Ultrasonic device and system, beamforming method, electronic device and storage medium

Abstract

Embodiments of the present application relate to the technical field of medical instruments, and disclose an ultrasonic device and system, a beamforming method, an electronic device and a storage medium. The ultrasonic device comprises a probe, the probe comprising an ultrasonic transducer array assembly, configured to emit ultrasonic waves and collect echo signals of human tissue; and a processing assembly, configured to: when the probe is stationary for more than one collection cycle, the processing assembly acquires echo collection information of a current window in a current collection cycle, and performs beamforming on the echo signals of the current window according to weight matrix parameters of a window corresponding to the current window in a previous collection cycle, and meanwhile, obtains the weight matrix parameters of the current window according to the echo collection information of the current window, so as to perform beamforming on echo signals of a window corresponding to a next collection cycle according to the weight matrix parameters of the current window in the next collection cycle, the collection cycle comprising at least one window. This is favorable to improving the accuracy and real-time performance of ultrasonic imaging of human tissue.

Description

Technical Field

[0001] This application relates to the field of medical device technology, and in particular to an ultrasound device and system, a beamforming wave method, an electronic device, and a storage medium. Background Technology

[0002] Medical ultrasound imaging is a technique that uses the reflection and diffraction of ultrasound waves in biological tissues to acquire image information. It has advantages such as being non-invasive, real-time, and low-cost, and is widely used in various clinical diagnoses and treatments. Taking intracardiac echocardiography (ICE) as an example, it typically involves installing a miniature transducer at the tip of a cardiac catheter, which is then delivered into the heart chambers via surgical pathways such as peripheral blood vessels. After the transducer emits sound waves, ultrasound imaging can be performed on the acquired echo signals, thereby enabling real-time, high-quality imaging and / or hemodynamic measurement of the heart and its adjacent tissues.

[0003] In ultrasound imaging, beamforming is central and plays a decisive role in image quality. Various beamforming methods have been proposed, such as Delay and Sum (DAS) beamforming and Minimum Variance (MV) beamforming.

[0004] However, human tissues are inevitably affected by factors such as movement, blood flow eddies, and friction, which gives the noise in the echo signal complex and non-stationary characteristics. If imaging is performed using the beamforming methods described above, the accuracy of the imaging will be low due to the difficulty in accurately estimating the channel covariance matrix or the noise spatial projection. Moreover, the beamforming methods described above are all designed for processing long-term echo signals, and their real-time performance is difficult to meet the real-time requirements of medical ultrasound imaging. Summary of the Invention

[0005] This application provides an ultrasound device and system, a beamforming method, an electronic device, and a storage medium, which at least helps to improve the accuracy and real-time performance of ultrasound imaging of human tissue.

[0006] According to some embodiments of this application, one aspect of this application provides an ultrasound device, including: a probe, the probe including an ultrasound transducer array assembly for emitting ultrasound and acquiring echo signals from human tissue; and a processing component configured to: when the probe is stationary for more than one acquisition cycle, the processing component acquires echo acquisition information of the current window within the current acquisition cycle, and beamforms the echo signal of the current window according to the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle, and simultaneously obtains the weight matrix parameters of the current window according to the echo acquisition information of the current window, so as to beamform the echo signal of the window corresponding to the next acquisition cycle according to the weight matrix parameters of the current window in the next acquisition cycle, wherein the acquisition cycle includes at least one window.

[0007] In some embodiments, the processing component is further configured to: when the probe is stationary for a period of time within the first acquisition cycle, obtain the weight matrix parameters of each window in the first acquisition cycle through the echo acquisition information of each window in the first acquisition cycle, and perform beamforming of the echo signal of each window according to the weight matrix parameters of each window in the first acquisition cycle.

[0008] In some embodiments, the processing component is further configured to: obtain the covariance matrix of the current window based on the echo acquisition information of the current window, so as to obtain the weight matrix parameters of the current window in combination with the array steering vector.

[0009] In some embodiments, a storage component is further included for storing echo acquisition information of multiple windows of multiple acquisition cycles prior to the current acquisition cycle. The number of acquisition cycles corresponding to the echo acquisition information stored in the storage component is less than the total number of acquisition cycles after the probe is stationary. The processing component is further configured to: take weights on the echo acquisition information of the current window in the current acquisition cycle and the echo acquisition information of the corresponding windows of the multiple acquisition cycles stored, and process them to obtain the covariance matrix of the current window in the current acquisition cycle.

[0010] In some embodiments, the weight of the acquisition period immediately adjacent to the current acquisition period is greater than the weight of the acquisition period far from the current acquisition period.

[0011] In some embodiments, the processing component is further configured to: convert the echo acquisition information within the current window to the time-frequency domain to obtain the time spectrum; process the time spectrum using a complex Gaussian mixture model to obtain the noise power spectral density; and obtain the weight matrix parameters of the current window based on the noise power spectral density and the array steering vector.

[0012] In some embodiments, the processing component is further configured to: when the probe movement time exceeds one window, the processing component beamforms the echo signal of the current window according to the weight matrix parameters of the previous window, and simultaneously obtains the weight matrix parameters of the current window according to the echo acquisition information of the current window, so as to beamform the echo signal of the next window according to the weight matrix parameters of the current window in the next window.

[0013] In some embodiments, the processing component is further configured such that the weight matrix parameters of the first window when the probe is moving are taken from the weight matrix parameters of the last window of the last acquisition cycle when the probe is stationary.

[0014] In some embodiments, a storage component is further included for storing echo acquisition information of multiple windows adjacent to the current window, wherein the number of windows adjacent to the current window stored by the storage component is less than the total number of windows after the probe moves; the processing component is further configured to: obtain the covariance matrix of the current window by weighted processing based on the stored echo acquisition information of multiple windows and the echo acquisition information of the current window, so as to obtain the weight matrix parameters of the current window.

[0015] In some embodiments, the processing component is further configured to: convert the echo acquisition information within the current window to the time-frequency domain to obtain the time spectrum; process the time spectrum using a complex Gaussian mixture model to obtain the noise power spectral density; and obtain the weight matrix parameters of the current window based on the noise power spectral density and the array steering vector.

[0016] In some embodiments, the probe further includes a position sensor for acquiring the movement of the probe.

[0017] According to some embodiments of this application, another aspect of this application provides an ultrasound system, including: an ultrasound device as described in any of the above embodiments; a monitoring device for detecting the activity cycle of human tissue, wherein the processing component of the ultrasound device adjusts the acquisition cycle according to the activity cycle of the human tissue.

[0018] According to some embodiments of this application, another aspect of this application provides a method for beamforming ultrasound echoes, comprising: when the probe is stationary for more than one acquisition cycle, acquiring echo acquisition information of human tissue in the current window within the current acquisition cycle; beamforming the echo signal of the current window according to the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle, and simultaneously obtaining the weight matrix parameters of the current window according to the echo acquisition information of the current window, so as to beamform the echo signal of the window corresponding to the next acquisition cycle according to the weight matrix parameters of the current window in the next acquisition cycle; the acquisition cycle includes at least one window.

[0019] According to some embodiments of this application, another aspect of this application provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the ultrasonic echo beamforming method as described in any of the above embodiments.

[0020] According to some embodiments of this application, another aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the ultrasonic echo beamforming method as described in any of the above embodiments.

[0021] The technical solution provided in this application has at least the following advantages:

[0022] When the probe remains stationary for more than one acquisition cycle, after acquiring the echo acquisition information of the current window within the current acquisition cycle, beamforming of the echo signal of the current window is performed based on the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle. Simultaneously, the weight matrix parameters of the current window are obtained based on the echo acquisition information of the current window, so that beamforming of the echo signal of the corresponding window in the next acquisition cycle can be performed based on the weight matrix parameters of the current window. In other words, by utilizing the periodic activity characteristics of human tissue and combining it with probe movement, the update method of the weight matrix can be accurately determined, thereby improving the accuracy of ultrasound imaging by improving the accuracy of the weight matrix parameters of the current window. Furthermore, due to the periodic activity characteristics of human tissue, when the probe remains stationary for more than one acquisition cycle, its noise also exhibits a periodic variation. Therefore, the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle can be used as the weight matrix parameters of the current window without affecting its beamforming accuracy, and there is no need to wait for the determination of the weight matrix parameters of the current window, thus improving the beamforming efficiency of the current window, reducing processing latency, and improving the real-time performance of ultrasound imaging. Attached Figure Description

[0023] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0024] Figure 1 This is a schematic diagram of the structure of an ultrasonic device provided in one embodiment of this application;

[0025] Figure 2This is a schematic diagram of the acquisition cycle and corresponding window of the ultrasonic device provided in one embodiment of this application;

[0026] Figure 3 This is another structural schematic diagram of the ultrasonic device provided in one embodiment of this application;

[0027] Figure 4 This is a schematic diagram of the window of the ultrasound device provided in one embodiment of this application during probe movement;

[0028] Figure 5 This is another structural schematic diagram of the ultrasonic device provided in one embodiment of this application;

[0029] Figure 6 This is a schematic diagram of the structure of an ultrasonic system provided in one embodiment of this application;

[0030] Figure 7 This is a flowchart of a beamforming method for ultrasonic echoes provided in one embodiment of this application;

[0031] Figure 8 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application. Detailed Implementation

[0032] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the various embodiments of this application will be described in detail below with reference to the accompanying drawings. However, those skilled in the art will understand that many technical details have been presented in the various embodiments of this application to enable readers to better understand this application. However, the technical solutions claimed in this application can be implemented even without these technical details and various changes and modifications based on the following embodiments.

[0033] The division of the following embodiments is for ease of description and should not constitute any limitation on the specific implementation of this application. The various embodiments can be combined with and referenced by each other without contradiction.

[0034] This application provides an ultrasonic device in one aspect. In some embodiments, such as Figure 1 As shown, the ultrasound device includes at least the following structures: a probe 100 and a processing component 200.

[0035] The probe 100 includes an ultrasonic transducer array 101 for emitting ultrasound and acquiring echo signals from human tissue. The processing component 200 is configured to: when the probe 100 is stationary for more than one acquisition cycle, acquire the echo acquisition information of the current window in the current acquisition cycle, and perform beamforming of the echo signal of the current window according to the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle; at the same time, obtain the weight matrix parameters of the current window according to the echo acquisition information of the current window, so as to perform beamforming of the echo signal of the corresponding window in the next acquisition cycle according to the weight matrix parameters of the current window; the acquisition cycle includes at least one window.

[0036] To facilitate a better understanding of the meanings of "acquisition period" and "window" in the above embodiments by those skilled in the art, the following explanation will use the intracardiac ultrasound echo beamforming process as an example. In the context of intracardiac ultrasound, the acquisition period can be the heartbeat cycle. (See reference...) Figure 2 The electrocardiogram shown uses the period between the peaks of two QRS complexes as the cardiac cycle. Specifically, Figure 2 The diagram illustrates two heartbeat cycles, corresponding to acquisition cycles T1 and T2. Acquisition cycle T1 occurs first, followed by acquisition cycle T2; that is, the preceding acquisition cycle of acquisition cycle T2 is acquisition cycle T1. The order of window C' within the windows contained in acquisition cycle T1 is the same as the order of window C within the windows contained in acquisition cycle T2. Therefore, the window corresponding to window C in acquisition cycle T2 in the preceding acquisition cycle T1 is window C'.

[0037] In some embodiments, the ultrasonic transducer array assembly 101 may include an ultrasonic transmitting unit and an echo receiving unit, wherein the ultrasonic transmitting unit is configured to transmit ultrasound, and the echo receiving unit is configured to receive and acquire the echo signal generated by the transmitted ultrasound reflected by human tissue. In this way, the transmission of ultrasound and the reception of the echo signal are implemented using independent units, avoiding mutual interference between the transmitted and echo signals, which helps to improve signal purity and thus improve the accuracy of ultrasound imaging.

[0038] In some embodiments, the ultrasonic transmitting unit (and / or echo receiving unit) may include multiple array elements that transmit (and / or receive signals) according to certain rules and timing, thereby changing the direction and focus of the ultrasonic beam to achieve spatial scanning and dynamic focusing. Therefore, the data acquired at each sampling point is a vector, and the dimension of the vector is consistent with the number of array elements in the echo receiving unit of the ultrasonic transducer array assembly 101.

[0039] The transmission according to certain rules and timing can be achieved through a controller. Specifically, in use, the probe 100 is inserted into the corresponding location in human tissue, such as inside the heart chamber. Then, the controller sets parameters such as the transmission frequency, pulse repetition frequency, and transmission voltage, causing the ultrasonic sound-emitting unit in the probe 100 to emit ultrasonic waves. The echo receiving unit then receives the echo signal reflected back from the human tissue. Thus, after analog-to-digital conversion and sampling processing, the echo signal is converted into a digital signal that can be processed by the processing component 200, i.e., echo acquisition information.

[0040] It should be noted that this embodiment does not limit parameters such as transmission frequency, pulse repetition frequency, and transmission voltage, which can be specifically set according to the actual application scenario.

[0041] Of course, the above is only an example of the ultrasonic transducer array assembly 101. In some embodiments, the ultrasonic transmitting unit and the echo receiving unit can also be integrated into one unit, that is, the same unit provides both transmitting and receiving functions, which will not be elaborated here.

[0042] It should be noted that the embodiments of this application do not limit the type or arrangement of the ultrasonic transducer array assembly 101. For example, the ultrasonic transducer array assembly 101 can be movably disposed within the probe. In some cases, the probe is located at the distal end of an insertable catheter. Similarly, this embodiment does not limit the arrangement of the processing assembly 200. In some embodiments, the processing assembly 200 is disposed within the catheter and communicatively connected to the ultrasonic transducer array assembly 101 in the probe 100 within the catheter; in some embodiments, the processing assembly 200 can also exist independently of the catheter, with the ultrasonic transducer array assembly 101 communicatively connected to the processing assembly 200 through an interface at the proximal end of the catheter. These will not be elaborated further here.

[0043] This embodiment does not limit the length of the window. The length of the window, i.e., the number of sampling points contained in the window, determines the temporal and frequency resolution of beamforming. The length of the window should not exceed the length of the acquisition period; that is, the acquisition period includes at least one window to ensure that beamforming is performed at least once within one acquisition period.

[0044] Generally, a larger window length results in better beamforming and thus better ultrasound imaging, but also increases computational complexity. Therefore, to ensure both ultrasound performance and real-time performance, the window length must be neither too large nor too small. In some embodiments, the window length is determined based on one or more factors such as imaging performance, signal-to-noise ratio, interference intensity, and desired signal direction.

[0045] Continuing with the example of intracardiac ultrasound imaging, the acquisition cycle can be the heartbeat cycle, and the length of the acquisition cycle can be determined based on the heart rate and sampling rate. For example, if the heart rate is 60 beats per minute, then one heartbeat cycle is 1 second, and if the ultrasound echo sampling rate is 10MHz, then the length of one acquisition cycle is 10M sampling points, meaning that 10M sampling points will be obtained within one acquisition cycle.

[0046] Furthermore, assuming the windows do not overlap, the sampling point data of the i-th window can be represented by the following expression:

[0047]

[0048] Among them, X i x(n) represents the echo acquisition information of the nth sampling point in the i-th window of the current acquisition period, and x(j) represents the echo acquisition information of the j-th sampling point in the current acquisition period, where j = n + (i-1)x. Indicates to Round down to the nearest integer. N is the length of the acquisition period, which is the total number of sampling points included in the acquisition period, and x is the length of the window.

[0049] For example, when the sampling frequency is 10MHz, N = 1.0 × 10 7 If a higher time and frequency resolution is desired, the window length can be set to 10,000, and x = 1.0 × 10⁻⁶. 4 This means that the data is acquired within a window of 10,000 sampling points. In this way, if the windows do not overlap, 1,000 windows can be included in one acquisition cycle, which can then be used for 1,000 beamforming processes, achieving high time and frequency resolution.

[0050] In other embodiments, the windows may also have a certain degree of overlap, meaning that some sampling points can appear in two or more windows, thereby increasing data utilization and spectral resolution. In this case, the sampling point data of the i-th window can be represented by the following expression:

[0051]

[0052] Among them, X i X(n) represents the echo acquisition information of the nth sampling point in the i-th window of the current acquisition period, and X(j) represents the echo acquisition information of the j-th sampling point in the current acquisition period, where j = n + (i-1)(1-r)x, and r is the window overlap rate. Indicates to Round down to the nearest integer. N is the length of the acquisition period, which is the total number of sampling points included in the acquisition period, and x is the length of the window.

[0053] In some embodiments, beamforming of the echo signal of the current window based on the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle can be represented by the following expression:

[0054]

[0055] Among them, Y i (n) represents the output obtained by beamforming the echo signal corresponding to the nth sampling point in the current window (i.e., the i-th window) during the current acquisition cycle. The weight matrix parameter W′ for the i-th window in the previous acquisition cycle. i The conjugate transpose of X i (n) represents the echo acquisition information of the nth sampling point in the current window (i.e., the i-th window) during the current acquisition cycle.

[0056] In some embodiments, the processing component 200 is further configured to: obtain the covariance matrix of the current window based on the echo acquisition information of the current window, and then combine it with the array steering vector to obtain the weight matrix parameters of the current window.

[0057] Specifically, the weight matrix parameters of each window are obtained through the covariance matrix and array steering vector of each window, and are achieved through the following expression:

[0058]

[0059] Among them, W i R represents the weight matrix parameters for the current window (i.e., the i-th window) in the current acquisition period; i Let be the covariance matrix of the current window (i.e., the i-th window) in the current acquisition cycle, d be the array steering vector, and R be the covariance matrix of the current window. i -1 The covariance matrix R i The inverse matrix, d H It is the conjugate transpose of the array guiding vector d.

[0060] Furthermore, the covariance matrix R of the i-th window in the current acquisition period i It can be determined by the following expression:

[0061]

[0062] Among them, X i (n) represents the echo acquisition information of the nth sampling point in the i-th window of the current acquisition cycle, which is an m-dimensional column vector; x is the length of the window; Let X be a matrix i The conjugate transpose of (n); R iis the covariance matrix of the current window (i.e. the i-th window) in the current acquisition cycle, and its size is m×m; m is the number of array elements included in the echo receiving unit of the ultrasonic transducer array component 101, and m is a positive integer.

[0063] Furthermore, the array steering vector d is determined by the following expression:

[0064] d = [1, e -j2πfDsinθ / c e -j2πf2Dsinθ / c , ..., e -j2πf(m-1)Dsinθ / c ] T ;

[0065] Where d is the array steering vector, f is the frequency of the ultrasonic echo, D is the spacing between the array elements in the echo receiving unit of the ultrasonic transducer array assembly 101, c is the propagation speed of the ultrasonic echo, and θ is the target direction angle. Here, the parameters affecting the array steering vector d are fixed values ​​or can be regarded as constant values ​​in this embodiment. Therefore, in this embodiment, the array steering vector d can be regarded as a fixed value and does not change with the window.

[0066] In other embodiments, such as Figure 3 As shown, the ultrasound assembly also includes a storage component 300. The storage component 300 is used to store echo acquisition information from multiple windows of multiple acquisition cycles prior to the current acquisition cycle. The number of acquisition cycles corresponding to the echo acquisition information stored in the storage component 300 is less than the total number of acquisition cycles after the probe has been stationary.

[0067] Accordingly, the processing component 200 is also configured to: take weights from the echo acquisition information of the current window in the current acquisition cycle and the echo acquisition information of the corresponding windows in multiple acquisition cycles stored, and process them to obtain the covariance matrix of the current window in the current acquisition cycle.

[0068] To help those skilled in the art better understand the implementation process described in the above embodiments, the corresponding processing procedures will be represented by relevant expressions below.

[0069] Specifically, the covariance matrix R of the current window (i.e., the i-th window) in the current acquisition period i It can be determined by the following expression:

[0070]

[0071] Among them, R i Let be the covariance matrix of the current window (i.e., the i-th window) in the current acquisition period. This refers to the echo acquisition information of the nth sampling point in the i-th window within the t-j+1 acquisition cycles prior to the current acquisition cycle, where j = t+1. This represents the echo acquisition information of the nth sampling point in the i-th window of the current acquisition cycle, where x is the length of the window and t represents the number of acquisition cycles prior to the current acquisition cycle stored in the storage component 300. for The conjugate transpose of the matrix, w ij The weighting coefficients are positive numbers and satisfy the condition. Its specific value can be determined according to different strategies, such as linear decay or exponential decay.

[0072] In other words, after the probe 100 has been stationary for more than one acquisition cycle, in each acquisition cycle, since the probe 100 did not move in the previous acquisition cycle, the previous acquisition cycle can provide a reference for the current acquisition cycle. Therefore, the covariance matrix of each window in the current acquisition cycle can refer to the covariance matrix of the corresponding window in the previous acquisition cycle. This is beneficial for more accurate and real-time determination of the covariance matrix of each window in the current acquisition cycle, which in turn is beneficial for accurately and real-time determination of the weight matrix parameters of each window in the current acquisition cycle, thereby improving the accuracy and real-time performance of ultrasound imaging.

[0073] In some embodiments, the weight of the acquisition period immediately adjacent to the current acquisition period is greater than the weight of the acquisition period farther away from the current acquisition period, i.e., w ij ≥w i(j+1) .

[0074] Understandably, acquisition cycles immediately adjacent to the current acquisition cycle are closer in time than those farther away, and generally have a stronger reference value and a greater impact on the current acquisition cycle. Therefore, assigning greater weight to acquisition cycles immediately adjacent to the current acquisition cycle is more consistent with reality. This results in a more accurate covariance matrix for the current window, which helps to further determine the weight matrix parameters of each window within the current acquisition cycle, thereby improving the accuracy of ultrasound imaging.

[0075] In some embodiments, the processing component 200 is further configured to: obtain the weight matrix parameters of each window in the first acquisition cycle by means of the echo acquisition information of each window in the first acquisition cycle when the probe 100 is stationary within the first acquisition cycle, and perform beamforming of the echo signal of each window according to the weight matrix parameters of each window in the first acquisition cycle.

[0076] In other words, for the first acquisition cycle, since the probe 100 was still moving in the previous acquisition cycle when the first acquisition cycle was stationary, the environmental changes caused by the probe 100 movement were significant. Therefore, the previous acquisition cycle could not provide a valid reference for the first acquisition cycle. Consequently, the weight matrix parameters for each window in the first acquisition cycle are obtained based on the echo acquisition information of each window within the first acquisition cycle, rather than using the echo acquisition information of the corresponding window from previous acquisition cycles. Simultaneously, the echo signal beamforming for each window also uses the weight matrix parameters of each window within the first acquisition cycle. This makes the weight matrix parameters for each window within the first acquisition cycle more accurate, improving the output quality within the first acquisition cycle and thus contributing to improved accuracy in ultrasound imaging.

[0077] Specifically, beamforming of the echo signal for each window based on the weight matrix parameters of each window within the first acquisition period can be represented by the following expression:

[0078]

[0079] Among them, Y i (n) represents the beamforming output of the echo signal from the nth sampling point in the i-th window during the first acquisition period. The weight matrix parameter W for the i-th window within the first acquisition period i The conjugate transpose of X i (n) represents the echo acquisition information of the nth sampling point in the i-th window within the first acquisition cycle.

[0080] The weight matrix parameters of each window can be obtained by using the covariance matrix and array steering vector of each window, which can be achieved by the following expression:

[0081]

[0082] Among them, W i R represents the weight matrix parameters for the i-th window in the first acquisition period. i Let be the covariance matrix of the i-th window in the first acquisition cycle, d be the array steering vector in the first acquisition cycle, and R be the... i -1 The covariance matrix R i The inverse matrix, d H This is the conjugate transpose of the array guide vector d. Similarly, here, the array guide vector d can also be considered a fixed value that does not change with the window.

[0083] Furthermore, the covariance matrix R of the i-th window in the first acquisition cycle i It can be determined by the following expression:

[0084]

[0085] Among them, X i (n) represents the echo acquisition information of the nth sampling point in the i-th window of the first acquisition cycle, which is an m-dimensional column vector; x is the length of the window; Let X be a matrix i The conjugate transpose of (n); m is the number of array elements included in the echo receiving unit of the ultrasonic transducer array assembly 101; R i Let m be the covariance matrix of the i-th window in the first acquisition cycle, and let m be a positive integer.

[0086] In some other embodiments, the processing component 200 is further configured to: convert the echo acquisition information in the current window to the time-frequency domain to obtain the time spectrum; process the time spectrum using a complex Gaussian mixture model to obtain the noise power spectral density; and obtain the weight matrix parameters of the current window based on the noise power spectral density and the array steering vector.

[0087] To help those skilled in the art better understand the above embodiments, the following will provide examples illustrating the specific implementation of the above process by the processing component 200.

[0088] First, within each window of the acquisition cycle, a Short-Time Fourier Transform (STFT) is performed on the echo acquisition information to obtain the time-spectrum matrix X representing the time spectrum. i (t, f), where t represents the time frame index, f represents the frequency channel index, and i represents the i-th window within the acquisition period. This embodiment does not impose any particular limitations on the method for obtaining the time-spectrum matrix. In some other embodiments, the time-spectrum matrix X... i (t, f) can also be implemented using algorithms such as Wavelet Transform (WT) and Wigner-Ville Distribution (WVD), which will not be elaborated on here.

[0089] Since the following iterative derivation process will occur within each window of each acquisition cycle, for clarity, X will be consistently referred to as X below. i (t, f) is written as X(t, f).

[0090] Next, a Complex Gaussian Mixture Model (CGMM) is used to model the target signal and noise for each frequency channel, obtaining the power spectral density (PSD) of the target signal and noise for each channel, i.e., the PSD of the target signal—φ. s (f) and PSD of noise — φ n (f).

[0091] Specifically, the power spectral density φ of the target signal and noise s (f) and φ n (f) Obtained by the following method:

[0092] The first step is to initialize φ using the following expression. s (f) and φ n (f):

[0093]

[0094] Where X(t, f) is the aforementioned time-frequency spectrum matrix, X H (t, f) is the conjugate transpose of X(t, f); T is the boundary of the time frame after time-frequency conversion, and t represents the t-th time frame.

[0095] Furthermore, initialize the prior probability P(s|f) of the target signal component in the f-th frequency channel of the t-th time frame, the prior probability P(n|f) of the noise component in the f-th frequency channel of the t-th time frame, and the mean vector μ of the target signal component in the f-th frequency channel of the t-th time frame. s (f) The mean vector μ of the noise component in the f-th frequency channel of the t-th time frame. n (f).

[0096] In this embodiment, the initial values ​​of the prior probabilities P(s|f) and P(n|f) are 0.5, and the mean vector μ is... s (f), μ n (f) The initial value is 0. In other embodiments, the above parameter may take other values.

[0097] The second step is to obtain the power spectral density φ of the noise through iterative EM method. n (f).

[0098] This embodiment does not impose any particular limitations on the specific steps of the EM method. The EM method includes an E-step and an M-step. In the E-step, the posterior probability P(s|X(t,f)) belonging to the target signal component in the f-th frequency channel of the t-th time frame and the posterior probability P(n|X(t,f)) = 1 - P(s|X(t,f)) belonging to the noise component in the f-th frequency channel of the t-th time frame are determined. In the M-step, based on the posterior probability obtained in the E-step, the prior probabilities P(n|f) and P(s|f), and the power spectral density φ are updated. s (f) and φ n (f) Mean vector μ s (f) and μ n (f) and other related parameters are used until convergence or the maximum number of iterations is reached. In this embodiment, the maximum number of iterations is 50.

[0099] Finally, the power spectral density φ of the noise n (f) is the covariance matrix of the current window in the f-th frequency channel at the t-th time frame, using the array steering vector d. f The power spectral density φ of the noise obtained in the second step n (f) Calculate the filter coefficients w of the Minimum Variance Distortionless Response (MVDR) beamformer. f That is, the weight matrix parameters of the t-th time frame and the f-th frequency channel.

[0100] Specifically, the weight matrix parameter w can be determined by the following expression. f :

[0101]

[0102] Where, d f The array guide vector for the current window. The array guide vector d for the current window f The conjugate transpose of φ; n (f) is the covariance matrix of the current window in the f-th frequency channel at the t-th time frame, w f The weight matrix parameters are for the echo acquisition information collected in the current window on the f-th frequency channel in the t-th time frame.

[0103] It should be noted that in this embodiment, the array guiding vector d f The method for obtaining is roughly the same as that for d mentioned above, and will not be repeated here. In other alternative embodiments, the spatial correlation matrix in the target direction can be calculated based on the time-frequency mask and the time-frequency spectrum, and the eigenvector of the spatial correlation matrix can be used as the array steering vector d. f Array guiding vector df This can maximize the signal in the target direction while minimizing the signal in other directions. For example, the posterior probability P(s|X(t,f)) of the target signal component in the f-th frequency channel of the t-th time frame obtained through M steps in the second step above can be used as a time-frequency mask, and the array steering vector d of the t-th time frame can be obtained through multiple iterations of the EM method. f Other methods can also be used, such as generating the corresponding time-frequency mask based on a neural network model. Similar to the above embodiment, the array steering vector d... f It can be considered a constant, and the array steering vector d is obtained in a certain time frame. f After that, the array steering vector d is no longer recalculated for each subsequent time frame. f To save on computational load.

[0104] Furthermore, based on the weight matrix parameters of all frequency channels in the t-th time frame, all frequency channels are processed and summed to obtain the beamformed output y. t .

[0105] Specifically, beamforming is performed on the components of all channels at time t of the current window according to the following expression, to obtain the output y of the echo signal of time frame t of the current window after beamforming. t :

[0106]

[0107] Among them, y t X(t, f) is the output obtained by beamforming in the t-th time frame of the current window; X(t, f) is the component of the echo acquisition information in the f-th frequency channel of the t-th time frame of the current window; F is the frequency boundary after time-frequency conversion; w′ f These are the weight matrix parameters for the window corresponding to the current window in the previous acquisition cycle, located on the f-th frequency channel of the t-th time frame. w′ f The conjugate transpose of .

[0108] Of course, the above are just specific examples of the MV beamforming algorithm. Other beamforming algorithms can be used in some embodiments, which will not be elaborated here.

[0109] In some embodiments, the processing component 200 is further configured to: when the probe 100 moves for more than a window period, the processing component 200 beamforms the echo signal of the current window according to the weight matrix parameters of the previous window, and obtains the weight matrix parameters of the current window according to the echo acquisition information of the current window, so as to beamform the echo signal of the next window according to the weight matrix parameters of the current window in the next window.

[0110] Similar to the above embodiments,

[0111]

[0112] Among them, Y i (n) represents the output of the nth sampling point in the current window (i.e., the i-th window) of the current acquisition cycle after beamforming. The weight matrix parameter W for the (i-1)th window i-1 The conjugate transpose of X i (n) represents the echo acquisition information of the nth sampling point in the current window (i.e., the i-th window).

[0113] Because the probe 100 is constantly moving, the environmental differences between the previous and current acquisition cycles are significant due to the movement of the probe 100. Therefore, the previous acquisition cycle cannot provide a valid reference for the current acquisition cycle. Consequently, the beamforming of the echo acquisition signal in each window is obtained based on the weight matrix parameters of adjacent windows, rather than using the weight matrix parameters of the window corresponding to the previous acquisition cycle. This effectively removes noise, improves the output quality of each window, and thus enhances the accuracy of ultrasound imaging.

[0114] In some embodiments, the processing component 200 is further configured such that the weight matrix parameters of the first window when the probe 100 is moving are taken from the weight matrix parameters of the last window of the last acquisition cycle when the probe 100 is stationary.

[0115] To help those skilled in the art better understand the movement of probe 100, the following will combine... Figure 4 Please provide an explanation.

[0116] like Figure 4 The electrocardiogram shown in the figure begins at time t, as indicated by the arrow. Probe 100 stops resting and starts moving, continuing for a certain period of time. At this time, the two dashed boxes after time t are the first and second windows, representing the start of the movement of probe 100 after it stops resting.

[0117] In some embodiments, such as Figure 3 As shown, the ultrasound device also includes a storage component 300 for storing echo acquisition information from multiple windows adjacent to the current window. The number of windows adjacent to the current window stored in the storage component 300 is less than the total number of windows after the probe 100 has moved. Clearly, "adjacent windows" here refers to windows preceding the current window.

[0118] In some embodiments, the processing component 200 is further configured to: obtain the covariance matrix of the current window by weighted post-processing based on the echo acquisition information of multiple stored windows and the echo acquisition information of the current window, and then obtain the weight matrix parameters of the current window.

[0119] The specific method is similar to the above embodiments:

[0120] The covariance matrix R of the current window i It can be determined by the following expression:

[0121]

[0122] Among them, R i Let X be the covariance matrix of the current window (i.e., the i-th window). j+it-1 (n) represents the echo acquisition information of the nth sampling point of the j+it-1th window preceding the i-th window, where x is the length of the window and t represents the number of windows preceding the current window stored in storage component 300. For X j+it-1 The conjugate transpose of (n), w j The weighting coefficients are positive numbers and satisfy the condition. Its specific value can be determined according to different strategies, such as linear decay or exponential decay.

[0123] The weight matrix parameter W of the current window i It can be obtained through the following expression:

[0124]

[0125] Among them, W i R represents the weight matrix parameters for the current window (i.e., the i-th window); i Let be the covariance matrix of the current window (i.e., the i-th window), d be the array steering vector, and R be the vector of the array. i -1 The covariance matrix R i The inverse matrix, d H It is the conjugate transpose of the array guiding vector d.

[0126] In some embodiments, the processing component 200 is further configured to: convert the echo acquisition information in the current window to the time-frequency domain to obtain the time spectrum; process the time spectrum using a complex Gaussian mixture model to obtain the noise power spectral density; and obtain the weight matrix parameters of the current window based on the noise power spectral density and the array steering vector.

[0127] Therefore, it can be seen that the process of determining the weight matrix parameters of the first acquisition cycle when the probe 100 is stationary for more than one acquisition cycle in this embodiment is roughly the same as that in the previous embodiments, and will not be described in detail here.

[0128] In some embodiments, such as Figure 5 As shown, the probe 100 also includes a position sensor 102, which is used to acquire the movement of the probe 100.

[0129] In this way, by setting a position sensor 102 in the probe 100, the movement of the probe 100 can be accurately sensed, providing a way to accurately determine the weight matrix parameters, which helps to improve the accuracy of the weight matrix parameters and thus improve the accuracy of ultrasound imaging.

[0130] When the probe 100 remains stationary for more than one acquisition cycle, the processing component 200, after acquiring the echo acquisition information of the current window within the current acquisition cycle, performs beamforming on the echo signal of the current window based on the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle. Simultaneously, it obtains the weight matrix parameters of the current window based on the echo acquisition information, so that in the next acquisition cycle, it can perform beamforming on the echo signal of the corresponding window in the next acquisition cycle based on the weight matrix parameters of the current window. In other words, by utilizing the periodic activity characteristics of human tissue and combining it with the motion state of the probe 100, more suitable weight matrix parameters are used to perform echo signal beamforming, thereby more effectively removing noise, improving the output quality of the current window, and further improving the accuracy of ultrasound imaging. Furthermore, due to the periodic activity of human tissues, when the probe 100 is stationary for more than one acquisition cycle, its noise also exhibits a specific periodic change. Therefore, the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle can be used as the weight matrix parameters of the current window, with minimal impact on the accuracy of beamforming. This eliminates the need to wait for the determination of the weight matrix parameters of the current window, thereby improving the beamforming efficiency of the current window, reducing processing latency, and enhancing the real-time performance of ultrasound imaging.

[0131] Another aspect of this application embodiment also provides an ultrasonic system, such as Figure 6 As shown, it includes: an ultrasonic device 1 and a monitoring device 2.

[0132] The ultrasound device 1 is the ultrasound device provided in any of the above embodiments; the monitoring device 2 is used to detect the activity cycle of human tissue, so that the processing component 200 of the ultrasound device 1 adjusts the acquisition cycle according to the activity cycle of human tissue.

[0133] It should be noted that the above embodiments do not limit the activity cycle of human tissues. It can be determined according to different human tissues. For example, when the human tissue is the heart chamber, the activity cycle of the human tissue can be the heartbeat cycle. When the human tissue is the lungs, the activity cycle of the human tissue can be the human respiratory cycle, etc., which will not be elaborated here.

[0134] It is not difficult to see that this embodiment is a system embodiment corresponding to the device embodiment, and this embodiment can be implemented in conjunction with the device embodiment. The relevant technical details mentioned in the device embodiment are still valid in this embodiment, and will not be repeated here to reduce repetition. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the device embodiment.

[0135] The above division of the composition structure of various devices and systems is only for clear description. In practice, they can be combined into one unit or component, or some units or components can be split into multiple sub-units or sub-components. As long as they include the same functional relationship and connection relationship, they are all within the scope of protection of this patent. Adding insignificant modifications or introducing insignificant designs to the ultrasonic device, but without changing its core design of function and structure, are all within the scope of protection of this patent.

[0136] Furthermore, in order to highlight the innovative aspects of this application, no units that are not closely related to solving the technical problems proposed in this application are introduced in this embodiment, but this does not mean that there are no other units in this embodiment.

[0137] Another aspect of this application embodiment provides a method for beamforming ultrasonic echoes, applied to the processing components involved in any of the above embodiments. The flow of the ultrasonic echo beamforming method is as follows: Figure 7 As shown, it includes:

[0138] Step 701: When the probe remains stationary for more than one acquisition cycle, acquire the echo acquisition information of the human tissue in the current window within the current acquisition cycle.

[0139] Step 702: Beamforming of the echo signal of the current window according to the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle, and obtaining the weight matrix parameters of the current window according to the echo acquisition information of the current window, so as to beamforming of the echo signal of the corresponding window in the next acquisition cycle according to the weight matrix parameters of the current window.

[0140] The acquisition cycle includes at least one window.

[0141] It is not difficult to see that this embodiment is a method embodiment corresponding to the device embodiment, and this embodiment can be implemented in conjunction with the device embodiment. The relevant technical details mentioned in the device embodiment are still valid in this embodiment, and will not be repeated here to reduce repetition. Correspondingly, the relevant technical details mentioned in this embodiment can also be applied to the device embodiment.

[0142] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the scope of protection of this patent. Adding insignificant modifications or introducing insignificant designs to the algorithm or process, but without changing the core design of the algorithm and process, are also within the scope of protection of this patent.

[0143] Another aspect of this application embodiment also provides an electronic device, such as... Figure 8 As shown, it includes: at least one processor 801; and a memory 802 communicatively connected to at least one processor 801; wherein the memory 802 stores instructions executable by at least one processor 801, the instructions being executed by at least one processor 801 to enable at least one processor 801 to perform the ultrasonic echo beamforming method described in any of the above method embodiments.

[0144] The memory 802 and processor 801 are connected via a bus, which can include any number of interconnecting buses and bridges. The bus connects various circuits of one or more processors 801 and memory 802 together. The bus can also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 801 is transmitted over a wireless medium via an antenna, which further receives data and transmits it to processor 801.

[0145] The processor 801 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. The memory 802 can be used to store data used by the processor 801 during operation.

[0146] Another aspect of this application provides a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the above-described method embodiments.

[0147] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, magnetic tapes, or optical discs.

[0148] Those skilled in the art will understand that the above embodiments are specific embodiments for implementing this application, and in practical applications, various changes can be made to them in form and detail without departing from the spirit and scope of this application.

Claims

1. An ultrasonic device, characterized in that, include: The probe includes an ultrasonic transducer array assembly for emitting ultrasound and acquiring echo signals from human tissue. The processing component is configured as follows: When the probe remains stationary for more than one acquisition cycle, the processing component acquires the echo acquisition information of the current window within the current acquisition cycle, and performs beamforming on the echo signal of the current window according to the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle. At the same time, it obtains the weight matrix parameters of the current window based on the echo acquisition information of the current window, so as to perform beamforming on the echo signal of the window corresponding to the next acquisition cycle according to the weight matrix parameters of the current window. The acquisition cycle includes at least one window.

2. The ultrasonic device according to claim 1, characterized in that, The processing component is further configured to: When the probe is stationary for a period of time within the first acquisition cycle, the weight matrix parameters of each window in the first acquisition cycle are obtained through the echo acquisition information of each window in the first acquisition cycle, and the echo signal beamforming of each window is performed according to the weight matrix parameters of each window in the first acquisition cycle.

3. The ultrasonic device according to claim 1, characterized in that, The processing component is further configured to: The covariance matrix of the current window is obtained based on the echo acquisition information of the current window, and the weight matrix parameters of the current window are obtained by combining the array steering vector.

4. The ultrasonic device according to claim 1, characterized in that, It also includes a storage component for storing echo acquisition information of multiple windows of multiple acquisition cycles prior to the current acquisition cycle. The number of acquisition cycles corresponding to the echo acquisition information stored in the storage component is less than the total number of acquisition cycles after the probe has been stationary. The processing component is further configured to: Weights are assigned to the echo acquisition information of the current window within the current acquisition period and the echo acquisition information of the corresponding windows of multiple acquisition periods stored in the memory, and the covariance matrix of the current window in the current acquisition period is obtained.

5. The ultrasonic device according to claim 4, characterized in that, The weight of the acquisition cycle that is immediately adjacent to the current acquisition cycle is greater than the weight of the acquisition cycle that is far away from the current acquisition cycle.

6. The ultrasonic device according to claim 1, characterized in that, The processing component is further configured to: The echo acquisition information within the current window is converted to the time-frequency domain to obtain the time spectrum; the time spectrum is processed using a complex Gaussian mixture model to obtain the noise power spectral density; and the weight matrix parameters of the current window are obtained based on the noise power spectral density and the array steering vector.

7. The ultrasonic device according to claim 1, characterized in that, The processing component is further configured to: When the probe moves for more than one window, the processing component beamforms the echo signal of the current window according to the weight matrix parameters of the previous window, and obtains the weight matrix parameters of the current window according to the echo acquisition information of the current window, so as to beamform the echo signal of the next window according to the weight matrix parameters of the current window in the next window.

8. The ultrasonic device according to claim 7, characterized in that, The processing component is further configured to: The weight matrix parameters of the first window when the probe is moving are taken from the weight matrix parameters of the last window of the last acquisition cycle when the adjacent probe is stationary.

9. The ultrasonic device according to claim 7, characterized in that, It also includes a storage component for storing echo acquisition information of multiple windows adjacent to the current window. The number of windows adjacent to the current window stored in the storage component is less than the total number of windows after the probe moves. The processing component is further configured to: The covariance matrix of the current window is obtained by weighting and processing the echo acquisition information from multiple stored windows and the echo acquisition information of the current window, so as to obtain the weight matrix parameters of the current window.

10. The ultrasonic device according to claim 7, characterized in that, The processing component is further configured to: The echo acquisition information within the current window is converted to the time-frequency domain to obtain the time spectrum; the time spectrum is processed using a complex Gaussian mixture model to obtain the noise power spectral density; and the weight matrix parameters of the current window are obtained based on the noise power spectral density and the array steering vector.

11. The ultrasonic device according to claim 1, characterized in that, The probe also includes a position sensor for acquiring the movement of the probe.

12. An ultrasonic system, characterized in that, include: The ultrasonic device as described in any one of claims 1 to 11; Monitoring devices are used to detect the activity cycles of human tissues. The processing component of the ultrasound device adjusts the acquisition cycle according to the activity cycle of the human tissue.

13. A method for beamforming ultrasonic echoes, characterized in that, include: When the probe remains stationary for more than one acquisition cycle, the echo acquisition information of the human tissue in the current window within the current acquisition cycle is obtained. Beamforming of the echo signal of the current window is performed based on the weight matrix parameters of the window corresponding to the current window in the previous acquisition cycle. At the same time, the weight matrix parameters of the current window are obtained based on the echo acquisition information of the current window, so as to perform beamforming of the echo signal of the window corresponding to the next acquisition cycle based on the weight matrix parameters of the current window in the next acquisition cycle. The acquisition cycle includes at least one window.

14. An electronic device, characterized in that, include: At least one processor; as well as, A memory communicatively connected to the at least one processor; wherein, The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the beamforming method for ultrasonic echoes as described in claim 13.

15. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the beamforming method for ultrasonic echoes as described in claim 13.

Images