WO2021077521A1 - 基于稀疏采样的全息磁感应胸腔成像方法及成像系统 - Google Patents

基于稀疏采样的全息磁感应胸腔成像方法及成像系统 Download PDF

Info

Publication number
WO2021077521A1
WO2021077521A1 PCT/CN2019/119937 CN2019119937W WO2021077521A1 WO 2021077521 A1 WO2021077521 A1 WO 2021077521A1 CN 2019119937 W CN2019119937 W CN 2019119937W WO 2021077521 A1 WO2021077521 A1 WO 2021077521A1
Authority
WO
WIPO (PCT)
Prior art keywords
magnetic field
module
image
scattered
chest
Prior art date
Application number
PCT/CN2019/119937
Other languages
English (en)
French (fr)
Inventor
王露露
Original Assignee
深圳技术大学
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 深圳技术大学 filed Critical 深圳技术大学
Publication of WO2021077521A1 publication Critical patent/WO2021077521A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/05—Detecting, measuring or recording for diagnosis by means of electric currents or magnetic fields; Measuring using microwaves or radio waves 
    • A61B5/0522—Magnetic induction tomography
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/0033—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room
    • A61B5/004—Features or image-related aspects of imaging apparatus classified in A61B5/00, e.g. for MRI, optical tomography or impedance tomography apparatus; arrangements of imaging apparatus in a room adapted for image acquisition of a particular organ or body part

Definitions

  • This application belongs to the field of magnetic induction imaging technology, and specifically relates to a holographic magnetic induction chest imaging method and imaging system based on sparse sampling.
  • MRI Magnetic resonance Imaging
  • Magnetic induction imaging (Electromagnetic Induction Imaging, EIT) is a non-contact imaging method that targets the conductivity of the object under the excitation of an alternating magnetic field. Because EIT has the advantages of non-invasive, non-contact, cheap, and convenient to carry, EIT has great application value in the field of biomedical imaging. However, the current magnetic induction tomography technology is faced with problems such as fast and slow imaging, severe noise interference, incomplete algorithms, and inability to three-dimensional imaging, which has led to slow progress in its clinical promotion.
  • compressed sensing can accurately recover compressible signals through measurement data much lower than the Nyquist sampling rate.
  • compressed sensing has the advantage of being able to greatly compress the number of measurements necessary to extract the signal by virtue of its sparsity.
  • the present application provides a holographic magnetic induction chest imaging method and imaging system based on sparse sampling.
  • the present application provides a holographic magnetic induction chest imaging method based on sparse sampling, which includes the following steps:
  • a holographic magnetic induction chest imaging system which includes a main control module, a magnetic field signal generation module, a switching module, a transceiver array module, an image processing module, and a display module; the transceiver array module adopts a transceiver integrated coil;
  • the main control module controls the transceiver array module to switch to the transmitting state through the switching module, and controls the magnetic field signal generation module to generate a single-frequency radio frequency signal.
  • the generated single-frequency radio frequency signal is transmitted to the transceiver array module through the switching module in the form of a sinusoidal alternating current.
  • the sinusoidal alternating current on the transceiver array module generates an excitation magnetic field in the target area, the excitation magnetic field generates eddy currents around the target area, and the eddy currents generate a scattered magnetic field;
  • the main control module controls the transceiver array module to switch to the receiving state through the switch module, and controls at least three coils to perform sparse sampling in a random scanning mode, and at the same time measure the scattered magnetic field signal from the chest cavity, and transmit the measured scattered magnetic field signal to Image processing module;
  • the image processing module sequentially compares the received scattered magnetic field signals measured by different coils, and combines the compressed sensing method to process the compared visible scattered magnetic field functions to obtain a reconstructed chest cavity image.
  • the reconstructed thoracic cavity image is transmitted to the image display module for display.
  • the eddy current is calculated by calculating the magnetic potential vector Obtain
  • ⁇ is the magnetic permeability
  • ⁇ is the angular frequency
  • ⁇ 2 ⁇ f
  • f is the transmission frequency of the signal
  • ⁇ is the conductivity
  • J s is the current density of the excitation coil.
  • M represents the magnetic field strength.
  • the image processing module sequentially compares the received scattered magnetic field signals measured by different coils, and combines the compressed sensing method to process the compared visible scattered magnetic field functions,
  • the specific process of obtaining the reconstructed chest cavity image is:
  • Modeling the thoracic cavity which includes establishing an electromagnetic model of the thoracic cavity and a nonlinear observation model between the electromagnetic properties of the thoracic cavity and the scattered magnetic field, wherein the nonlinear observation model includes an internal magnetic field model and an external magnetic field model;
  • the internal magnetic field model and the external magnetic field model are used to reconstruct the image of the thoracic cavity in the target area.
  • the electromagnetic model of the chest cavity is:
  • j is the complex imaginary part
  • f is the working frequency of the imaging system
  • ⁇ 0 is the permeability of the free space
  • ⁇ is the conductivity of the thoracic cavity tissue
  • ⁇ 0 is the dielectric constant of the free space
  • ⁇ r is the pleural cavity tissue Permittivity
  • ⁇ r ⁇ ′ r -j ⁇ / ⁇ 0
  • ⁇ ′ r is the real part of the relative permittivity of the thoracic cavity
  • G is the Green's function
  • Is the position vector from the field source point to the scattered magnetic field Is the position vector from the field source point to any point in the chest cavity
  • k 0 is the wave number in free space
  • Is the magnetic current density Is the magnetic current density
  • ⁇ r is the permeability of the thoracic cavity tissue
  • Is the induced current density Is the total electric field
  • Is the scattered magnetic field Is the unit vector from the source point of the field to any point in the field
  • R is the distance from the source point of the field to any point in the scattering field
  • the corresponding change value and curve are extracted from the established nonlinear observation model, and the two-dimensional image of the chest cavity is reconstructed according to the change value;
  • the total visible scattered magnetic field function R of the N coils is the sum of the visible scattered magnetic field functions of N(N-1) coils:
  • the process of performing signal processing based on the iterative algorithm of compressed sensing technology on the total visible scattered magnetic field detected by the at least three coils arranged in a non-uniform random arrangement is:
  • ⁇ is the weight of l 1 norm consistency
  • ⁇ is the weight of l TV norm consistency
  • l 1 norm represents the sum of the absolute values of the elements of the vector
  • L 2 norm represents the sum of the squares of each element of the vector and then find the square root
  • l TV norm represents the total variation of each element of the vector
  • TV represents the two-dimensional isotropic operator
  • ⁇ represents the non-subsampled non Uniform random k-space data
  • A represents the acquired measurement matrix reflecting the under-sampling data, and is a sparse matrix that transforms the image into a sparse representation
  • ⁇ represents the accuracy
  • ⁇ represents the observation matrix
  • the measurement matrix A is:
  • U represents a binary matrix, used for random location selection under random sampling
  • I represents the visibility intensity function of the target area, Represents the two-dimensional inverse Fourier transform.
  • the reconstructed image of the thoracic cavity is:
  • R represents the total visible scattering magnetic field function
  • ⁇ is the angle between the line of the origin O and any point P in space and the positive z axis
  • ⁇ is the difference between the xoz plane and the half-plane passing through any point P in space If point P is on the z-axis, the angle ⁇ is uncertain; ⁇ b represents the operating wavelength, v b represents the background speed, and f represents the operating frequency.
  • the present application also provides a holographic magnetic induction chest imaging based on sparse sampling, which includes a main control module, a magnetic field signal generation module, a switching module, a transceiver array module, an image processing module, and a display Module; the transceiver array module adopts a coil integrated with the transceiver;
  • the main control module is used to control the magnetic field signal generating module to transmit radio frequency signals, and is also used to control the transceiver array module to switch between the transmitting state and the receiving state through the switching module;
  • the transceiver array module When the transceiver array module is in the transmitting state, it is used to generate an alternating magnetic field according to the alternating current, so that the human body in the target area generates electromagnetic scattering signals in an electromagnetic field environment; when the transceiver array module is in the receiving state, it is used to Measure the electromagnetic scattering signal of the target area;
  • the image processing module is used to reconstruct the thoracic cavity image according to the received electromagnetic scattering signal in combination with the compressed sensing method, and transmit the reconstructed thoracic cavity image to the display module for display;
  • the transceiver array module includes N transceiving integrated coils, where N is a natural number and N ⁇ 3;
  • the transmitting and receiving module adopts a two-dimensional microwave antenna array, and the two-dimensional microwave antenna array includes N transceiver integrated microwave antennas, where N is a natural number and N ⁇ 3;
  • the frequency of the radio frequency signal is 900 Hz-20 GHz.
  • the holographic magnetic induction chest imaging system based on sparse sampling in the present application uses a magnetic field signal generating module to continuously generate a single-frequency radio frequency signal, and the single-frequency radio frequency signal is a sinusoidal alternating current.
  • the transceiver array module transmits sinusoidal alternating current to the target area in the detection bed to generate an excitation magnetic field, which generates eddy currents around the target area, and the eddy current generates a scattered magnetic field.
  • the transceiver array module is in the receiving state Measure the scattered magnetic field generated by the target area, the image processing module reconstructs the chest cavity image based on the received scattered magnetic field data, combined with the compressed sensing method, and displays the reconstructed chest cavity image.
  • This application is based on sparse sampling holographic magnetic induction chest imaging
  • the system applies methods based on compressed sensing and magnetic induction imaging to the specific problem of thoracic imaging to detect tumors, which can quickly image the thoracic cavity and automatically detect lung tumors.
  • the holographic magnetic induction chest imaging method based on sparse sampling in this application uses the scattered magnetic field signals collected by the transceiver integrated coils non-uniformly randomly distributed around the chest cavity to perform pairwise comparison of the visible scattered magnetic field function, combined with compressed sensing technology for signal processing to re- This method can obtain high-quality and clear images more quickly through less sampling data, thereby greatly reducing imaging costs and time, and improving image quality.
  • This application is widely used in the fields of non-destructive testing, medical imaging and target detection.
  • FIG. 1 is a block diagram of a holographic magnetic induction chest imaging system based on sparse sampling according to an embodiment of the application.
  • FIG. 2 is a schematic diagram of an initial working state of a holographic magnetic induction chest imaging system based on sparse sampling according to an embodiment of the application.
  • FIG. 3 is a schematic diagram of a non-uniform random distribution arrangement of 16 coils in a holographic magnetic induction chest imaging system based on sparse sampling according to an embodiment of the application.
  • FIG. 4 is a schematic diagram of the geometric arrangement of two of the at least three coils of a holographic magnetic induction chest imaging system based on sparse sampling according to an embodiment of the application.
  • FIG. 5 is a flowchart of a holographic magnetic induction chest imaging method based on sparse sampling according to an embodiment of the application.
  • Figure 6(a) is the real part of the image of the chest cavity model to be reconstructed
  • Figure 6(b) is the imaginary part of the image of the chest cavity model to be reconstructed.
  • Figure 7(a) is the real part of the two-dimensional reconstructed chest image obtained by the imaging method of the present application.
  • Fig. 7(b) is the imaginary part of the two-dimensional reconstructed thoracic cavity image obtained by the imaging method of this application.
  • Magnetic induction imaging performs image reconstruction through the detection of the electromagnetic field distribution inside and around the target organism under the action of an excitation magnetic field to obtain the dielectric constant distribution, electrical conductivity distribution, temperature distribution and blood oxygen content of certain biological tissues.
  • bioimaging and diagnosis such as thoracic imaging to detect lung tumors, etc.
  • this application provides a holographic magnetic induction chest imaging system based on sparse sampling, which includes a main control module 1, a magnetic field signal generation module 2, a switching module 3, a transceiver array module 4, and image processing. Module 5 and display module 6.
  • the main control module 1 is connected to the magnetic field signal generating module 2 and the switching module 3, the magnetic field signal generating module 2 is connected to the transceiver array module 4 through the switching module 3, and the transceiver array module 4 is connected to the image processing module 5 through the switching module 3.
  • the processing module 5 is connected to the display module 6.
  • the magnetic field signal generating module 2 adopts a vector network analyzer, which can generate a radio frequency signal with a frequency of 900 Hz-20 GHz.
  • the switching module 3 adopts a multi-channel switch circuit board.
  • the transceiver array module 4 includes N transceiving integrated coils 41, where N is a natural number and N ⁇ 3.
  • N is a natural number and N ⁇ 3.
  • One or more of solenoid coils, Helmholtz coils and patch coils can be used for the integrated transmitting and receiving coils.
  • the integrated transceiver coil can be used as a signal transmitter or as a signal detector.
  • the gaps between the coils of the transceivers are filled with dielectric materials, where the dielectric constant of the dielectric material is the same or that of the human lung tissue. approximate.
  • the number of coils 41 integrated with the transceiver are set to 16, centered on the thoracic cavity of the human body, the 16 coils are non-uniformly randomly distributed around the thoracic cavity and are located in the same plane.
  • Each coil acts as a signal transmitter to transmit radio frequency signals to the chest cavity, and at the same time as a signal detector to collect scattered magnetic field signals from the chest cavity.
  • the coil 41 integrated with the transceiver is a solenoid coil.
  • the holographic magnetic induction chest imaging system based on sparse sampling proposed in this application is also provided with a detection bed 7.
  • the human body is located on the detection bed 7, so that the chest cavity is in the target area on the detection bed 7.
  • N transmitter-receiver integrated coils surround the thoracic cavity, and are distributed non-uniformly and randomly on the same plane.
  • a target area is set on the detection bed 7, and the human lungs are located in the target area during detection.
  • the transmitter-receiver integrated coil sequentially transmits radio frequency signals to the target area in the form of sparse sampling.
  • the transmitter-receiver integrated coil also sequentially measures the magnetic field changes and the distribution status of the dielectric constant, temperature, and conductivity of the target area in the form of sparse sampling.
  • the working frequency of the holographic magnetic induction chest imaging system based on sparse sampling in this application is a single frequency, and its optimal working frequency range is 10 MHz.
  • the holographic magnetic induction chest imaging system based on sparse sampling proposed in the present application further includes a storage module, which is connected to the image processing module 5 and is used to store original chest image data and reconstructed chest image data.
  • the main control module 1 controls the magnetic field signal generating module 2 to continuously generate magnetic field signals, and switches the transceiver array module 4 to the transmitting state through the switching module 3, and the magnetic field signals are
  • the form of alternating current is applied to the transceiver array module 4, which scans the human chest in the target area in a horizontal direction.
  • the alternating current generates an alternating magnetic field, and the human body generates electromagnetic scattering signals in an electromagnetic field environment.
  • the main control module 1 switches the transceiver array module 4 to the receiving state through the switching module 3, and the transceiver array module 4 measures electromagnetic scattering signals and transmits the measured signals to the image processing module 5.
  • the image processing module 5 reconstructs the thoracic cavity image according to the received electromagnetic scattering signal in combination with the compressed sensing method, and transmits the reconstructed thoracic cavity image to the display module 6 for display.
  • the present application also provides a holographic magnetic induction chest imaging method based on sparse sampling, which includes the following steps:
  • a holographic magnetic induction chest imaging system which includes a main control module 1, a magnetic field signal generation module 2, a switching module 3, a transceiver array module 4, an image processing module 5 and a display module 6.
  • the transceiver array module 4 adopts a coil integrating transceiver.
  • the main control module 1 controls the magnetic field signal generation module 2 to continuously generate a single-frequency radio frequency signal.
  • the radio frequency signal is transmitted to the transceiver array module 4 in the form of a sinusoidal alternating current.
  • the sinusoidal alternating current generates excitation in the target area of the detection bed 7 Magnetic field, the excitation magnetic field generates eddy currents around the target area, and the eddy currents generate scattered magnetic fields.
  • the main control module 1 controls at least three coils to perform sparse sampling in a random scanning mode, simultaneously measures the scattered magnetic field signals from the chest cavity, and transmits the measured scattered magnetic field signals to the image processing module 5.
  • the image processing module 5 sequentially compares the received scattered magnetic field signals measured by different coils, and combines the compressed sensing method to process the compared visible scattered magnetic field functions to obtain a reconstructed chest cavity image.
  • step S2 in the process of transmitting the radio frequency signal to the target area:
  • S21 Establish a rectangular coordinate system of the target area where the thoracic cavity is located, and determine the distance between the thoracic cavity and the coil, the position coordinates of the coil, and the number of image sampling points.
  • the distance from the coil to the target area is far greater than a working wavelength (d ⁇ b ), which belongs to the far field.
  • At least three coils simultaneously apply uninterrupted sinusoidal alternating current to the chest cavity.
  • the sinusoidal alternating current generates an excitation magnetic field near the chest cavity.
  • the excitation magnetic field can be regarded as a time harmonic electromagnetic field. When the excitation magnetic field passes through the chest cavity, it is caused by electromagnetic induction. The effect produces eddy currents.
  • the center of the transceiver array module 4 is taken as the origin, the direction perpendicular to the paper surface is the X-axis direction, the horizontal right direction is the Y-axis direction, and the vertical upward direction is the Z-axis.
  • Direction establish a rectangular coordinate system OXYZ, then the coordinates of a point P on the object in the area to be detected are P(x,y,z), and the plane passing through the P point and parallel to the plane OXY is the plane npq.
  • the point A i (x i , y i , z i ) and the point A j (x j , y j , z j ) are both provided with a transmitter-receiver integrated coil.
  • the eddy current is calculated by calculating the magnetic potential vector Obtain,
  • ⁇ is the magnetic permeability
  • ⁇ is the angular frequency
  • ⁇ 2 ⁇ f
  • f is the transmission frequency of the signal
  • ⁇ is the conductivity
  • J s is the current density of the excitation coil.
  • M represents the magnetic field strength.
  • step S4 the specific process for the image processing module 5 to reconstruct the thoracic cavity image is as follows:
  • j is the complex imaginary part
  • f is the working frequency of the imaging system
  • ⁇ 0 is the permeability of the free space
  • ⁇ is the conductivity of the thoracic cavity tissue
  • ⁇ 0 is the dielectric constant of the free space
  • ⁇ r is the pleural cavity tissue Permittivity
  • ⁇ r ⁇ ′ r -j ⁇ / ⁇ 0
  • ⁇ ′ r is the real part of the relative permittivity of the thoracic cavity
  • the nonlinear observation model includes an internal magnetic field model and an external magnetic field model.
  • the internal magnetic field model is:
  • Is the incident magnetic field G is the Green's function
  • Is the position vector from the field source point to the scattered magnetic field Is the position vector from the field source point to any point in the chest cavity
  • k 0 is the wave number in free space
  • Is the magnetic current density Is the magnetic current density
  • ⁇ r is the permeability of the thoracic cavity tissue
  • Is the induced current density Is the total electric field
  • Is the scattered magnetic field Is the unit vector from the source point of the field to any point in the field
  • R is the distance from the source point of the field to any point in the scattering field.
  • S421 sequentially compare the scattered magnetic fields detected by any two of the at least three integrated transceiver coils
  • Means located at with The visible scattered magnetic field of the two coils at which contains phase delay and/or amplitude difference information, Represents the distance vector from any point in the target area to the i-th coil, Represents the distance vector from any point in the target area to the j-th coil, Means located at The scattered magnetic field detected by the coil at Means located at The conjugate of the scattered magnetic field measured by the coil at the position, ⁇ > represents the average time.
  • S422 Obtain information that can reflect the amplitude and phase of the electromagnetic property distribution of the thoracic cavity tissue according to the differences obtained by the pairwise comparison in turn.
  • S423 According to the continuously detected scattered magnetic field distribution information, extract corresponding change values and curves from the nonlinear observation model established by using the MATLAB platform or other computer languages, and reconstruct the two-dimensional image of the thoracic cavity according to the changed values.
  • S424 Perform signal processing based on the compressed sensing technology of the iterative algorithm on the total visible scattered magnetic field detected by the at least three non-uniformly randomly arranged coils to obtain compressed sensing processed signals:
  • ⁇ is the weight of the consistency of l 1 norm
  • ⁇ is the weight of the consistency of l TV norm.
  • l 1 norm represents the sum of the absolute values of each element of a vector
  • l 2 norm represents the sum of the squares of each element of the vector and then find the square root
  • l TV norm represents the total variation of each element of the vector
  • TV represents two-dimensional Isotropic operator
  • ⁇ represents the under-sampled non-uniform random k-space data
  • A represents the acquired measurement matrix reflecting the under-sampled data, and is a sparse matrix that transforms the image into a sparse representation.
  • ⁇ represents accuracy
  • ⁇ represents the observation matrix.
  • the measurement matrix A is:
  • U represents a binary matrix, which is used to select random locations under random sampling
  • I represents the visibility intensity function of the target area, Represents the two-dimensional inverse Fourier transform.
  • R represents the total visible scattering magnetic field function
  • R represents the unit vectors along the x, y, and z axes in the positive space-time direction, Respectively for any coil
  • ⁇ is the angle between the line of the origin O and any point P in space and the positive z axis
  • ⁇ is the difference between the xoz plane and the half-plane passing through any point P in space If point P is on the z-axis, the angle ⁇ is uncertain
  • ⁇ b represents the operating wavelength
  • v b represents the background speed
  • f represents the operating frequency.
  • Equation (10) and (11) p represents the number of iterations, The regularization parameters that determine the trade-off between measurement consistency and sparsity in area A and the finite difference domain are given. Represents the Hamiltonian operator,
  • Equation (12) Represents the Hamiltonian in the x direction
  • d x , d y , c x , c y and c w are auxiliary variables.
  • a H represents the conjugate transpose of A
  • ⁇ H represents the conjugate transpose of ⁇
  • step S3 at least three coils are used to continuously transmit single-frequency radio frequency signals to the chest cavity; in the above step S4, at least three coils are used to measure the scattered magnetic field around the chest cavity, wherein the distance between all coils and the target area is much greater than A working wavelength (d ⁇ b ) belongs to the far field.
  • a time series of at least one electromagnetic property of the thoracic cavity is formed based on the scattered magnetic fields measured by at least two of the at least three coils, and the difference in the scattered magnetic fields measured by the at least two coils is calculated to reproduce Construct an image of the chest cavity.
  • FIG. 6 is a diagram of a chest cavity model, in which Figure 6(a) is the real part of the chest cavity model to be reconstructed, showing the distribution of dielectric constant, and Figure 6(b) is the imaginary part of the chest cavity model to be reconstructed , manifested as the distribution of electrical conductivity.
  • Figure 7 is a two-dimensional reconstructed chest diagram obtained by the imaging method of this application, wherein Figure 7(a) is the real part of the two-dimensional reconstructed chest diagram obtained by the imaging method of this application, and Figure 7(b) is the real part Apply for the imaginary part of the two-dimensional reconstruction of the chest image obtained by the imaging method.
  • the experimental results show that the two-dimensional reconstruction of the chest image under the radio frequency signal with a frequency of 900Hz-20GHz can clearly show the different tissues of the chest cavity, which contains tumor cells.
  • this application has the following advantages: fast imaging, can significantly reduce imaging costs, increase scanning speed, model robustness, can achieve rapid imaging of the chest cavity and automatic detection of lung tumors, which will be based on compression Perception and magnetic induction imaging methods are applied to the specific problem of thoracic imaging to detect tumors, which can effectively improve image quality and imaging speed.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Radiology & Medical Imaging (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Pathology (AREA)
  • Engineering & Computer Science (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Physics & Mathematics (AREA)
  • Biophysics (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Magnetic Resonance Imaging Apparatus (AREA)
  • Measurement And Recording Of Electrical Phenomena And Electrical Characteristics Of The Living Body (AREA)
  • Image Processing (AREA)

Abstract

一种基于稀疏采样的全息磁感应胸腔成像方法及成像系统，其包括：设置一基于稀疏采样的全息磁感应胸腔成像系统；成像系统中的主控模块（1）通过切换模块（3）控制收发器阵列模块（4）切换发射和接收状态，并控制磁场信号发生模块（2）产生单频射频信号，单频射频信号以正弦交变电流的形式通过切换模块（3）传输至收发器阵列模块（4）上，在目标区域周围产生涡流，涡流产生散射磁场；主控模块（1）控制至少三个线圈（41）以随机的扫描模式进行稀疏采样，同时测量来自胸腔的散射磁场信号，并将散射磁场信号传输至图像处理模块（5）；图像处理模块（5）对散射磁场信号依次进行两两比较，并结合压缩感知方法对比较所得的可见散射磁场函数进行处理，得到重构的胸腔图像，并进行显示。

Description

基于稀疏采样的全息磁感应胸腔成像方法及成像系统 技术领域
本申请属于磁感应成像技术领域，具体涉及一种基于稀疏采样的全息磁感应胸腔成像方法及成像系统。
背景技术
肺癌、慢性阻塞性肺疾病等肺部重大疾病严重威胁人类健康，且随空气污染的加剧而日益严重。MRI是一种重要的临床医学影像学技术，与胸透、CT和PET等技术相比具有无放射性的优点。然而，肺部大部分是空腔组织，导致肺部成为常规MRI的盲区。众所周知，不同肺疾病的生理学和病理生理学特征会导致不同的通气模式，通过观测肺的通气模式能够有效地解释肺通气缺陷的病因。然而，现有技术难以精准刻画肺部通气的动态过程。因此，亟需发展对肺部通气的动态可视化的新技术。
磁感应成像(Electromagnetic Induction Imaging，EIT)是一种在交变磁场激励下，以被测物体内电导率为目标的非接触式成像方法。由于EIT具有非侵入、非接触、价格便宜、携带方便等优点，因此EIT在生物医学成像领域中具有很大的应用价值。然而，当前磁感应断层成像技术面临着成像快速慢、噪音干扰严重、算法不完备、不能三维成像等问题，导致其临床推广的进展缓慢。
采用压缩感知能够通过远低于Nyquist采样率的测量数据对可压缩信号进行精确地恢复。压缩感知作为一个新的采样理论，其优势是能凭借被测信号的稀疏性大幅度压缩提取该信号所必需的测量数。该理论一经提出，即在图像处理、医疗成像、模式识别、地质勘探、光学/雷达成像、无线通信等多个信号处理领域得到应用，并被美国科技评论评为2007年度十大科技进展。尤其在各种成像应用中，利用压缩感知技术可以用极少的线性测量得到无模 糊的目标图像，从而降低成像系统的测量消耗和系统复杂度。
发明内容
为至少在一定程度上克服相关技术中存在的问题，本申请提供了一种基于稀疏采样的全息磁感应胸腔成像方法及成像系统。
根据本申请实施例的第一方面，本申请提供了一种基于稀疏采样的全息磁感应胸腔成像方法，其包括以下步骤：
设置一基于稀疏采样的全息磁感应胸腔成像系统，其包括主控模块、磁场信号发生模块、切换模块、收发器阵列模块、图像处理模块和显示模块；所述收发器阵列模块采用收发一体的线圈；
主控模块通过切换模块控制收发器阵列模块切换至发射状态，并控制磁场信号发生模块产生单频射频信号，产生的单频射频信号以正弦交变电流的形式通过切换模块传输至收发器阵列模块上，收发器阵列模块上的正弦交变电流在目标区域产生激励磁场，激励磁场在目标区域周围产生涡流，涡流产生散射磁场；
主控模块通过切换模块控制收发器阵列模块切换至接收状态，并控制至少三个线圈以随机的扫描模式进行稀疏采样，同时测量来自胸腔的散射磁场信号，并将测量到的散射磁场信号传输至图像处理模块；
图像处理模块对接收到的不同线圈测量的散射磁场信号依次进行两两比较，并结合压缩感知方法对比较所得的可见散射磁场函数进行处理，得到重构的胸腔图像。
将重构的胸腔图像传输至图像显示模块进行显示。
上述基于稀疏采样的全息磁感应胸腔成像方法中，所述涡流通过计算磁势矢量
Figure PCTCN2019119937-appb-000001
获取，
Figure PCTCN2019119937-appb-000002
式中，μ为磁导率，ω为角频率，ω＝2πf，f为信号的发射频率，σ为电导率，J s为激励线圈的电流密度。
Figure PCTCN2019119937-appb-000003
表示哈密尔顿算子，M表示磁场强度。
上述基于稀疏采样的全息磁感应胸腔成像方法中，所述图像处理模块对 接收到的不同线圈测量的散射磁场信号依次进行两两比较，并结合压缩感知方法对比较所得的可见散射磁场函数进行处理，得到重构的胸腔图像的具体过程为：
对胸腔进行建模，其包括建立胸腔的电磁模型以及胸腔的电磁属性和散射磁场之间的非线性观测模型，其中，非线性观测模型包括内部磁场模型和外部磁场模型；
利用内部磁场模型和外部磁场模型重构目标区域中胸腔的图像。
进一步地，所述胸腔的电磁模型为：
Figure PCTCN2019119937-appb-000004
式中，j为复数虚部，
Figure PCTCN2019119937-appb-000005
ω＝2πf为工作角频率，f为成像系统的工作频率，μ 0为自由空间的磁导率，σ为胸腔组织的电导率，ε 0为自由空间的介电常数，ε r为胸腔组织的介电常数，ε r＝ε′ r-jσ/ωε 0，ε′ r为胸腔组织相对介电常数的实部，
Figure PCTCN2019119937-appb-000006
为总磁场，
Figure PCTCN2019119937-appb-000007
进一步地，所述内部磁场模型为：
Figure PCTCN2019119937-appb-000008
式中，
Figure PCTCN2019119937-appb-000009
为入射磁场，G为格林函数，
Figure PCTCN2019119937-appb-000010
为从场源点到散射磁场的位置矢量，
Figure PCTCN2019119937-appb-000011
为从场源点到胸腔内任意一点的位置矢量，k 0为自由空间的波数，
Figure PCTCN2019119937-appb-000012
为磁电流密度，
Figure PCTCN2019119937-appb-000013
μ r为胸腔组织的磁导率，
Figure PCTCN2019119937-appb-000014
为感应电流密度，
Figure PCTCN2019119937-appb-000015
为总电场，
Figure PCTCN2019119937-appb-000016
外部磁场模型为：
Figure PCTCN2019119937-appb-000017
式中，
Figure PCTCN2019119937-appb-000018
为散射磁场，
Figure PCTCN2019119937-appb-000019
为从场源点到场域内任一点的单位向量，
Figure PCTCN2019119937-appb-000020
R为从场 源点到散射场内任意一点的距离；
Figure PCTCN2019119937-appb-000021
进一步地，所述利用内部磁场模型和外部磁场模型重构目标区域中胸腔的图像的过程为：
依次对至少三个收发一体的线圈中的任意两个线圈所探测到的散射磁场进行比较；
依次根据两两比较得到的差异获得能够反映胸腔组织电磁属性分布的幅值和相位的信息；
根据连续探测到的散射磁场分布信息，从建立的非线性观测模型中提取出相应的变化数值和曲线，并根据变化数值重建胸腔二维图像；
对非均匀随机排布的至少三个线圈探测到的总可见散射磁场进行基于迭代算法的压缩感知技术的信号处理，获取压缩感知处理后的信号；
通过对非均匀排布的所有线圈获得的总可见散射磁场函数进行处理，得到胸腔重构图像。
更进一步地，所述N个线圈的总可见散射磁场函数R为N(N-1)个线圈的可见散射磁场函数之和：
Figure PCTCN2019119937-appb-000022
其中，
Figure PCTCN2019119937-appb-000023
表示位于
Figure PCTCN2019119937-appb-000024
和
Figure PCTCN2019119937-appb-000025
处的两个线圈的可见散射磁场，
Figure PCTCN2019119937-appb-000026
其包含相位延迟和/或振幅差异信息；
Figure PCTCN2019119937-appb-000027
表示目标区域中任意点到第i个线圈的距离矢量，
Figure PCTCN2019119937-appb-000028
表示目标区域中任意点到第j个线圈的距离矢量，
Figure PCTCN2019119937-appb-000029
表示位于
Figure PCTCN2019119937-appb-000030
处的线圈探测到的散射磁场，
Figure PCTCN2019119937-appb-000031
表示位于
Figure PCTCN2019119937-appb-000032
处的线圈测量到的散射磁场的共轭，<>表示平均时间。
更进一步地，所述对非均匀随机排布的至少三个线圈探测到的总可见散射磁场进行基于迭代算法的压缩感知技术的信号处理的过程为：
Figure PCTCN2019119937-appb-000033
服从于
Figure PCTCN2019119937-appb-000034
式中，
Figure PCTCN2019119937-appb-000035
是基于非均匀随机分布的线圈阵列的全息磁感应图像，α为l 1范数一致性的权重，β为l TV范数一致性的权重；l 1范数表示向量各个元素绝对值之和，l 2范数表示向量各个元素的平方求和然后求平方根，l TV范数表示向量各个元素的总变异量，|| || TV表示二维各向同性算子，γ表示欠采样的非均匀随机k空间数据，A表示获取的反映欠采样数据的测量矩阵，并且是将图像变换为稀疏表示的稀疏矩阵；δ表示精度，Ψ表示观测矩阵；
其中，测量矩阵A为：
Figure PCTCN2019119937-appb-000036
式中，U表示二值矩阵，用于随机抽样下随机位置的选择；I表示目标区域的可见度强度函数，
Figure PCTCN2019119937-appb-000037
表示二维傅里叶逆变换。
更进一步地，所述胸腔重构图像为：
Figure PCTCN2019119937-appb-000038
式中，R表示总可见散射磁场函数，
Figure PCTCN2019119937-appb-000039
Figure PCTCN2019119937-appb-000040
Figure PCTCN2019119937-appb-000041
分别为沿x，y，z轴正时空方向的单位矢量,
Figure PCTCN2019119937-appb-000042
分别为任一线圈
Figure PCTCN2019119937-appb-000043
在直角坐标系中沿x，y，z轴的位置，
Figure PCTCN2019119937-appb-000044
分别为任一线圈
Figure PCTCN2019119937-appb-000045
在直角坐标系中沿x，y，z轴的位置，θ是原点O和空间任一点P的连线与正向z轴的夹角，φ为xoz平面与通过空间任一点P的半平面之间的夹角，若P点在z轴上则φ角是不确定的；λ b表示工作波长，v b表示背景的速度，f表示工作频率。
根据本申请实施例的第二方面，本申请还提供了一种基于稀疏采样的全息磁感应胸腔成像，其包括主控模块、磁场信号发生模块、切换模块、收发器阵列模块、图像处理模块和显示模块；所述收发器阵列模块采用收发一体的线圈；
所述主控模块用于控制所述磁场信号发生模块发射射频信号，还用于通过所述切换模块控制收发器阵列模块在发射状态和接收状态间进行切换；
所述收发器阵列模块在发射状态时，用于根据交变电流产生交变磁场，使目标区域的人体在电磁场环境下产生电磁散射信号；所述收发器阵列模块在接收状态时，用于对目标区域的电磁散射信号进行测量；
所述图像处理模块用于根据接收到的电磁散射信号，结合压缩感知方法对胸腔图像进行重构处理，并将重构的胸腔图像传输至显示模块进行显示；
根据接收到的散射电场数据进行脑图像重构，并将重构的脑图像传输至显示模块进行显示；
所述收发器阵列模块包括N个收发一体的线圈，其中，N为自然数且N≥3；
所述发射接收模块采用二维微波天线阵列，所述二维微波天线阵列包括N个收发一体的微波天线，其中，N为自然数且N≥3；
所述射频信号的频率为900Hz-20GHz。
根据本申请的上述具体实施方式可知，至少具有以下有益效果：本申请基于稀疏采样的全息磁感应胸腔成像系统利用磁场信号发生模块不间断地产生单频射频信号，单频射频信号以正弦交变电流的形式传输到收发器阵列模块；收发器阵列模块传输正弦交变电流到检测床中的目标区产生激励磁场，激励磁场使目标区域周围产生涡流，涡流产生散射磁场，收发器阵列模块在接收状态测量目标区域产生的散射磁场，图像处理模块根据接收到的散射磁场数据，结合压缩感知方法对胸腔图像进行重构，并对重构的胸腔图像进行显示，本申请基于稀疏采样的全息磁感应胸腔成像系统将基于压缩感知和磁感应成像的方法应用到胸腔成像检测肿瘤这一具体问题上，能够对胸腔进行快速成像，并对肺肿瘤进行自动检测。
本申请基于稀疏采样的全息磁感应胸腔成像方法利用非均匀随机分布在胸腔周围的收发一体的线圈采集的散射磁场信号进行两两对比获的可见散射磁场函数，结合压缩感知技术用于信号处理从而重构胸腔图像，该方法能够通过更少的采样数据、更快速地获得优质、清晰的图像，从而大幅减少成像成本和时间，提高图像质量。本申请广泛应用于无损检测、医学成像和 目标探测等领域。
应了解的是，上述一般描述及以下具体实施方式仅为示例性及阐释性的，其并不能限制本申请所欲主张的范围。
附图说明
下面的所附附图是本申请的说明书的一部分，其示出了本申请的实施例，所附附图与说明书的描述一起用来说明本申请的原理。
图1为本申请实施例提供的一种基于稀疏采样的全息磁感应胸腔成像系统的框图。
图2为本申请实施例提供的一种基于稀疏采样的全息磁感应胸腔成像系统的初始工作状态示意图。
图3为本申请实施例提供的一种基于稀疏采样的全息磁感应胸腔成像系统中16个线圈的非均匀随机分布排列的示意图。
图4为本申请实施例提供的一种基于稀疏采样的全息磁感应胸腔成像系统的至少三个线圈中的两个线圈的几何排列示意图。
图5为本申请实施例提供的一种基于稀疏采样的全息磁感应胸腔成像方法的流程图。
图6(a)为待重构的胸腔模型图像的实部；
图6(b)为待重构的胸腔模型图像的虚部。
图7(a)为采用本申请成像方法得到的二维重构胸腔图像的实部；
图7(b)为采用本申请成像方法得到的二维重构胸腔图像的虚部。
附图标记说明：
1、主控模块；2、磁场信号发生模块；3、切换模块；4、收发器阵列模块；41、线圈；5、图像处理模块；6、显示模块；7、检测床。
具体实施方式
为使本申请实施例的目的、技术方案和优点更加清楚明白，下面将以附图及详细叙述清楚说明本申请所揭示内容的精神，任何所属技术领域技术人员在了解本申请内容的实施例后，当可由本申请内容所教示的技术，加以改 变及修饰，其并不脱离本申请内容的精神与范围。
本申请的示意性实施例及其说明用于解释本申请，但并不作为对本申请的限定。另外，在附图及实施方式中所使用相同或类似标号的元件/构件是用来代表相同或类似部分。
关于本文中所使用的“第一”、“第二”、…等，并非特别指称次序或顺位的意思，也非用以限定本申请，其仅为了区别以相同技术用语描述的元件或操作。
关于本文中所使用的方向用语，例如：上、下、左、右、前或后等，仅是参考附图的方向。因此，使用的方向用语是用来说明并非用来限制本创作。
关于本文中所使用的“包含”、“包括”、“具有”、“含有”等等，均为开放性的用语，即意指包含但不限于。
关于本文中所使用的“及/或”，包括所述事物的任一或全部组合。
关于本文中的“多个”包括“两个”及“两个以上”；关于本文中的“多组”包括“两组”及“两组以上”。
关于本文中所使用的用语“大致”、“约”等，用以修饰任何可以细微变化的数量或误差，但这些微变化或误差并不会改变其本质。一般而言，此类用语所修饰的细微变化或误差的范围在部分实施例中可为20％，在部分实施例中可为10％，在部分实施例中可为5％或是其他数值。本领域技术人员应当了解，前述提及的数值可依实际需求而调整，并不以此为限。
某些用以描述本申请的用词将于下或在此说明书的别处讨论，以提供本领域技术人员在有关本申请的描述上额外的引导。
不同类型生物组织的介电常数和电导率差异明显，此差异能够为磁感应成像检测活体生物组织的生理病理状态提供可行的物理基础。磁感应成像通过对激励磁场作用下目标生物体内部和周围电磁场分布的探测进行图像重构，获取某些生物组织的介电常数分布、电导率分布、温度分布和血液含氧量等重要特征，以便于在生物成像和诊断方面进行应用，例如进行胸腔成像检测肺肿瘤等。
下面结合附图和实施例对本申请进行详细的说明。
如图1和图2所示，本申请提供了一种基于稀疏采样的全息磁感应胸腔成像系统，其包括主控模块1、磁场信号发生模块2、切换模块3、收发器阵列模块4、图像处理模块5和显示模块6。
主控模块1与磁场信号发生模块2和切换模块3连接，磁场信号发生模块2通过切换模块3与收发器阵列模块4连接，收发器阵列模块4通过切换模块3与图像处理模块5连接，图像处理模块5与显示模块6连接。
其中，磁场信号发生模块2采用矢量网络分析仪，其可以产生频率为900Hz-20GHz的射频信号。
切换模块3采用多通道开关电路板。
收发器阵列模块4包括N个收发一体的线圈41，N为自然数且N≥3。收发一体的线圈可以采用螺线管线圈、亥姆霍兹线圈和贴片线圈中的一种或多种。收发一体的线圈可以作为信号发射器使用，也可以作为信号探测器使用。
为了减少信号耦合，提高系统的成像质量和肿瘤检测的灵敏度，各收发一体的线圈之间的间隙中填充有介质材料，其中，介质材料的介电常数与人体肺部组织的介电常数相同或近似。
在一个具体的实施例中，收发一体的线圈41设置为16个，以人体的胸腔为中心，16个线圈围绕胸腔呈非均匀随机分布，且位于同一平面内。每个线圈作为信号发射器发射射频信号到胸腔，同时作为信号探测器采集来自胸腔的散射磁场信号。具体地，收发一体的线圈41采用螺线管线圈。
如图2所示，本申请提出的基于稀疏采样的全息磁感应胸腔成像系统中还设置有检测床7，成像系统工作时，人体位于检测床7上，使得胸腔处于检测床7上的目标区域内。N个收发一体的线圈环绕于胸腔的周围，且在同一平面上呈非均匀随机分布。检测床7上设置有目标区域，检测时，人体肺部位于该目标区域。收发一体的线圈以稀疏采样的形式依次发射射频信号到目标区域，收发一体的线圈还以稀疏采样的形式依次测量目标区域的磁场变 化以及介电常数、温度、电导率的分布状态。
本申请基于稀疏采样的全息磁感应胸腔成像系统的工作频率为单频率，其最佳工作频率范围为10MHz。
本申请提出的基于稀疏采样的全息磁感应胸腔成像系统中还包括存储模块，存储模块与图像处理模块5连接，其用于存储原始的胸腔图像数据和重构的胸腔图像数据。
本申请基于稀疏采样的全息磁感应胸腔成像系统工作时，主控模块1控制磁场信号发生模块2不间断地产生磁场信号，并通过切换模块3将收发器阵列模块4切换至发射状态，磁场信号以交变电流的形式施加在收发器阵列模块4上，收发器阵列模块4对目标区域的人体胸部进行水平方向的扫描，交变电流产生交变磁场，人体在电磁场环境下产生电磁散射信号。
主控模块1通过切换模块3将收发器阵列模块4切换至接收状态，收发器阵列模块4对电磁散射信号进行测量，并将测量到的信号传输至图像处理模块5。图像处理模块5根据接收到的电磁散射信号，结合压缩感知方法对胸腔图像进行重构处理，并将重构的胸腔图像传输至显示模块6进行显示。
根据上述基于稀疏采样的全息磁感应胸腔成像系统，如图5所示，本申请还提供了一种基于稀疏采样的全息磁感应胸腔成像方法，其包括以下步骤：
S1、设置一基于稀疏采样的全息磁感应胸腔成像系统，其包括主控模块1、磁场信号发生模块2、切换模块3、收发器阵列模块4、图像处理模块5和显示模块6。其中，收发器阵列模块4采用收发一体的线圈。
S2、向目标区域发射射频信号；
主控模块1控制磁场信号发生模块2不间断地产生单频射频信号，射频信号以正弦交变电流的形式传输至收发器阵列模块4上，正弦交变电流在检测床7的目标区域产生激励磁场，激励磁场在目标区域周围产生涡流，涡流产生散射磁场。
S3、对散射磁场信号进行测量与传输；
主控模块1控制至少三个线圈以随机的扫描模式进行稀疏采样，同时测 量来自胸腔的散射磁场信号，并将测量到的散射磁场信号传输至图像处理模块5。
S4、图像处理模块5对接收到的不同线圈测量的散射磁场信号依次进行两两比较，并结合压缩感知方法对比较所得的可见散射磁场函数进行处理，得到重构的胸腔图像。
S5、将重构的胸腔图像传输至图像显示模块6进行显示。
上述步骤S2中，向目标区域发射射频信号的过程中：
S21、建立胸腔所在目标区域的直角坐标系，确定胸腔与线圈的距离、线圈的位置坐标、以及图像采样点数。
其中，线圈到目标区域的距离远远大于一个工作波长(d≥λ b)，属于远场。
S22、至少三个线圈同时向胸腔施加不间断的正弦交变电流，该正弦交变电流在胸腔附近产生激励磁场，该激励磁场可以视为一个时间谐波电磁场，激励磁场通过胸腔时因电磁感应作用产生涡流。
如图4所示，以收发器阵列模块4的中心为原点，以垂直于纸面向外的方向为X轴方向，以水平向右的方向为Y轴方向，以竖直向上的方向为Z轴方向，建立直角坐标系OXYZ，则待检测区域的物体上的一点P的坐标为P(x,y,z)，过P点且与平面OXY平行的平面为平面npq。其中，点A i(x i，y i，z i)处和点A j(x j，y j，z j)处均设置有收发一体的线圈。
涡流通过计算磁势矢量
Figure PCTCN2019119937-appb-000046
获取，
Figure PCTCN2019119937-appb-000047
式(1)中，μ为磁导率，ω为角频率，ω＝2πf，f为信号的发射频率，σ为电导率，J s为激励线圈的电流密度。
Figure PCTCN2019119937-appb-000048
表示哈密尔顿算子，M表示磁场强度。
上述步骤S4中，图像处理模块5进行胸腔图像重构的具体过程为：
S41、对胸腔进行建模；
建立胸腔的电磁模型：
Figure PCTCN2019119937-appb-000049
式(2)中，j为复数虚部，
Figure PCTCN2019119937-appb-000050
ω＝2πf为工作角频率，f为成像系统的工作频率，μ 0为自由空间的磁导率，σ为胸腔组织的电导率，ε 0为自由空间的介电常数，ε r为胸腔组织的介电常数，ε r＝ε′ r-jσ/ωε 0，ε′ r为胸腔组织相对介电常数的实部，
Figure PCTCN2019119937-appb-000051
为总磁场，
Figure PCTCN2019119937-appb-000052
建立胸腔的电磁属性和散射磁场之间的非线性观测模型，其中，非线性观测模型包括内部磁场模型和外部磁场模型。
其中，内部磁场模型为：
Figure PCTCN2019119937-appb-000053
式(3)中，
Figure PCTCN2019119937-appb-000054
为入射磁场，G为格林函数，
Figure PCTCN2019119937-appb-000055
为从场源点到散射磁场的位置矢量，
Figure PCTCN2019119937-appb-000056
为从场源点到胸腔内任意一点的位置矢量，k 0为自由空间的波数，
Figure PCTCN2019119937-appb-000057
为磁电流密度，
Figure PCTCN2019119937-appb-000058
μ r为胸腔组织的磁导率，
Figure PCTCN2019119937-appb-000059
为感应电流密度，
Figure PCTCN2019119937-appb-000060
为总电场，
Figure PCTCN2019119937-appb-000061
外部磁场模型为：
Figure PCTCN2019119937-appb-000062
式(4)中，
Figure PCTCN2019119937-appb-000063
为散射磁场，
Figure PCTCN2019119937-appb-000064
为从场源点到场域内任一点的单位向量，
Figure PCTCN2019119937-appb-000065
R为从场源点到散射场内任意一点的距离。
Figure PCTCN2019119937-appb-000066
S42、利用内部磁场模型和外部磁场模型重构目标区域中胸腔的图像；
S421、依次对至少三个收发一体的线圈中的任意两个线圈所探测到的散射磁场进行比较；
其中，任意两个线圈的可见散射磁场为：
Figure PCTCN2019119937-appb-000067
式(5)中，
Figure PCTCN2019119937-appb-000068
表示位于
Figure PCTCN2019119937-appb-000069
和
Figure PCTCN2019119937-appb-000070
处的两个线圈的可见散射磁场，其包含相位延迟和/或振幅差异信息，
Figure PCTCN2019119937-appb-000071
表示目标区域中任意点到第i个线圈的距离矢量，
Figure PCTCN2019119937-appb-000072
表示目标区域中任意点到第j个线圈的距离矢量，
Figure PCTCN2019119937-appb-000073
表示位于
Figure PCTCN2019119937-appb-000074
处的线圈探测到的散射磁场，
Figure PCTCN2019119937-appb-000075
表示位于
Figure PCTCN2019119937-appb-000076
处的线圈测量到的散射磁场的共轭，<>表示平均时间。
S422、依次根据两两比较得到的差异获得能够反映胸腔组织电磁属性分布的幅值和相位的信息。
S423、根据连续探测到的散射磁场分布信息，从利用MATLAB平台或其他计算机语言所建立的非线性观测模型中提取出相应的变化数值和曲线，并根据变化数值重建胸腔二维图像。
计算N个线圈的总可见散射磁场函数R，N为自然数且N≥3，总可见散射磁场函数R为N(N-1)个线圈的可见散射磁场函数之和：
Figure PCTCN2019119937-appb-000077
S424、对非均匀随机排布的至少三个线圈探测到的总可见散射磁场进行基于迭代算法的压缩感知技术的信号处理，获取压缩感知处理后的信号：
Figure PCTCN2019119937-appb-000078
式(7)中，
Figure PCTCN2019119937-appb-000079
是基于非均匀随机分布的线圈阵列的全息磁感应图像，α为l 1范数一致性的权重，β为l TV范数一致性的权重。l 1范数表示向量各个元素绝对值之和，l 2范数表示向量各个元素的平方求和然后求平方根，l TV范数表示向量各个元素的总变异量，|| || TV表示二维各向同性算子，γ表示欠采样的非均匀随机k空间数据，A表示获取的反映欠采样数据的测量矩阵，并且是将图像变换为稀疏表示的稀疏矩阵。δ表示精度，Ψ表示观测矩阵。
测量矩阵A为：
Figure PCTCN2019119937-appb-000080
式(8)中，U表示二值矩阵，用于随机抽样下随机位置的选择；I表示目标区域的可见度强度函数，
Figure PCTCN2019119937-appb-000081
表示二维傅里叶逆变换。
S425、通过对非均匀排布的所有线圈获得的总可见散射磁场函数进行处理，得到胸腔重构图像：
Figure PCTCN2019119937-appb-000082
式(9)中，R表示总可见散射磁场函数，
Figure PCTCN2019119937-appb-000083
Figure PCTCN2019119937-appb-000084
Figure PCTCN2019119937-appb-000085
分别为沿x，y，z轴正时空方向的单位矢量,
Figure PCTCN2019119937-appb-000086
分别为任一线圈
Figure PCTCN2019119937-appb-000087
在直角坐标系中沿x，y，z轴的位置，
Figure PCTCN2019119937-appb-000088
分别为任一线圈
Figure PCTCN2019119937-appb-000089
在直角坐标系中沿x，y，z轴的位置，θ是原点O和空间任一点P的连线与正向z轴的夹角，φ为xoz平面与通过空间任一点P的半平面之间的夹角，若P点在z轴上则φ角是不确定的；λ b表示工作波长，v b表示背景的速度，f表示工作频率。
进一步地，式(7)表示的约束问题转化为以下无约束问题：
Figure PCTCN2019119937-appb-000090
Figure PCTCN2019119937-appb-000091
式(10)和式(11)中，p表示迭代次数，
Figure PCTCN2019119937-appb-000092
给出了确定区域A和有限差分域中测量一致性和稀疏性之间的折衷的正则化参数。
Figure PCTCN2019119937-appb-000093
表示哈密尔顿算子，
Figure PCTCN2019119937-appb-000094
进一步地，在忽略p的情况下，式(7)可以简化为：
Figure PCTCN2019119937-appb-000095
式(12)中，
Figure PCTCN2019119937-appb-000096
表示x方向上的哈密尔顿算子，
Figure PCTCN2019119937-appb-000097
表示y方向上的哈密尔顿算子。d x，d y，c x，c y和c w均为辅助变量。w，d x，d y的最优值可由软阈值公式(x，t)＝max(0，1-t/|x|)x获得。其中，根据软阈值公式获得w，d x，d y的最优值，是本领域的常规计算，在此不再赘述。
进一步地，将公式
Figure PCTCN2019119937-appb-000098
Figure PCTCN2019119937-appb-000099
进一步优化为：
Figure PCTCN2019119937-appb-000100
式(13)中，A H表示A的共轭转置，Ψ H表示Ψ的共轭转置，
Figure PCTCN2019119937-appb-000101
表示x方向上的哈密尔顿算子，
Figure PCTCN2019119937-appb-000102
表示在x方向的哈密尔顿算子的导数，
Figure PCTCN2019119937-appb-000103
表示y方向上的哈密尔顿算子，
Figure PCTCN2019119937-appb-000104
表示在y方向的哈密尔顿算子的导数。
上述步骤S3中，采用至少三个线圈不间断地向胸腔发射单频射频信号；上述步骤S4中，利用至少三个线圈测量胸腔周围的散射磁场，其中，所有线圈到目标区域的距离远远大于一个工作波长(d≥λ b)，属于远场。
上述步骤S4中，基于至少三个线圈中的至少两个线圈所测量的散射磁场来形成胸腔的至少一项电磁属性的时间序列，并计算出至少两个线圈测量到的散射磁场差异，从而重构胸腔图像。
为验证本申请所提出的基于稀疏采样的全息磁感应胸腔成像方法，建立胸腔所在待成像区域的直角坐标系，确定线圈的位置坐标、胸腔与线圈的距离以及图像采样点数；通过MATLAB平台建立了仿真系统模型，以模拟当肺肿瘤发生时胸腔组织的散射磁场分布。图6为胸腔模型图，其中，图6(a)是待重构的胸腔模型图的实部，表现为介电常数的分布，图6(b)是待重构 的胸腔模型图的虚部，表现为电导率的分布。图7是采用本申请成像方法得到的二维重构胸腔图，其中，图7(a)是采用本申请成像方法得到的二维重构胸腔图的实部，图7(b)是采用本申请成像方法得到的二维重构胸腔图的虚部。
实验结果表明，在频率为900Hz-20GHz的射频信号下二维重构胸腔图能够清晰地显示胸腔部的不同组织，其中包含肿瘤细胞。
与现有的磁感应成像技术相比，本申请具有如下优点：快速成像、能够显著降低成像成本，提高扫描速度、模型鲁棒性强，能够实现对胸腔快速成像和肺肿瘤自动检测，将基于压缩感知和磁感应成像的方法应用到胸腔成像检测肿瘤这一具体问题，能够有效提高图像质量和成像速度。
以上所述仅为本申请示意性的具体实施方式，在不脱离本申请的构思和原则的前提下，任何本领域的技术人员所做出的等同变化与修改，均应属于本申请保护的范围。

Claims (10)

  1. 一种基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，包括以下步骤：
    设置一基于稀疏采样的全息磁感应胸腔成像系统，其包括主控模块、磁场信号发生模块、切换模块、收发器阵列模块、图像处理模块和显示模块；所述收发器阵列模块采用收发一体的线圈；
    主控模块通过切换模块控制收发器阵列模块切换至发射状态，并控制磁场信号发生模块产生单频射频信号，产生的单频射频信号以正弦交变电流的形式通过切换模块传输至收发器阵列模块上，收发器阵列模块上的正弦交变电流在目标区域产生激励磁场，激励磁场在目标区域周围产生涡流，涡流产生散射磁场；
    主控模块通过切换模块控制收发器阵列模块切换至接收状态，并控制至少三个线圈以随机的扫描模式进行稀疏采样，同时测量来自胸腔的散射磁场信号，并将测量到的散射磁场信号传输至图像处理模块；
    图像处理模块对接收到的不同线圈测量的散射磁场信号依次进行两两比较，并结合压缩感知方法对比较所得的可见散射磁场函数进行处理，得到重构的胸腔图像。
    将重构的胸腔图像传输至图像显示模块进行显示。
  2. 根据权利要求1所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述涡流通过计算磁势矢量
    Figure PCTCN2019119937-appb-100001
    获取，
    Figure PCTCN2019119937-appb-100002
    式中，μ为磁导率，ω为角频率，ω＝2πf，f为信号的发射频率，σ为电导率，J s为激励线圈的电流密度。
    Figure PCTCN2019119937-appb-100003
    表示哈密尔顿算子，M表示磁场强度。
  3. 根据权利要求1所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述图像处理模块对接收到的不同线圈测量的散射磁场信号依次进行两两比较，并结合压缩感知方法对比较所得的可见散射磁场函数进行处理，得到重构的胸腔图像的具体过程为：
    对胸腔进行建模，其包括建立胸腔的电磁模型以及胸腔的电磁属性和散 射磁场之间的非线性观测模型，其中，非线性观测模型包括内部磁场模型和外部磁场模型；
    利用内部磁场模型和外部磁场模型重构目标区域中胸腔的图像。
  4. 根据权利要求3所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述胸腔的电磁模型为：
    Figure PCTCN2019119937-appb-100004
    式中，j为复数虚部，
    Figure PCTCN2019119937-appb-100005
    ω＝2πf为工作角频率，f为成像系统的工作频率，μ 0为自由空间的磁导率，σ为胸腔组织的电导率，ε 0为自由空间的介电常数，ε r为胸腔组织的介电常数，ε r＝ε′ r-jσ/ωε 0，ε′ r为胸腔组织相对介电常数的实部，
    Figure PCTCN2019119937-appb-100006
    为总磁场，
    Figure PCTCN2019119937-appb-100007
  5. 根据权利要求3所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述内部磁场模型为：
    Figure PCTCN2019119937-appb-100008
    式中，
    Figure PCTCN2019119937-appb-100009
    为入射磁场，G为格林函数，
    Figure PCTCN2019119937-appb-100010
    为从场源点到散射磁场的位置矢量，
    Figure PCTCN2019119937-appb-100011
    为从场源点到胸腔内任意一点的位置矢量，k 0为自由空间的波数，
    Figure PCTCN2019119937-appb-100012
    为磁电流密度，
    Figure PCTCN2019119937-appb-100013
    μ r为胸腔组织的磁导率，
    Figure PCTCN2019119937-appb-100014
    为感应电流密度，
    Figure PCTCN2019119937-appb-100015
    为总电场，
    Figure PCTCN2019119937-appb-100016
    外部磁场模型为：
    Figure PCTCN2019119937-appb-100017
    式中，
    Figure PCTCN2019119937-appb-100018
    为散射磁场，
    Figure PCTCN2019119937-appb-100019
    为从场源点到场域内任一点的单位向量，
    Figure PCTCN2019119937-appb-100020
    R为从场源点到散射场内任意一点的距离；
    Figure PCTCN2019119937-appb-100021
  6. 根据权利要求3所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述利用内部磁场模型和外部磁场模型重构目标区域中胸腔的图像的过程为：
    依次对至少三个收发一体的线圈中的任意两个线圈所探测到的散射磁场进行比较；
    依次根据两两比较得到的差异获得能够反映胸腔组织电磁属性分布的幅值和相位的信息；
    根据连续探测到的散射磁场分布信息，从建立的非线性观测模型中提取出相应的变化数值和曲线，并根据变化数值重建胸腔二维图像；
    对非均匀随机排布的至少三个线圈探测到的总可见散射磁场进行基于迭代算法的压缩感知技术的信号处理，获取压缩感知处理后的信号；
    通过对非均匀排布的所有线圈获得的总可见散射磁场函数进行处理，得到胸腔重构图像。
  7. 根据权利要求6所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述N个线圈的总可见散射磁场函数R为N(N-1)个线圈的可见散射磁场函数之和：
    Figure PCTCN2019119937-appb-100022
    其中，
    Figure PCTCN2019119937-appb-100023
    表示位于
    Figure PCTCN2019119937-appb-100024
    和
    Figure PCTCN2019119937-appb-100025
    处的两个线圈的可见散射磁场，
    Figure PCTCN2019119937-appb-100026
    其包含相位延迟和/或振幅差异信息；
    Figure PCTCN2019119937-appb-100027
    表示目标区域中任意点到第i个线圈的距离矢量，
    Figure PCTCN2019119937-appb-100028
    表示目标区域中任意点到第j个线圈的距离矢量，
    Figure PCTCN2019119937-appb-100029
    表示位于
    Figure PCTCN2019119937-appb-100030
    处的线圈探测到的散射磁场，
    Figure PCTCN2019119937-appb-100031
    表示位于
    Figure PCTCN2019119937-appb-100032
    处的线圈测量到的散射磁场的共轭，<>表示平均时间。
  8. 根据权利要求6所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述对非均匀随机排布的至少三个线圈探测到的总可见散射磁场进行基于迭代算法的压缩感知技术的信号处理的过程为：
    Figure PCTCN2019119937-appb-100033
    服从于
    Figure PCTCN2019119937-appb-100034
    式中，
    Figure PCTCN2019119937-appb-100035
    是基于非均匀随机分布的线圈阵列的全息磁感应图像，α为
    Figure PCTCN2019119937-appb-100036
    范数一致性的权重，β为
    Figure PCTCN2019119937-appb-100037
    范数一致性的权重；
    Figure PCTCN2019119937-appb-100038
    范数表示向量各个元素绝对值之和，
    Figure PCTCN2019119937-appb-100039
    范数表示向量各个元素的平方求和然后求平方根，
    Figure PCTCN2019119937-appb-100040
    范数表示向量各个元素的总变异量，|| || TV表示二维各向同性算子，γ表示欠采样的非均匀随机k空间数据，A表示获取的反映欠采样数据的测量矩阵，并且是将图像变换为稀疏表示的稀疏矩阵；δ表示精度，Ψ表示观测矩阵；
    其中，测量矩阵A为：
    Figure PCTCN2019119937-appb-100041
    式中，U表示二值矩阵，用于随机抽样下随机位置的选择；I表示目标区域的可见度强度函数，
    Figure PCTCN2019119937-appb-100042
    表示二维傅里叶逆变换。
  9. 根据权利要求6所述的基于稀疏采样的全息磁感应胸腔成像方法，其特征在于，所述胸腔重构图像为：
    Figure PCTCN2019119937-appb-100043
    式中，R表示总可见散射磁场函数，
    Figure PCTCN2019119937-appb-100044
    Figure PCTCN2019119937-appb-100045
    l＝sinθcosφ，m＝sinθsinφ；
    Figure PCTCN2019119937-appb-100046
    分别为沿x，y，z轴正时空方向的单位矢量,
    Figure PCTCN2019119937-appb-100047
    分别为任一线圈
    Figure PCTCN2019119937-appb-100048
    在直角坐标系中沿x，y，z轴的位置，
    Figure PCTCN2019119937-appb-100049
    分别为任一线圈
    Figure PCTCN2019119937-appb-100050
    在直角坐标系中沿x，y，z轴的位置，θ是原点O和空间任一点P的连线与正向z轴的夹角，φ为xoz平面与通过空间任一点P的半平面之间的夹角，若P点在z轴上则φ角是不确定的；λ b表示工作波长，v b表示背景的速度，f表示工作频率。
  10. 一种基于稀疏采样的全息磁感应胸腔成像，其特征在于，包括主控模块、磁场信号发生模块、切换模块、收发器阵列模块、图像处理模块和显示模块；所述收发器阵列模块采用收发一体的线圈；
    所述主控模块用于控制所述磁场信号发生模块发射射频信号，还用于通过所述切换模块控制收发器阵列模块在发射状态和接收状态间进行切换；
    所述收发器阵列模块在发射状态时，用于根据交变电流产生交变磁场，使目标区域的人体在电磁场环境下产生电磁散射信号；所述收发器阵列模块在接收状态时，用于对目标区域的电磁散射信号进行测量；
    所述图像处理模块用于根据接收到的电磁散射信号，结合压缩感知方法对胸腔图像进行重构处理，并将重构的胸腔图像传输至显示模块进行显示；
    根据接收到的散射电场数据进行脑图像重构，并将重构的脑图像传输至显示模块进行显示；
    所述收发器阵列模块包括N个收发一体的线圈，其中，N为自然数且N≥3；
    所述发射接收模块采用二维微波天线阵列，所述二维微波天线阵列包括N个收发一体的微波天线，其中，N为自然数且N≥3；
    所述射频信号的频率为900Hz-20GHz。
PCT/CN2019/119937 2019-10-25 2019-11-21 基于稀疏采样的全息磁感应胸腔成像方法及成像系统 WO2021077521A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911021126.6 2019-10-25
CN201911021126.6A CN110720914A (zh) 2019-10-25 2019-10-25 基于稀疏采样的全息磁感应胸腔成像方法及成像系统

Publications (1)

Publication Number Publication Date
WO2021077521A1 true WO2021077521A1 (zh) 2021-04-29

Family

ID=69222056

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/119937 WO2021077521A1 (zh) 2019-10-25 2019-11-21 基于稀疏采样的全息磁感应胸腔成像方法及成像系统

Country Status (2)

Country Link
CN (1) CN110720914A (zh)
WO (1) WO2021077521A1 (zh)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111223285A (zh) * 2020-03-13 2020-06-02 昆明理工大学 一种基于稀疏表示的无线涡流检测系统及方法
CN113670952A (zh) * 2021-07-19 2021-11-19 宁波大学 基于稀疏感应电流的介质材料实时微波成像方法

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102499682A (zh) * 2011-10-19 2012-06-20 中国人民解放军第四军医大学 用于磁感应断层成像的激励测量复用线圈组及数据采集
CN102955159A (zh) * 2011-08-30 2013-03-06 中国科学院电子学研究所 一种基于压缩感知的电磁逆散射成像方法
CN105997070A (zh) * 2016-06-15 2016-10-12 合肥工业大学 一种非接触式磁感应成像系统及其成像方法
CN106108899A (zh) * 2016-06-15 2016-11-16 合肥工业大学 一种全息微波成像系统及其成像方法
JP2018130142A (ja) * 2017-02-13 2018-08-23 株式会社日立製作所 医用画像診断装置
CN108577840A (zh) * 2018-02-11 2018-09-28 南京邮电大学 一种动态磁共振稳健pca成像方法
CN109188431A (zh) * 2018-09-11 2019-01-11 合肥工业大学 一种基于压缩感知的全息微波快速成像方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1981700A (zh) * 2005-12-15 2007-06-20 中国科学院电工研究所 一种磁感应式磁共振电阻率成像方法及装置
CN109199381B (zh) * 2018-09-11 2021-11-02 合肥工业大学 一种全息微波弹性成像系统及其成像方法
CN109283530B (zh) * 2018-09-14 2020-08-14 浙江大学 一种利用压缩感知提高微波成像线性度的方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102955159A (zh) * 2011-08-30 2013-03-06 中国科学院电子学研究所 一种基于压缩感知的电磁逆散射成像方法
CN102499682A (zh) * 2011-10-19 2012-06-20 中国人民解放军第四军医大学 用于磁感应断层成像的激励测量复用线圈组及数据采集
CN105997070A (zh) * 2016-06-15 2016-10-12 合肥工业大学 一种非接触式磁感应成像系统及其成像方法
CN106108899A (zh) * 2016-06-15 2016-11-16 合肥工业大学 一种全息微波成像系统及其成像方法
JP2018130142A (ja) * 2017-02-13 2018-08-23 株式会社日立製作所 医用画像診断装置
CN108577840A (zh) * 2018-02-11 2018-09-28 南京邮电大学 一种动态磁共振稳健pca成像方法
CN109188431A (zh) * 2018-09-11 2019-01-11 合肥工业大学 一种基于压缩感知的全息微波快速成像方法

Also Published As

Publication number Publication date
CN110720914A (zh) 2020-01-24

Similar Documents

Publication Publication Date Title
JP7397067B2 (ja) 医療用イメージングのための装置及び処理
Bevacqua et al. Millimeter-waves breast cancer imaging via inverse scattering techniques
CN109199381B (zh) 一种全息微波弹性成像系统及其成像方法
US8977340B2 (en) System and method for collection and use of magnetic resonance data and microwave data to identify boundaries of interest
Guo et al. Adaptive clustering distorted born iterative method for microwave brain tomography with stroke detection and classification
Guardiola et al. 3-D microwave magnitude combined tomography for breast cancer detection using realistic breast models
Dachena et al. Microwave imaging for the diagnosis of cervical diseases: A feasibility analysis
US10768214B2 (en) Multi-view planar near-field scattering tomography system
Hosseinzadegan et al. Discrete dipole approximation-based microwave tomography for fast breast cancer imaging
Elloian et al. Continuous wave simulations on the propagation of electromagnetic fields through the human head
US20230108098A1 (en) Magnetic Induction Molecular Imaging Method and Magnetic Induction Molecular Imaging System for Biological Tissue Detection
WO2021077521A1 (zh) 基于稀疏采样的全息磁感应胸腔成像方法及成像系统
Wang et al. Imaging of 3-D dielectric objects using far-field holographic microwave imaging technique
Zhang et al. A microwave thorax imaging system based on symmetrical dipole antenna and one-step supervised descent method
JP6678985B2 (ja) 診断装置
Zhou et al. Magnetoacoustic tomography with magnetic induction (MAT-MI) for breast tumor imaging: numerical modeling and simulation
Bisio et al. Brain stroke detection by microwave imaging systems: Preliminary two-dimensional numerical simulations
WO2021077546A1 (zh) 磁感应成像方法和系统
Jalilvand Application-specific broadband antennas for microwave medical imaging
Rosa et al. Urinary bladder volume monitoring using magnetic induction tomography: A rotational simulation model for anatomical slices within the pelvic region
CN110960216A (zh) 多频全息微波脑成像系统及成像方法
Flores et al. Machine learning approach to microwave imaging for cancer detection
AU2020103702A4 (en) Holographic electromagnetic induction chest imaging method based on sparse sampling and imaging system thereof
Rubæk et al. A contrast source inversion algorithm formulated using the log-phase formulation
WO2020043115A1 (zh) 融合触觉传感和光断层扫描成像的检测装置与检测方法

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: 19949867

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: 19949867

Country of ref document: EP

Kind code of ref document: A1