A method of measuring a parameter of an activity of a tissue and related apparatus

Abstract

The application discloses a kind of tissue activity parameter measurement method and related equipment, the method includes: obtaining the image structure information of target tissue, establishes tissue contraction activity model according to image structure information, converts linear instantaneous model into tissue contraction activity model;Obtain observation matrix in linear instantaneous model, obtain multiple sub-observation matrices by segmentation to observation matrix, carry out whitening processing to multiple sub-observation matrices, obtain multiple independent activity signal sources;According to the classification method of preestablished, multiple independent activity signal sources are classified and screened, obtain target signal, obtain tissue activity parameter according to target signal, according to tissue activity parameter, the contraction activity of the target tissue is evaluated;The application is combined with independent component analysis, and the contraction activity of tissue is quantitatively analyzed, can accurately extract tissue activity parameter from ultrasonic video, and can be realized in ordinary imaging platform, and the cost is low.

Description

Technical Field

[0001] This application relates to the field of image processing technology, and in particular to a method, system, terminal, and computer-readable storage medium for measuring tissue activity parameters. Background Technology

[0002] Tissue activity parameter measurement typically refers to the automated measurement and analysis of changes in muscle tissue during contraction using technologies such as image processing, machine learning, and deep learning. This includes measuring changes in muscle thickness, pendant angle, muscle bundle length, and other related parameters. Tissue activity parameter measurement provides detailed and accurate data on muscle contractile performance, which is crucial for understanding muscle health and function. In the fields of sports and rehabilitation, the introduction of muscle information helps athletes assess their muscle performance and optimize training plans to prevent sports injuries; it also helps those undergoing rehabilitation to evaluate the recovery process and the effectiveness of rehabilitation training after injury.

[0003] However, the current methods for measuring relevant activity parameters of muscle tissue are diverse. Non-invasive instruments and equipment for measuring tissue physical properties are expensive, usually large and difficult to carry, and have high requirements for image quality, resulting in inaccurate measurement of tissue activity parameters.

[0004] Therefore, existing technologies still need to be improved and developed. Summary of the Invention

[0005] The main objective of this application is to provide a method, system, terminal, and computer-readable storage medium for measuring tissue activity parameters, aiming to solve the problems in the prior art where quantitative analysis of tissues is sensitive to noise in images, requires expensive equipment, and has high requirements for image quality, resulting in inaccurate measurement of tissue activity parameters.

[0006] To achieve the above objectives, this application provides a method for measuring organizational activity parameters, which includes the following steps:

[0007] Obtain image structural information of the target tissue, establish a tissue contraction activity model based on the image structural information, and transform the tissue contraction activity model into a linear transient model;

[0008] Obtain the signal matrix in the linear instantaneous model, divide the signal matrix into sub-signal matrices, and expand the sub-signal matrices to obtain the observation matrix of the sub-signal matrices;

[0009] The observation matrix is ​​whitened, a second-order covariance matrix is ​​constructed based on the whitened observation matrix, a mixture matrix is ​​obtained based on the second-order covariance matrix, and the mixture matrix is ​​separated to obtain multiple independent active signal sources;

[0010] Multiple independent activity signal sources are classified and filtered according to a preset classification method to obtain target signals. Tissue activity parameters are obtained based on the target signals, and the contraction activity of the target tissue is evaluated based on the tissue activity parameters.

[0011] Optionally, the method for measuring tissue activity parameters, wherein acquiring image structural information of the target tissue and establishing a tissue contraction activity model based on the image structural information specifically includes:

[0012] The structural information of the target tissue is obtained by acquiring images or videos of the tissue through an imaging system. This image structural information is represented as multiple window functions. A tissue contraction activity model is obtained by modeling the sum of the convolutions of the multiple window functions with their respective impulse responses.

[0013]

[0014] Where, x i (t) represents the evolution of the axial displacement of the i-th pixel in the image with time t, where t represents time, n represents the number of activity sources causing target tissue contraction due to physiological response, k represents the number of sparse sources of non-target tissue activity, and s j (tl) represents the j-th source as a function of time tl, h ij (l) represents the duration of the impulse response of the i-th pixel and the j-th source, ω. i (t) is the additional white noise at the i-th pixel.

[0015] Optionally, in the method for measuring tissue activity parameters, the step of converting the tissue contraction activity model into a linear instantaneous model specifically includes:

[0016] Obtain the delayed sample for each window function, and based on each window function and the corresponding delayed sample, transform the tissue contraction activity model into a linear transient model:

[0017]

[0018] in,

[0019]

[0020]

[0021] in, A matrix representation representing the signals received by all channels. Represents a mixture matrix. This represents all sources of organizational activity signals to be estimated. x represents the additional white noise on the source to be estimated. m (t) represents the matrix representation on the m-th source, ω m (t) represents the additional white noise on the m-th source, L refers to the delay unit of the signal source, R refers to the delay unit of the measurement, and T represents the matrix transpose.

[0022] Optionally, the method for measuring organizational activity parameters, wherein obtaining the signal matrix in the linear instantaneous model, segmenting the signal matrix to obtain sub-signal matrices, and expanding the sub-signal matrices to obtain the observation matrix, specifically includes:

[0023] Obtain the signal matrix in the linear instantaneous model. The signal matrix Segmentation is performed, and the segmented signal matrix is ​​processed. Vectorization and normalization are performed to obtain several sub-signal matrices;

[0024] The sub-signal matrix is ​​expanded to obtain multiple delayed versions of the observation matrix.

[0025] Optionally, in the method for measuring organizational activity parameters, the whitening process of the observation matrix specifically includes:

[0026] Calculate the covariance matrix of the observation matrix, and perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalue matrix and the first eigenvector;

[0027] Multiplying the inverse square root of the eigenvalue matrix by the first eigenvector yields the whitening transformation matrix. Multiplying the observation matrix by the whitening transformation matrix yields the whitened observation matrix.

[0028] Optionally, the method for measuring tissue activity parameters, wherein constructing a second-order covariance matrix based on the whitened observation matrix, obtaining a mixture matrix based on the second-order covariance matrix, and separating the mixture matrix to obtain independent activity signal sources, specifically includes:

[0029] Calculate the second-order covariance matrix of the whitened observation matrix, and perform eigenvalue decomposition on the second-order covariance matrix to obtain the second eigenvector;

[0030] The mixing matrix is ​​calculated based on the second eigenvector, and the mixing matrix is ​​separated by an independent component analysis algorithm to obtain multiple independent active signal sources.

[0031] Optionally, the method for measuring organizational activity parameters, wherein classifying and filtering multiple independent activity signal sources according to a preset classification method to obtain a target signal specifically includes:

[0032] The multiple independent active signal sources are converted into multiple time-domain signals and multiple frequency signals;

[0033] The multiple time-domain signals and multiple frequency signals are compared with a preset standard, and the independent active signal sources corresponding to the time-domain signals and frequency signals that conform to the preset standard are separated from the multiple independent active signal sources to obtain the target signal.

[0034] Furthermore, to achieve the above objectives, this application also provides an organizational activity parameter measurement system, wherein the organizational activity parameter measurement system includes:

[0035] The tissue model building module is used to acquire image structural information of the target tissue, build a tissue contraction activity model based on the image structural information, and convert the tissue contraction activity model into a linear transient model.

[0036] The observation matrix acquisition module is used to acquire the signal matrix in the linear instantaneous model, divide the signal matrix into sub-signal matrices, and expand the sub-signal matrices to obtain the observation matrix of the sub-signal matrices;

[0037] The activity signal source acquisition module is used to whiten the observation matrix, construct a second-order covariance matrix based on the whitened observation matrix, obtain a mixing matrix based on the second-order covariance matrix, and separate the mixing matrix to obtain multiple independent activity signal sources.

[0038] The tissue activity parameter measurement module is used to classify and filter multiple independent activity signal sources according to a preset classification method to obtain target signals, obtain tissue activity parameters based on the target signals, and evaluate the contraction activity of the target tissue based on the tissue activity parameters.

[0039] In addition, to achieve the above objectives, this application also provides a terminal, wherein the terminal includes: a memory, a processor, and a tissue activity parameter measurement program stored in the memory and executable on the processor, wherein when the tissue activity parameter measurement program is executed by the processor, it implements the steps of the tissue activity parameter measurement method as described above.

[0040] In addition, to achieve the above objectives, this application also provides a computer-readable storage medium, wherein the computer-readable storage medium stores an organizational activity parameter measurement program, which, when executed by a processor, implements the steps of the organizational activity parameter measurement method as described above.

[0041] In this application, image structural information of the target tissue is obtained, a tissue contraction activity model is established based on the image structural information, and the tissue contraction activity model is transformed into a linear transient model; the signal matrix in the linear transient model is obtained, the signal matrix is ​​segmented to obtain a sub-signal matrix, and the sub-signal matrix is ​​expanded to obtain an observation matrix of the sub-signal matrix; the observation matrix is ​​whitened to obtain multiple independent activity signal sources; the multiple independent activity signal sources are classified and filtered according to a preset classification method to obtain the target signal, tissue activity parameters are obtained based on the target signal, and the contraction activity of the target tissue is evaluated based on the tissue activity parameters; this application combines independent component analysis to perform quantitative analysis of tissue contraction activity, can accurately extract tissue activity parameters from ultrasound video, and can be implemented on a common imaging platform at a low cost. Attached Figure Description

[0042] Figure 1 This is a flowchart of a preferred embodiment of the method for measuring organizational activity parameters according to this application;

[0043] Figure 2 This is a schematic diagram of a preferred embodiment of the activity parameter measurement system of this application;

[0044] Figure 3 This is a schematic diagram of the operating environment of a preferred embodiment of the terminal of this application. Detailed Implementation

[0045] This application provides a method and related equipment for measuring organizational activity parameters. To make the purpose, technical solution, and effects of this application clearer and more explicit, the following detailed description is provided with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only for explaining this application and are not intended to limit this application.

[0046] To address the issues mentioned in the background section regarding the sensitivity to noise in images, the need for expensive equipment, and the high image quality requirements in tissue quantitative analysis, which lead to inaccurate measurements of tissue activity parameters, this application provides a method for measuring tissue activity parameters. This method, based on independent component analysis, proposes a way to obtain parameter information such as feather angle, muscle thickness, and muscle strength from ultrasound videos of tissue contraction activity. This achieves stable and smooth parameter quantification with high real-time performance. By obtaining several ultrasound signals related to tissue contraction activity, it can provide information on the state changes of the tissue during contraction, thereby enabling automatic quantitative analysis of tissue contraction activity. This method can be implemented on a common ultrasound imaging platform, is simple to operate, low in cost, and requires no external equipment. Therefore, it solves the problems of inaccurate tissue activity parameter measurements caused by the sensitivity to noise in images, the need for expensive equipment, and the high image quality requirements in related technologies for tissue quantitative analysis.

[0047] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless specifically defined as herein.

[0048] Furthermore, if the embodiments of this application involve descriptions such as "first" or "second," these descriptions are for descriptive purposes only and should not be construed as indicating or implying their relative importance or implicitly specifying the number of technical features indicated. Therefore, features defined with "first" or "second" may explicitly or implicitly include at least one of those features. Additionally, the technical solutions of various embodiments can be combined with each other, but this must be based on the ability of those skilled in the art to implement them. If the combination of technical solutions is contradictory or impossible to implement, it should be considered that such a combination of technical solutions does not exist and is not within the scope of protection claimed in this application.

[0049] The method for measuring tissue activity parameters described in the preferred embodiment of this application, such as Figure 1 As shown, the method for measuring tissue activity parameters includes the following steps:

[0050] Step S100: Obtain image structure information of the target tissue, establish a tissue contraction activity model based on the image structure information, and convert the tissue contraction activity model into a linear instantaneous model.

[0051] It is understood that this application utilizes an ultrasound imaging system (any ordinary ultrasound imaging system will suffice) to acquire images or videos of tissues (in addition to video analysis, images of consecutive frames are also applicable to this application), and extracts image structural information of the tissues from the images or videos. The reason for employing an ultrasound imaging system is that its high resolution and real-time imaging capabilities are highly compatible with those of this application, enabling precise analysis of tissue structure and function. Furthermore, the non-invasiveness and flexibility of ultrasound imaging further enhance the application potential of this application, providing in-depth insights into dynamic changes in tissues.

[0052] Furthermore, the step of acquiring the image structural information of the target tissue and establishing a tissue contraction activity model based on the image structural information specifically includes:

[0053] The structural information of the target tissue is obtained by acquiring images or videos of the tissue through an imaging system. This image structural information is represented as multiple window functions. A tissue contraction activity model is obtained by modeling the sum of the convolutions of the multiple window functions with their respective impulse responses.

[0054]

[0055] Where, x i (t) represents the evolution of the axial displacement of the i-th pixel in the image with time t, where t represents time, n represents the number of activity sources causing target tissue contraction due to physiological response, k represents the number of sparse sources of non-target tissue activity (e.g., periodic activity of other organs, such as vascular pulsation, interference from other muscles, etc.), and s j (tl) represents the j-th source as a function of time tl, h ij (l) represents the duration of the impulse response of the i-th pixel and the j-th source, ω. i (t) is the additional white noise at the i-th pixel.

[0056] It is understood that the images or videos in the target tissue show tissue activities caused by physiological responses (such as muscle twitching, muscle fiber elongation, etc.), which can be modeled as a series of window functions (including impulse functions and other special cases). The obtained ultrasound data can be considered as the sum of the convolutions of these unit window functions with their impulse responses. The tissue contraction activity of physiological responses can be modeled based on the sum of the convolutions of a series of window functions with the impulse responses corresponding to each window function.

[0057] Furthermore, in the considered tissue contraction activities, the signal sources of these n target tissue contraction activities are represented by a series of unit window functions, which are represented as continuous signals in the time domain. Meanwhile, for the same tissue contraction activity, we assume that the window functions are the same for each tissue contraction activity; in order to identify the n+k possible tissue activity signal sources in the tissue contraction activity model (where n represents the number of activity sources causing target tissue contraction due to physiological responses, and k is the number of sparse sources of non-target tissue activities), the tissue contraction activity model needs to be reconstructed into a standard linear independent component analysis model.

[0058] In order to enable independent component analysis to be used on the tissue contraction activity model, this application considers the process of reconstructing a standard linear independent component analysis model as a linear instantaneous combination consisting of a new set of window functions for tissue contraction activities, which includes the window functions of the original tissue contraction activities and their delayed versions.

[0059] The reason for adding a delayed version of the window function is that the duration of the tissue contraction process exceeds the average interval within the tissue contraction activity interval, which directly affects the recognition effect. Furthermore, the signal-to-noise ratio of actual tissue contraction activity is relatively high.

[0060] Therefore, assuming that the window function of tissue contraction activity is delayed by L samples (L samples refer to the number of other tissue contraction activity window functions initiated within the duration of the window function of a single tissue contraction activity), and the observation data is delayed by at most R samples, the linear instantaneous model can be obtained.

[0061] The transformation of the tissue contraction activity model into a linear transient model specifically includes:

[0062] Obtain the delayed sample for each window function, and based on each window function and the corresponding delayed sample, transform the tissue contraction activity model into a linear transient model:

[0063]

[0064] in,

[0065]

[0066]

[0067] in, A matrix representation representing the signals received by all channels. Represents a mixture matrix. This represents all sources of organizational activity signals to be estimated. x represents the additional white noise on the source to be estimated. m (t) represents the matrix representation on the m-th source, ω m (t) represents the additional white noise on the m-th source, L refers to the delay unit of the signal source, R refers to the delay unit of the measurement, and T represents the matrix transpose.

[0068] Step S200: Obtain the signal matrix in the linear instantaneous model, divide the signal matrix into sub-signal matrices, and expand the sub-signal matrices to obtain the observation matrix of the sub-signal matrices.

[0069] Specifically, the signal matrix in the linear instantaneous model is obtained. The signal matrix Segmentation is performed, and the segmented signal matrix is ​​processed. Vectorization and normalization are performed to obtain several sub-signal matrices.

[0070] The sub-signal matrix is ​​expanded to obtain multiple delayed versions of the observation matrix.

[0071] It is understandable that the application requests the signal matrix in the linear instantaneous model. The process is divided into multiple networks (sub-matrices). The specific segmentation process needs to be determined based on the actual situation. For example, in a typical ultrasound case, the process can be divided into sub-matrices of size 10×10 pixels, corresponding to 3×3mm. The sub-matrices are then vectorized into 2D and standardized.

[0072] The signal matrix in the linear instantaneous model After segmenting into multiple sub-matrices, all subsequent processing is performed within the reduced map region. This approach offers two advantages: first, it reduces the number of expected sources in the region, thus simplifying subsequent source separation; second, considering that the next step is to expand the observations by a factor of R, limiting the number of observations reduces the size of the expanded matrix, thereby reducing the computational load.

[0073] Furthermore, according to the formula given above:

[0074] It can be seen that the sub-signal matrix This can be extended to obtain multiple delayed versions of the observation matrix, which includes x1(t), x1(t-1), ..., x1(t-R+1), ..., x m (t),x m (t-1),…,x m (t-R+1)] T .

[0075] Step S300: Whiten the observation matrix, construct a second-order covariance matrix based on the whitened observation matrix, obtain a mixing matrix based on the second-order covariance matrix, separate the mixing matrix to obtain multiple independent active signal sources.

[0076] Specifically, the covariance matrix of the observation matrix is ​​calculated, and the covariance matrix is ​​decomposed into eigenvalues ​​to obtain the eigenvalue matrix and the first eigenvector.

[0077] Multiplying the inverse square root of the eigenvalue matrix by the first eigenvector yields the whitening transformation matrix. Multiplying the observation matrix by the whitening transformation matrix yields the whitened observation matrix.

[0078] It is understood that the whitening process in this application is as follows: calculate the covariance matrix of the data, perform eigenvalue decomposition on the covariance matrix, obtain an eigenvalue matrix and a first eigenvector, multiply the square root of the inverse of the eigenvalue matrix by the first eigenvector to obtain the whitening transformation matrix, and finally multiply the observation matrix (original data) by the whitening transformation matrix to obtain the whitened observation matrix. Whitening can reduce data redundancy, increase information content, and help to better capture patterns and structures in the data.

[0079] Furthermore, the step of constructing a second-order covariance matrix based on the whitened observation matrix, obtaining a mixing matrix based on the second-order covariance matrix, and separating the mixing matrix to obtain independent active signal sources specifically includes:

[0080] Calculate the second-order covariance matrix of the whitened observation matrix, and perform eigenvalue decomposition on the second-order covariance matrix to obtain the second eigenvector.

[0081] The mixing matrix is ​​calculated based on the second eigenvector, and the mixing matrix is ​​separated by an independent component analysis algorithm to obtain multiple independent active signal sources.

[0082] Understandably, based on the whitened observation matrix, a second-order covariance matrix of the whitened observation matrix is ​​constructed. By performing eigenvalue decomposition on the second-order covariance matrix, a second eigenvector of the second-order covariance matrix can be obtained. Then, based on the second eigenvector, the mixing matrix is ​​estimated. This mixing matrix describes the mixing pattern of signals in the observation data. Using the estimated mixing matrix, an independent component analysis algorithm (in signal processing, independent component analysis is a computational method used to separate multivariate signals into additive components) is applied to separate the mixing matrix, obtaining the original multiple independent active signal sources.

[0083] Furthermore, in this application, after applying the independent component analysis algorithm to separate the mixture matrix and obtain the original multiple independent active signal sources, another independent component analysis can be set up. This independent component analysis can further clarify the estimation of the tissue signal source in an iterative form, thereby improving the accuracy of the estimation.

[0084] It should be noted that this application considers ultrasound video as a collection of data from multiple channels. Therefore, it is possible to combine independent component analysis to separate signals with actual physical meaning from the video and use them for the quantification of tissue contraction activity.

[0085] Furthermore, this application can also use Principal Component Analysis (PCA). Compared to Independent Component Analysis, PCA focuses more on reducing the dimensionality of the data and retaining the most important components. PCA can serve as an alternative method for processing medical images and extracting key features, and is particularly suitable for high-dimensional datasets.

[0086] Step S400: Classify and filter multiple independent activity signal sources according to a preset classification method to obtain target signals, obtain tissue activity parameters based on the target signals, and evaluate the contraction activity of the target tissue based on the tissue activity parameters.

[0087] Specifically, the multiple independent active signal sources are converted into multiple time-domain signals and multiple frequency signals.

[0088] The multiple time-domain signals and multiple frequency signals are compared with a preset standard, and the independent active signal sources corresponding to the time-domain signals and frequency signals that conform to the preset standard are separated from the multiple independent active signal sources to obtain the target signal.

[0089] It is understandable that the multiple independent activity signal sources are further screened and classified. First, the multiple independent activity signal sources are converted into multiple time-domain signals and multiple frequency signals. The output obtained in the application is a continuous time signal. For integrable time series, the independent activity signal sources are converted into time-domain signals and frequency-domain signals through corresponding methods (e.g., fast Fourier transform).

[0090] Multiple time-domain signals and multiple frequency signals are compared with a preset standard. Here, the multiple time-domain signals and multiple frequency signals are compared with the industry gold standard. In the time-domain signals, kurtosis and waveform are used to determine the classification effect, and in the frequency-domain signals, power spectral density and frequency components are used to determine the classification effect, thus achieving the desired classification effect. The independent active signal sources corresponding to the time-domain signals and frequency signals that meet the preset standard are separated from the multiple independent active signal sources to obtain the target signal.

[0091] Furthermore, tissue activity parameters are obtained based on the target signal, and the contractile activity of the target tissue is evaluated based on the tissue activity parameters. For example, the final separated signal is used as an automatic quantification of tissue contractile activity. The final output signal includes time-series signals including structural parameters (muscle thickness changes, feather angle changes, etc.) and other parameters (such as muscle strength, joint angles), and these signals are used for the evaluation of tissue contractile activity.

[0092] As can be seen, this application uses independent component analysis to quantify tissue contraction activity. Specifically, it extracts temporal signals related to tissue contraction activity, including changes in feather angle, muscle thickness, muscle fiber length, and muscle strength, from ultrasound video information for quantification. The method in this application provides a high-precision, automated solution capable of effectively extracting multiple temporal signals related to tissue contraction from ultrasound video. This method not only improves the accuracy and reliability of the analysis but also separates signals with practical physical meaning from multi-channel data, offering higher interpretability compared to deep learning-related methods. The extraction of motion unit information using this method is of significant value for sports medicine, rehabilitation therapy, and biomechanical research. Through this method, physicians can obtain more accurate information on tissue activity, thereby developing more effective treatment plans. It also provides researchers with a powerful tool for understanding muscle tissue movement and changes, thus possessing broad application potential in medical and scientific research. Furthermore, compared to existing methods, the method in this application has relatively lower requirements for image quality, and can be used with ordinary ultrasound equipment for the quantitative analysis of tissue contraction activity while maintaining accuracy.

[0093] Furthermore, such as Figure 2 As shown, based on the above-described method for measuring organizational activity parameters, this application also provides a system for measuring organizational activity parameters, wherein the system includes:

[0094] The tissue model building module 51 is used to acquire image structural information of the target tissue, build a tissue contraction activity model based on the image structural information, and convert the tissue contraction activity model into a linear transient model.

[0095] The observation matrix acquisition module 52 is used to acquire the signal matrix in the linear instantaneous model, divide the signal matrix to obtain sub-signal matrices, and expand the sub-signal matrices to obtain the observation matrix of the sub-signal matrices;

[0096] The activity signal source acquisition module 53 is used to whiten the observation matrix, construct a second-order covariance matrix based on the whitened observation matrix, obtain a mixing matrix based on the second-order covariance matrix, and separate the mixing matrix to obtain multiple independent activity signal sources.

[0097] The tissue activity parameter measurement module 54 is used to classify and filter multiple independent activity signal sources according to a preset classification method to obtain target signals, obtain tissue activity parameters based on the target signals, and evaluate the contraction activity of the target tissue based on the tissue activity parameters.

[0098] Furthermore, such as Figure 3 As shown, based on the above-mentioned method and system for measuring organizational activity parameters, this application also provides a terminal, which includes a processor 10, a memory 20 and a display 30. Figure 3 Only some of the terminal components are shown; however, it should be understood that it is not required to implement all of the components shown, and more or fewer components may be implemented instead.

[0099] In some embodiments, the memory 20 may be an internal storage unit of the terminal, such as a hard disk or memory. In other embodiments, the memory 20 may be an external storage device of the terminal, such as a plug-in hard disk, smart media card (SMC), secure digital card (SD), flash card, etc. Further, the memory 20 may include both internal and external storage devices. The memory 20 is used to store application software and various types of data installed on the terminal, such as program code installed on the terminal. The memory 20 can also be used to temporarily store data that has been output or will be output. In one embodiment, the memory 20 stores an organizational activity parameter measurement program 40, which can be executed by the processor 10 to implement the organizational activity parameter measurement method of this application.

[0100] In some embodiments, the processor 10 may be a central processing unit (CPU), a microprocessor, or other data processing chip, used to run program code stored in the memory 20 or process data, such as executing the tissue activity parameter measurement method.

[0101] In some embodiments, the display 30 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. The display 30 is used to display information on the terminal and to display a visual user interface. The components 10-30 of the terminal communicate with each other via a system bus.

[0102] In one embodiment, when the processor 10 executes the tissue activity parameter measurement program 40 in the memory 20, the following steps are performed:

[0103] Obtain image structural information of the target tissue, establish a tissue contraction activity model based on the image structural information, and transform the tissue contraction activity model into a linear transient model;

[0104] Obtain the signal matrix in the linear instantaneous model, divide the signal matrix into sub-signal matrices, and expand the sub-signal matrices to obtain the observation matrix of the sub-signal matrices;

[0105] The observation matrix is ​​whitened, a second-order covariance matrix is ​​constructed based on the whitened observation matrix, a mixture matrix is ​​obtained based on the second-order covariance matrix, and the mixture matrix is ​​separated to obtain multiple independent active signal sources;

[0106] Multiple independent activity signal sources are classified and filtered according to a preset classification method to obtain target signals. Tissue activity parameters are obtained based on the target signals, and the contraction activity of the target tissue is evaluated based on the tissue activity parameters.

[0107] The step of acquiring image structural information of the target tissue and establishing a tissue contraction activity model based on the image structural information specifically includes:

[0108] The structural information of the target tissue is obtained by acquiring images or videos of the tissue through an imaging system. This image structural information is represented as multiple window functions. A tissue contraction activity model is obtained by modeling the sum of the convolutions of the multiple window functions with their respective impulse responses.

[0109]

[0110] Where, x i(t) represents the evolution of the axial displacement of the i-th pixel in the image with time t, where t represents time, n represents the number of activity sources causing target tissue contraction due to physiological response, k represents the number of sparse sources of non-target tissue activity, and s j (tl) represents the j-th source as a function of time tl, h ij (l) represents the duration of the impulse response of the i-th pixel and the j-th source, ω. i (t) is the additional white noise at the i-th pixel.

[0111] Specifically, transforming the tissue contraction activity model into a linear transient model includes:

[0112] Obtain the delayed sample for each window function, and based on each window function and the corresponding delayed sample, transform the tissue contraction activity model into a linear transient model:

[0113]

[0114] in,

[0115]

[0116]

[0117] in, A matrix representation representing the signals received by all channels. Represents a mixture matrix. This represents all sources of organizational activity signals to be estimated. x represents the additional white noise on the source to be estimated. m (t) represents the matrix representation on the m-th source, ω m (t) represents the additional white noise on the m-th source, L refers to the delay unit of the signal source, R refers to the delay unit of the measurement, and T represents the matrix transpose.

[0118] Specifically, obtaining the signal matrix in the linear instantaneous model, dividing the signal matrix into sub-signal matrices, and expanding the sub-signal matrices to obtain the observation matrix includes:

[0119] Obtain the signal matrix in the linear instantaneous model. The signal matrix Segmentation is performed, and the segmented signal matrix is ​​processed. Vectorization and normalization are performed to obtain several sub-signal matrices;

[0120] The sub-signal matrix is ​​expanded to obtain multiple delayed versions of the observation matrix.

[0121] Specifically, the whitening process for the observation matrix includes:

[0122] Calculate the covariance matrix of the observation matrix, and perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalue matrix and the first eigenvector;

[0123] Multiplying the inverse square root of the eigenvalue matrix by the first eigenvector yields the whitening transformation matrix. Multiplying the observation matrix by the whitening transformation matrix yields the whitened observation matrix.

[0124] Specifically, the step of constructing a second-order covariance matrix based on the whitened observation matrix, obtaining a mixing matrix based on the second-order covariance matrix, and separating the mixing matrix to obtain independent active signal sources includes:

[0125] Calculate the second-order covariance matrix of the whitened observation matrix, and perform eigenvalue decomposition on the second-order covariance matrix to obtain the second eigenvector;

[0126] The mixing matrix is ​​calculated based on the second eigenvector, and the mixing matrix is ​​separated by an independent component analysis algorithm to obtain multiple independent active signal sources.

[0127] The step of classifying and filtering multiple independent active signal sources according to a preset classification method to obtain the target signal specifically includes:

[0128] The multiple independent active signal sources are converted into multiple time-domain signals and multiple frequency signals;

[0129] The multiple time-domain signals and multiple frequency signals are compared with a preset standard, and the independent active signal sources corresponding to the time-domain signals and frequency signals that conform to the preset standard are separated from the multiple independent active signal sources to obtain the target signal.

[0130] This application also provides a computer-readable storage medium storing an organization activity parameter measurement program, which, when executed by a processor, implements the steps of the organization activity parameter measurement method as described above.

[0131] In summary, this application provides a method and related equipment for measuring tissue activity parameters. The method includes: acquiring image structural information of a target tissue; establishing a tissue contraction activity model based on the image structural information; converting the tissue contraction activity model into a linear transient model; acquiring the signal matrix in the linear transient model; segmenting the signal matrix to obtain sub-signal matrices; expanding the sub-signal matrices to obtain observation matrices of the sub-signal matrices; whitening the observation matrices to obtain multiple independent activity signal sources; classifying and filtering the multiple independent activity signal sources according to a preset classification method to obtain a target signal; obtaining tissue activity parameters based on the target signal; and evaluating the contraction activity of the target tissue based on the tissue activity parameters. This application combines independent component analysis to quantitatively analyze tissue contraction activity, accurately extracting tissue activity parameters from ultrasound video, and can be implemented on a common imaging platform at a low cost.

[0132] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or terminal. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or terminal that includes that element.

[0133] Of course, those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0134] It should be understood that the application of this application is not limited to the examples above. Those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A method for measuring tissue activity parameters, characterized in that, The method for measuring the tissue activity parameters includes: Obtain image structural information of the target tissue, establish a tissue contraction activity model based on the image structural information, and transform the tissue contraction activity model into a linear transient model; Obtain the signal matrix in the linear instantaneous model, divide the signal matrix into sub-signal matrices, and expand the sub-signal matrices to obtain the observation matrix of the sub-signal matrices; The observation matrix is ​​whitened, a second-order covariance matrix is ​​constructed based on the whitened observation matrix, a mixture matrix is ​​obtained based on the second-order covariance matrix, and the mixture matrix is ​​separated to obtain multiple independent active signal sources; Multiple independent activity signal sources are classified and filtered according to a preset classification method to obtain target signals. Tissue activity parameters are obtained based on the target signals, and the contraction activity of the target tissue is evaluated based on the tissue activity parameters. The step of acquiring image structural information of the target tissue and establishing a tissue contraction activity model based on the image structural information specifically includes: The structural information of the target tissue is obtained by acquiring images or videos of the tissue through an imaging system. This image structural information is represented as multiple window functions. A tissue contraction activity model is obtained by modeling the sum of the convolutions of the multiple window functions with their respective impulse responses. ； in, Indicates the first in the image The axial displacement of each pixel over time The evolution, Indicates time, This indicates the number of active sources that cause target tissue contraction in response to physiological reactions. This represents the number of sparse sources of non-target organizational activities. Indicates the first Individual source as time The function, Indicates the first The pixel and the The duration of the impulse response of each source is , It is the additional white noise at the i-th pixel; The process of transforming the tissue contraction activity model into a linear transient model specifically includes: Obtain the delayed sample for each window function, and based on each window function and the corresponding delayed sample, transform the tissue contraction activity model into a linear transient model: ； ； ； ； in, A matrix representation representing the signals received by all channels. Represents a mixture matrix. This represents all sources of organizational activity signals to be estimated. This represents the additional white noise on the source to be estimated. Indicates the first Matrix representation over each source Indicates the first Added white noise on the source, This refers to the unit of delay of the signal source. This refers to the unit of delay in the measurement. This indicates the matrix transpose.

2. The method for measuring tissue activity parameters according to claim 1, characterized in that, The process of obtaining the signal matrix from the linear instantaneous model, dividing the signal matrix into sub-signal matrices, and expanding the sub-signal matrices to obtain the observation matrix specifically includes: Obtain the signal matrix in the linear instantaneous model. The signal matrix Segmentation is performed, and the segmented signal matrix is ​​processed. Vectorization and normalization are performed to obtain several sub-signal matrices; The sub-signal matrix is ​​expanded to obtain multiple delayed versions of the observation matrix.

3. The method for measuring tissue activity parameters according to claim 1, characterized in that, The whitening process for the observation matrix specifically includes: Calculate the covariance matrix of the observation matrix, and perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalue matrix and the first eigenvector; Multiplying the inverse square root of the eigenvalue matrix by the first eigenvector yields the whitening transformation matrix. Multiplying the observation matrix by the whitening transformation matrix yields the whitened observation matrix.

4. The method for measuring tissue activity parameters according to claim 1, characterized in that, The process of constructing a second-order covariance matrix based on the whitened observation matrix, obtaining a mixing matrix based on the second-order covariance matrix, and separating the mixing matrix to obtain independent active signal sources specifically includes: Calculate the second-order covariance matrix of the whitened observation matrix, and perform eigenvalue decomposition on the second-order covariance matrix to obtain the second eigenvector; The mixing matrix is ​​calculated based on the second eigenvector, and the mixing matrix is ​​separated by an independent component analysis algorithm to obtain multiple independent active signal sources.

5. The method for measuring tissue activity parameters according to claim 1, characterized in that, The step of classifying and filtering multiple independent active signal sources according to a preset classification method to obtain the target signal specifically includes: The multiple independent active signal sources are converted into multiple time-domain signals and multiple frequency signals; The multiple time-domain signals and multiple frequency signals are compared with a preset standard, and the independent active signal sources corresponding to the time-domain signals and frequency signals that conform to the preset standard are separated from the multiple independent active signal sources to obtain the target signal.

6. A system for measuring organizational activity parameters, characterized in that, The tissue activity parameter measurement system is used to implement the tissue activity parameter measurement method according to any one of claims 1-5, and the tissue activity parameter measurement system comprises: The tissue model building module is used to acquire image structural information of the target tissue, build a tissue contraction activity model based on the image structural information, and convert the tissue contraction activity model into a linear transient model. The observation matrix acquisition module is used to acquire the signal matrix in the linear instantaneous model, divide the signal matrix into sub-signal matrices, and expand the sub-signal matrices to obtain the observation matrix of the sub-signal matrices; The activity signal source acquisition module is used to whiten the observation matrix, construct a second-order covariance matrix based on the whitened observation matrix, obtain a mixing matrix based on the second-order covariance matrix, and separate the mixing matrix to obtain multiple independent activity signal sources. The tissue activity parameter measurement module is used to classify and filter multiple independent activity signal sources according to a preset classification method to obtain target signals, obtain tissue activity parameters based on the target signals, and evaluate the contraction activity of the target tissue based on the tissue activity parameters.

7. A terminal, characterized in that, The terminal includes: a memory, a processor, and a tissue activity parameter measurement program stored in the memory and executable on the processor, wherein the tissue activity parameter measurement program, when executed by the processor, implements the steps of the tissue activity parameter measurement method as described in any one of claims 1-5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a program for measuring tissue activity parameters, which, when executed by a processor, implements the steps of the method for measuring tissue activity parameters as described in any one of claims 1-5.

Images