Heart sound signal processing method, device, equipment, system and program product

Abstract

The embodiment of the invention provides a heart sound signal processing method, which comprises the following steps of: sequentially performing wavelet decomposition on a heart sound signal to be processed according to the number of decomposition layers to obtain a wavelet coefficient corresponding to each decomposition layer number; according to the decomposition layer number, the denoising threshold value is adjusted to obtain an adjustment threshold value corresponding to each decomposition layer number; the wavelet coefficients of the corresponding decomposition layers are screened based on the adjustment threshold value, and a processing coefficient is obtained; and performing wavelet inverse transformation reconstruction according to the processing coefficient to obtain a target heart sound signal. Therefore, the dynamic heart sound signal de-noising processing process is achieved, due to the fact that dynamic de-noising threshold value setting is carried out based on different decomposition layer numbers in the de-noising process, the problem that effective heart sound components are excessively reduced is avoided, more effective heart sound components are reserved, and the accuracy of the heart sound signals is improved.

Description

Technical Field

[0001] The present application relates to the field of computer technology, and in particular to a method, device, equipment, system and program product for processing heart sound signals. Background Art

[0002] With the development of smart medicine, research and analysis of collected heart sound signals can help doctors understand the patient's heart condition more conveniently and accurately, avoiding misdiagnosis due to lack of experience or interference from environmental noise.

[0003] Heart sound signals are easily contaminated by noise during collection, such as lung sounds, power frequency interference noise, noise generated by the friction between the stethoscope head and clothing, and voices of people in the surrounding environment. The presence of these noises will drown out the effective pathological components in the heart sound signals, which is not conducive to the subsequent processing and analysis of the heart sound signals, such as heart sound segmentation and feature extraction.

[0004] Therefore, how to effectively reduce the noise of the collected heart sound signals and improve the quality of the heart sound signals becomes a difficult problem. Summary of the invention

[0005] In view of this, the present application provides a method for processing a heart sound signal, which can effectively denoise the heart sound signal.

[0006] According to a first aspect of an embodiment of the present application, a method for processing a heart sound signal is provided, which can be applied to a system or program including a heart sound signal processing function in a terminal device, and specifically includes:

[0007] Acquire a heart sound signal to be processed, and a wavelet function and a decomposition layer number corresponding to the heart sound signal to be processed;

[0008] Based on the wavelet function, the heart sound signal to be processed is sequentially subjected to wavelet decomposition according to the decomposition levels to obtain wavelet coefficients corresponding to each decomposition level;

[0009] Determine a denoising threshold corresponding to unbiased simulated estimation of the heart sound signal to be processed, so as to adjust the denoising threshold according to the number of decomposition levels to obtain adjustment thresholds corresponding to each number of decomposition levels;

[0010] Based on the adjustment threshold, the wavelet coefficients of the corresponding decomposition layer are screened to obtain processing coefficients;

[0011] The inverse wavelet transform is performed according to the processing coefficients to reconstruct the target heart sound signal.

[0012] Optionally, in some possible implementations, determining a denoising threshold corresponding to unbiased simulated estimation of the heart sound signal to be processed, and adjusting the denoising threshold according to the number of decomposition layers to obtain adjustment thresholds corresponding to each number of decomposition layers, includes:

[0013] Determine a denoising threshold corresponding to unbiased likelihood estimation of the heart sound signal to be processed;

[0014] According to the decomposition layer number, a layer number adaptive factor corresponding to each decomposition layer number is configured respectively, and the layer number adaptive factor increases as the decomposition layer number increases;

[0015] The denoising threshold is adjusted based on the layer adaptive factor to obtain an adjustment threshold corresponding to each decomposition layer number, and the adjustment threshold decreases as the layer adaptive factor increases.

[0016] Optionally, in some possible implementations, configuring the layer number adaptive factors corresponding to each decomposition layer number according to the decomposition layer number respectively includes:

[0017] Determining the number of processed layers corresponding to the number of decomposed layers;

[0018] configuring a layer number coefficient based on the decomposed layer number and the processed layer number;

[0019] The layer number adaptive factors corresponding to each decomposition layer number are determined according to the layer number coefficients.

[0020] Optionally, in some possible implementations, determining a denoising threshold corresponding to unbiased likelihood estimation of the heart sound signal to be processed includes:

[0021] Determining a similarity index between the heart sound signal to be processed and the wavelet function to obtain a signal element;

[0022] Performing square calculation on the signal elements, and arranging the calculation results from large to small to obtain an estimation sequence;

[0023] Calculate the risk estimation value corresponding to each signal element in the estimation sequence according to the unbiased quasi-estimation formula;

[0024] The square root of the signal element with the smallest risk estimation value is used as the denoising threshold.

[0025] Optionally, in some possible implementations, the acquiring the heart sound signal to be processed, and the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed include:

[0026] Acquire and collect heart sound signals;

[0027] According to the effective frequency range corresponding to the collected heart sound signal, the filter order, cutoff frequency and passband edge frequency in the Butterworth filter are configured;

[0028] Processing the collected heart sound signal based on the configured Butterworth filter to obtain the heart sound signal to be processed;

[0029] Obtain the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed.

[0030] Optionally, in some possible implementations, the obtaining the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed includes:

[0031] Comparing the heart sound signal to be processed with a preset function for similarity to obtain comparison information;

[0032] determining the wavelet function based on the comparison information;

[0033] Determine the sampling frequency range corresponding to the heart sound signal to be processed;

[0034] Performing range decomposition in sequence based on the sampling frequency range to approach the effective frequency range;

[0035] The number of decompositions corresponding to the effective frequency range is determined to obtain the number of decomposition layers.

[0036] Optionally, in some possible implementations, the performing wavelet decomposition on the heart sound signal to be processed in sequence according to the decomposition levels based on the wavelet function to obtain wavelet coefficients corresponding to each decomposition level includes:

[0037] Performing wavelet decomposition on the heart sound signal to be processed based on the wavelet function to obtain an input signal;

[0038] The input signal is divided by using the frequency median of the input signal as a dividing line to obtain an approximate component indicating a low frequency band and a detail component indicating a high frequency band;

[0039] The approximate components are sequentially subjected to wavelet decomposition according to the decomposition levels to obtain wavelet coefficients corresponding to each decomposition level.

[0040] Optionally, in some possible implementations, the method further includes:

[0041] Determining a valid frequency range corresponding to the heart sound signal to be processed;

[0042] screening the detail components based on the effective frequency range;

[0043] If the frequency range indicated by the detail component is outside the effective frequency range, the wavelet coefficients corresponding to the detail component are set to zero.

[0044] Optionally, in some possible implementations, the method further includes:

[0045] Determining a comparison index between the heart sound signal to be processed and the target heart sound signal, wherein the comparison index includes a root mean square error and a signal-to-noise ratio;

[0046] Quality assessment information is generated based on the comparison indicator.

[0047] Optionally, in some possible implementations, the method further includes:

[0048] Adjusting the layer number adaptation factor according to the quality assessment information;

[0049] The heart sound signal to be processed is processed based on the adjusted layer adaptive factor to obtain an adjusted heart sound signal.

[0050] According to a second aspect of an embodiment of the present application, a device for processing a heart sound signal is provided, comprising:

[0051] An acquisition unit, used for acquiring a heart sound signal to be processed, and a wavelet function and a decomposition layer number corresponding to the heart sound signal to be processed;

[0052] A processing unit, configured to perform wavelet decomposition on the heart sound signal to be processed in sequence according to the decomposition levels based on the wavelet function, to obtain wavelet coefficients corresponding to each decomposition level;

[0053] The processing unit is further used to determine a denoising threshold corresponding to unbiased simulated estimation of the heart sound signal to be processed, so as to adjust the denoising threshold according to the number of decomposition levels to obtain an adjustment threshold corresponding to each number of decomposition levels;

[0054] The processing unit is further used to screen the wavelet coefficients of the corresponding decomposition layer based on the adjustment threshold to obtain processing coefficients;

[0055] The processing unit is further used to perform inverse wavelet transform reconstruction according to the processing coefficients to obtain a target heart sound signal.

[0056] Optionally, in some possible implementations, the processing unit determines the denoising threshold corresponding to the unbiased simulated estimation of the heart sound signal to be processed, and adjusts the denoising threshold according to the number of decomposition layers to obtain the adjustment threshold corresponding to each decomposition layer, which is specifically used to determine the denoising threshold corresponding to the unbiased simulated estimation of the heart sound signal to be processed; configures the layer number adaptive factor corresponding to each decomposition layer according to the number of decomposition layers, and the layer number adaptive factor increases with the increase of the decomposition layer number; adjusts the denoising threshold based on the layer number adaptive factor to obtain the adjustment threshold corresponding to each decomposition layer, and the adjustment threshold decreases with the increase of the layer number adaptive factor.

[0057] Optionally, in some possible implementations, when the processing unit configures the layer adaptive factors corresponding to each decomposition layer number according to the decomposition layer number, it is specifically used to determine the number of processed layers corresponding to the decomposition layer number; configure the layer coefficient based on the decomposition layer number and the processed layer number; and determine the layer adaptive factors corresponding to each decomposition layer number according to the layer coefficient.

[0058] Optionally, in some possible implementations, when the processing unit determines the denoising threshold corresponding to the unbiased probabilistic estimation of the heart sound signal to be processed, it is specifically used to determine the similarity index corresponding to the heart sound signal to be processed and the wavelet function to obtain a signal element; square the signal element, and arrange the calculation results from large to small to obtain an estimation sequence; calculate the risk estimation value corresponding to each signal element in the estimation sequence according to the unbiased probabilistic estimation formula; and use the square root of the signal element with the smallest risk estimation value as the denoising threshold.

[0059] Optionally, in some possible implementations, when the acquisition unit acquires the heart sound signal to be processed, as well as the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed, it is specifically used to acquire the collected heart sound signal; configure the filter order, cutoff frequency and passband edge frequency in the Butterworth filter according to the effective frequency range corresponding to the collected heart sound signal; process the collected heart sound signal based on the configured Butterworth filter to obtain the heart sound signal to be processed; and acquire the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed.

[0060] Optionally, in some possible implementations, when the acquisition unit acquires the wavelet function and decomposition layer number corresponding to the heart sound signal to be processed, it is specifically used to compare the heart sound signal to be processed with a preset function for similarity to obtain comparison information; determine the wavelet function based on the comparison information; determine the sampling frequency range corresponding to the heart sound signal to be processed; perform range decomposition in sequence based on the sampling frequency range to approach the effective frequency range; determine the number of decompositions corresponding to approaching the effective frequency range to obtain the decomposition layer number.

[0061] Optionally, in some possible implementations, the processing unit performs wavelet decomposition on the heart sound signal to be processed in sequence according to the decomposition number based on the wavelet function to obtain the wavelet coefficients corresponding to each decomposition number, and is specifically used to perform wavelet decomposition on the heart sound signal to be processed based on the wavelet function to obtain an input signal; divide the input signal using the frequency median of the input signal as a dividing line to obtain an approximate component indicating a low frequency band and a detail component indicating a high frequency band; and perform wavelet decomposition on the approximate components in sequence according to the decomposition number to obtain the wavelet coefficients corresponding to each decomposition number.

[0062] Optionally, in some possible implementations, the processing unit is also used to determine the effective frequency range corresponding to the heart sound signal to be processed; screen the detail component based on the effective frequency range; if the frequency range indicated by the detail component is outside the effective frequency range, setting the wavelet coefficient corresponding to the detail component to zero.

[0063] Optionally, in some possible implementations, the processing unit is further used to determine comparison indicators between the heart sound signal to be processed and the target heart sound signal, the comparison indicators including root mean square error and signal-to-noise ratio; and generate quality assessment information based on the comparison indicators.

[0064] Optionally, in some possible implementations, the processing unit is further used to adjust the layer adaptive factor according to the quality assessment information; and process the heart sound signal to be processed based on the adjusted layer adaptive factor to obtain an adjusted heart sound signal.

[0065] According to a third aspect of an embodiment of the present application, there is provided a heart sound signal processing device, including an input and output component and a processor;

[0066] The input and output components are used to obtain the heart sound signal to be processed;

[0067] The processor is used to perform denoising on the heart sound signal to be processed acquired by the input and output component by executing the heart sound signal processing method as described in the first aspect or any one of the first aspects.

[0068] According to a fourth aspect of an embodiment of the present application, there is provided a heart sound signal processing system, including an interactive client and a server;

[0069] The interactive client is used to obtain the heart sound signal to be processed and send the heart sound signal to be processed to the server, and display the processing result of the heart sound signal output by the server;

[0070] The server is used to perform denoising on the heart sound signal to be processed obtained by the interactive client by executing the heart sound signal processing method described in the first aspect or any one of the first aspects.

[0071] According to a fifth aspect of an embodiment of the present application, a computer program product is provided, comprising: a computer program, which, when executed by a processor, implements the method for processing heart sound signals described in the first aspect or any one of the first aspects.

[0072] It can be seen from the above technical solutions that the embodiments of the present application have the following advantages:

[0073] By obtaining the heart sound signal to be processed, as well as the wavelet function and decomposition level corresponding to the heart sound signal to be processed; then based on the wavelet function, the heart sound signal to be processed is sequentially decomposed by wavelet according to the decomposition level, and the wavelet coefficients corresponding to each decomposition level are obtained; and the denoising threshold corresponding to the heart sound signal to be processed is determined when the unbiased quasi-estimation is performed, so as to adjust the denoising threshold according to the decomposition level to obtain the adjustment threshold corresponding to each decomposition level; then based on the adjustment threshold, the wavelet coefficients of the corresponding decomposition level are screened to obtain the processing coefficient; and then the inverse wavelet transform is reconstructed according to the processing coefficient to obtain the target heart sound signal. Thus, the dynamic heart sound signal denoising process is realized. Since the dynamic denoising threshold is set based on different decomposition levels during the denoising process, the problem of excessive reduction of effective heart sound components is avoided, more effective heart sound components are retained, and the accuracy of the heart sound signal is improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0075] Figure 1 A network architecture diagram for the heart sound signal processing system;

[0076] Figure 2 A flow chart of processing a heart sound signal provided in an embodiment of the present application;

[0077] Figure 3 A flowchart of a method for processing heart sound signals provided in an embodiment of the present application;

[0078] Figure 4 A filtering effect diagram of a Butterworth bandpass filter provided in an embodiment of the present application;

[0079] Figure 5 A processing effect diagram of a wavelet decomposition provided in an embodiment of the present application;

[0080] Figure 6 A schematic diagram of a scenario of a method for processing heart sound signals provided in an embodiment of the present application;

[0081] Figure 7 A signal processing effect diagram of a heart sound signal provided in an embodiment of the present application;

[0082] Figure 8 A processing effect diagram of an unbiased likelihood estimation threshold method provided in an embodiment of the present application;

[0083] Fig. 9 A processing effect diagram of another unbiased likelihood estimation threshold method provided in an embodiment of the present application;

[0084] Fig.10 A signal processing effect diagram of another heart sound signal provided in an embodiment of the present application;

[0085] Fig.11 A schematic diagram of the structure of a heart sound signal processing device provided in an embodiment of the present application;

[0086] Fig.12 A schematic diagram of the structure of a terminal device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0087] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0088] It should be understood that the heart sound signal processing method provided in the present application can be applied to a system or program in a terminal device that includes a heart sound signal processing function, such as a teaching aid application. Specifically, the heart sound signal processing system can be run on a system such as Figure 1 In the network architecture shown in Figure 1As shown in FIG. 1 , it is a network architecture diagram of the operation of the heart sound signal processing system. As can be seen from the figure, the heart sound signal processing system can provide a processing process of heart sound signals from multiple information sources, that is, the heart sound signal to be processed is determined through the acquisition operation on the terminal side, and then sent to the server to perform targeted denoising on the heart sound signal; it can be understood that Figure 1 A variety of terminal devices are shown in FIG. 1 . The terminal device may be a computer device. In actual scenarios, more or fewer types of terminal devices may be involved in the processing of heart sound signals. The specific number and type depend on the actual scenario and are not limited here. In addition, Figure 1 One server is shown in the figure, but in actual scenarios, multiple servers may be involved, especially in multi-disciplinary scenarios, and the specific number of servers depends on the actual scenario.

[0089] In this embodiment, the server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, tablet computer, laptop computer, desktop computer, smart speaker, smart watch, etc., but is not limited to this. The terminal and the server can be directly or indirectly connected through wired or wireless communication, and the terminal and the server can be connected to form a blockchain network, which is not limited in this application.

[0090] It can be understood that the above-mentioned heart sound signal processing system can run on a personal mobile terminal, for example: as an application such as smart medical care, it can also run on a server, and it can also run on a third-party device to provide heart sound signal processing to obtain the processing result of the heart sound signal of the information source; the specific heart sound signal processing system can be run in the above-mentioned device in the form of a program, and it can also be run as a system component in the above-mentioned device, and it can also be used as a cloud service program. The specific operation mode depends on the actual scenario and is not limited here.

[0091] With the development of smart medicine, research and analysis of collected heart sound signals can help doctors understand the patient's heart condition more conveniently and accurately, avoiding misdiagnosis due to lack of experience or interference from environmental noise.

[0092] Heart sound signals are easily contaminated by noise during collection, such as lung sounds, power frequency interference noise, noise generated by the friction between the stethoscope head and clothing, and voices of people in the surrounding environment. The presence of these noises will drown out the effective pathological components in the heart sound signals, which is not conducive to the subsequent processing and analysis of the heart sound signals, such as heart sound segmentation and feature extraction.

[0093] Therefore, how to effectively reduce the noise of the collected heart sound signals and improve the quality of the heart sound signals becomes a difficult problem.

[0094] In order to solve the above problems, the present application proposes a method for processing heart sound signals, which is applied to Figure 2 In the process framework of processing the heart sound signal shown in FIG. Figure 2 As shown, it is a process architecture diagram of processing a heart sound signal provided in an embodiment of the present application. The heart sound signal is collected by the collection device on the terminal side and then sent to the server side. The heart sound signal is denoised by using the wavelet threshold method, and the threshold rule is improved according to the characteristics of the heart sound signal, so as to remove noise as much as possible while retaining the characteristic information of the heart sound.

[0095] It can be understood that the processing method of the heart sound signal provided in the present application can be written as a program to serve as a processing logic in a hardware system, or can be used as a processing device for heart sound signals, and the above processing logic is implemented in an integrated or external manner. As an implementation method, the processing device for the heart sound signal obtains the heart sound signal to be processed, as well as the wavelet function and decomposition level corresponding to the heart sound signal to be processed; then, based on the wavelet function, the heart sound signal to be processed is sequentially decomposed by wavelet according to the decomposition level to obtain the wavelet coefficients corresponding to each decomposition level; and the denoising threshold corresponding to the unbiased quasi-estimation of the heart sound signal to be processed is determined, so as to adjust the denoising threshold according to the decomposition level to obtain the adjustment threshold corresponding to each decomposition level; then, based on the adjustment threshold, the wavelet coefficients of the corresponding decomposition level are screened to obtain the processing coefficient; and then, the inverse wavelet transform is reconstructed according to the processing coefficient to obtain the target heart sound signal. Thereby, a dynamic heart sound signal denoising process is realized. Since the dynamic denoising threshold is set based on different decomposition levels during the denoising process, the problem of excessive attenuation of effective heart sound components is avoided, more effective heart sound components are retained, and the accuracy of the heart sound signal is improved.

[0096] Combined with the above process architecture, the following will introduce the processing method of the central tone signal of this application. Figure 3 , Figure 3 A flowchart of a method for processing a heart sound signal provided in an embodiment of the present application, the embodiment of the present application at least includes the following steps:

[0097] 301. Obtain a heart sound signal to be processed, and a wavelet function and a decomposition layer number corresponding to the heart sound signal to be processed.

[0098] In this embodiment, the heart sound signal to be processed may be collected in real time, such as a signal collected during a heart examination; or it may be data stored in a server, such as a heart sound signal obtained from a user's heart examination stored in a physical examination center.

[0099] Specifically, the heart sound signal denoising process of this embodiment is performed using wavelet transform, i.e. wavelet threshold heart sound denoising processing, which includes two steps: heart sound signal denoising preprocessing and wavelet threshold denoising processing. Among them, the heart sound signal denoising preprocessing is mainly to eliminate the noise components below 25Hz and above 400Hz, which are eliminated by using a Butterworth bandpass filter; the wavelet threshold denoising processing is to eliminate the noise components overlapping with the heart sound components, and the denoised heart sound signal is obtained through wavelet transform, wavelet threshold processing, and wavelet inverse transform reconstruction.

[0100] In a possible scenario, the denoising preprocessing process can be based on the Butterworth filter, that is, first obtain the collected heart sound signal; then configure the filter order, cutoff frequency and passband edge frequency in the Butterworth filter according to the effective frequency range corresponding to the collected heart sound signal; and process the collected heart sound signal based on the configured Butterworth filter to obtain the heart sound signal to be processed; and then obtain the wavelet function and decomposition layer number corresponding to the heart sound signal to be processed. This is considering that the denoising preprocessing of the heart sound signal mainly retains the effective heart sound components of 25Hz to 400Hz through the Butterworth filter. Among them, the frequency response curve of the Butterworth filter in the entire passband is maximally flat, and gradually decreases to zero in the stopband. As the filter order increases, the better its amplitude-frequency characteristics are, the higher the fidelity of the low-frequency signal. In addition, regardless of its order, the amplitude-to-diagonal frequency curve of the Butterworth filter can maintain the same shape, which is not available in other filters. Its formula is shown as follows:

[0101]

[0102] Where n is the order of the filter, here n = 5; ω c It is called the cut-off frequency, which is 500Hz here, ω is the passband edge frequency, and the design range is [25,400].

[0103] Specifically, the filtering effect of the Butterworth bandpass filter using this design is as follows Figure 4 shown. Figure 4 This is a filtering effect diagram of a Butterworth bandpass filter provided in an embodiment of the present application, and it can be seen that the denoising effect is obvious.

[0104] Furthermore, for the determination of the wavelet function and the number of decomposition layers in the wavelet transform process, since the wavelet transform is a very effective time-frequency analysis tool, it takes into account both the time domain and frequency domain information of the signal. When using the wavelet transform for signal analysis, for the same signal, the results of wavelet decomposition and reconstruction are different depending on the selected wavelet basis. When actually selecting the wavelet basis for wavelet transform, it is usually selected according to the similarity between the wavelet function and the signal to be analyzed. Therefore, the heart sound signal to be processed can be first compared with the preset function for similarity to obtain comparison information; then the wavelet function can be determined based on the comparison information.

[0105] In a possible scenario, db6 is a relatively suitable wavelet function in the analysis of heart sound signals, so db6 can be used as the basis function of wavelet decomposition. The processing based on db6 is as follows: Figure 5 As shown, Figure 5 A processing effect diagram of a wavelet decomposition provided in an embodiment of the present application.

[0106] In addition, for the selection of the number of wavelet decomposition layers, since the wavelet transform performs scaling transformation according to the scale function and the translation factor, it can decompose the signal into different frequency ranges while maintaining the time sequence, which is a very effective time-frequency transformation method. The selection of the number of decomposition layers is mainly based on the frequency characteristics of the heart sound signal. It is necessary to select a suitable number of decomposition layers to separate the noise from the original heart sound signal as much as possible. If the number of decomposition layers is too large, more time will be consumed during decomposition, and the useful heart sound components will be removed during denoising; if the number of decomposition layers is too small, the noise cannot be separated from the heart sound signal, resulting in unsatisfactory denoising effect. Therefore, it is necessary to match the number of layers in combination with the sampling frequency range of the heart sound signal, that is, first determine the sampling frequency range corresponding to the heart sound signal to be processed; then perform range decomposition in sequence based on the sampling frequency range to approach the effective frequency range; and determine the number of decompositions corresponding to the effective frequency range to obtain the number of decomposition layers. For example, the sampling frequency of the processed heart sound signal is 1000Hz. According to the Nyquist sampling theorem, the highest frequency of the signal is 500Hz. According to the frequency characteristics of the heart sound signal, the lowest frequency of the effective component of the heart sound signal is 10Hz, and the highest component is at 220Hz. In addition, as the number of wavelet decomposition layers increases, the amount of calculation of the wavelet transform also increases. According to the principle of the Mallat algorithm, this embodiment selects 5 layers for wavelet decomposition, which can maximize the approximation of the lowest effective frequency of the heart sound signal.

[0107] 302. Based on the wavelet function, the heart sound signal to be processed is sequentially decomposed by wavelet according to the number of decomposition levels to obtain wavelet coefficients corresponding to each number of decomposition levels.

[0108] In this embodiment, wavelet decomposition is used to process the noise components overlapping with the effective heart sound components, that is, the heart sound signal processed in the first step is first decomposed into different frequency components through wavelet transform, and then the different frequency components are subjected to threshold denoising, and finally the signal is reconstructed through inverse wavelet transform to obtain the denoised heart sound signal.

[0109] Specifically, the wavelet decomposition process is carried out layer by layer, that is, first, the heart sound signal to be processed is decomposed by wavelet based on the wavelet function to obtain the input signal; then the frequency median of the input signal is used as the dividing line to divide the input signal to obtain an approximate component indicating a low-frequency band and a detail component indicating a high-frequency band; next, the approximate components are decomposed by wavelet in turn according to the number of decomposition layers to obtain the wavelet coefficients corresponding to each decomposition layer.

[0110] In one possible scenario, the wavelet decomposition process is as follows Figure 6 As stated, Figure 6 A schematic diagram of a scenario of a method for processing heart sound signals provided in an embodiment of the present application; the wavelet decomposition uses the Mallat algorithm to decompose the original signal S with a frequency of F into N layers, that is, firstly, the signal is decomposed into an approximate component and a detail component with the middle value of the frequency as the dividing line, the approximate component represents the low frequency band of the original signal S, the frequency range is 0-500, and the detail component is the high frequency band of S, the frequency range is 500-1000; then, the approximate component continues to be decomposed, and each decomposition is carried out by halving the current signal frequency. The schematic diagram of the wavelet decomposition of 5 layers is shown in FIG. Figure 6 As shown, S is the original signal, the approximate component is represented by CA, the detail component is represented by CD, and the following number represents the number of decomposition levels.

[0111] It can be understood that, considering that the heart sound signal has a valid range, in order to reduce the amount of data processing, the signals outside the valid frequency range can be discarded, that is, first determine the valid frequency range corresponding to the heart sound signal to be processed; then screen the detail component based on the valid frequency range; if the frequency range indicated by the detail component is outside the valid frequency range, the wavelet coefficient corresponding to the detail component is set to zero.

[0112] Based on the above process, that is, through the processing of 5 layers of decomposition, we can get Figure 7 The results shown, Figure 7This is a signal processing effect diagram of a heart sound signal provided in an embodiment of the present application; and according to the decomposition principle of the Mallat algorithm, it can be known that the main components of heart sounds are concentrated in CD2, CD3 and CD4, and CD5 and CA5 also contain effective components of heart sounds, while the lowest frequency of CD1 is higher than the effective frequency of heart sounds, and it is regarded as a noise component. During denoising, the coefficient is directly set to zero, and the other components are processed using appropriate thresholds and threshold functions. Finally, the inverse wavelet transform is used to reconstruct the signal to obtain the denoised heart sound signal S=CA5+CD5+CD4+CD3+CD2. Next, the process of determining the threshold value for dynamic adjustment based on each decomposition layer is introduced, that is, after selecting the appropriate wavelet function and the number of wavelet decomposition layers, the wavelet coefficients of each layer after decomposition need to be processed to achieve the purpose of denoising. Figure 6 In the example shown, the high-frequency components of the first layer of wavelet decomposition are regarded as noise components, and their coefficients are directly set to zero or discarded, and the remaining wavelet coefficients of each layer are processed using appropriate thresholds and threshold functions.

[0113] 303. Determine a denoising threshold corresponding to unbiased simulated estimation of the heart sound signal to be processed, so as to adjust the denoising threshold according to the number of decomposition layers to obtain an adjustment threshold corresponding to each number of decomposition layers.

[0114] In this embodiment, among the wavelet coefficients of each layer decomposed by wavelet transform, the wavelet coefficients of noise signal are relatively small, while the wavelet coefficients corresponding to effective signal are relatively large. According to this characteristic, the wavelet coefficients below the threshold can be set to zero or smoothed by threshold processing to suppress the influence of high-frequency noise.

[0115] Therefore, the selection of the threshold has a great influence on the denoising effect. If the threshold is selected too small, too much noise component is retained and the denoising effect is poor. On the contrary, if the threshold is selected too large, it will lead to excessive denoising and the loss of effective components of heart sounds. Among them, the Stein unbiased likelihood estimation threshold method is an adaptive threshold selection method, but this method mainly calculates the threshold based on the variance of the high-frequency coefficients of the first layer of wavelet decomposition to estimate the noise, and uses this threshold to process the wavelet coefficients of other layers. It does not take into account the problem of reducing high-frequency components and is easy to eliminate useful heart sound components. Therefore, this embodiment improves the Stein unbiased likelihood estimation threshold method on the basis of it, effectively alleviating the problem of excessive reduction of effective heart sound components and retaining more effective heart sound components.

[0116] Specifically, for the process of determining the denoising threshold based on the Stein unbiased likelihood estimation threshold method, first determine the similarity index between the heart sound signal to be processed and the wavelet function to obtain the signal element; then square the signal element and arrange the calculation results from large to small to obtain an estimation sequence; and calculate the risk estimation value corresponding to each signal element in the estimation sequence according to the unbiased likelihood estimation formula; and then use the square root of the signal element with the smallest risk estimation value as the denoising threshold. Specifically, corresponding to Figure 6 In the example, the square values ​​of each element in the signal S are first calculated, and then they are sorted from large to small to obtain a new sequence M. Finally, the risk estimate is calculated for each element in M ​​according to the following formula:

[0117]

[0118] Where R is the risk estimate, L is the length of the signal, k represents the index in M ​​corresponding to the element currently being calculated, and M represents the estimation sequence.

[0119] Furthermore, the square root of the element in M ​​that minimizes the risk estimate is selected as the threshold, and its formula is expressed as:

[0120]

[0121] Among them, λ3 is the denoising threshold used in the first layer decomposition process, that is, the denoising threshold; σ3 represents the noise signal, and M represents the estimated sequence.

[0122] It is understandable that since the heart sound signal is a low-frequency signal, the noise component is mainly distributed in the high-frequency range. As the number of wavelet decomposition layers increases, the high-frequency components contained in each layer of wavelet coefficients are decreasing. The Stein unbiased likelihood estimation threshold method mainly estimates the variance of the noise based on the high-frequency coefficients of the first layer of decomposition to calculate the threshold. This threshold is used to process the wavelet coefficients of other layers without considering the problem of reducing high-frequency components. It is easy to eliminate useful heart sound components. Its effect is as follows Figure 8 As shown, Figure 8 The figure is a processing effect diagram of an unbiased likelihood estimation threshold method provided in an embodiment of the present application. To this end, this embodiment introduces a layer number adaptive factor based on the denoising threshold, and according to the low-frequency characteristics of the heart sound signal and the Stein unbiased likelihood estimation threshold method, which only relies on the first layer decomposition coefficient as the noise component to calculate the threshold, the problem of over-attenuation occurs on the heart sound signal. The threshold of each layer is adaptively adjusted based on the number of decomposition layers to reduce the problem of over-attenuation.

[0123] Specifically, for the process of dynamic threshold adjustment according to the layer number adaptive factor, first determine the denoising threshold corresponding to the unbiased quasi-estimation of the processed heart sound signal; then configure the layer number adaptive factors corresponding to each decomposition layer number according to the number of decomposition layers, and the layer number adaptive factor increases with the increase of the number of decomposition layers; and adjust the denoising threshold based on the layer number adaptive factor to obtain the adjustment threshold corresponding to each decomposition layer number, and the adjustment threshold decreases with the increase of the layer number adaptive factor. Therefore, the formula for determining the adjustment threshold is as follows:

[0124]

[0125] Among them, λ3 is the denoising threshold used in the first layer decomposition process, λ j is the threshold of the jth layer, α j is the layer number adaptation factor.

[0126] Specifically, as for the composition of the layer adaptive factor, as the number of layers of wavelet decomposition increases, the high-frequency components contained in the wavelet coefficients of each layer are reduced. In order to eliminate the proportional relationship between the frequencies of heart sound signals and more intuitively simulate the frequency distribution of high-frequency components when they are reduced, the layer adaptive factor can be dynamically determined by configuring the layer coefficient; that is, first determine the number of processed layers corresponding to the number of decomposition layers; then configure the layer coefficient based on the number of decomposition layers and the number of processed layers; determine the layer adaptive factor corresponding to each decomposition layer according to the layer coefficient, that is, take the logarithm to eliminate the proportional relationship between the frequencies of heart sound signals. The specific formula is as follows:

[0127]

[0128] Among them, α j is the adaptive factor of the number of layers; j is the number of decomposed layers; J is the number of processed layers, that is, the number of decomposed layers.

[0129] From the above formula, we can see that α j It is related to the number of wavelet decomposition layers. As the number of decomposition layers increases, α j Increase, λ j Reduce, so as to retain the low-frequency components as much as possible, thereby achieving the effect of retaining more effective heart sound signal components.

[0130] In one possible scenario, after obtaining the thresholds of each layer using the improved Stein unbiased likelihood estimation threshold method, the above-mentioned dynamic adjustment threshold is used to denoise the wavelet coefficients of each layer. The result is as follows: Fig. 9 As shown, Fig. 9This is a processing effect diagram of another unbiased likelihood estimation threshold method provided in an embodiment of the present application; as can be seen from the figure, the unbiased likelihood estimation threshold method provided in this embodiment retains more useful signal components, significantly improves the signal-to-noise ratio, and has a good denoising effect; that is, through the adaptive threshold rule, the denoising effect is guaranteed while greatly retaining the details of the heart sound signal components.

[0131] 304. Based on the adjustment threshold, the wavelet coefficients of the corresponding decomposition layer are screened to obtain processing coefficients.

[0132] In this embodiment, in the process of screening the wavelet coefficients of the corresponding decomposition level based on the adjustment threshold, the wavelet coefficients below the adjustment threshold are set to zero or smoothed by adjusting the threshold to suppress the influence of high-frequency noise.

[0133] 305. Perform inverse wavelet transform reconstruction according to the processing coefficients to obtain a target heart sound signal.

[0134] In this embodiment, the inverse wavelet transform reconstruction is a process of synthesizing the approximate coefficients and detail coefficients obtained by decomposition into the original signal through the inverse wavelet transform (IDWT). In the inverse wavelet transform reconstruction, the inverse wavelet transform is needed to synthesize the approximate coefficients and detail coefficients into the original signal. The inverse wavelet transform process is to realize the synthesis of the signal through a series of filters and upsampling operations. The specific process is to filter the approximate coefficients and detail coefficients through upsampling and filtering, and add the filtered signals to obtain the reconstructed signal.

[0135] In a possible scenario, the reconstructed signal obtained based on this embodiment is as follows: Fig.10 As shown, Fig.10 Another signal processing effect diagram of a heart sound signal provided in an embodiment of the present application shows the effect of denoising and reconstruction through inverse wavelet transform. It can be seen that more useful signal components are retained, the signal-to-noise ratio is significantly improved, and the denoising effect is good.

[0136] Furthermore, in order to evaluate the performance of the improved wavelet threshold heart sound denoising, quality assessment may be performed, that is, firstly determining a comparison index between the heart sound signal to be processed and the target heart sound signal, and then generating quality assessment information based on the comparison index.

[0137] The comparison index may include root mean square error and signal-to-noise ratio, that is, the root mean square error RMSE (Root-Mean-Square Error) and signal-to-noise ratio SNR (Signal-to-Noise Ratio) are used to evaluate the denoising effect of the heart sound signal.

[0138] Specifically, the root mean square error (RMSE) is mainly used to measure the deviation between the denoised signal and the original signal. The smaller the deviation, the better the denoising effect. The formula is as follows:

[0139]

[0140] Among them, S' represents the signal after denoising, S represents the original signal, and N is the length of the data.

[0141] In addition, the signal-to-noise ratio (SNR) is the ratio of the sum of squares of the original signal to the sum of squares of the difference between the original signal and the denoised signal. The larger the value, the better the denoising effect. The formula is as follows:

[0142]

[0143] Among them, S' represents the signal after denoising, S represents the original signal, and N is the length of the data.

[0144] Therefore, the denoising performance of the improved wavelet threshold method proposed in this embodiment is evaluated. As shown in Table 1, it can be seen that the denoising effect of the improved method is significantly better than that before the improvement.

[0145] Table 1 Comparison of evaluation indicators before and after improvement of this embodiment

[0146]

[0147] In addition, if the effect during the quality assessment process does not meet expectations, the key factor (layer adaptive factor) in this embodiment can be adjusted, that is, the layer adaptive factor can be adjusted according to the quality assessment information, for example, the layer coefficient of the layer adaptive factor can be fine-tuned (increase, decrease or add a weighting coefficient); then the processed heart sound signal is processed based on the adjusted layer adaptive factor to obtain an adjusted heart sound signal, thereby realizing a dynamic layer adaptive factor adjustment process based on quality assessment.

[0148] In combination with the above embodiments, it can be known that by obtaining the heart sound signal to be processed, and the wavelet function and decomposition level corresponding to the heart sound signal to be processed; then based on the wavelet function, the heart sound signal to be processed is sequentially decomposed by wavelet according to the decomposition level, and the wavelet coefficients corresponding to each decomposition level are obtained; and the denoising threshold corresponding to the heart sound signal to be processed is determined when the unbiased quasi-estimation is performed, so as to adjust the denoising threshold according to the decomposition level to obtain the adjustment threshold corresponding to each decomposition level; then the wavelet coefficients of the corresponding decomposition level are screened based on the adjustment threshold to obtain the processing coefficient; and then the inverse wavelet transform is reconstructed according to the processing coefficient to obtain the target heart sound signal. Thus, a dynamic heart sound signal denoising process is realized. Since the dynamic denoising threshold is set based on different decomposition levels during the denoising process, the problem of excessive reduction of effective heart sound components is avoided, more effective heart sound components are retained, and the accuracy of the heart sound signal is improved.

[0149] In order to better implement the above solution of the embodiment of the present application, the following also provides related devices for implementing the above solution. Fig.11 , Fig.11 This is a structural diagram of a heart sound signal processing device provided in an embodiment of the present application. The heart sound signal processing device 1100 includes:

[0150] An acquisition unit 1101 is used to acquire a heart sound signal to be processed, and a wavelet function and a decomposition layer number corresponding to the heart sound signal to be processed;

[0151] The processing unit 1102 is configured to perform wavelet decomposition on the heart sound signal to be processed in sequence according to the decomposition levels based on the wavelet function to obtain wavelet coefficients corresponding to each decomposition level;

[0152] The processing unit 1102 is further configured to determine a denoising threshold corresponding to unbiased simulated estimation of the heart sound signal to be processed, so as to adjust the denoising threshold according to the number of decomposition layers to obtain an adjustment threshold corresponding to each number of decomposition layers;

[0153] The processing unit 1102 is further configured to screen the wavelet coefficients of the corresponding decomposition layer based on the adjustment threshold to obtain processing coefficients;

[0154] The processing unit 1102 is further configured to perform inverse wavelet transform reconstruction according to the processing coefficients to obtain a target heart sound signal.

[0155] Optionally, in some possible implementations, the processing unit 1102 determines the denoising threshold corresponding to the unbiased simulated estimation of the heart sound signal to be processed, and adjusts the denoising threshold according to the number of decomposition layers to obtain the adjustment threshold corresponding to each decomposition layer, which is specifically used to determine the denoising threshold corresponding to the unbiased simulated estimation of the heart sound signal to be processed; configures the layer number adaptive factor corresponding to each decomposition layer according to the number of decomposition layers, and the layer number adaptive factor increases with the increase of the decomposition layer number; adjusts the denoising threshold based on the layer number adaptive factor to obtain the adjustment threshold corresponding to each decomposition layer, and the adjustment threshold decreases with the increase of the layer number adaptive factor.

[0156] Optionally, in some possible implementations, when the processing unit 1102 configures the layer adaptive factors corresponding to each decomposition layer number according to the decomposition layer number, it is specifically used to determine the number of processed layers corresponding to the decomposition layer number; configure the layer coefficient based on the decomposition layer number and the processed layer number; and determine the layer adaptive factors corresponding to each decomposition layer number according to the layer coefficient.

[0157] Optionally, in some possible implementations, when the processing unit 1102 determines the denoising threshold corresponding to the unbiased probabilistic estimation of the heart sound signal to be processed, it is specifically used to determine the similarity index corresponding to the heart sound signal to be processed and the wavelet function to obtain a signal element; square the signal element, and arrange the calculation results from large to small to obtain an estimation sequence; calculate the risk estimation value corresponding to each signal element in the estimation sequence according to the unbiased probabilistic estimation formula; and use the square root of the signal element with the smallest risk estimation value as the denoising threshold.

[0158] Optionally, in some possible implementations, when the acquisition unit 1101 acquires the heart sound signal to be processed, as well as the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed, it is specifically used to acquire the collected heart sound signal; configure the filter order, cutoff frequency and passband edge frequency in the Butterworth filter according to the effective frequency range corresponding to the collected heart sound signal; process the collected heart sound signal based on the configured Butterworth filter to obtain the heart sound signal to be processed; and acquire the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed.

[0159] Optionally, in some possible implementations, when the acquisition unit 1101 acquires the wavelet function and decomposition layer number corresponding to the heart sound signal to be processed, it is specifically used to compare the heart sound signal to be processed with a preset function for similarity to obtain comparison information; determine the wavelet function based on the comparison information; determine the sampling frequency range corresponding to the heart sound signal to be processed; perform range decomposition in sequence based on the sampling frequency range to approach the effective frequency range; determine the number of decompositions corresponding to approaching the effective frequency range to obtain the decomposition layer number.

[0160] Optionally, in some possible implementations, the processing unit 1102 performs wavelet decomposition on the heart sound signal to be processed according to the decomposition level in sequence based on the wavelet function to obtain wavelet coefficients corresponding to each decomposition level. Specifically, it is used to perform wavelet decomposition on the heart sound signal to be processed based on the wavelet function to obtain an input signal; divide the input signal using the frequency median of the input signal as a dividing line to obtain an approximate component indicating a low frequency band and a detail component indicating a high frequency band; and perform wavelet decomposition on the approximate components in sequence according to the decomposition level to obtain wavelet coefficients corresponding to each decomposition level.

[0161] Optionally, in some possible implementations, the processing unit 1102 is further used to determine a valid frequency range corresponding to the heart sound signal to be processed; screen the detail component based on the valid frequency range; if the frequency range indicated by the detail component is outside the valid frequency range, setting the wavelet coefficient corresponding to the detail component to zero.

[0162] Optionally, in some possible implementations, the processing unit 1102 is further used to determine comparison indicators between the heart sound signal to be processed and the target heart sound signal, the comparison indicators including root mean square error and signal-to-noise ratio; and generate quality assessment information based on the comparison indicators.

[0163] Optionally, in some possible implementations, the processing unit 1102 is further used to adjust the layer adaptive factor according to the quality assessment information; and process the heart sound signal to be processed based on the adjusted layer adaptive factor to obtain an adjusted heart sound signal.

[0164] The processing device of the heart sound signal proposed in the embodiment of the present application obtains the heart sound signal to be processed, and the wavelet function and decomposition level corresponding to the heart sound signal to be processed; then, based on the wavelet function, the heart sound signal to be processed is sequentially decomposed by wavelet according to the decomposition level to obtain the wavelet coefficients corresponding to each decomposition level; and the corresponding denoising threshold when the heart sound signal to be processed is unbiasedly estimated is determined, so as to adjust the denoising threshold according to the decomposition level to obtain the adjustment threshold corresponding to each decomposition level; then, based on the adjustment threshold, the wavelet coefficients of the corresponding decomposition level are screened to obtain the processing coefficient; and then, the inverse wavelet transform is reconstructed according to the processing coefficient to obtain the target heart sound signal. Thus, a dynamic heart sound signal denoising process is realized. Since the dynamic denoising threshold is set based on different decomposition levels during the denoising process, the problem of excessive reduction of effective heart sound components is avoided, more effective heart sound components are retained, and the accuracy of the heart sound signal is improved.

[0165] The heart sound signal processing device provided in this embodiment belongs to the same application concept as the method provided in the above embodiments of this application, and can execute the heart sound signal processing method provided in any of the above embodiments of this application, and has the corresponding functional modules and beneficial effects of executing the heart sound signal processing method. For technical details not fully described in this embodiment, please refer to the specific processing content of the heart sound signal processing method provided in the above embodiments of this application, and will not be repeated here.

[0166] The functions implemented by the above acquisition unit 1101 and processing unit 1102 may be implemented by the same or different processors, respectively, and are not limited in the embodiments of the present application.

[0167] It should be understood that each unit in the above device can be implemented in the form of a processor calling software. For example, the device includes a processor, the processor is connected to a memory, and instructions are stored in the memory. The processor calls the instructions stored in the memory to implement any of the above methods or realize the functions of each unit of the device, wherein the processor can be a general-purpose processor, such as a CPU or a microprocessor, etc., and the memory can be a memory in the device or a memory outside the device. Alternatively, the unit in the device can be implemented in the form of a hardware circuit, and the functions of some or all units can be realized by designing the hardware circuit. The hardware circuit can be understood as one or more processors; for example, in one implementation, the hardware circuit is an ASIC, and the functions of some or all units above are realized by designing the logical relationship of the components in the circuit; for another example, in another implementation, the hardware circuit can be implemented by PLD, taking FPGA as an example, which can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured by a configuration file, so as to realize the functions of some or all units above. All units of the above device can be implemented in the form of a processor calling software, or in the form of a hardware circuit, or in the form of a processor calling software, and the remaining part can be implemented in the form of a hardware circuit.

[0168] In an embodiment of the present application, a processor is a circuit with the ability to process signals. In one implementation, the processor may be a circuit with the ability to read and run instructions, such as a CPU, a microprocessor, a GPU, or a DSP; in another implementation, the processor may implement certain functions through the logical relationship of a hardware circuit, and the logical relationship of the hardware circuit is fixed or reconfigurable, such as a hardware circuit implemented by an ASIC or PLD, such as an FPGA. In a reconfigurable hardware circuit, the process of the processor loading a configuration document to implement the hardware circuit configuration can be understood as the process of the processor loading instructions to implement the functions of some or all of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as an NPU, TPU, DPU, etc.

[0169] It can be seen that each unit in the above device can be one or more processors (or processing circuits) configured to implement the above method, such as: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0170] In addition, all or part of the units in the above device can be integrated together, or can be implemented independently. In one implementation, these units are integrated together and implemented in the form of a SOC. The SOC may include at least one processor for implementing any of the above methods or implementing the functions of each unit of the device. The type of the at least one processor may be different, for example, including a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0171] The embodiment of the present application also provides a heart sound signal processing device; the heart sound signal processing device includes a processor and an input-output component. The input-output component is used to obtain the heart sound signal to be processed;

[0172] The processor is used to perform denoising on the heart sound signal to be processed acquired by the input-output component by executing any one of the heart sound signal processing methods in the above embodiments.

[0173] The above-mentioned interface circuit can be any interface circuit that can realize the data communication function, for example, it can be a USB interface circuit, a Type-C interface circuit, a serial port circuit, a PCIE circuit, etc.

[0174] Optionally, the embodiment of the present application further provides a heart sound signal processing system, including an interactive client and a server. The interactive client is used to obtain the heart sound signal to be processed and send the heart sound signal to be processed to the server, and display the processing result of the heart sound signal output by the server;

[0175] The server is used to perform denoising on the heart sound signal to be processed obtained by the interactive client by executing the heart sound signal processing method described in any one of the above embodiments.

[0176] Another embodiment of the present application also provides a heart sound signal processing device, see Fig.12 As shown, the device includes:

[0177] Memory 1200 and processor 1210;

[0178] The memory 1200 is connected to the processor 1210 and is used to store programs;

[0179] The processor 1210 is used to implement the heart sound signal processing method disclosed in any of the above embodiments by running the program stored in the memory 1200.

[0180] Specifically, the heart sound signal processing device may further include: a bus, a communication interface 1220 , an input device 1230 and an output device 1240 .

[0181] The processor 1210, the memory 1200, the communication interface 1220, the input device 1230 and the output device 1240 are connected to each other via a bus.

[0182] A bus may include a pathway that transfers information between components of a computer system.

[0183] The processor 1210 may be a general-purpose processor, such as a general-purpose central processing unit (CPU), a microprocessor, etc., or an application-specific integrated circuit (ASIC), or one or more integrated circuits for controlling the execution of the program of the scheme of the present invention. It may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0184] The processor 1210 may include a main processor, and may also include a baseband chip, a modem, and the like.

[0185] The memory 1200 stores a program for executing the technical solution of the present invention, and may also store an operating system and other key services. Specifically, the program may include a program code, and the program code includes a computer operation instruction. More specifically, the memory 1200 may include a read-only memory (ROM), other types of static storage devices that can store static information and instructions, a random access memory (RAM), other types of dynamic storage devices that can store information and instructions, a disk storage, a flash, and the like.

[0186] The input device 1230 may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.

[0187] Output device 1240 may include devices that allow information to be output to a user, such as a display screen, printer, speaker, etc.

[0188] The communication interface 1220 may include any transceiver or the like to communicate with other devices or communication networks, such as Ethernet, a radio access network (RAN), a wireless local area network (WLAN), etc.

[0189] The processor 1210 executes the program stored in the memory 1200 and calls other devices, which can be used to implement each step of any heart sound signal processing method provided in the above embodiments of the present application.

[0190] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the method for processing heart sound signals according to various embodiments of the present application described in any of the above embodiments of this specification.

[0191] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages, such as Java, C++, etc., and conventional procedural programming languages, such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as an independent software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0192] In addition, the embodiment of the present application may also be a storage medium on which a computer program is stored. The computer program is executed by a processor to execute the steps of the method for processing heart sound signals according to various embodiments of the present application described in any of the above embodiments of this specification, and specifically the following steps may be implemented:

[0193] Acquire a heart sound signal to be processed, and a wavelet function and a decomposition layer number corresponding to the heart sound signal to be processed;

[0194] Based on the wavelet function, the heart sound signal to be processed is sequentially subjected to wavelet decomposition according to the decomposition levels to obtain wavelet coefficients corresponding to each decomposition level;

[0195] Determine a denoising threshold corresponding to unbiased simulated estimation of the heart sound signal to be processed, so as to adjust the denoising threshold according to the number of decomposition levels to obtain adjustment thresholds corresponding to each number of decomposition levels;

[0196] Based on the adjustment threshold, the wavelet coefficients of the corresponding decomposition layer are screened to obtain processing coefficients;

[0197] The inverse wavelet transform is performed according to the processing coefficients to reconstruct the target heart sound signal.

[0198] For the aforementioned method embodiments, for the sake of simplicity, they are all described as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the order of the actions described, because according to the present application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0199] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0200] The steps in the methods of each embodiment of the present application can be adjusted in order, combined and deleted according to actual needs, and the technical features recorded in each embodiment can be replaced or combined.

[0201] The modules and sub-modules in the devices and terminals of the various embodiments of the present application can be combined, divided and deleted according to actual needs.

[0202] In the several embodiments provided in the present application, it should be understood that the disclosed terminals, devices and methods can be implemented in other ways. For example, the terminal embodiments described above are only schematic, for example, the division of modules or submodules is only a logical function division, and there may be other division methods in actual implementation, for example, multiple submodules or modules can be combined or integrated into another module, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0203] The modules or submodules described as separate components may or may not be physically separated, and the components of the modules or submodules may or may not be physical modules or submodules, that is, they may be located in one place, or they may be distributed on multiple network modules or submodules. Some or all of the modules or submodules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0204] In addition, each functional module or submodule in each embodiment of the present application may be integrated into one processing module, or each module or submodule may exist physically separately, or two or more modules or submodules may be integrated into one module. The above-mentioned integrated modules or submodules may be implemented in the form of hardware or in the form of software functional modules or submodules.

[0205] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in this embodiment can be implemented with electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in terms of function in the above description. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0206] The steps of the method or algorithm described in the embodiments disclosed in this embodiment can be directly implemented by hardware, software units executed by a processor, or a combination of the two. The software unit can be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the technical field.

[0207] Finally, it should be noted that, in the present embodiment, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply that there is any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements inherent to such process, method, article or equipment. In the absence of more restrictions, the elements defined by the statement "comprise one..." do not exclude the existence of other identical elements in the process, method, article or equipment including the elements.

[0208] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined in the present embodiment may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown in the present embodiment, but will conform to the widest scope consistent with the principles and novel features disclosed in the present embodiment.

Claims

1. A method for processing heart sound signals, characterized in that: include: Acquire a heart sound signal to be processed, and a wavelet function and a decomposition layer number corresponding to the heart sound signal to be processed; Based on the wavelet function, the heart sound signal to be processed is sequentially subjected to wavelet decomposition according to the decomposition levels to obtain wavelet coefficients corresponding to each decomposition level; Determine a denoising threshold corresponding to unbiased simulated estimation of the heart sound signal to be processed, so as to adjust the denoising threshold according to the number of decomposition levels to obtain adjustment thresholds corresponding to each number of decomposition levels; Based on the adjustment threshold, the wavelet coefficients of the corresponding decomposition layer are screened to obtain processing coefficients; An inverse wavelet transform is performed according to the processing coefficients to reconstruct the target heart sound signal.

2. The method according to claim 1, characterized in that The step of determining a denoising threshold corresponding to an unbiased simulated estimation of the heart sound signal to be processed, and adjusting the denoising threshold according to the number of decomposition layers to obtain an adjustment threshold corresponding to each number of decomposition layers, includes: Determine a denoising threshold corresponding to unbiased likelihood estimation of the heart sound signal to be processed; According to the decomposition layer number, a layer number adaptive factor corresponding to each decomposition layer number is configured respectively, and the layer number adaptive factor increases as the decomposition layer number increases; The denoising threshold is adjusted based on the layer adaptive factor to obtain an adjustment threshold corresponding to each decomposition layer number, and the adjustment threshold decreases as the layer adaptive factor increases.

3. The method according to claim 2, characterized in that The configuring the layer number adaptive factors corresponding to each decomposition layer number respectively according to the decomposition layer number includes: Determining the number of processed layers corresponding to the number of decomposed layers; configuring a layer number coefficient based on the decomposed layer number and the processed layer number; The layer number adaptive factors corresponding to each decomposition layer number are determined according to the layer number coefficients.

4. The method according to claim 2, characterized in that: The step of determining a denoising threshold corresponding to unbiased likelihood estimation of the heart sound signal to be processed includes: Determining a similarity index between the heart sound signal to be processed and the wavelet function to obtain a signal element; Performing square calculation on the signal elements, and arranging the calculation results from large to small to obtain an estimation sequence; Calculate the risk estimation value corresponding to each signal element in the estimation sequence according to the unbiased quasi-estimation formula; The square root of the signal element with the smallest risk estimation value is used as the denoising threshold.

5. The method according to claim 1, characterized in that The obtaining of the heart sound signal to be processed, and the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed, comprises: Acquire and collect heart sound signals; According to the effective frequency range corresponding to the collected heart sound signal, the filter order, cutoff frequency and passband edge frequency in the Butterworth filter are configured; Processing the collected heart sound signal based on the configured Butterworth filter to obtain the heart sound signal to be processed; Obtain the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed.

6. The method according to claim 5, characterized in that The step of obtaining the wavelet function and the number of decomposition layers corresponding to the heart sound signal to be processed includes: Comparing the heart sound signal to be processed with a preset function for similarity to obtain comparison information; determining the wavelet function based on the comparison information; Determine the sampling frequency range corresponding to the heart sound signal to be processed; Performing range decomposition in sequence based on the sampling frequency range to approach the effective frequency range; The number of decompositions corresponding to the effective frequency range is determined to obtain the number of decomposition layers.

7. The method according to claim 1, characterized in that The step of sequentially performing wavelet decomposition on the heart sound signal to be processed according to the number of decomposition levels based on the wavelet function to obtain wavelet coefficients corresponding to each number of decomposition levels includes: Performing wavelet decomposition on the heart sound signal to be processed based on the wavelet function to obtain an input signal; The input signal is divided by using the frequency median of the input signal as a dividing line to obtain an approximate component indicating a low frequency band and a detail component indicating a high frequency band; The approximate components are sequentially subjected to wavelet decomposition according to the decomposition levels to obtain wavelet coefficients corresponding to each decomposition level.

8. The method according to claim 7, characterized in that The method further comprises: Determining a valid frequency range corresponding to the heart sound signal to be processed; screening the detail components based on the effective frequency range; If the frequency range indicated by the detail component is outside the effective frequency range, the wavelet coefficients corresponding to the detail component are set to zero.

9. The method according to any one of claims 1 to 8, characterized in that: The method further comprises: Determining a comparison index between the heart sound signal to be processed and the target heart sound signal, wherein the comparison index includes a root mean square error and a signal-to-noise ratio; Quality assessment information is generated based on the comparison indicator.

10. The method according to claim 9, characterized in that The method further comprises: Adjusting the layer number adaptation factor according to the quality assessment information; The heart sound signal to be processed is processed based on the adjusted layer adaptive factor to obtain an adjusted heart sound signal.

11. A heart sound signal processing device, characterized in that: include: An acquisition unit, used for acquiring a heart sound signal to be processed, and a wavelet function and a decomposition layer number corresponding to the heart sound signal to be processed; A processing unit, configured to perform wavelet decomposition on the heart sound signal to be processed in sequence according to the decomposition levels based on the wavelet function, to obtain wavelet coefficients corresponding to each decomposition level; The processing unit is further used to determine a denoising threshold corresponding to the unbiased simulated estimation of the heart sound signal to be processed, so as to adjust the denoising threshold according to the number of decomposition layers to obtain an adjustment threshold corresponding to each number of decomposition layers; The processing unit is further used to screen the wavelet coefficients of the corresponding decomposition layer based on the adjustment threshold to obtain processing coefficients; The processing unit is further used to perform inverse wavelet transform reconstruction according to the processing coefficients to obtain a target heart sound signal.

12. A heart sound signal processing device, comprising an input and output component and a processor; The input and output components are used to obtain the heart sound signal to be processed; The processor is used to perform denoising on the heart sound signal to be processed acquired by the input and output component by executing the heart sound signal processing method as described in any one of claims 1 to 10.

13. A heart sound signal processing system, comprising an interactive client and a server; The interactive client is used to obtain the heart sound signal to be processed and send the heart sound signal to be processed to the server, and display the processing result of the heart sound signal output by the server; The server is used to perform denoising on the heart sound signal to be processed obtained by the interactive client by executing the heart sound signal processing method according to any one of claims 1 to 10.

14. A computer program product, characterized in that include: A computer program, wherein when the computer program is executed by a processor, the method for processing a heart sound signal according to any one of claims 1 to 10 is implemented.

Images