CN111265245B - Passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation - Google Patents
Passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation Download PDFInfo
- Publication number
- CN111265245B CN111265245B CN202010066167.3A CN202010066167A CN111265245B CN 111265245 B CN111265245 B CN 111265245B CN 202010066167 A CN202010066167 A CN 202010066167A CN 111265245 B CN111265245 B CN 111265245B
- Authority
- CN
- China
- Prior art keywords
- apodization
- signals
- passive
- imaging
- array element
- Prior art date
- Legal status (The legal status 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 status listed.)
- Active
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/48—Diagnostic techniques
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B8/00—Diagnosis using ultrasonic, sonic or infrasonic waves
- A61B8/52—Devices using data or image processing specially adapted for diagnosis using ultrasonic, sonic or infrasonic waves
Landscapes
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Medical Informatics (AREA)
- Pathology (AREA)
- Radiology & Medical Imaging (AREA)
- Biophysics (AREA)
- Physics & Mathematics (AREA)
- Heart & Thoracic Surgery (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- General Health & Medical Sciences (AREA)
- Public Health (AREA)
- Veterinary Medicine (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Investigating Or Analyzing Materials By The Use Of Ultrasonic Waves (AREA)
Abstract
The invention discloses a passive cavitation imaging method and a system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation, which comprises the following steps: delaying cavitation signals passively received by the linear array, constructing a covariance matrix, solving an optimal solution implicit simplified equation of double-constraint robust Capon beam forming by utilizing a Newton iteration method, then calculating an optimal weighting vector and weighting each array element delay signal; constructing respective sub-apodization functions of a positive parent cross apodization mode and a negative parent cross apodization mode, apodizing the delay weighted signals of each array element, calculating Pearson correlation coefficients between the obtained positive apodization synthetic signal and the obtained negative apodization synthetic signal, and averaging the correlation coefficients; and weighting the double-constraint robust Capon beam-forming signals by using the average value of the correlation coefficient, finally calculating the acoustic energy, and obtaining a passive cavitation image after normalization and logarithm. The invention can effectively eliminate the interference artifact of passive cavitation imaging.
Description
Technical Field
The invention belongs to the technical field of ultrasonic detection and ultrasonic imaging, and particularly relates to a time domain passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation.
Background
Focused ultrasound can realize noninvasive treatment due to the advantages of low cost, strong tissue penetrability, small side effect and the like, and has been widely and deeply researched and obtained positive research results in the aspects of tumor thermal ablation, ultrasonic thrombolysis, drug release and the like. Research has shown that cavitation activity is an important physical mechanism of focused ultrasound therapy regardless of the treatment modality. Cavitation itself is a random physical phenomenon that can lead to unpredictability of the area of treatment, and therefore reliable image guidance and monitoring must be developed to ensure that focused ultrasound can be safely and effectively used in the clinic.
The dynamic process of cavitation from nucleation, vibration to final collapse of microbubbles in a medium is observed by means such as high-speed photography and sonoluminescence, but the optical method can only be used for basic physical research of cavitation and cannot be used for monitoring the treatment process of opaque biological tissues. Magnetic resonance imaging is sensitive to tissue temperature changes during treatment and can be used for tumor thermal ablation monitoring, but is not suitable for cavitation transient physical studies due to poor time resolution, and on the other hand magnetic resonance equipment is costly and may not be suitable for some patients. The active ultrasonic imaging is generally used for detecting damage after treatment because the emitted imaging pulse of the active ultrasonic imaging can interfere with a focused ultrasonic treatment signal, and cannot carry out real-time imaging on cavitation in the treatment process. The passive cavitation detection technology can carry out real-time quantitative monitoring on cavitation components (stable cavitation or inertial cavitation) and the intensity thereof by using a single-array-element ultrasonic transducer to passively sense cavitation signals, but cannot obtain the cavitation space distribution in treatment.
The ultrasonic passive cavitation imaging technology is an image monitoring technology developed from single-array-element passive cavitation detection, and is characterized in that a clinically common ultrasonic diagnostic imaging transducer (such as a linear array transducer) or a custom transducer (such as a sparse hemispherical array transducer) is used for passively receiving cavitation signals, and then a passive beam synthesis algorithm is used for processing, so that two-dimensional or even three-dimensional spatial distribution of cavitation is obtained, the defects of the single-array-element passive cavitation detection technology are overcome, and the real-time quantitative monitoring can be effectively carried out on the ultrasonic treatment process. Since the birth of the ultrasonic passive cavitation imaging technology, extensive research is carried out on the application of the ultrasonic passive cavitation imaging technology in ultrasonic therapy. The current common use is a passive cavitation imaging method based on a time exposure acoustic algorithm. Research shows that for the hemispherical array transducer, the algorithm can obtain good image quality; however, when the linear array transducer is used as a receiving array, the algorithm can generate serious interference artifacts to the image on the basis of the limitation of the linear array diffraction mode: on one hand, when the imaging transducer is used for a long time, the positions and the sensitivities of some array elements can be deviated, so that the received signals are distorted, and artifacts are generated; on the other hand, the non-uniformity of the biological tissue makes the propagation speed of the ultrasonic wave in the medium inconsistent, and an error occurs in the beam forming process, which also causes a certain degree of interference. Furthermore, cavitation often occurs as a cloud of microbubbles in a small area, and the uneven distribution of such microbubbles and their interaction can also contribute to image interference artifacts. In fact, there is not any real cavitation source in the artifact region, and when the energy of the artifact is high, the position of the cavitation source and the number of the cavitation sources are misjudged, so that the cavitation spatial-temporal evolution in the focused ultrasound treatment process cannot be accurately monitored. Because the linear array transducer can be used for focus observation and treatment monitoring at the same time, and has a better application prospect in clinic, a key problem to be solved urgently is how to eliminate the interference artifact of linear array passive cavitation imaging so as to improve the image quality.
Disclosure of Invention
The invention aims to provide a passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation.
In order to achieve the purpose, the invention adopts the following technical scheme:
a passive cavitation imaging method, comprising the steps of:
the method comprises the following steps: acquiring cavitation signals (generated after focused ultrasonic waves emitted by a focused ultrasonic transducer act on a medium) received by a passive receiving ultrasonic transducer (for example, an ultrasonic linear array transducer) by using parallel channel data acquisition equipment to obtain original channel signals; then setting an imaging area of passive cavitation imaging and dividing a pixel grid;
step two: and (2) delaying the original channel signal obtained in the step one aiming at any pixel point position x in the imaging area, constructing a covariance matrix R (x) according to each array element delay signal, solving a double-constraint robust Capon beam synthesis optimal solution implicit equation (the reduced double-constraint robust Capon beam synthesis optimal solution implicit equation is an optimal solution implicit simplified equation) which is decomposed and simplified according to the characteristics of the covariance matrix by using a Newton iteration method, estimating a direction vector of double-constraint robust Capon beam synthesis according to the Lagrange factor obtained by solving, calculating to obtain an optimal weighting vector, and weighting each array element delay signal by using the weighting vector to obtain each array element delay signal.
Preferably, the second step further comprises the following steps: and summing the delay weighted signals of the array elements to obtain a double-constraint robust Capon beam forming signal at the position of the corresponding pixel point in the imaging area.
Preferably, the passive cavitation imaging method further comprises the following steps three to five:
step three: according to the continuous N in the passive receiving ultrasonic transducerEPArray element clusters formed by array elements respectively construct N corresponding to positive-mother cross apodization mode and negative-mother cross apodization modeEPA sub-apodization function; utilizing the obtained sub-apodization function to apodize the delay weighted signals of the array elements obtained in the step two, and then calculating N obtained by apodizationEPA positive apodization combined signal and NEPPearson correlation coefficients between every two inverse trace synthesis signals;
step four: weighting the double-constraint robust Capon beam-forming signals by using the average value of the Pearson correlation coefficient obtained in the step three to obtain correlation synthesis signals, and integrating the correlation synthesis signals in the acquisition time period of the original channel signals to obtain the acoustic energy of the corresponding pixel point position in the imaging region;
step five: and carrying out normalization and logarithm processing on the sound energy at the positions of all the pixel points in the imaging area to obtain a passive cavitation imaging result (namely a passive cavitation image).
Preferably, in the first step, the passive receiving ultrasonic transducer is selected from an ultrasonic linear array transducer, and the number of array elements of the linear array transducer is 128 or 256 so as to match the number of channels of the existing ultrasonic imaging platform.
Preferably, in the second step, the cost function of the doubly constrained robust Capon beamforming is constructed as a Lagrange doubly constrained function by introducing a Lagrange factor λ (x), and an implicit equation of an optimal solution of the doubly constrained robust Capon beamforming is obtained by minimizing the doubly constrained function:
wherein I is an identity matrix having the same dimension as the covariance matrix R (x),
for the assumed direction vector in the cost function, ε is the direction vector uncertainty parameter.
Preferably, in the step two, the estimated direction vector of the doubly constrained robust Capon beam forming is represented as:
wherein the content of the first and second substances,
the direction vector is synthesized for the estimated dual-constrained robust Capon beam.
Preferably, the uncertainty parameter epsilon of the direction vector is N/16-N/4.
Preferably, in the third step, NEP<N/2, N is the array element number of the passive receiving ultrasonic transducer, NEPIs generally 2χ(χ＝1,2,3,4)。
Preferably, in the third step, the positive apodization synthetic signal and the inverse apodization synthetic signal are respectively expressed as:
wherein, swi(x, t) is the ith array element delay weighted signal, t represents the original channel signal acquisition time, spak(x, t) is the kth positive apodized composite signal, snak(x, t) is the k-th inverse trace synthesis signal, k being 1,2EP。
Preferably, in the fourth step, NEPA positive apodization combined signal and NEPThe average value of Pearson correlation coefficients between every two inverse apodization synthetic signals is calculated according to the following formula so as to fully utilize the correlation between a plurality of groups of positive and negative apodization synthetic signals:
wherein, PearCorrm,n(x) Is the Pearson correlation coefficient between the mth positive apodized composite signal and the nth inverse apodized composite signal.
A passive cavitation imaging system comprises a digital ultrasonic imaging platform, a passive receiving ultrasonic transducer and a parallel channel data acquisition device, wherein the passive receiving ultrasonic transducer and the parallel channel data acquisition device are connected with the ultrasonic imaging platform;
the parallel channel data acquisition equipment is used for acquiring cavitation signals (generated after focused ultrasonic waves emitted by a focused ultrasonic transducer act on a medium) received by a passive receiving ultrasonic transducer (such as an ultrasonic linear array transducer) (so as to obtain original channel signals);
the ultrasonic imaging platform is used for processing an original channel signal obtained by the parallel channel data acquisition equipment according to the first step, the second step (and calculating acoustic energy, namely integrating the square of the double-constraint robust Capon beam forming signal in the acquisition time period of the original channel signal to obtain the acoustic energy of a corresponding pixel point position in an imaging area) and the fifth step so as to obtain a passive cavitation imaging result based on the double-constraint robust Capon beam forming, or is used for processing the original channel signal obtained by the parallel channel data acquisition equipment according to the first step to the fifth step so as to obtain the passive cavitation imaging result based on the double-constraint robust Capon beam forming and the multiple apodization cross correlation.
The invention has the beneficial effects that:
aiming at the problem of serious interference artifact of the traditional linear array passive cavitation imaging, the invention delays the passively received original channel signals and constructs a covariance matrix, obtains Lagrange factors by using a Newton iteration method after simplifying an optimal solution implicit equation of double-constraint robust Capon beam synthesis by using the covariance matrix, further calculates direction vectors and optimal weighting vectors, and weights each array element delay signal by using the optimal weighting vectors, thereby inhibiting the interference artifact of the passive cavitation imaging.
Furthermore, the invention eliminates the residual interference artifact by constructing a complementary positive-parent cross apodization mode and an inverse-parent cross apodization mode and sub apodization functions of the positive-parent cross apodization mode and the inverse-parent cross apodization mode, averaging Pearson correlation coefficients between a positive apodization synthesis signal and an inverse apodization synthesis signal which are obtained after apodization processing of delay weighted signals of each array element, and weighting the double-constraint robust Capon the beam synthesis signal by utilizing the obtained correlation coefficient average value.
Drawings
Fig. 1 is a flowchart of a linear array passive cavitation imaging method based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation in an embodiment of the present invention.
FIG. 2 is a positive parent cross-apodization pattern and method constructed in a multiple apodization cross-correlation embodiment of the inventionAnti-mother cross apodization mode (array element number N of array element cluster)EPIs 4).
FIG. 3 is a graph of four sub-apodization functions (a, b, c, d) derived from a positive parent cross-apodization mode and four sub-apodization functions (e, f, g, h) derived from an inverse parent cross-apodization mode in a multiple apodization cross-correlation embodiment of the present invention, wherein: number N of array elements of array element clusterEPIs 4.
FIG. 4 shows four positive apodization composite signals (a, b, c, d) and four inverse apodization composite signals (e, f, g, h) resulting from multiple apodization cross correlations according to embodiments of the present invention, where: number N of array elements of array element clusterEPIs 4.
Fig. 5 shows the linear array passive cavitation imaging results obtained by using the conventional method (a) and the proposed method (b) in the embodiment of the present invention.
Fig. 6 is a cross-sectional curve (a) and an axial sectional curve (b) of the linear array passive cavitation imaging result obtained according to the conventional method and the proposed method in the embodiment of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
Referring to fig. 1, the linear array passive cavitation imaging method based on the double-constraint robust Capon beam synthesis and the multiple apodization cross-correlation comprises the following specific steps:
(1) the method comprises the steps of closing the emission of each array element of an ultrasonic linear array transducer (for example, the number N of the array elements is 128, and the array element spacing is 0.3mm) on a digital ultrasonic imaging platform through programming, enabling each array element of the linear array transducer to passively receive a cavitation signal generated after focused ultrasonic waves act on a medium, collecting the signal by using a parallel channel data collection device (for example, the data sampling rate is 40MHz), and recording the collected original channel signal as chi(t), wherein i ═ 1, 2.., N, with an acquisition period of t ═ t ·start～tend(e.g., t)end-tstart＝100μs)；
(2) Setting an imaging area for passive cavitation imaging, wherein the width of the imaging area is generally the aperture size of an ultrasonic linear array transducer, and the height is generally set to be a cavitation sourceThe upper part and the lower part of the center are respectively in the range of 20 mm; partitioning a grid of pixels P for passive cavitation imaging in an imaging regionL×PA(PLAnd PAThe number of pixels along the transverse direction and the axial direction of the ultrasonic linear array transducer respectively), the transverse size of the spatial position x (called the pixel position x for short) of the pixel can be generally set as the array element interval (for example, 0.3mm), and the axial size (for example, 0.5mm) is the height of the imaging area divided by the number of pixels P along the axial directionA；
(3) Calculating the relative time delay tau of each array element of the ultrasonic linear array transducer according to the distance difference from the pixel point position x to each array element of the ultrasonic linear array transduceri(x)：
τi(x)＝||x-ei||/c
Wherein, | | · | represents the Euclidean norm, c is the propagation speed of the ultrasonic wave in the medium (e.g., c is 1540m/s), eiThe spatial position of the ith array element of the ultrasonic linear array transducer is shown;
(4) using the relative time delay tau as described in step (3)i(x) For the original channel signal ch in the step (1)i(t) delaying to obtain delay signals s of each array elementi(x,t)：
si(x,t)＝chi[t+τi(x)]
(5) Utilizing the delay signal s of each array element obtained in the step (4)i(x, t) during the acquisition period t of step (1)start～tendIntra-constructed covariance matrix r (x):
wherein [ ·]TRepresents a transpose of the matrix;
(6) constructing a cost function of the double-constraint robust Capon beam forming according to the covariance matrix R (x) obtained in the step (5):
wherein a isThe direction vector to be solved for is,
for the purpose of the assumed direction vector,
for the first constraint, Re (-) represents the real part of the complex number, ε is the uncertainty parameter of the direction vector, | a (x) | calculation2N is the second constraint [ ·]-1Represents inverting the matrix;
(7) introducing Lagrange factor lambda (x), constructing a Lagrange dual-constraint function, and obtaining an optimum solution implicit equation of the dual-constraint robust Capon beam synthesis by minimizing the dual-constraint function:
wherein I is an identity matrix with the same dimension as the covariance matrix R (x);
(8) performing eigen decomposition on the covariance matrix obtained in step (5), wherein R (x) is equal to U (x) is Λ (x) is U (x)TOn the basis, the optimal solution implicit equation in the step (7) is simplified to obtain the following optimal solution implicit simplified equation:
wherein z isiIs a vector
U (x) is a feature vector obtained by decomposing the features of R (x), gammaiIs the ith diagonal element in the eigenvalue matrix Lambda (x) obtained after the characteristic decomposition of R (x);
(9) solving the optimal solution implicit simplified equation obtained in the step (8) by using a Newton iterative method to obtain a solution of the Lagrange factor lambda (x) introduced in the step (7):
wherein the content of the first and second substances,
(-) represents the derivation of the function, n represents the number of iterations;
(10) and then estimating the direction vector according to the Lagrange factor lambda (x) obtained by the solution in the step (9):
(11) estimating a direction vector according to step (10)
Calculating an optimal weight vector w (x) of the double-constrained robust Capon beam synthesis:
(12) using the optimal weight vector w (x) obtained in the step (11) to delay the signal s of each array element obtained in the step (4)i(x, t) are weighted to obtain delay weighted signals sw of each array elementi(x,t)：
swi(x,t)＝wi(x)si(x,t)
Wherein, wi(x) The ith element of the optimal weight vector w (x) obtained in the step (11);
(13) by successive NEP(e.g., N)EP4) array elements are an array element cluster, starting from the 1 st array element and every N array elements from the N array elementsEPOne array element cluster is extracted from each array element, and N/(2N) is the total of the extracted array elementsEP) Positive and negative cross apodization mode PAP constructed by single array element clusteri：
From NEP+1 array elements from N-NEPEvery N in each array elementEPOne array element cluster is extracted from each array element, and N/(2N) is the total of the extracted array elementsEP) An array element cluster is constructed to form an anti-mother cross apodization mode NAPi：
NEPPositive-mother cross apodization mode PAP with 4-hour structureiAnd anti-parent-crossing apodization mode NAPiAs shown in fig. 2(a) and 2 (b);
(14) extracting the kth (k is 1, 2.. multidot.n.multidot.n) from each array element cluster corresponding to the positive mother cross apodization mode in the step (13)EP) Setting the function value corresponding to the kth array element as 1 and setting the function values corresponding to other array elements of the linear array transducer as 0 to obtain the kth sub-apodization function of the positive-matrix cross apodization mode
Extracting the kth (k is 1, 2.. multidot.n.multidot.n) from each array element cluster corresponding to the anti-mother cross apodization mode in the step (13)EP) Setting the function value corresponding to the kth array element as 1 and setting the function values corresponding to other array elements of the linear array transducer as 0 to obtain the kth sub-apodization function of the anti-parent cross apodization mode
Referring to FIG. 3, N is shown in FIGS. 3(a) to 3(d), respectivelyEPApodization from positive parent cross when 4-Of the formula PAPiThe obtained 1 st, 2 nd, 3 rd and 4 th sub-apodization functions
FIG. 3(e) to FIG. 3(h) are each NEP4-hour from anti-mother cross apodization mode NAPiThe obtained 1 st, 2 nd, 3 rd and 4 th sub-apodization functions
(15) And (5) delaying the weighted signal sw of each array element obtained in the step (12) by using the kth sub-apodization function of the positive parent cross apodization mode in the step (14)i(x, t) apodization until N is obtainedEPA positive apodization combined signal spak(x,t)：
Apodizing each array element delay weighted signal obtained in the step (12) by using the kth sub-apodization function of the anti-parent cross apodization mode in the step (14) until N is obtainedEPInverse trace synthesized signal snak(x,t)：
Referring to fig. 4, taking a cavitation source position (0mm,40mm) as an example, fig. 4(a) to 4(d) are respectively the 1 st, 2 nd, 3 rd and 4 th positive apodization synthetic signals obtained when a pixel point position x is (0mm,40mm), and fig. 4(e) to 4(h) are respectively the 1 st, 2 nd, 3 rd and 4 th inverse apodization synthetic signals obtained when the pixel point position x is (0mm,40 mm); wherein a direction vector uncertainty parameter epsilon in the double-constraint robust Capon beam forming is set to 15;
(16) calculating N obtained in step (15)EPA positive apodization combined signal spak(x, t) and NEPInverse trace synthesized signal snak(x, t) Pearson between twoCorrelation coefficient PearCorrm,n(x)：
Wherein, m is 1,2EP,n＝1,2,...,NEP，Cov[·]Representing the covariance between the two signals, Std [ ·]Represents the standard deviation of the signal;
(17) calculating the total N obtained in the step (16)EP×NEPAverage of individual Pearson correlation coefficients
(18) And (3) adding the array element delay weighted signals obtained in the step (12) to obtain a double-constraint robust Capon beam forming signal BFS (x, t):
(19) weighting the double-constraint robust Capon beam-forming signals obtained in the step (18) by using the average value of the Pearson correlation coefficient obtained in the step (17) to obtain final correlation composite signals CBFS (x, t):
(20) calculating the acoustic energy E (x) at the pixel position x according to the square of the correlation synthesis signal CBFS (x, t) obtained in the step (19):
(21) repeating the steps (3) to (20) until the step (2) is setP in the imaging area of the deviceL×PAAnd (4) after the acoustic energy at the position of each pixel point is calculated, carrying out normalization processing and logarithm processing on the calculated acoustic energy values at the positions of all the pixel points to obtain a final linear array passive cavitation image.
Taking cavitation source position (0mm,40mm) as an example, an imaging result obtained according to a traditional linear array passive cavitation imaging method (a passive cavitation imaging method based on a time exposure acoustic algorithm) is shown in fig. 5(a), an imaging result obtained according to a linear array passive cavitation imaging method based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation is shown in fig. 5(b), and the imaging results are displayed in a dynamic range of 30 dB; wherein a direction vector uncertainty parameter epsilon in the double-constraint robust Capon beam forming is set to be 15, and the array element number N of the array element cluster in the multiple apodization cross correlation is setEPIs 4. As can be seen from fig. 5(a), the imaging result obtained by the conventional method has severe interference artifacts, and the image contrast is very poor; comparing fig. 5(a) and fig. 5(b), it can be seen that the linear array passive cavitation imaging method based on the double-constraint robust Capon beam synthesis and the multiple apodization cross-correlation provided by the present invention significantly reduces the interference artifact, so that the image contrast is greatly improved.
With respect to fig. 5(a) and 5(b), a cross-sectional curve obtained when the axial distance is 40mm is shown in fig. 6(a), and an axial cross-sectional curve obtained when the lateral distance is 0mm is shown in fig. 6 (b). As can be seen from fig. 6, compared with the conventional linear array passive cavitation imaging method, the linear array passive cavitation imaging method based on the double-constrained robust Capon beam synthesis and the multiple apodization cross-correlation is utilized, so that the lateral side lobe level of the passive cavitation image is reduced by about 50dB, and the axial side lobe level is reduced by about 45 dB.
Further, the artifact suppression ratio for passive cavitation imaging is defined to be 20lg (ME)in/MEout) Wherein MEinThe mean value of pixel values greater than half the maximum pixel value before the logarithmic process, MEoutThe average value of the pixel values which are less than half of the maximum pixel value before the logarithmic processing, wherein the pixel value refers to the normalized acoustic energy at the position of the pixel point. Through calculation, the traditional methodThe artifact inhibition rates of the passive cavitation image obtained by the method are respectively 16.15dB and 51.12dB, which shows that the linear array passive cavitation imaging method based on the double-constraint robust Capon beam synthesis and the multiple apodization cross correlation can effectively inhibit interference artifacts, thereby improving the image quality.
The invention has the following advantages:
(1) in the beam synthesis process of the traditional linear array passive cavitation imaging, the weighting coefficient of each array element delay signal is independent of a received signal and the weighting vector of each pixel point position in the passive cavitation imaging result is also fixed; the double-constraint robust Capon beam synthesis method provided by the invention performs double constraints of a fixed norm and a spherical uncertain set on the direction vector, effectively compensates the error between the actual direction vector and the assumed direction vector, and the weighted vector calculated on the basis is continuously updated in a self-adaptive manner at the position of each pixel point in the passive cavitation imaging result, thereby effectively inhibiting the interference artifact;
(2) on the basis of the double-constraint robust Capon beam synthesis, the invention continuously utilizes the multiple apodization cross-correlation to process the delay weighted signal; because the positive apodization synthetic signal and the inverse apodization synthetic signal have higher correlation at the position of the pixel point positioned at the cavitation source, and the correlation is lower at the positions of other pixel points far away from the cavitation source, the interference artifact of the linear array passive cavitation imaging can be further inhibited by calculating the Pearson correlation coefficient and taking the Pearson correlation coefficient as a weighting factor;
(3) under extreme conditions that a large number of array elements with reduced sensitivity or a plurality of array elements in a passive receiving ultrasonic transducer (such as a linear array transducer) are damaged, the interaction between cavitation microbubbles is severe under the action of media (such as ribs and skull) with sharply changed sound velocity in the process of cavitation signal propagation or high-intensity ultrasonic waves, and the like, the single use of the multiple apodization cross-correlation in the method for processing the original channel signal can not effectively resist severe interference artifacts caused by the extreme conditions; the error between the actual direction vector and the assumed direction vector under the extreme conditions can be compensated by using the double-constraint robust Capon beam forming in the method, the interference artifact can be inhibited by the self-adaptive calculation weighting vector (for example, epsilon is N/16-N/4), and the interference artifact can be more effectively removed by further combining a multiple apodization cross-correlation method.
(4) Because the resolution performance of passive cavitation imaging does not depend on the time length of a cavitation signal, the linear array passive cavitation imaging method based on the double-constraint robust Capon beam forming and the multiple apodization cross-correlation can be used for monitoring the continuous wave focusing ultrasonic treatment process and the pulse wave focusing ultrasonic treatment process under a single pulse parameter (pulse length, pulse repetition frequency and the like) or a composite pulse parameter (namely, a plurality of different pulse lengths or pulse repetition frequencies exist in a treatment sequence) in real time.
(5) Although the passive cavitation imaging method based on the double-constraint robust Capon beam forming and the multiple apodization cross correlation only relates to the case that the linear array transducer is used as a passive receiving array, the method is also suitable for other transducers such as a phased array transducer, a sparse hemispherical array transducer or a sparse boundary array transducer.
Claims (10)
1. A passive cavitation imaging method, characterized by: the passive cavitation imaging method comprises the following steps:
1) acquiring cavitation signals received by a passive receiving ultrasonic transducer by using parallel channel data acquisition equipment to obtain original channel signals; setting an imaging area of passive cavitation imaging and dividing a pixel grid;
2) delaying the original channel signals obtained in the step 1) aiming at any pixel point position x in the imaging area, constructing a covariance matrix R (x) according to the obtained array element delayed signals, solving a double-constraint robust Capon beam synthesis optimal solution implicit equation simplified according to the characteristic decomposition of the covariance matrix by using a Newton iteration method, estimating a direction vector of double-constraint robust Capon beam synthesis according to the Lagrange factor obtained by solving, calculating to obtain an optimal weighted vector, weighting the array element delayed signals by using the weighted vector to obtain array element delayed weighted signals, and summing the array element delayed weighted signals to obtain double-constraint robust Capon beam synthesis signals at corresponding pixel point positions in the imaging area;
3) according to the continuous N in the passive receiving ultrasonic transducerEPArray element clusters formed by array elements respectively construct N corresponding to positive-mother cross apodization mode and negative-mother cross apodization modeEPA sub-apodization function; utilizing the sub-apodization function to apodize the delay weighted signals of each array element obtained in the step 2), and then calculating N obtained by apodizationEPA positive apodization combined signal and NEPPearson correlation coefficients between every two inverse trace synthesis signals;
4) weighting the double-constraint robust Capon beam synthesis signals at the corresponding pixel positions in the imaging region by using the average value of the Pearson correlation coefficients obtained in the step 3) to obtain correlation synthesis signals, and integrating the correlation synthesis signals in the acquisition time period of the original channel signals to obtain the acoustic energy at the corresponding pixel positions in the imaging region.
2. A passive cavitation imaging method as claimed in claim 1, characterized by: the passive cavitation imaging method further comprises the following steps:
5) and carrying out normalization and logarithm processing on the acoustic energy at the positions of all the pixel points in the imaging area to obtain a passive cavitation imaging result.
3. A passive cavitation imaging method as claimed in claim 1, characterized by: in the step 2), constructing a cost function of the dual-constraint robust Capon beam synthesis as a Lagrange dual-constraint function by introducing Lagrange factors, and obtaining an optimal solution implicit equation of the dual-constraint robust Capon beam synthesis by minimizing the dual-constraint function:
wherein I is the same dimension as the covariance matrix R (x)The matrix of the unit is formed by a matrix of units,
for the assumed direction vector in the cost function, epsilon is a direction vector uncertainty parameter, lambda (x) is a Lagrange factor, and N is the array element number of the passive receiving ultrasonic transducer.
4. A passive cavitation imaging method as claimed in claim 1, characterized by: in the step 2), the estimated direction vector of the double-constraint robust Capon beam forming is represented as:
wherein the content of the first and second substances,
i is an identity matrix with the same dimension as the covariance matrix r (x),
for the assumed direction vector in the cost function, epsilon is the uncertainty parameter of the direction vector, lambda (x) is Lagrange factor, and N is the array element number of the passive receiving ultrasonic transducer.
5. A passive cavitation imaging method as claimed in claim 3, characterized by: and the uncertainty parameter epsilon of the direction vector is N/16-N/4.
6. A passive cavitation imaging method as claimed in claim 1, characterized by: in the step 3), NEP<N/2, wherein N is the array element number of the passive receiving ultrasonic transducer.
7. A passive cavitation imaging method as claimed in claim 1, characterized by: in the step 3), the positive apodization synthetic signal and the inverse apodization synthetic signal are respectively expressed as:
wherein, swi(x, t) is the ith array element delay weighted signal, t represents the original channel signal acquisition time, spak(x, t) is the kth positive apodized composite signal, snak(x, t) is the k-th inverse trace synthesis signal, k being 1,2EP，PFi kThe kth sub-apodization function, NF, for a positive parent cross-apodization modei kAnd N is the array element number of the passive receiving ultrasonic transducer.
8. A passive cavitation imaging method as claimed in claim 1, characterized by: in the step 4), NEPA positive apodization combined signal and NEPThe average value of Pearson correlation coefficients between every two inverse trace synthesis signals is calculated according to the following formula:
wherein, PearCorrm,n(x) For Pearson correlation between the mth positive apodized composite signal and the nth inverse apodized composite signalAnd (4) the coefficient.
9. A passive cavitation imaging system, characterized by: the system comprises a digital ultrasonic imaging platform, a passive receiving ultrasonic transducer and a parallel channel data acquisition device, wherein the passive receiving ultrasonic transducer and the parallel channel data acquisition device are connected with the ultrasonic imaging platform;
the parallel channel data acquisition equipment is used for acquiring cavitation signals received by the passive receiving ultrasonic transducer;
the ultrasonic imaging platform is used for processing an original channel signal obtained by the parallel channel data acquisition equipment according to a passive cavitation imaging method based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation, so that a passive cavitation imaging result is obtained;
the passive cavitation imaging method based on the double-constraint robust Capon beam synthesis and the multiple apodization cross-correlation comprises the following steps:
1) acquiring cavitation signals received by a passive receiving ultrasonic transducer by using parallel channel data acquisition equipment to obtain original channel signals; setting an imaging area of passive cavitation imaging and dividing a pixel grid;
2) delaying the original channel signals obtained in the step 1) aiming at any pixel point position x in the imaging area, constructing a covariance matrix R (x) according to the obtained array element delayed signals, solving a double-constraint robust Capon beam synthesis optimal solution implicit equation simplified according to the characteristic decomposition of the covariance matrix by using a Newton iteration method, estimating a direction vector of double-constraint robust Capon beam synthesis according to the Lagrange factor obtained by solving, calculating to obtain an optimal weighted vector, weighting the array element delayed signals by using the weighted vector to obtain array element delayed weighted signals, and summing the array element delayed weighted signals to obtain double-constraint robust Capon beam synthesis signals at corresponding pixel point positions in the imaging area;
3) according to the continuous N in the passive receiving ultrasonic transducerEPArray element clusters formed by array elements respectively construct N corresponding to positive-mother cross apodization mode and negative-mother cross apodization modeEPA sub-apodization function; using said sub-transformersApodizing the delay weighted signals of each array element obtained in the step 2) by the apodization function, and then calculating N obtained by apodizationEPA positive apodization combined signal and NEPPearson correlation coefficients between every two inverse trace synthesis signals;
4) weighting the double-constraint robust Capon beam synthesis signals at the corresponding pixel positions in the imaging region by using the average value of the Pearson correlation coefficients obtained in the step 3) to obtain correlation synthesis signals, and integrating the correlation synthesis signals in the acquisition time period of the original channel signals to obtain the acoustic energy at the corresponding pixel positions in the imaging region.
10. A passive cavitation imaging system according to claim 9, characterized by: the passive cavitation imaging method based on the double-constraint robust Capon beam synthesis and the multiple apodization cross-correlation further comprises the following steps: 5) and carrying out normalization and logarithm processing on the acoustic energy at the positions of all the pixel points in the imaging area to obtain a passive cavitation imaging result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010066167.3A CN111265245B (en) | 2020-01-20 | 2020-01-20 | Passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010066167.3A CN111265245B (en) | 2020-01-20 | 2020-01-20 | Passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111265245A CN111265245A (en) | 2020-06-12 |
| CN111265245B true CN111265245B (en) | 2021-02-09 |
Family
ID=70992192
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010066167.3A Active CN111265245B (en) | 2020-01-20 | 2020-01-20 | Passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111265245B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112023283B (en) * | 2020-08-03 | 2021-09-07 | 西安交通大学 | Annular multi-array ultrasonic passive imaging method and system based on high-order aperture autocorrelation |
| CN114098799B (en) * | 2021-10-27 | 2023-06-27 | 西安交通大学 | Rapid low-artifact real-time dynamic imaging method and system for ultrasonic cavitation in single pulse |
| CN115969411B (en) * | 2023-01-16 | 2023-07-11 | 西安交通大学 | Super-resolution ultrasonic passive cavitation imaging method and system based on single cavitation source separation and positioning |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007115200A2 (en) * | 2006-03-31 | 2007-10-11 | University Of Florida Research Foundation, Inc. | Method and system for adaptive and robust detection |
| CN107260217A (en) * | 2017-07-17 | 2017-10-20 | 西安交通大学 | The three-dimensional passive imaging method and system monitored in real time for brain focused ultrasonic cavitation |
| CN109171811A (en) * | 2018-09-25 | 2019-01-11 | 西安交通大学 | The passive cavitation imaging of frequency domain and frequency multiplexed imaging method based on the synthesis of feature space adaptive beam |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US6894642B2 (en) * | 2003-10-01 | 2005-05-17 | University Of Florida Research Foundation, Inc. | Doubly constrained robust capon beamformer |
| GB201216455D0 (en) * | 2012-09-14 | 2012-10-31 | Isis Innovation | Passive ultrasound imaging with sparse transducer arrays |
-
2020
- 2020-01-20 CN CN202010066167.3A patent/CN111265245B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2007115200A2 (en) * | 2006-03-31 | 2007-10-11 | University Of Florida Research Foundation, Inc. | Method and system for adaptive and robust detection |
| CN107260217A (en) * | 2017-07-17 | 2017-10-20 | 西安交通大学 | The three-dimensional passive imaging method and system monitored in real time for brain focused ultrasonic cavitation |
| CN109171811A (en) * | 2018-09-25 | 2019-01-11 | 西安交通大学 | The passive cavitation imaging of frequency domain and frequency multiplexed imaging method based on the synthesis of feature space adaptive beam |
Non-Patent Citations (4)
| Title |
|---|
| Passive acoustic mapping of cavitation using eigenspace-based robust Capon beamformer in ultrasound therapy;Shukuan Lu 等;《Ultrasonics-Sonochemistry》;20171020;670-679 * |
| Passive acoustic mapping utilizing optimal beamforming in ultrasound therapy monitoring;Christian Coviello等;《Acoustical Society of America》;20150531;第137卷(第5期);2573-2583 * |
| Passive cavitation mapping using dual apodization with cross-correlation in ultrasound therapy monitoring;Shukuan Lu等;《Ultrasonics-Sonochemistry》;20190222;18-31 * |
| Real-time monitoring of controllable cavitation erosion in a vessel phantom with passive acoustic mapping;Shukuan Lu 等;《Ultrasonics-Sonochemistry》;20170428;291-300 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111265245A (en) | 2020-06-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111265245B (en) | Passive cavitation imaging method and system based on double-constraint robust Capon beam synthesis and multiple apodization cross-correlation | |
| Mehdizadeh et al. | Eigenspace based minimum variance beamforming applied to ultrasound imaging of acoustically hard tissues | |
| CN104272134B (en) | Clutter recognition in ultrasonic image-forming system | |
| Vignon et al. | Capon beamforming in medical ultrasound imaging with focused beams | |
| CN111134719B (en) | Active and passive ultrasonic composite imaging method and system for phase-change nano liquid drops through focused ultrasonic irradiation | |
| US8818064B2 (en) | Time-domain estimator for image reconstruction | |
| Xu et al. | Spatio-temporally smoothed coherence factor for ultrasound imaging [correspondence] | |
| CN109513123B (en) | High-resolution three-dimensional passive cavitation imaging method based on hemispherical array | |
| US7867166B2 (en) | Statistical estimation of ultrasonic propagation parameters for aberration correction | |
| CN108836389B (en) | Plane wave correlation point coherent self-adaptive beam forming imaging method | |
| EP3548920A1 (en) | Methods and systems for filtering ultrasound image clutter | |
| CN104688271A (en) | Ultrasonic imaging method and ultrasonic imaging device by synthetic focusing | |
| Lu et al. | Complex convolutional neural networks for ultrafast ultrasound imaging reconstruction from in-phase/quadrature signal | |
| Du et al. | A novel transcranial ultrasound imaging method with diverging wave transmission and deep learning approach | |
| Kim et al. | An efficient motion estimation and compensation method for ultrasound synthetic aperture imaging | |
| WO2018099867A1 (en) | Methods and systems for filtering ultrasound image clutter | |
| Xu et al. | Adaptive minimum variance beamforming combined with phase coherence imaging for ultrasound imaging | |
| Pan et al. | Improving axial resolution based on the deconvolution recovery method combined with adaptive weighting techniques for ultrasound imaging | |
| CN112927145B (en) | Correction method for planar ultrasonic craniocerebral imaging | |
| Wang et al. | A mixed transmitting–receiving beamformer with a robust generalized coherence factor: Enhanced resolution and contrast | |
| CN114098799A (en) | Rapid low-artifact real-time dynamic imaging method and system for ultrasonic cavitation in single pulse | |
| Al Mukaddim et al. | Photoacoustic delay-and-sum beamforming with spatiotemporal coherence factor | |
| Wang et al. | High frame rate adaptive imaging using coherence factor weighting and the MVDR method | |
| Ossenkoppele et al. | Improving Lateral Resolution in 3-D Imaging With Micro-beamforming Through Adaptive Beamforming by Deep Learning | |
| Nili et al. | Field of View and Resolution Improvement in Coprime Sparse Synthetic Aperture Ultrasound Imaging |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| CB03 | Change of inventor or designer information | ||
| CB03 | Change of inventor or designer information |
Inventor after: Lu Shukuan Inventor after: Wan Mingxi Inventor after: Zhao Yan Inventor after: Zhang Bo Inventor before: Wan Mingxi Inventor before: Lu Shukuan Inventor before: Zhao Yan Inventor before: Zhang Bo |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |