,b排列成的矩阵，
Figure PCTCN2020130626-appb-000002
a＝1,2,…,M·N，b＝1,2,…,p·q，τ a＝k τ(a)＝vect(repmat(a,1,N))，n a＝k n(a)＝vect(repmat(a,M,1))，
Figure PCTCN2020130626-appb-000003
D b＝k D(b)＝vect(repmat(b,1,q))，repmat()表示用于复制和平铺矩阵的函数，vect(repmat())表示repmat()的向量形式，s＝vect(S)，S表示S(D,T 2)，vect(S)表示S(D,T 2)的向量形式；
根据所述矩阵公式，将求解S(D,T 2)的问题转化为如下公式的求解最优化问题：
Figure PCTCN2020130626-appb-000004
式中，F(s)表示关于变量s的函数，|||| 2表示向量的二范数，λ表示约束项系数；
根据所述求解最优化问题求解得到最小的且大于0的拟合向量s；
根据求解得到的拟合向量s，还原得到所述ADC-T2二维图谱的求解结果。
基于上述发明内容，可以在具有极不均匀的磁场或者是无法实现极短的回波时间的核磁共振***中，通过获取的多个采用不同回波间隔的CPMG序列采集的回波信号，从多组回波信号中拟合出ADC系数和T2值，从而能测量出ADC-T2图谱，使得不再需要复杂的扩散加权 序列，具有算法简单且对***要求低的优点，可以降低对对谱仪设备、射频功放和射频线圈等硬件***的成本。同时所述测量方法还具有算法稳定的特点，不易受流动液体影响，对于T1/T2较小的物质同样适用。即测量序列简单快速，无需扩散梯度编码模块，相对于现有技术更容易实现；对于极不均匀的磁场，或者是无法实现极短的回波时间的核磁共振***，该方法仍然能测量出ADC-T2图谱；基于前述两点优势，在设计用于测量ADC-T2的非均匀场核磁共振***时可以降低***硬件要求，从而降低硬件成本；所述ADC-T2测量方法，不易受流动液体影响，对于T1/T2较小的物质同样适用；可以较为准确得区分被检物体中的水和脂肪成分。
在一个可能的设计中，所述扩散敏感系数b值与(t EE) 3成正比，其中，t EE表示在单组回波信号采集过程中首个90°激发脉冲的中心点时刻至首个回波信号的中心点时刻的时长。
在一个可能的设计中，根据所述求解最优化问题求解得到最小的且大于0的拟合向量s，包括：
对矩阵K进行如下的且基于奇异值分解的降维处理：
Figure PCTCN2020130626-appb-000005
式中，U M·N×M·N表示M·N×M·N阶酉矩阵，
Figure PCTCN2020130626-appb-000006
表示p·q×p·q阶酉矩阵V p·q×p·q的共轭转置矩阵，∑ M·N×p·q对角线上的元素即为奇异值，k′表示降维后的奇异值个数，U M·N×k′表示U M·N×M·N的前k′列矩阵；
按照如下公式将向量s′从M×N维度降维到k′维度：
Figure PCTCN2020130626-appb-000007
式中，s′ svd表示将向量s′从M×N维度降维到k′维度时的向量，
Figure PCTCN2020130626-appb-000008
为矩阵U M·N×k′的转置矩阵；
将所述求解最优化问题转换为如下形式后，再求解得到所述拟合向量s：
Figure PCTCN2020130626-appb-000009
式中，
Figure PCTCN2020130626-appb-000010
表示将矩阵K从M×N维度降维到k′维度时的矩阵。
在一个可能的设计中，在对所述多组回波信号进行积分公式拟合前，还包括对所述多组回波信号进行如下的预处理：
在单次读出数据的采样点数维度上，对所述多组回波信号进行快速傅立叶变换，得到频域数据，然后保留低于预设频率阈值的低频部分并进行均值化处理，得到在所述采样点数维度上无差异的多组回波信号。
在一个可能的设计中，在对所述多组回波信号进行积分公式拟合前，还包括对所述多组回波信号进行如下的预处理：
在平均次数维度上，对所述多组回波信号进行均值化处理，得到在所述平均次数维度上无差异的多组回波信号。
在一个可能的设计中，在对所述多组回波信号进行积分公式拟合前，还包括对所述多组回波信号进行如下的预处理：
在回波链长度维度上，对所述多组回波信号进行基于奇异值分解的滤波处理，得到在所述回波链长度维度上无噪声影响的多组回波信号。
第二方面，本发明提供了一种ADC-T2二维图谱的测量装置，包括依次通信连接的回波信号获取模块、积分公式拟合模块、矩阵公式转换模块、求解问题转化模块、拟合向量求解模块和求解结果还原模块；
所述回波信号获取模块，用于获取基于CPMG核磁共振序列采集得到的多组回波信号，其中，所述CPMG核磁共振序列用于进行扩散梯度编码和各组回波信号的读出，并针对各组回波信号采用有不同的回波间隔；
所述积分公式拟合模块，用于对所述多组回波信号进行如下积分公式的拟合：
Figure PCTCN2020130626-appb-000011
式中，S′(i,j)表示在所述多组回波信号中与变量i和j对应的回波信号，i＝1,2,…,M，j＝1,2,…,N，τ i表示在第i组回波信号采集过程中所采用的回波间隔，M表示所述多组回波信号的总组数，n j表示在单组回波信号中第j个回波信号，N表示单组回波信号的回波链长度，D max表示表观扩散系数ADC的最大值边界，D min表示表观扩散系数ADC的最小值边界，T 2-max表示横向磁化矢量衰减时间常数T2的最大值边界，T 2-min表示横向磁化矢量衰减时间常数T2的最小值边界，S(D,T 2)表示待求解的ADC-T2二维图谱，D表示表观扩散系数ADC，T 2表示横向磁化矢量衰减时间常数T2，e表示自然对数的底数；
所述矩阵公式转换模块，用于将所述待求解的ADC-T2二维图谱的维度设定为p×q，然后将所述积分公式转换为如下的矩阵公式：
s′＝K·s
式中，s′＝vect(S′)，S′表示S′(i,j)，vect(S′)表示S′(i,j)的向量形式，K表示由M·N×p·q个元素K a,b排列成的矩阵，
Figure PCTCN2020130626-appb-000012
a＝1,2,…,M·N，b＝1,2,…,p·q，τ a＝k τ(a)＝vect(repmat(a,1,N))，n a＝k n(a)＝vect(repmat(a,M,1))，
Figure PCTCN2020130626-appb-000013
D b＝k D(b)＝vect(repmat(b,1,q))，repmat()表示用于复制和平铺矩阵的函数，vect(repmat())表示repmat()的向量形式，s＝vect(S)，S表示S(D,T 2)，vect(S)表示S(D,T 2)的向量形式；
所述求解问题转化模块，用于根据所述矩阵公式，将求解S(D,T 2)的问题转化为如下公式的求解最优化问题：
Figure PCTCN2020130626-appb-000014
式中，F(s)表示关于变量s的函数，|||| 2表示向量的二范数，λ表示约束项系数；
所述拟合向量求解模块，用于根据所述求解最优化问题求解得到最小的且大于0的拟合向量s；
所述求解结果还原模块，用于根据求解得到的拟合向量s，还原得到所述ADC-T2二维图谱的求解结果。
第三方面，本发明提供了一种计算机设备，包括通信相连的存储器和处理器，其中，所述存储器用于存储计算机程序，所述处理器用于读取所述计算机程序，执行如第一方面或第一方面中任意一种可能设计所述的测量方法。
第四方面，本发明提供了一种非均匀场磁共振***，包括有控制台、核磁共振谱仪、磁体和射频子***；
所述控制台，通信连接所述核磁共振谱仪，用于向所述核磁共振谱仪发送指令，以便控制测量序列的参数选择和感兴趣区域定位，并接收由所述核磁共振谱仪采集到的磁共振信号，完成数据处理，其中，所述数据处理包括执行如第一方面或第一方面中任意一种可能设计所述的测量方法；
所述核磁共振谱仪，通信连接所述射频子***，用于执行来自所述控制台的指令，并通过所述射频子***发射测量序列的射频激励信号和接收所述磁共振信号；
所述磁体，用于布置在被检物体和所述射频子***中信号收发部件的正上方；
所述射频子***，用于在所述核磁共振谱仪的控制下，发射测量序列的射频激励信号和接收所述磁共振信号。
在一个可能的设计中，所述射频子***包括有射频功率放大器、前置放大器、收发转换开关和射频线圈；
所述射频功率放大器的信号输入端电连接所述核磁共振谱仪的信号输出端，所述射频功率放大器的信号输出端电连接所述收发转换开关的第一切换端；
所述前置放大器的信号输入端电连接所述收发转换开关的第二切换端，所述前置放大器的信号输出端电连接所述核磁共振谱仪的信号输入端；
所述收发转换开关的受控端通信连接所述核磁共振谱仪的输出端，所述收发转换开关的 切换公共端电连接所述射频线圈；
所述射频线圈，用于作为所述射频子***的信号收发部件，向被检物体发射所述射频激励信号和接收来自被检物体的磁共振信号。
第五方面，本发明提供了一种计算机可读存储介质，所述计算机可读存储介质上存储有指令，当所述指令在计算机上运行时，执行如上第一方面或第一方面中任意一种可能设计的所述测量方法。
第六方面，本发明提供了一种包含指令的计算机程序产品，当所述指令在计算机上运行时，使所述计算机执行如上第一方面或第一方面中任意一种可能设计的所述测量方法。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是现有技术在非均匀场磁共振***中所采用的ADC测量脉冲序列的示例图，其中，(a)采用SE-CPMG序列；(b)采用DSE-CPMG序列；(c)采用STE-CPMG序列。
图2是本发明提供的测量方法的流程示意图。
图3是本发明提供的在非均匀场磁共振***中所采用的ADC-T2测量脉冲序列的示例图。
图4是本发明提供的当被检物质为0.5mmol/L MnCl2溶液和花生油时针对不同回波间隔τ＝TE所获取的CPMG测量数据示例图。
图5是本发明提供的当被检物质为0.5mmol/L MnCl2溶液和花生油时测量得到的ADC-T2二维图谱示例图。
图6是本发明提供的测量装置的结构示意图。
图7是本发明提供的计算机设备的结构示意图。
图8是本发明提供的非均匀场磁共振***的结构示意图。
具体实施方式
下面结合附图及具体实施例来对本发明作进一步阐述。在此需要说明的是，对于这些实施例方式的说明虽然是用于帮助理解本发明，但并不构成对本发明的限定。本文公开的特定结构和功能细节仅用于描述本发明的示例实施例。然而，可用很多备选的形式来体现本发明，并且不应当理解为本发明限制在本文阐述的实施例中。
应当理解，尽管本文可能使用术语第一、第二等等来描述各种单元，但是这些单元不应当受到这些术语的限制。这些术语仅用于区分一个单元和另一个单元。例如可以将第一单元称作第二单元,并且类似地可以将第二单元称作第一单元，同时不脱离本发明的示例实施例的范围。
应当理解，对于本文中可能出现的术语“和/或”，其仅仅是一种描述关联对象的关联关系，表示可以存在三种关系，例如，A和/或B，可以表示：单独存在A，单独存在B，同时存在A和B三种情况；对于本文中可能出现的术语“/和”，其是描述另一种关联对象关系，表示可以存在两种关系，例如，A/和B，可以表示：单独存在A，单独存在A和B两种情况；另外，对于本文中可能出现的字符“/”，一般表示前后关联对象是一种“或”关系。
应当理解，在本文中若将单元称作与另一个单元“连接”、“相连”或“耦合”时，它可以与另一个单元直相连接或耦合，或中间单元可以存在。相対地，在本文中若将单元称作与另一个单元“直接相连”或“直接耦合”时，表示不存在中间单元。另外，应当以类似方式来解释用于描述单元之间的关系的其他单词(例如，“在……之间”对“直接在……之间”,“相邻”对“直接相邻”等等)。
应当理解，本文使用的术语仅用于描述特定实施例，并不意在限制本发明的示例实施例。若本文所使用的，单数形式“一”、“一个”以及“该”意在包括复数形式，除非上下文明确指示相反意思。还应当理解，若术语“包括”、“包括了”、“包含”和/或“包含了”在本文中被使用时,指定所声明的特征、整数、步骤、操作、单元和/或组件的存在性,并且不排除一个或多个其他特征、数量、步骤、操作、单元、组件和/或他们的组合存在性或增加。
应当理解，还应当注意到在一些备选可能设计中，所出现的功能/动作可能与附图出现的顺序不同。例如,取决于所涉及的功能/动作,实际上可以实质上并发地执行,或者有时可以以相反的顺序来执行连续示出的两个图。
应当理解，在下面的描述中提供了特定的细节,以便于对示例实施例的完全理解。然而,本领域普通技术人员应当理解可以在没有这些特定细节的情况下实现示例实施例。例如可以在框图中示出***，以避免用不必要的细节来使得示例不清楚。在其他实例中，可以不以非必要的细节来示出众所周知的过程、结构和技术，以避免使得示例实施例不清楚。
如图2～5所示，本实施例第一方面提供的所述ADC-T2二维图谱的测量方法,可以但不限于由在具有极不均匀的磁场或者是无法实现极短回波时间的核磁共振***中的控制台(其与核磁共振谱仪连接，以便控制测量序列的参数选择和感兴趣区域定位，并接收由所述核磁共振谱仪采集到的磁共振信号，完成数据处理)执行，以便得到准确的ADC-T2二维图谱。所述ADC-T2二维图谱的测量方法，包括但不限于有如下步骤S101～S106。
S101.获取基于CPMG核磁共振序列采集得到的多组回波信号，其中，所述CPMG核磁共振序列用于进行扩散梯度编码和各组回波信号的读出，并针对各组回波信号采用有不同的回波间隔。
在所述步骤S101中，如图3所示，即采用一个典型的θ-2θ-2θ-2θ……射频脉冲序列：首个激发脉冲翻转角为θ，其后跟随若干个回聚脉冲，翻转角为2θ；首个激发脉冲和首个回聚脉冲之间的相位差为90度，首个激发脉冲与首个回聚脉冲之间的时间间隔为τ/2,首个回聚脉冲到首个采样窗之间的时间间隔为τ/2；回聚脉冲之间的时间间隔均为τ，称为回波间隔。在核磁共振***中，恒定梯度场为磁体的天然梯度场，不需要控制。此外，虽然一次激发即可采集N个回波信号(其中，N为正整数)，但也需要重复多次，以便通过平均信号提升信噪比，以及为了后续能够拟合得到ADC-T2二维图谱，需要通过改变回波间隔τ来采集多组回波信号。
S102.对所述多组回波信号进行如下积分公式的拟合：
Figure PCTCN2020130626-appb-000015
式中，S′(i,j)表示在所述多组回波信号中与变量i和j对应的回波信号，i＝1,2,…,M，j＝1,2,…,N，τ i表示在第i组回波信号采集过程中所采用的回波间隔，M表示所述多组回波信号的总组数，n j表示在单组回波信号中第j个回波信号，N表示单组回波信号的回波链长度，D max表示表观扩散系数ADC的最大值边界，D min表示表观扩散系数ADC的最小值边界，T 2-max表示横向磁化矢量衰减时间常数T2的最大值边界，T 2-min表示横向磁化矢量衰减时间常数T2的最小值边界，S(D,T 2)表示待求解的ADC-T2二维图谱，D表示表观扩散系数ADC，T 2表示横向磁化矢量衰减时间常数T2，e表示自然对数的底数。
在所述步骤S102中，采集获取的所述多组回波信号可用一个4维数组S(m,n,a,q)来表示，其中，第一维数组S(m)基于不同的回波间隔，长度为正整数M，即对应M个不同长度的回波间隔τ；第二维数组S(n)基于不同的回波链长度，长度为正整数N；第三维数组S(a)基于在进行重复采集时的不同平均次数，长度为正整数A；第四维数组S(q)基于单次读出数据的不同采样点数，长度为正整数Q。
在所述步骤S102中，考虑所述多组回波信号是一个4维数组S(m,n,a,q)，维度过多且存在噪声影响，会增加不必要的积分计算量以及带来误差，由此有必要先进行降维和/或去噪处理，即在对所述多组回波信号进行积分公式拟合前，还包括但不限于对所述多组回波信号进行如下方式(A)至(C)中任意一项或任意组合的预处理：(A)在单次读出数据的采样 点数维度上，对所述多组回波信号进行快速傅立叶变换，得到频域数据，然后保留低于预设频率阈值的低频部分并进行均值化处理，得到在所述采样点数维度上无差异的多组回波信号；(B)在平均次数维度上，对所述多组回波信号进行均值化处理，得到在所述平均次数维度上无差异的多组回波信号；(C)在回波链长度维度上，对所述多组回波信号进行基于奇异值分解(Singular Value Decompositionm,简称SVD，是在机器学习领域应用较为广泛的算法之一，也是学习机器学习算法绕不开的基石之一。SVD算法主要用在降维算法中的特征分解、推荐***、自然语言处理计算机视觉等领域，通俗一点讲就是将一个线性变换分解为两个线性变换，一个线性变换代表旋转，一个线性变换代表拉伸)的滤波处理，得到在所述回波链长度维度上无噪声影响的多组回波信号。由此通过前述方式(A)和(B)的处理，可将四维数组S(m,n,a,q)降维成二维数组S(m,n)，利于减少后续计算。
S103.将所述待求解的ADC-T2二维图谱的维度设定为p×q，然后将所述积分公式转换为如下的矩阵公式：
s′＝K·s
式中，s′＝vect(S′)，S′表示S′(i,j)，vect(S′)表示S′(i,j)的向量形式，K表示由M·N×p·q个元素K a,b排列成的矩阵，
Figure PCTCN2020130626-appb-000016
a＝1,2,…,M·N，b＝1,2,…,p·q，τ a＝k τ(a)＝vect(repmat(a,1,N))，n a＝k n(a)＝vect(repmat(a,M,1))，
Figure PCTCN2020130626-appb-000017
D b＝k D(b)＝vect(repmat(b,1,q))，repmat()表示用于复制和平铺矩阵的函数，vect(repmat())表示repmat()的向量形式，s＝vect(S)，S表示S(D,T 2)，vect(S)表示S(D,T 2)的向量形式。
S104.根据所述矩阵公式，将求解S(D,T 2)的问题转化为如下公式的求解最优化问题：
Figure PCTCN2020130626-appb-000018
式中，F(s)表示关于变量s的函数，|||| 2表示向量的二范数，λ表示约束项系数。
S105.根据所述求解最优化问题求解得到最小的且大于0的拟合向量s。
在所述步骤S105中，根据所述求解最优化问题求解所述拟合向量s的具体过程为现有常规求解方式。优化的，为了降低在求解过程中的计算量，可以但不限于按照如下步骤S501～S503进行降维求解。
S501.对矩阵K进行如下的且基于奇异值分解的降维处理：
Figure PCTCN2020130626-appb-000019
式中，U M·N×M·N表示M·N×M·N阶酉矩阵，
Figure PCTCN2020130626-appb-000020
表示p·q×p·q阶酉矩阵V p·q×p·q的共轭转置矩阵，∑ M·N×p·q对角线上的元素即为奇异值，k′表示降维后的奇异值个数，U M·N×k′表示U M·N×M·N的前k′列矩阵；
S502.按照如下公式将向量s′从M×N维度降维到k′维度：
Figure PCTCN2020130626-appb-000021
式中，s′ svd表示将向量s′从M×N维度降维到k′维度时的向量，
Figure PCTCN2020130626-appb-000022
为矩阵U M·N×k′的转置矩阵。
S503.将所述求解最优化问题转换为如下形式后，再求解得到所述拟合向量s：
Figure PCTCN2020130626-appb-000023
式中，
Figure PCTCN2020130626-appb-000024
表示将矩阵K从M×N维度降维到k′维度时的矩阵。
S106.根据求解得到的拟合向量s，还原得到所述ADC-T2二维图谱的求解结果。
在所述步骤S106中，由于拟合向量s＝vect(S)＝vect(S(D,T 2))，因此可通过常规方式还原得到所述ADC-T2二维图谱的求解结果。即考虑非均匀场中存在的强梯度场，在CPMG序列中既可以起频率编码的作用，同时也始终存在扩散编码的作用。也就是说如果CPMG序列的回波间隔越大，则信号越低，其中，扩散效应对CPMG信号的影响，可以用如下公式描述：
Figure PCTCN2020130626-appb-000025
式中，γ表示原子核的磁旋比，D′表示物质的ADC系数，G表示梯度磁场大小，τ表示CPMG序列的回波间隔。由该公式可以看出，CPMG信号和物质的ADC系数、梯度场和回波间隔有关，由此可以通过采用不同回波间隔的CPMG信号估计出ADC系数和T2值，得到所述ADC-T2二维图谱。
如图4～5所示，下面以在非均匀场核磁共振***上对0.5mmol/L MnCl 2溶液和花生油进行检测为例，来具体说明本实施例的实验结果。实验参数主要如下：感兴趣区域(Region of Interest，ROI)的恒定磁场为0.07T，梯度场为110Gauss/cm，CPMG序列扩散梯度编码的回波间隔τ分别为0.5ms、1.2ms、1.5ms、1.7ms、1.85ms、2.0ms、2.5ms、6.0ms、7.5ms、8.5ms、9.0ms和10.0ms共12个(即M＝12)，CPMG序列的回波链长度为256(即N＝256)，平均次数为64次(即A＝64)，回波采样点数为64(即Q＝64)。
根据图4可见具有不同回波间隔的CPMG序列采集信号衰减程度不一样，这是由于分子扩散的影响；根据图5可见0.5mmol/L MnCl 2溶液在ADC-T2二维图谱上为一均匀峰，而花生油在ADC-T2二维图谱为双峰，两种物质的ADC系数和T2值均与理论值接近。此外，还可以看到，MnCl 2溶液和花生油在图谱上的峰区分明显：MnCl 2溶液的ADC较大，而花生油具有更小的ADC。
由此通过上述步骤S101～S106所详细描述的测量方案，可以在具有极不均匀的磁场或者是无法实现极短的回波时间的核磁共振***中，通过获取的多个采用不同回波间隔的CPMG序列采集的回波信号，从多组回波信号中拟合出ADC系数和T2值，从而能测量出ADC-T2图谱，使得不再需要复杂的扩散加权序列，具有算法简单且对***要求低的优点，可以降低对对谱仪设备、射频功放和射频线圈等硬件***的成本。同时所述测量方法还具有算法稳定的特点，不易受流动液体影响，对于T1/T2较小的物质同样适用。即测量序列简单快速，无需扩散梯度编码模块，相对于现有技术更容易实现；对于极不均匀的磁场，或者是无法实现极短的回波时间的核磁共振***，该方法仍然能测量出ADC-T2图谱；基于前述两点优势，在设计用于测量ADC-T2的非均匀场核磁共振***时可以降低***硬件要求，从而降低硬件成本；所述ADC-T2测量方法，不易受流动液体影响，对于T1/T2较小的物质同样适用；可以较为准确得区分被检物体中的水和脂肪成分。
如图6所示，本实施例第二方面提供了一种实现第一方面或第一方面中任意一种可能设计的所述测量方法的虚拟装置，包括依次通信连接的回波信号获取模块、积分公式拟合模块、矩阵公式转换模块、求解问题转化模块、拟合向量求解模块和求解结果还原模块；
所述回波信号获取模块，用于获取基于CPMG核磁共振序列采集得到的多组回波信号，其中，所述CPMG核磁共振序列用于进行扩散梯度编码和各组回波信号的读出，并针对各组回波信号采用有不同的回波间隔；
所述积分公式拟合模块，用于对所述多组回波信号进行如下积分公式的拟合：
Figure PCTCN2020130626-appb-000026
式中，S′(i,j)表示在所述多组回波信号中与变量i和j对应的回波信号，i＝1,2,…,M，j＝1,2,…,N，τ i表示在第i组回波信号采集过程中所采用的回波间隔，M表示所述多组回波信号的总组数，n j表示在单组回波信号中第j个回波信号，N表示单组回波信号的回波链长度，D max表示表观扩散系数ADC的最大值边界，D min表示表观扩散系数ADC的最小值边界，T 2-max表示横向磁化矢量衰减时间常数T2的最大值边界，T 2-min表示横向磁化矢量衰减时间常数T2的最小值边界，S(D,T 2)表示待求解的ADC-T2二维图谱，D表示表观扩散系数ADC，T 2表示横向磁化矢量衰减时间常数T2，e表示自然对数的底数；
所述矩阵公式转换模块，用于将所述待求解的ADC-T2二维图谱的维度设定为p×q，然后将所述积分公式转换为如下的矩阵公式：
s′＝K·s
式中，s′＝vect(S′)，S′表示S′(i,j)，vect(S′)表示S′(i,j)的向量形式，K表示由M·N×p·q个元素K a,b排列成的矩阵，
Figure PCTCN2020130626-appb-000027
a＝1,2,…,M·N，b＝1,2,…,p·q，τ a＝k τ(a)＝vect(repmat(a,1,N))，n a＝k n(a)＝vect(repmat(a,M,1))，
Figure PCTCN2020130626-appb-000028
D b＝k D(b)＝vect(repmat(b,1,q))，repmat()表示用于复制和平铺矩阵的函数，vect(repmat())表示repmat()的向量形式，s＝vect(S)，S表示S(D,T 2)，vect(S)表示S(D,T 2)的向量形式；
所述求解问题转化模块，用于根据所述矩阵公式，将求解S(D,T 2)的问题转化为如下公式的求解最优化问题：
Figure PCTCN2020130626-appb-000029
式中，F(s)表示关于变量s的函数，|||| 2表示向量的二范数，λ表示约束项系数；
所述拟合向量求解模块，用于根据所述求解最优化问题求解得到最小的且大于0的拟合向量s；
所述求解结果还原模块，用于根据求解得到的拟合向量s，还原得到所述ADC-T2二维图谱的求解结果。
本实施例第二方面提供的前述装置的工作过程、工作细节和技术效果，可以参见第一方面或第一方面中任意一种可能设计所述的测量方法，于此不再赘述。
如图7所示，本实施例第三方面提供了一种执行第一方面或第一方面中任意一种可能设计的所述测量方法的计算机设备，包括通信相连的存储器和处理器，其中，所述存储器用于存储计算机程序，所述处理器用于读取所述计算机程序，执行如第一方面或第一方面中任意一种可能设计所述的测量方法。具体举例的，所述存储器可以但不限于包括随机存取存储器(Random-Access Memory，RAM)、只读存储器(Read-Only Memory，ROM)、闪存(Flash Memory)、先进先出存储器(First Input First Output，FIFO)和/或先进后出存储器(First Input Last Output，FILO)等等；所述处理器可以不限于采用型号采用STM32F105系列的微处理器。此外，所述计算机设备还可以但不限于包括有电源模块、显示屏和其它必要的部件。
本实施例第三方面提供的前述计算机设备的工作过程、工作细节和技术效果，可以参见第一方面或第一方面中任意一种可能设计所述的测量方法，于此不再赘述。
如图8所示，本实施例第四方面提供了一种执行第一方面或第一方面中任意一种可能设计的所述测量方法的非均匀场磁共振***，包括有控制台、核磁共振谱仪、磁体和射频子***；
所述控制台，通信连接所述核磁共振谱仪，用于向所述核磁共振谱仪发送指令，以便控制测量序列的参数选择和感兴趣区域定位，并接收由所述核磁共振谱仪采集到的磁共振信号，完成数据处理，其中，所述数据处理包括执行如第一方面或第一方面中任意一种可能设计所述的测量方法；
所述核磁共振谱仪，通信连接所述射频子***，用于执行来自所述控制台的指令，并通过所述射频子***发射测量序列的射频激励信号和接收所述磁共振信号；
所述磁体，用于布置在被检物体和所述射频子***中信号收发部件的正上方；
所述射频子***，用于在所述核磁共振谱仪的控制下，发射测量序列的射频激励信号和接收所述磁共振信号。
如图8所示，在所述非均匀场磁共振***的具体结构中，所述磁体一般设计为永磁体，例如单边永磁体，使得在感兴趣区域内仍然具有高度不均匀的磁场。
在一种可能设计中，所述射频子***包括有射频功率放大器、前置放大器、收发转换开关和射频线圈；
所述射频功率放大器的信号输入端电连接所述核磁共振谱仪的信号输出端，所述射频功率放大器的信号输出端电连接所述收发转换开关的第一切换端；
所述前置放大器的信号输入端电连接所述收发转换开关的第二切换端，所述前置放大器的信号输出端电连接所述核磁共振谱仪的信号输入端；
所述收发转换开关的受控端通信连接所述核磁共振谱仪的输出端，所述收发转换开关的切换公共端电连接所述射频线圈；
所述射频线圈，用于作为所述射频子***的信号收发部件，向被检物体发射所述射频激励信号和接收来自被检物体的磁共振信号。
如图8所示，在所述射频子***的具体结构中，所述射频功率放大器用于放大待发射的所述射频激励信号；所述前置放大器用于放大接收的所述磁共振信号。所述收发转换开关用于通过切换控制，使所述射频线圈即可发射所述射频激励信号，也能接收所述磁共振信号。
本实施例第四方面提供的前述非均匀场磁共振***的工作过程、工作细节和技术效果，可以参见第一方面或第一方面中任意一种可能设计所述的测量方法，于此不再赘述。
本实施例第五方面提供了一种存储包含第一方面或第一方面中任意一种可能设计的所述测量方法的指令的计算机可读存储介质，即所述计算机可读存储介质上存储有指令，当所述指令在计算机上运行时，执行如第一方面或第一方面中任意一种可能设计所述的测量方法。其中，所述计算机可读存储介质是指存储数据的载体，可以但不限于包括软盘、光盘、硬盘、闪存、优盘和/或记忆棒(Memory Stick)等，所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。
本实施例第五方面提供的前述计算机可读存储介质的工作过程、工作细节和技术效果，可以参见第一方面或第一方面中任意一种可能设计所述的测量方法，于此不再赘述。
本实施例第六方面提供了一种包含指令的计算机程序产品，当所述指令在计算机上运行时，使所述计算机执行如第一方面或第一方面中任意一种可能设计所述的测量方法。其中，所述计算机可以是通用计算机、专用计算机、计算机网络、或者其他可编程装置。
以上所描述的实施例仅仅是示意性的，若涉及到作为分离部件说明的单元，其可以是或者也可以不是物理上分开的；若涉及到作为单元显示的部件，其可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。本领域普通技术人员在不付出创造性的劳动的情况下，即可以理解并实施。
以上实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换。而这些修改或者替换，并不使相应技术方案的本质脱离本发明实施例技术方案的精神和范围。
最后应说明的是，本发明不局限于上述可选的实施方式，任何人在本发明的启示下都可得出其他各种形式的产品。上述具体实施方式不应理解成对本发明的保护范围的限制，本发明的保护范围应当以权利要求书中界定的为准，并且说明书可以用于解释权利要求书。

Claims (10)

  1. 一种ADC-T2二维图谱的测量方法，其特征在于，包括：
    获取基于CPMG核磁共振序列采集得到的多组回波信号，其中，所述CPMG核磁共振序列用于进行扩散梯度编码和各组回波信号的读出，并针对各组回波信号采用有不同的回波间隔；
    对所述多组回波信号进行如下积分公式的拟合：
    Figure PCTCN2020130626-appb-100001
    式中，S′(i,j)表示在所述多组回波信号中与变量i和j对应的回波信号，i＝1,2,…,M，j＝1,2,…,N，τ i表示在第i组回波信号采集过程中所采用的回波间隔，M表示所述多组回波信号的总组数，n j表示在单组回波信号中第j个回波信号，N表示单组回波信号的回波链长度，D max表示表观扩散系数ADC的最大值边界，D min表示表观扩散系数ADC的最小值边界，T 2-max表示横向磁化矢量衰减时间常数T2的最大值边界，T 2-min表示横向磁化矢量衰减时间常数T2的最小值边界，S(D,T 2)表示待求解的ADC-T2二维图谱，D表示表观扩散系数ADC，T 2表示横向磁化矢量衰减时间常数T2，e表示自然对数的底数；
    将所述待求解的ADC-T2二维图谱的维度设定为p×q，然后将所述积分公式转换为如下的矩阵公式：
    s′＝K·s
    式中，s′＝vect(S′)，S′表示S′(i,j)，vect(S′)表示S′(i,j)的向量形式，K表示由M·N×p·q个元素K a,b排列成的矩阵，
    Figure PCTCN2020130626-appb-100002
    a＝1,2,…,M·N，b＝1,2,…,p·q，τ a＝k τ(a)＝vect(repmat(a,1,N))，n a＝k n(a)＝vect(repmat(a,M,1))，
    Figure PCTCN2020130626-appb-100003
    D b＝k D(b)＝vect(repmat(b,1,q))，repmat()表示用于复制和平铺矩阵的函数，vect(repmat())表示repmat()的向量形式，s＝vect(S)，S表示S(D,T 2)，vect(S)表示S(D,T 2)的向量形式；
    根据所述矩阵公式，将求解S(D,T 2)的问题转化为如下公式的求解最优化问题：
    Figure PCTCN2020130626-appb-100004
    式中，F(s)表示关于变量s的函数，|| || 2表示向量的二范数，λ表示约束项系数；
    根据所述求解最优化问题求解得到最小的且大于0的拟合向量s；
    根据求解得到的拟合向量s，还原得到所述ADC-T2二维图谱的求解结果。
  2. 如权利要求1所述的测量方法，其特征在于，根据所述求解最优化问题求解得到最小的且大于0的拟合向量s，包括：
    对矩阵K进行如下的且基于奇异值分解的降维处理：
    Figure PCTCN2020130626-appb-100005
    式中，U M·N×M·N阶表示M·N×M·N阶酉矩阵，
    Figure PCTCN2020130626-appb-100006
    表示p·q×p·q阶酉矩阵V p·q×p·q的共轭转置矩阵，∑ M·N×p·q对角线上的元素即为奇异值，k′表示降维后的奇异值个数，U M·N×k′表示U M·N×M·N的前k′列矩阵；
    按照如下公式将向量s′从M×N维度降维到k′维度：
    Figure PCTCN2020130626-appb-100007
    式中，s′ svd表示将向量s′从M×N维度降维到k′维度时的向量，
    Figure PCTCN2020130626-appb-100008
    为矩阵U M·N×k′的转置矩阵；
    将所述求解最优化问题转换为如下形式后，再求解得到所述拟合向量s：
    Figure PCTCN2020130626-appb-100009
    式中，
    Figure PCTCN2020130626-appb-100010
    表示将矩阵K从M×N维度降维到k′维度时的矩阵。
  3. 如权利要求1所述的测量方法，其特征在于，在对所述多组回波信号进行积分公式拟合前，还包括对所述多组回波信号进行如下的预处理：
    在单次读出数据的采样点数维度上，对所述多组回波信号进行快速傅立叶变换，得到频域数据，然后保留低于预设频率阈值的低频部分并进行均值化处理，得到在所述采样点数维度上无差异的多组回波信号。
  4. 如权利要求1所述的测量方法，其特征在于，在对所述多组回波信号进行积分公式拟合前，还包括对所述多组回波信号进行如下的预处理：
    在平均次数维度上，对所述多组回波信号进行均值化处理，得到在所述平均次数维度上无差异的多组回波信号。
  5. 如权利要求1所述的测量方法，其特征在于，在对所述多组回波信号进行积分公式拟合前，还包括对所述多组回波信号进行如下的预处理：
    在回波链长度维度上，对所述多组回波信号进行基于奇异值分解的滤波处理，得到在所述回波链长度维度上无噪声影响的多组回波信号。
  6. 一种ADC-T2二维图谱的测量装置，其特征在于，包括依次通信连接的回波信号获取 模块、积分公式拟合模块、矩阵公式转换模块、求解问题转化模块、拟合向量求解模块和求解结果还原模块；
    所述回波信号获取模块，用于获取基于CPMG核磁共振序列采集得到的多组回波信号，其中，所述CPMG核磁共振序列用于进行扩散梯度编码和各组回波信号的读出，并针对各组回波信号采用有不同的回波间隔；
    所述积分公式拟合模块，用于对所述多组回波信号进行如下积分公式的拟合：
    Figure PCTCN2020130626-appb-100011
    式中，S′(i,j)表示在所述多组回波信号中与变量i和j对应的回波信号，i＝1,2,…,M，j＝1,2,…,N，τ i表示在第i组回波信号采集过程中所采用的回波间隔，M表示所述多组回波信号的总组数，n j表示在单组回波信号中第j个回波信号，N表示单组回波信号的回波链长度，D max表示表观扩散系数ADC的最大值边界，D min表示表观扩散系数ADC的最小值边界，T 2-max表示横向磁化矢量衰减时间常数T2的最大值边界，T 2-min表示横向磁化矢量衰减时间常数T2的最小值边界，S(D,T 2)表示待求解的ADC-T2二维图谱，D表示表观扩散系数ADC，T 2表示横向磁化矢量衰减时间常数T2，e表示自然对数的底数；
    所述矩阵公式转换模块，用于将所述待求解的ADC-T2二维图谱的维度设定为p×q，然后将所述积分公式转换为如下的矩阵公式：
    s′＝K·s
    式中，s′＝vect(S′)，S′表示S′(i,j)，vect(S′)表示S′(i,j)的向量形式，K表示由M·N×p·q个元素K a,b排列成的矩阵，
    Figure PCTCN2020130626-appb-100012
    a＝1,2,…,M·N，b＝1,2,…,p·q，τ a＝k τ(a)＝vect(repmat(a,1,N))，n a＝k n(a)＝vect(repmat(a,M,1))，
    Figure PCTCN2020130626-appb-100013
    D b＝k D(b)＝vect(repmat(b,1,q))，repmat()表示用于复制和平铺矩阵的函数，vect(repmat())表示repmat()的向量形式，s＝vect(S)，S表示S(D,T 2)，vect(S)表示S(D,T 2)的向量形式；
    所述求解问题转化模块，用于根据所述矩阵公式，将求解S(D,T 2)的问题转化为如下公式的求解最优化问题：
    Figure PCTCN2020130626-appb-100014
    式中，F(s)表示关于变量s的函数，|| || 2表示向量的二范数，λ表示约束项系数；
    所述拟合向量求解模块，用于根据所述求解最优化问题求解得到最小的且大于0的拟合向量s；
    所述求解结果还原模块，用于根据求解得到的拟合向量s，还原得到所述ADC-T2二维图谱的求解结果。
  7. 一种计算机设备，其特征在于,包括通信相连的存储器和处理器，其中，所述存储器用于存储计算机程序，所述处理器用于读取所述计算机程序，执行如权利要求1～5中任意一项所述的测量方法。
  8. 一种非均匀场磁共振***，其特征在于，包括有控制台、核磁共振谱仪、磁体和射频子***；
    所述控制台，通信连接所述核磁共振谱仪，用于向所述核磁共振谱仪发送指令，以便控制测量序列的参数选择和感兴趣区域定位，并接收由所述核磁共振谱仪采集到的磁共振信号，完成数据处理，其中，所述数据处理包括执行如权利要求1～5中任意一项所述的测量方法；
    所述核磁共振谱仪，通信连接所述射频子***，用于执行来自所述控制台的指令，并通过所述射频子***发射测量序列的射频激励信号和接收所述磁共振信号；
    所述磁体，用于布置在被检物体和所述射频子***中信号收发部件的正上方；
    所述射频子***，用于在所述核磁共振谱仪的控制下，发射测量序列的射频激励信号和接收所述磁共振信号。
  9. 如权利要求8所述的一种非均匀场磁共振***，其特征在于，所述射频子***包括有射频功率放大器、前置放大器、收发转换开关和射频线圈；
    所述射频功率放大器的信号输入端电连接所述核磁共振谱仪的信号输出端，所述射频功率放大器的信号输出端电连接所述收发转换开关的第一切换端；
    所述前置放大器的信号输入端电连接所述收发转换开关的第二切换端，所述前置放大器的信号输出端电连接所述核磁共振谱仪的信号输入端；
    所述收发转换开关的受控端通信连接所述核磁共振谱仪的输出端，所述收发转换开关的切换公共端电连接所述射频线圈；
    所述射频线圈，用于作为所述射频子***的信号收发部件，向被检物体发射所述射频激励信号和接收来自被检物体的磁共振信号。
  10. 一种计算机可读存储介质，其特征在于,所述计算机可读存储介质上存储有指令，当所述指令在计算机上运行时，执行如权利要求1～5中任意一项所述的测量方法。
PCT/CN2020/130626 2020-11-09 2020-11-20 一种adc-t2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振*** WO2022095142A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202011241126.X 2020-11-09
CN202011241126.XA CN112415454B (zh) 2020-11-09 2020-11-09 一种adc-t2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振***

Publications (1)

Publication Number Publication Date
WO2022095142A1 true WO2022095142A1 (zh) 2022-05-12

Family

ID=74780809

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/130626 WO2022095142A1 (zh) 2020-11-09 2020-11-20 一种adc-t2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振***

Country Status (2)

Country Link
CN (1) CN112415454B (zh)
WO (1) WO2022095142A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113197566B (zh) * 2021-04-23 2022-07-08 无锡鸣石峻致医疗科技有限公司 一种便携式核磁共振检测***的体内定位方法、装置、计算机设备及核磁共振检测***
CN114236442B (zh) * 2021-12-14 2022-09-16 无锡鸣石峻致医疗科技有限公司 一种对核磁共振信号进行运动不敏感采集的方法、装置、计算机设备及核磁共振检测***
CN117310581B (zh) * 2023-10-11 2024-05-10 安徽峻德医疗科技有限公司 一种核磁共振信号衰减拟合方法、***、设备及存储介质

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100264920A1 (en) * 2009-04-20 2010-10-21 The Trustees Of The University Of Pennslyvania Spin locked balanced steady-state free precession (slssfp)
CN102711602A (zh) * 2011-01-11 2012-10-03 株式会社东芝 磁共振成像装置和磁共振成像方法
WO2017132182A1 (en) * 2016-01-29 2017-08-03 The Regents Of The University Of California Systems and methods for joint reconstruction of quantitative t2 and adc maps
CN111351813A (zh) * 2020-03-17 2020-06-30 无锡鸣石峻致医疗科技有限公司 一种基于非均匀场磁共振***的表观扩散系数测量方法
CN111721795A (zh) * 2020-06-29 2020-09-29 无锡鸣石峻致医疗科技有限公司 一种基于核磁共振***的物质测量方法及***

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPS63184048A (ja) * 1986-09-10 1988-07-29 Matsushita Electric Works Ltd 樹脂硬化度評価方法
US10393839B2 (en) * 2012-02-03 2019-08-27 Arizona Board Of Regents On Behalf Of The University Of Arizona System and method for image processing with highly undersampled imaging data
CN104198970B (zh) * 2014-09-05 2017-01-18 厦门大学 一种在不均匀磁场下获取高分辨核磁共振谱图的方法
CN104282007B (zh) * 2014-10-22 2017-12-19 长春理工大学 基于非采样的轮廓波变换自适应的医学图像融合方法
CN106226748A (zh) * 2016-06-29 2016-12-14 南京理工大学 一种线性正则变换联合s变换的微多普勒信号分析方法
CN113466280B (zh) * 2018-02-27 2022-07-26 华东师范大学 便于扩展分子信息库的仿真核磁共振波谱分析方法、分析***及其应用
GB2581168B (en) * 2019-02-06 2021-03-31 Siemens Healthcare Ltd A method and apparatus for generating a T1/T2 map
CN111537928B (zh) * 2020-03-17 2021-07-23 无锡鸣石峻致医疗科技有限公司 一种基于扩散效应的磁共振***梯度场测量方法

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20100264920A1 (en) * 2009-04-20 2010-10-21 The Trustees Of The University Of Pennslyvania Spin locked balanced steady-state free precession (slssfp)
CN102711602A (zh) * 2011-01-11 2012-10-03 株式会社东芝 磁共振成像装置和磁共振成像方法
WO2017132182A1 (en) * 2016-01-29 2017-08-03 The Regents Of The University Of California Systems and methods for joint reconstruction of quantitative t2 and adc maps
CN111351813A (zh) * 2020-03-17 2020-06-30 无锡鸣石峻致医疗科技有限公司 一种基于非均匀场磁共振***的表观扩散系数测量方法
CN111721795A (zh) * 2020-06-29 2020-09-29 无锡鸣石峻致医疗科技有限公司 一种基于核磁共振***的物质测量方法及***

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
FAN YI-REN, WU FEI, LI HU, HUO NING-NING, WANG YAO-SEN, DENG SHAO-GUI, YANG PEI-QIANG: "A modified design of pulse sequence and inversion method for D-T2 two-dimensional NMR", ACTA PHYSICA SINICA, vol. 64, no. 9, 1 January 2015 (2015-01-01), pages 099301, XP055927605, ISSN: 1000-3290, DOI: 10.7498/aps.64.099301 *
SONG GONGPU, WU LEI;CHENG JINGJING: "Study of Sequence and Inversion Algorithm in 2D NMR Well logging ", FOREIGN ELECTRONIC MEASUREMENT TECHNOLOGY, vol. 32, no. 9, 30 September 2013 (2013-09-30), pages 38 - 41, XP055927603, DOI: 10.19652/j.cnki.femt.2013.09.014 *

Also Published As

Publication number Publication date
CN112415454A (zh) 2021-02-26
CN112415454B (zh) 2021-08-03

Similar Documents

Publication Publication Date Title
WO2022095142A1 (zh) 一种adc-t2二维图谱的测量方法、装置、计算机设备及非均匀场磁共振***
Kim et al. Diffusion‐relaxation correlation spectroscopic imaging: a multidimensional approach for probing microstructure
CN112462311B (zh) 一种横向磁化矢量衰减时间常数的校正测量方法、装置、计算机设备及非均匀场磁共振***
Aganj et al. Reconstruction of the orientation distribution function in single‐and multiple‐shell q‐ball imaging within constant solid angle
Edzes et al. Quantitative T2 imaging of plant tissues by means of multi-echo MRI microscopy
EP3105606B1 (en) Method for quantifying isotropic diffusion and/or anisotropic diffusion in a sample
WO2021184471A1 (zh) 一种基于非均匀场磁共振***的表观扩散系数测量方法
WO2022000948A1 (zh) 基于核磁共振***的物质测量方法及***
US20110204892A1 (en) Spin echo spi methods for quantitative analysis of fluids in porous media
WO2021184470A1 (zh) 一种基于扩散效应的磁共振***梯度场测量方法
CN107561112B (zh) 一种获取岩石渗透率剖面的核磁共振方法
US5652516A (en) Spectroscopic magnetic resonance imaging using spiral trajectories
US8810244B2 (en) Method and system for magnetic resonance imaging, and use thereof
Caruyer et al. Optimal Design of Multiple Q-shells experiments for Diffusion MRI
CN102654569B (zh) 用于处理复数图像数据的方法和装置
Sharma et al. Accelerated water–fat imaging using restricted subspace field map estimation and compressed sensing
JP6762284B2 (ja) 磁気共鳴イメージング装置およびノイズ除去方法
Chun et al. Mean squared error based excitation pattern design for parallel transmit and receive SENSE MRI image reconstruction
Wörtge et al. Quantitative stray-field T1 relaxometry with the matrix pencil method
Willis et al. Fundamentals of diffusion measurements using NMR
Nugent et al. T1ρ imaging using magnetization‐prepared projection encoding (MaPPE)
Cai Diffusion-based mr methods for measuring water exchange
Jiang et al. An Optimized Multi-Component Imaging Method on a Homebuilt 0.5 T MRI System: Combing Intra-and Inter-Voxel Constraints
Yepes-Calderon et al. Magnetic resonance image enhancement by reducing receptors' effective size and enabling multiple channel acquisition
Claeser Fast mapping of the T1-relaxation in magnetic resonance imaging with advanced spiral k-space sampling and interleaving methods at 3 Tesla and 7 Tesla

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 20960599

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 20960599

Country of ref document: EP

Kind code of ref document: A1