Vital Sign Signal Extraction Method, Device, Electronic Device and Storage Medium

Abstract

The present disclosure provides a method, apparatus, electronic device, and storage medium for extracting vital sign signals, relating to the technical field of life signal detection. The method includes: obtaining initial echo data to be processed and decomposing the initial echo data into multiple sub-modules; extracting feature data of the sub-modules; obtaining a vital sign signal recognition result of the sub-modules based on the feature data, and determining that the sub-module with the vital sign signal recognition result indicating the presence of vital signs is the target sub-module; and extracting the vital sign data of the target sub-module as the vital sign signal of the initial echo data. Thus, by dividing the initial echo data into multiple sub-modules and then processing the sub-modules separately to obtain the vital sign signal, the processing efficiency of the initial echo data can be improved, and at the same time, the accuracy of the initial echo data processing can be improved, increasing the success rate of rescue.

Description

Technical Field

[0001] The present disclosure relates to the technical field of life signal detection, and in particular, to a method, apparatus, electronic device, and storage medium for extracting vital sign signals. Background Art

[0002] Non-contact life detection technology is a new technology that has emerged in recent years with the needs of medical engineering, military, and social development. It breaks through the current detection methods and technologies and explores living bodies by using human vital parameters such as breathing, heartbeat, and human motion signals without contacting the target body.

[0003] In the actual disaster rescue process, life detectors have played a certain role, but it is undeniable that there are still many problems, and the actual test results are not very satisfactory. Especially when the clutter and noise interference are large and the vital sign signals are very weak, it is difficult to successfully detect. Therefore, there is an urgent need for an intelligent detection method that is very sensitive to vital sign signals, so as to better detect weak life signals in various complex environments, find trapped survivors more accurately and quickly, and make life detectors play a more important role in disaster rescue. Summary of the Invention

[0004] The present disclosure aims to solve at least one of the technical problems in the related art to some extent.

[0005] To this end, an object of the present disclosure is to propose a method for extracting vital sign signals.

[0006] A second object of the present disclosure is to propose an apparatus for extracting vital sign signals.

[0007] A third object of the present disclosure is to propose an electronic device.

[0008] A fourth object of the present disclosure is to propose a non-transitory computer-readable storage medium.

[0009] A fifth object of the present disclosure is to propose a computer program product.

[0010] To achieve the above object, a first aspect embodiment of the present disclosure proposes a method for extracting vital sign signals, including: obtaining initial echo data to be processed, and decomposing the initial echo data into a plurality of sub-modules; extracting feature data of the sub-modules; obtaining a vital sign signal recognition result of the sub-modules based on the feature data, and determining the sub-modules with the vital sign signal recognition result being of vital characteristics as target sub-modules; and extracting vital sign data of the target sub-modules as the vital sign signals of the initial echo data.

[0011] According to an embodiment of the present disclosure, before extracting the vital sign signal of the target sub-module as the vital sign signal of the initial echo data, it further includes: drawing the spatial coordinates of the target sub-module; removing the mis-identified target sub-module based on the K-Means algorithm and the spatial coordinates.

[0012] According to an embodiment of the present disclosure, extracting the vital sign data of the target sub-module as the vital sign signal of the initial echo data further includes: performing wavelet transform on the vital sign data; reconstructing the transformed vital sign data based on the Mallet algorithm; processing the reconstructed vital sign signal based on the method of linear trend suppression to generate the vital sign signal.

[0013] According to an embodiment of the present disclosure, extracting the respiration signal from the vital sign signal based on the Butterworth filter.

[0014] According to an embodiment of the present disclosure, performing MTI harmonic suppression on the vital sign signal and extracting the heartbeat signal from the vital sign signal through the Butterworth filter.

[0015] According to an embodiment of the present disclosure, extracting the feature data of the sub-module includes: extracting the singular entropy, wavelet packet scale entropy and sample entropy of the sub-module.

[0016] According to an embodiment of the present disclosure, obtaining the vital sign signal recognition result of the sub-module based on the feature data includes: inputting the feature data into the vital sign signal extraction model to obtain the vital sign signal recognition result.

[0017] To achieve the above object, an embodiment of the second aspect of the present disclosure provides a vital sign signal extraction device, including: an acquisition module, configured to acquire the initial echo data to be processed and decompose the initial echo data into multiple sub-modules; an extraction module, configured to extract the feature data of the sub-module; a determination module, configured to obtain the vital sign signal recognition result of the sub-module based on the feature data and determine the sub-module with the vital sign signal recognition result of having vital characteristics as the target sub-module; a generation module, configured to extract the vital sign data of the target sub-module as the vital sign signal of the initial echo data.

[0018] To achieve the above object, an embodiment of the third aspect of the present disclosure 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, and the instructions are executed by the at least one processor to implement the vital sign signal extraction method as described in the embodiment of the first aspect of the present disclosure.

[0019] To achieve the above object, an embodiment of the fourth aspect of the present disclosure provides a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to implement the vital sign signal extraction method as described in the embodiment of the first aspect of the present disclosure.

[0020] To achieve the above object, an embodiment of the fifth aspect of the present disclosure provides a computer program product, including a computer program, where the computer program is used to implement the vital sign signal extraction method as described in the embodiment of the first aspect of the present disclosure when executed by a processor.

[0021] By dividing the initial echo data into multiple sub-modules and then processing the sub-modules separately to obtain vital sign signals, the processing efficiency of the initial echo data can be improved, and at the same time, the accuracy of the initial echo data processing can be improved, increasing the success rate of rescue. Description of the Drawings

[0022] Figure 1 is a schematic diagram of a vital sign signal extraction method according to an embodiment of the present disclosure;

[0023] Figure 2 is a sampling diagram of the initial echo data of a vital sign signal extraction method according to an embodiment of the present disclosure;

[0024] Figure 3 is a spectrogram of a target sub-module of a vital sign signal extraction method according to an embodiment of the present disclosure;

[0025] Figure 4 is a spectrogram of the target sub-module after screening by the K-Means algorithm of a vital sign signal extraction method according to an embodiment of the present disclosure;

[0026] Figure 5 is a schematic diagram of another vital sign signal extraction method according to an embodiment of the present disclosure;

[0027] Figure 6 is a waveform diagram of the vital sign signal of a vital sign signal extraction method according to an embodiment of the present disclosure;

[0028] Figure 7 is a waveform diagram of the respiratory signal of a vital sign signal extraction method according to an embodiment of the present disclosure;

[0029] Figure 8 is a waveform diagram of the heartbeat signal of a vital sign signal extraction method according to an embodiment of the present disclosure;

[0030] Figure 9 is a schematic diagram of a vital sign signal extraction device according to an embodiment of the present disclosure;

[0031] Figure 10 It is a schematic diagram of an electronic device according to an embodiment of the present disclosure. Detailed implementation manners

[0032] Embodiments of the present disclosure will be described in detail below. Examples of the embodiments are shown in the accompanying drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to explain the present disclosure and should not be construed as limiting the present disclosure.

[0033] In the actual disaster rescue process, life detectors have played a certain role, but the effect is not very ideal when processing the actual test results. Especially when the clutter and noise interference are large and the vital sign signals are very weak, it is difficult to detect successfully. Therefore, there is an urgent need for an intelligent detection method that is very sensitive to vital sign signals, so as to better detect weak life signals in various complex environments, find trapped survivors more accurately and quickly, and make life detectors play a more important role in disaster rescue.

[0034] The present invention proposes a method for extracting vital sign signals, which solves the problems that in a complex environment, weak vital sign signals are masked by a large amount of noise and clutter and are difficult to be detected, and can accurately locate a living body and extract heartbeat and respiration signals, and has high accuracy, timeliness, robustness and signal-to-noise ratio.

[0035] Figure 1 It is a schematic diagram of an exemplary embodiment of a method for extracting vital sign signals proposed by the present disclosure, as Figure 1 shown. The method for extracting vital sign signals includes the following steps:

[0036] S101, obtain initial echo data to be processed, and decompose the initial echo data into multiple sub-modules.

[0037] It should be noted that the initial echo data may include various data. For example, it may include vital sign data, environmental data, waveform data, etc., and there is no limitation here, and it specifically depends on the actual situation.

[0038] The initial echo data may be real-time data actually collected by a life detector device, and the life detector device may be an ultra-wideband radar life detector, an infrared life detector, an audio life detector, etc., and there is no limitation here.

[0039] In the embodiments of the present disclosure, an ultra-wideband radar vital sign detector is used to transmit a continuous periodic pulse sequence for detection. The signal sampling frequency FS and the continuous pulse sequence frequency FPR are set. Echo signals {X1, X2, X3, …, Xn} containing vital sign signals are received in multiple consecutive groups, and the received echo signals of each pulse sequence are arranged row by row. For example, as Figure 2 shown in the waveform image, an initial echo data matrix Xmxn is formed, as shown in the following table:

[0040]

[0041] In the embodiments of the present disclosure, after obtaining the initial echo data matrix, a constant longitudinal step p and a transverse step q can be taken to read consecutive M×N data, so as to divide the entire matrix Rmxn into several M×N sub-modules Si,j (15≥M≥5, 20≥N≥10). That is:

[0042]

[0043] Among them, It should be noted that the longitudinal step p and the transverse step q can be set according to actual needs and are not limited here.

[0044] S102, extract the characteristic data of the sub-module.

[0045] In the embodiments of the present disclosure, the characteristic data of the sub-module can include various types. For example, it can include singular entropy, wavelet packet scale entropy of the first-order right singular vector, sample entropy, etc. There is no limitation here, and specific extraction needs to be carried out according to actual needs.

[0046] In the embodiments of the present disclosure, the extraction process of singular entropy includes: determining a singular value matrix based on the data matrix, and determining the singular entropy of the data matrix based on the singular value matrix.

[0047] Among them, the process of performing singular value decomposition on the vital sign data S is as follows:

[0048]

[0049] The singular value matrix Σ k×k has singular values only on the main diagonal, and other elements are all 0. The main diagonal elements σ 1 , σ 2 , σ 3 , …, σ k are k singular values, and satisfy σ1≥σ2≥σ3≥…≥σk. ui represents the i-th column vector of the matrix U M×k , which is called the i-th order left singular vector. vi represents the matrix V N×kThe i-th column vector is called the i-th order right singular vector. σi represents the i-th element of the singular value spectrum. For the present invention, the first k (min(M,N)≥k≥3) order singular values σ 1 , σ 2 , σ 3 , …, σ k , left singular vectors u 1 , u 2 , u 3 , …, u k and right singular vectors v 1 , v 2 , v 3 , …, v k are used as the decomposition results.

[0050] The formula for calculating the singular entropy of the vital sign data S is as follows:

[0051]

[0052] In the embodiments of the present disclosure, extracting the wavelet packet scale entropy includes:

[0053] Calculating the wavelet packet scale entropy for the first-order right singular vectors of all data in the vital sign data samples and the environmental data samples;

[0054] Among them, the process of performing wavelet packet decomposition on the first-order right singular vector v1 of the vital sign data S is as follows:

[0055] Select the 'db6' wavelet, set the wavelet packet decomposition layer number to 2 for wavelet packet decomposition, and the wavelet packet decomposition method is as follows:

[0056]

[0057] Among them, W 0 0 is the original signal v 1 , W is the wavelet packet decomposition coefficient vector of each layer, n and j are wavelet packet nodes, k is the wavelet packet decomposition layer number, h is the high-pass filter coefficient, and g is the low-pass filter coefficient.

[0058] The formula for calculating the wavelet packet scale entropy of each node is:

[0059]

[0060] Among them, k is the wavelet packet decomposition layer number, j is the wavelet packet node (j = 0, 1, …, 2k - 1) (in the present invention, k = 2 can be taken). is the wavelet packet decomposition coefficient vector, and wi (i = 1, 2, …, x) are the wavelet coefficients in Wkj.

[0061] In the embodiments of the present disclosure, extracting the sample entropy of the data matrix includes:

[0062] Calculating the sample entropy for all the data in the vital sign data samples and the environmental data samples;

[0063] Among them, the process of calculating the sample entropy for the vital sign data S is as follows:

[0064] Expand the vital sign data S into a one-dimensional sequence {x1, x2, …, xn}, take the sample entropy reconstruction dimension m = 2, the threshold is r, calculate the standard deviation of the sequence as δ, and form a vector sequence {X m (1), X m (2), … X m (n - m + 1)} according to the serial numbers of the above one-dimensional sequence.

[0065] Among them, X m (i) = [x(i), x(i + 1), … x(i + m - 1)], 1 ≤ i ≤ n - m + 1 represents m consecutive x values starting from the i-th point.

[0066] Define the distance between the vector X m (i) and X m (j) as the absolute value of the maximum difference among their corresponding elements, that is:

[0067] d[X m (i), X m (j)] = max k＝0,…,m-1 (|x(i + k) - x(j + k)|)

[0068] For a given X m (i), count the number of j (1 ≤ j ≤ n - m, j ≠ i) whose distance between X m (i) and X m (j) is less than r·δ, and denote it as Bi. For 1 ≤ i ≤ n - m, define:

[0069]

[0070] Increase the dimension to m + 1, calculate the number of j (1 ≤ j ≤ n - m - 1, j ≠ i) whose distance between X m+1 (i) and X m+1 (j) is less than r·δ, and denote it as Ai. Then define:

[0071]

[0072] In this way, B m (r) is the probability that two sequences match m points under the similarity tolerance r, and A m(r) is the probability that two sequences match at m + 1 points under the similarity tolerance r. The sample entropy calculation formula is as follows:

[0073]

[0074] S103. Obtain the vital sign signal recognition result of the feature data acquisition sub-module, and determine the sub-module whose vital sign signal recognition result is a vital sign as the target sub-module.

[0075] In the embodiment of the present disclosure, determine the sample signature value of the vital sign data, and judge whether the element data in the vital sign data meets the vital sign condition; in response to the element data in the vital sign data meeting the vital sign condition, determine that the label value of the element data is 1; in response to the site data in the vital sign data not meeting the preset condition, determine that the label value of the element data is 0; generate the sample signature value of the vital sign data based on the label values of all the element data.

[0076] For example, when the sub-module is as follows,

[0077]

[0078] The label value can be:

[0079]

[0080] S104. Extract the vital sign data of the target sub-module as the vital sign signal of the initial echo data.

[0081] In the embodiment of the present disclosure, first obtain the initial echo data to be processed, decompose the initial echo data into multiple sub-modules, then extract the feature data of the sub-modules, and then obtain the vital sign signal recognition result of the sub-modules based on the feature data, and determine the sub-module whose vital sign signal recognition result is a vital sign as the target sub-module, and finally extract the vital sign data of the target sub-module as the vital sign signal of the initial echo data. Thus, by dividing the initial echo data into multiple sub-modules and then processing the sub-modules separately to obtain the vital sign signal, the processing efficiency of the initial echo data can be improved, and at the same time, the processing accuracy of the initial echo data can be improved, and the success rate of rescue can be increased.

[0082] It should be noted that the feature data can also be input into the vital sign signal extraction model to obtain the vital sign signal recognition result. The vital sign signal extraction model can be pre-trained and stored in the storage space of the electronic device for convenient retrieval and use when needed.

[0083] In one possible implementation, after obtaining the target sub-module, the initial echo data can be intercepted based on the position information of the target sub-module to obtain complete echo data with vital sign signals, and then the echo data is processed to obtain vital sign signals.

[0084] In the above embodiment, before extracting the vital sign signal of the target sub-module as the vital sign signal of the initial echo data, it is also necessary to draw the spatial coordinates of the target sub-module, and then remove the mis-identified target sub-module based on the K-Means algorithm and the spatial coordinates.

[0085] As Figure 3 shown, extract the central position coordinates of the target sub-module, draw the coordinate information in space, and construct the position input sample {x 1 , x 2 , …, x m}.

[0086] Determine the number of clustering categories k and the maximum number of iterations N according to the position distribution map of the position input sample. Then complete the K-Means algorithm process according to the following steps:

[0087] ① Select k initialized centroids μ j (j = 1, 2,... k) from the position input sample, and ensure that the distances between the k initialized centroids are greater than the threshold Th1.

[0088] ② For each sample xi, calculate the distance between each sample xi and each centroid vector μ j (j = 1, 2,... k):

[0089]

[0090] ③ Mark the smallest d ij corresponding to the category labeli of xi:

[0091]

[0092] ④ After completing the classification of all samples, obtain the cluster partition C = {C 1 , C 2 ,..., C k} of the current iteration, and recalculate the new centroids for all sample points in all categories j = 1, 2,..., k:

[0093]

[0094] ⑤ If the number of iterations ≥ N, or all k centroid vectors μ j do not change compared with the previous iteration, then terminate the iteration process and output the cluster partition C = {C1 , C 2 ,..., C k}, and the centroid {μ 1 , μ 2 ,..., μ k}. Otherwise, repeat steps ② to ⑤ to continue the iteration.

[0095] ⑥ Determine the interference point rejection threshold Th2. For all sample points in all categories j = 1, 2,..., k of the cluster partition C = {C 1 , C 2 ,..., C k}, calculate the distance from the data to its respective centroid {μ 1 , μ 2 ,..., μ k}. If it is greater than Th2, it is determined as interference data and then deleted from the original data, and the coordinate information after removing the interference points is plotted in the space.

[0096] After removing all interference points, recalculate the centroid of the sample points in each category j = 1, 2,..., k of the cluster partition C = {C 1 , C 2 , …, C k}:

[0097]

[0098] The re - partitioned cluster class partitions and the centroid positions of each class of samples are as Figure 4 shown. Thus, by screening the target sub - module through the K - Means algorithm, the size of the processed data can be reduced, the accuracy and efficiency of data processing can be improved, and the processing cost can be reduced.

[0099] In the above - mentioned embodiment, extracting the vital sign data of the target sub - module as the vital sign signal of the initial echo data can also be further explained by Figure 5 : The method includes:

[0100] S501, perform wavelet transform on the vital sign data.

[0101] In the embodiments of the present disclosure, first select an appropriate wavelet basis and decomposition level, and then perform wavelet transform:

[0102]

[0103] Among them, WT f (i, k) is the wavelet coefficient, and i is the wavelet decomposition level.

[0104] S502, reconstruct the transformed vital sign data based on the Mallet algorithm.

[0105] To simplify the cumbersome process, with the help of the two-scale equation, namely the Mallet algorithm, a recursive implementation method of wavelet transform is obtained:

[0106] S f (0,k) = v j (k)

[0107] S f (i + 1,k) = S f (i,k) * h(i,k)

[0108] WT f (i + 1,k) = S f (i,k) * g(i,k)

[0109] Where h and g are the low-pass and high-pass filters corresponding to the scaling function δ(x) and the wavelet function ψ(x) respectively, S f (i,k) is the scaling coefficient, and WT f (i,k) is the wavelet coefficient. Correspondingly, the wavelet transform reconstruction formula is:

[0110]

[0111] According to the above steps, the wavelet transform completes the process of denoising each vital sign signal v j and performs the denoising process.

[0112] S503, Process the reconstructed vital sign signal based on the method of linear trend suppression to generate the vital sign signal.

[0113] The method of linear trend suppression is to use a fourth-order polynomial to fit the data after wavelet decomposition and denoising to obtain the linear trend term Ω of the data. Then subtract this trend term from the original data to obtain the denoised vital sign signal sj:

[0114] s j = v j -Ω

[0115] The denoised vital sign signal sj, and the signal spectrum of this signal is as Figure 6 shown.

[0116] After obtaining the vital sign signal, the respiratory signal can be extracted from the vital sign signal based on the Butterworth filter. A second-order Butterworth filter can be constructed, and the filter type is set to a band-pass filter. For example, the filter pass frequency can be set to 0.2 Hz and the cut-off frequency can be set to 0.6 Hz. Filter the denoised vital sign signal si to obtain the respiratory signal bi, and the obtained respiratory signal is as Figure 7 shown in the signal spectrum.

[0117] After obtaining the vital sign signals, MTI harmonic suppression can also be performed on the vital sign signals, and the heartbeat signals can be extracted from the vital sign signals through a Butterworth filter. For the denoised vital sign signal s i The heartbeat signals are extracted through MTI harmonic suppression and a Butterworth filter. Then, the method of MTI harmonic suppression is as follows: according to the frequency F of the respiration signal r Design a double-delay harmonic eliminator, and the delay parameter T of this system r is:

[0118]

[0119] The dynamic equation of the double-delay harmonic eliminator system is:

[0120] h(t) = δ(t) - 2δ(t - T r ) + δ(t - 2T r )

[0121] The transfer function of the double-delay harmonic eliminator system is:

[0122]

[0123] The denoised vital sign signal s is processed through the above system j to eliminate the harmonics of the respiration signal, and the signal c after eliminating the harmonics is obtained i . Then, set the filter type as a band-pass filter, set the pass frequency of the filter as 0.9 Hz, and the cut-off frequency as 2.0 Hz. Filter the signal after MTI harmonic suppression processing to obtain the heartbeat signal hi, as Figure 8 shown

[0124] Figure 9 is a schematic diagram of a vital sign signal extraction device proposed by the present disclosure, as Figure 9 shown. The vital sign signal extraction device 900 includes: an acquisition module 910, an extraction module 920, a determination module 930, and a generation module 940

[0125] Among them, the acquisition module 910 is used to acquire the initial echo data to be processed and decompose the initial echo data into multiple sub-modules

[0126] The extraction module 920 is used to extract the feature data of the sub-modules

[0127] The determination module 930 is used to obtain the vital sign signal recognition result of the sub-module based on the feature data, and determine the sub-module with the vital sign signal recognition result of having vital characteristics as the target sub-module

[0128] A generation module 940 is configured to extract the vital sign data of the target sub-module as the vital sign signal of the initial echo data.

[0129] In one embodiment of the present disclosure, the generation module 940 is further configured to: draw the spatial coordinates of the target sub-module; remove the mis-identified target sub-module based on the K-Means algorithm and the spatial coordinates.

[0130] In one embodiment of the present disclosure, the generation module 940 is further configured to: perform wavelet transform on the vital sign data; reconstruct the transformed vital sign data based on the Mallet algorithm; process the reconstructed vital sign signal based on the method of linear trend suppression to generate the vital sign signal.

[0131] In one embodiment of the present disclosure, the generation module 940 is further configured to: extract the respiration signal from the vital sign signal based on a Butterworth filter.

[0132] In one embodiment of the present disclosure, the generation module 940 is further configured to: perform MTI harmonic suppression on the vital sign signal, and extract the heartbeat signal from the vital sign signal through a Butterworth filter.

[0133] In one embodiment of the present disclosure, the extraction module 920 is further configured to: extract the singular entropy, wavelet packet scale entropy, and sample entropy of the sub-module.

[0134] In one embodiment of the present disclosure, the determination module 930 is further configured to: input the feature data into a vital sign signal extraction model to obtain the vital sign signal recognition result.

[0135] To implement the above embodiments, the embodiments of the present disclosure also propose an electronic device 1000, as Figure 10 shown. The electronic device 1000 includes: a processor 1001 and a memory 1002 communicatively connected to the processor. The memory 1002 stores instructions executable by at least one processor. The instructions are executed by at least one processor 1001 to implement the vital sign signal extraction method according to the embodiments of the first aspect of the present disclosure.

[0136] To implement the above embodiments, the embodiments of the present disclosure also propose a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause a computer to implement the vital sign signal extraction method according to the embodiments of the first aspect of the present disclosure.

[0137] To implement the above embodiments, the embodiments of the present disclosure also propose a computer program product, including a computer program, where the computer program, when executed by a processor, implements the vital sign signal extraction method according to the embodiments of the first aspect of the present disclosure.

[0138] In the description of the present disclosure, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", "axial", "radial", "circumferential", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present disclosure and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present disclosure.

[0139] In addition, the terms "first" and "second" are only used for descriptive purposes and should not be construed as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present disclosure, "a plurality" means two or more, unless otherwise specifically defined.

[0140] In the description of this specification, the description with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described may be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0141] Although the embodiments of the present disclosure have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as a limitation to the present disclosure. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A method for extracting vital sign signals, characterized in that, it includes: Obtain the initial echo data to be processed, construct an initial echo data matrix, and decompose the initial echo data matrix into multiple sub-modules; Extract the characteristic data of the sub-module, and the characteristic data includes singular entropy, wavelet packet scale entropy of the first-order right singular vector, and sample entropy; where: The formula for calculating the singular entropy is: The formula for calculating the wavelet packet scale entropy is: The formula for calculating the sample entropy is: Based on the characteristic data, obtain the vital sign signal recognition result of the sub-module, and determine the sub-module with the vital sign signal recognition result as having vital characteristics as the target sub-module; Plot the spatial coordinates of the target sub-module, remove the mis-identified target sub-module based on the K-Means algorithm and the spatial coordinates, extract the vital sign data of the target sub-module after removing the identification, perform wavelet transform on the vital sign data, reconstruct the transformed vital sign data based on the Mallet algorithm, and process the reconstructed vital sign signal based on the method of linear trend suppression to generate the vital sign signal of the initial echo data; where: the formula for reconstructing the transformed vital sign data based on the Mallet algorithm is: Among them, S f (i, k) is the scaling coefficient, h and g are the low-pass and high-pass filters corresponding to the scaling function δ(x) and the wavelet function ψ(x) respectively, i is the wavelet decomposition level, WT f (i, k) is the wavelet coefficient, 2. The method according to claim 1, characterized in that, the method further includes: Extract the respiration signal from the vital sign signal based on a Butterworth filter.

3. The method according to claim 1, characterized in that, the method further includes: Perform MTI harmonic suppression on the vital sign signal, and extract the heartbeat signal from the vital sign signal through a Butterworth filter.

4. The method according to claim 1, characterized in that, obtaining the vital sign signal recognition result of the sub-module based on the characteristic data includes: Input the characteristic data into a vital sign signal extraction model to obtain the vital sign signal recognition result.

5. A vital sign signal extraction device, characterized in that, it includes: An acquisition module for acquiring the initial echo data to be processed, constructing an initial echo data matrix, and decomposing the initial echo data matrix into multiple sub-modules; An extraction module for extracting the characteristic data of the sub-module, and the characteristic data includes singular entropy, wavelet packet scale entropy of the first-order right singular vector, and sample entropy; where: The formula for calculating the singular entropy is: The formula for calculating the wavelet packet scale entropy is: The formula for calculating the sample entropy is: A determination module for obtaining the vital sign signal recognition result of the sub-module based on the characteristic data, and determining the sub-module with the vital sign signal recognition result as having vital characteristics as the target sub-module; A generation module is configured to draw the spatial coordinates of the target sub-module, remove mis-identified target sub-modules based on the K-Means algorithm and the spatial coordinates, extract the vital sign data of the target sub-module after removal of identification, perform wavelet transform on the vital sign data, reconstruct the transformed vital sign data based on the Mallet algorithm, and process the reconstructed vital sign signal based on the method of linear trend suppression to generate the vital sign signal of the initial echo data; wherein: the formula for reconstructing the transformed vital sign data based on the Mallet algorithm is: Among them, S f (i, k) is the scaling coefficient, h and g are the low-pass and high-pass filters corresponding to the scaling function δ(x) and the wavelet function ψ(x) respectively, i is the wavelet decomposition level, WT f (i, k) is the wavelet coefficient, 6. An electronic device, characterized in that, it includes a memory and a processor; wherein, the processor runs a program corresponding to the executable program code by reading the executable program code stored in the memory, so as to implement the method according to any one of claims 1-4.

7. A computer-readable storage medium, on which a computer program is stored, characterized in that, when the program is executed by a processor, it implements the method according to any one of claims 1-4.

Images