Heart-sound signal quality index assessment method, and electronic device and readable storage medium

Abstract

A heart-sound signal quality index assessment method, and an electronic device and a readable storage medium. The method comprises: collecting a heart sound signal (S10); performing feature extraction on the heart sound signal, so as to obtain a signal feature, wherein the signal feature comprises one or more of a local variance feature, a local peak feature, a signal energy feature and a zero-crossing rate feature (S20); and inputting the signal feature into a pre-trained signal quality index assessment model, and outputting a signal quality index assessment result (S30).

Description

Heart sound signal quality evaluation method, electronic device and readable storage medium TECHNICAL FIELD

[0001] The present application relates to the technical field of signal processing, in particular to a heart sound signal quality evaluation method, an electronic device and a readable storage medium. BACKGROUND

[0002] Heart sound refers to the sound caused by the vibration of myocardial contraction, heart valve closure and blood impact on the ventricular wall and aortic wall. It can be heard by a stethoscope at a certain part of the chest wall. Now it can also be heard by an electronic stethoscope, which converts heart sound into digital audio signal and saves it.

[0003] Due to the complexity of the clinical environment, human or environmental factors will inevitably introduce interference and noise during the collection of heart sound signals. For this problem, the existing method often uses filtering and other denoising methods to reduce the influence of interference and noise. However, since the noise interference may be mixed with the heart sound in the time domain, frequency domain or other transform domain, the denoising method is not good for removing some noise interference. In order to ensure the accuracy of the results based on heart sound signal analysis, the quality of the signal must be evaluated.

[0004] However, there is still a lack of an effective quality evaluation method for heart sound signals to evaluate the quality (Signal Quality Index, SQI) of heart sound signals.

[0005] The disclosure of the above background art content is only used to assist in understanding the inventive concept and technical solutions of the present application, which does not necessarily belong to the prior art of the present patent application, nor does it necessarily provide technical teaching, in order to provide general background information, and does not necessarily constitute prior art. SUMMARY

[0006] The main purpose of the present application is to provide a heart sound signal quality evaluation method, device, electronic device and readable storage medium, which aims to solve the technical problem of how to effectively evaluate the signal quality of heart sound signals.

[0007] To achieve the above purpose, the present application provides a heart sound signal quality evaluation method, which comprises:

[0008] Collecting heart sound signals;

[0009] Extracting features of the heart sound signals to obtain signal features, wherein the signal features include one or more of local variance features, local peak features, signal energy features and zero-crossing rate features;

[0010] inputting the signal feature into a pre-trained signal quality evaluation model to output a signal quality evaluation result.

[0011] Optionally, the step of extracting the signal feature from the heart sound signal comprises:

[0012] performing sliding window processing on the heart sound signal, and taking the heart sound signal in the sliding window as a sub-heart sound signal;

[0013] For each sub-heart sound signal, performing feature extraction on the sub-heart sound signal to obtain a signal feature corresponding to the sub-heart sound signal.

[0014] Optionally, in the case where the signal feature comprises a local variance feature, the step of extracting the signal feature from the sub-heart sound signal comprises:

[0015] performing sliding window processing on the sub-heart sound signal, and taking the sub-heart sound signal in the sliding window as a secondary sub-heart sound signal;

[0016] determining a signal value standard deviation of each secondary sub-heart sound signal, and taking a variance of all the signal value standard deviations as a first variance feature;

[0017] taking a mean value of all the signal value standard deviations as a first mean value, and sorting all the signal value standard deviations to obtain a standard deviation sequence, wherein the standard deviation sequence comprises the signal value standard deviations sorted in descending order based on numerical values;

[0018] determining a first preset number of signal value standard deviations in the standard deviation sequence;

[0019] taking a mean value of the first preset number of signal value standard deviations as a second mean value, and taking a ratio of the first mean value to the second mean value as a second variance feature;

[0020] taking a variance of all the window start sampling point positions as a third variance feature;

[0021] taking one or more of the first variance feature, the second variance feature, and the third variance feature as a local variance feature.

[0022] Optionally, in the case where the signal feature comprises a local peak value feature, the step of extracting the signal feature from the sub-heart sound signal comprises:

[0023] determining a signal peak value of the sub-heart sound signal, and adjusting the signal peak value based on a first preset adjustment coefficient to obtain a first threshold value, wherein the first threshold value is less than the signal peak value;

[0024] determining a first target signal value greater than the first threshold value in the signal values of the sub-heart sound signal, taking a sampling point number of a signal sampling point corresponding to the first target signal value as a first sampling point number;

[0025] adjusting the signal peak value to obtain a second threshold value based on a second preset adjustment coefficient, wherein the first threshold value is less than the signal peak value, and the second preset adjustment coefficient is less than the first preset adjustment coefficient;

[0026] determining a second target signal value less than the second threshold value in the signal values of the sub-heart sound signal, taking a sampling point number of a signal sampling point corresponding to the second target signal value as a second sampling point number;

[0027] taking the first sampling point number and / or the second sampling point number as a local peak value feature.

[0028] Optionally, in the case where the signal feature includes a signal energy feature, the step of extracting the signal feature corresponding to the sub-heart sound signal includes:

[0029] performing Fourier transformation on the sub-heart sound signal, and taking the sub-heart sound signal after Fourier transformation as a target signal;

[0030] determining a signal energy corresponding to each frequency of the target signal, and determining a total signal energy of all the signal energies;

[0031] determining a first frequency less than a first preset frequency among all the frequencies included in the target signal, determining a signal energy corresponding to each of the first frequencies, and taking a total energy of all the signal energies corresponding to the first frequencies as a low-frequency energy distribution proportion;

[0032] determining a second frequency greater than a second preset frequency among all the frequencies included in the target signal, determining a signal energy corresponding to each of the second frequencies, and taking a total energy of all the signal energies corresponding to the second frequencies as a high-frequency energy distribution proportion, wherein the second preset frequency is greater than or equal to the first preset frequency;

[0033] inputting the target signal into a preset logarithmic energy feature extraction model to output a logarithmic energy;

[0034] taking one or more of the total signal energy, the low-frequency energy distribution proportion, the high-frequency energy distribution proportion and the logarithmic energy as a signal energy feature.

[0035] Optionally, in the case that the signal feature comprises a log energy feature, in the case that the signal feature comprises a zero-crossing rate feature, the step of performing feature extraction on the sub-heart sound signal to obtain the signal feature corresponding to the sub-heart sound signal comprises:

[0036] determining a zero-crossing rate of each frame of the sub-heart sound signal, wherein the zero-crossing rate comprises a ratio of the number of signal sampling points of zero-crossing to the total number of signal sampling points of a frame of the heart sound signal;

[0037] determining a maximum zero-crossing rate among all the zero-crossing rates and a minimum zero-crossing rate among all the zero-crossings;

[0038] differentially processing all the zero-crossing rates to obtain a first-order differential zero-crossing rate;

[0039] determining a maximum first-order differential zero-crossing rate among all the first-order differential zero-crossing rates and a variance of all the first-order differential zero-crossing rates;

[0040] differentially processing all the first-order differential zero-crossing rates to obtain a second-order differential zero-crossing rate;

[0041] determining a maximum second-order differential zero-crossing rate among all the second-order differential zero-crossing rates and a standard deviation of all the second-order differential zero-crossing rates;

[0042] taking one or more of the maximum zero-crossing rate, the minimum zero-crossing rate, the maximum first-order differential zero-crossing rate, the variance, the maximum second-order differential zero-crossing rate, and the standard deviation as a zero-crossing rate feature.

[0043] Optionally, the step of inputting the signal feature into a pre-trained signal quality evaluation model to output a signal quality evaluation result comprises:

[0044] inputting the signal feature corresponding to each sub-heart sound signal into a pre-trained signal quality evaluation model to output a signal quality evaluation result of each sub-heart sound signal.

[0045] Optionally, before the step of performing feature extraction on the heart sound signal to obtain a signal feature, the method further comprises:

[0046] performing signal filtering processing on the heart sound signal to obtain a filtered heart sound signal, wherein the signal filtering comprises zero-phase Butterworth filtering;

[0047] performing signal clipping processing on the filtered heart sound signal to obtain a pre-processed heart sound signal, wherein the signal clipping comprises clipping a first-end heart sound signal of a preset time length and / or clipping a tail-end heart sound signal of a preset time length;

[0048] Based on the pre-processed heart sound signal, the following is performed: feature extraction is performed on the heart sound signal to obtain signal features.

[0049] In addition, to achieve the above-mentioned purpose, the present application also provides a heart sound signal quality evaluation device, which comprises:

[0050] A signal acquisition module is configured to acquire a heart sound signal.

[0051] A feature extraction module is configured to perform feature extraction on the heart sound signal to obtain signal features, wherein the signal features comprise one or more of local variance features, local peak value features, signal energy features, and zero-crossing rate features.

[0052] A quality evaluation module is configured to input the signal features into a pre-trained signal quality evaluation model to output signal quality evaluation results.

[0053] The present application also provides an electronic device, which is a physical device. The electronic device comprises at least one processor and a memory communicatively connected to the at least one processor. The memory stores instructions executable by the at least one processor. The instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the heart sound signal quality evaluation method.

[0054] The present application also provides a readable storage medium, which is a computer readable storage medium. The computer readable storage medium stores a program for implementing a heart sound signal quality evaluation method. The program is executed by a processor to implement the steps of the heart sound signal quality evaluation method.

[0055] The present application also provides a computer program product comprising a computer program. The computer program is executed by a processor to implement the steps of the heart sound signal quality evaluation method.

[0056] In the present application, a heart sound signal is acquired, and feature extraction is performed on the heart sound signal to obtain signal features, wherein the signal features comprise one or more of local variance features, local peak value features, signal energy features, and zero-crossing rate features. The signal features are input into a pre-trained signal quality evaluation model to output signal quality evaluation results. In this way, the present application extracts local variance features, local peak value features, signal energy features, and zero-crossing rate features of a heart sound signal, and evaluates the signal quality of the heart sound signal based on the multi-dimensional signal features of the heart sound signal, thereby effectively evaluating the signal quality of the heart sound signal. BRIEF DESCRIPTION OF DRAWINGS

[0057] The accompanying drawings, which are incorporated into and form a part of the specification, illustrate preferred embodiments of the present application and, together with the description, serve to explain the principles of the application.

[0058] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the accompanying drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, for those of ordinary skill in the art, based on the drawings, other drawings can be obtained without creative labor.

[0059] Fig. 1 is a flowchart of a first embodiment of the heart sound signal quality evaluation method of the present application;

[0060] Fig. 2 is a flowchart of a specific application scenario of the heart sound signal quality evaluation method of the present application;

[0061] Fig. 3 is another flowchart of a specific application scenario of the heart sound signal quality evaluation method of the present application;

[0062] Fig. 4 is a device module diagram of the heart sound signal quality evaluation device of the present application;

[0063] Fig. 5 is a device structure diagram of the hardware operating environment involved in the heart sound signal quality evaluation device of the present application.

[0064] The object, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0065] In order to make the above-mentioned objects, features and advantages of the present application more apparent and easy to understand, the technical solutions in the embodiments of the present application will be described clearly and completely with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor are within the scope of protection of the present application.

[0066] Embodiment One

[0067] The present application proposes a first embodiment of a heart sound signal quality evaluation method, please refer to Fig. 1, the heart sound signal quality evaluation method comprises:

[0068] Step S10, collecting heart sound signals;

[0069] As a feasible implementation, the heart sound signal can be collected by a sensor, such as one or more of a VPU (Voice Pick Up, bone conduction) sensor, a microphone sensor, and the like, and the present embodiment does not make specific limitations thereon. Further, the heart sound signal can be collected at a preset signal sampling rate, which can be any sampling rate set in advance, such as 1000 Hz, and the like, and the present embodiment does not make specific limitations thereon.

[0070] Further, before extracting the signal features, in order to improve the accuracy of the extracted signal features, the heart sound signal can be preprocessed, such as filtering, noise reduction, normalization, and the like. The signal features are extracted based on the preprocessed heart sound signal.

[0071] In addition, the acceleration signal of the user can also be collected by a 6-axis signal collection module. The acceleration and gyroscope signals are mainly collected by the 6-axis signal collection module, which is mainly composed of a 6-axis sensor integrated inside the device. Before signal collection, the 6-axis signal collection module can calculate the current Euler angle to assist the user in positioning the collection position and posture. During the signal collection process, the 6-axis signal collection module can monitor the movement of the user's arm, combine with the heart sound signal, and perform signal denoising, which can improve the signal-to-noise ratio of the above signals during the collection process.

[0072] In step S20, the heart sound signal is subjected to feature extraction to obtain signal features, wherein the signal features include one or more of a local variance feature, a local peak value feature, a signal energy feature, and a zero-crossing rate feature.

[0073] The signal features include but are not limited to a local variance feature, a local peak value feature, a signal energy feature, and a zero-crossing rate feature, and can also include kurtosis and skewness. In the present embodiment, preferably, the signal features include a local variance feature, a local peak value feature, a signal energy feature, and a zero-crossing rate feature, so as to evaluate the signal quality of the heart sound signal from multiple angles, and the accuracy of the heart sound signal quality evaluation can be improved.

[0074] Let the kurtosis be ssqi, and the kurtosis be ksqi, then wherein N represents the length of the signal, Xi represents the signal value of the i-th signal sampling point, μ represents the mean of the signal value, and σ represents the standard deviation of the signal value. It should be noted that the length of the signal can be the time length or the number of sampling points of the signal, that is, N can be the number of all sampling points included in the signal, or the total time length of the signal. In the present embodiment, preferably, the number of sampling points included in the signal is taken as the length of the signal.

[0075] In a possible implementation, the step of performing feature extraction on the heart sound signal to obtain signal features comprises:

[0076] In step S201, the heart sound signal is subjected to sliding window processing, and the heart sound signal in the sliding window is taken as a sub-heart sound signal.

[0077] The heart sound signal can be subjected to sliding window processing with a preset window length and a preset step length. The preset window length and the preset step length can be set in advance, for example, the length of five seconds and the step length of two seconds are used for sliding window processing. In this embodiment, the window length is preferably greater than the step length, so that there is an overlap between adjacent two windows, thereby improving the accuracy of signal quality evaluation and reducing the discarded signals (for example, in this embodiment, the signals with poor signal quality evaluation results are discarded). For example, assuming that the step length of the sliding window is q and the window length is q*3, the heart sound signal is divided into

q1, q2, q3, q4

q1, q2, q3

q2, q3, q4

[0078] In step S202, for each sub-heart sound signal, feature extraction is performed on the sub-heart sound signal to obtain the signal features corresponding to the sub-heart sound signal.

[0079] For each sub-heart sound signal, feature extraction is performed, so that the signal quality of the sub-heart sound signal can be evaluated based on the signal features corresponding to each sub-heart sound signal.

[0080] In step S30, the signal features are input into a pre-trained signal quality evaluation model to output a signal quality evaluation result.

[0081] The signal quality evaluation model is a pre-trained signal quality evaluation model using a training data set. Specifically, the signal quality evaluation model can use a machine learning algorithm to achieve nonlinear, efficient, and lightweight heart sound signal quality evaluation. The signal quality evaluation result can include quality levels, such as good, medium, and poor quality levels.

[0082] As one of the implementations, the signal features of the entire heart sound signal (such as the complete heart sound signal collected) can be extracted, and the extracted signal features are input into the pre-trained signal quality evaluation model to output the quality evaluation result of the entire heart sound signal.

[0083] As another implementation, the step of inputting the signal features into the pre-trained signal quality evaluation model and outputting a signal quality evaluation result comprises:

[0084] In step S301, the signal features corresponding to each sub-heart sound signal are input into the pre-trained signal quality evaluation model, and the signal quality evaluation result of each sub-heart sound signal is output.

[0085] Based on the respective signal features of each sub-heart sound signal, the signal quality of the sub-heart sound signal is evaluated, so that the signal quality distribution of different parts in the entire signal is obtained, and the accuracy of the heart sound signal quality evaluation is improved.

[0086] In this embodiment, heart sound signals are collected, and signal features are extracted from the heart sound signals, wherein the signal features include one or more of local variance features, local peak features, signal energy features, and zero-crossing rate features. The signal features are input into a pre-trained signal quality evaluation model, and a signal quality evaluation result is output. In this way, the local variance features, local peak features, signal energy features, and zero-crossing rate features of the heart sound signals are extracted, and the signal quality of the heart sound signals is evaluated based on the multi-dimensional signal features of the heart sound signals, thereby effectively evaluating the signal quality of the heart sound signals.

[0087] Embodiment Two

[0088] Based on the first embodiment of the present application, in another embodiment of the present application, the same or similar contents as the above-mentioned embodiment one can refer to the above introduction, and the subsequent will not be described in detail. On this basis, in the case where the signal features include local variance features, the step of extracting features from the sub-heart sound signals to obtain the signal features corresponding to the sub-heart sound signals comprises:

[0089] In step A10, the sub-heart sound signal in the sliding window is processed as a secondary sub-heart sound signal.

[0090] The sub-heart sound signal can be processed by a preset window length and a preset step length. The preset window length and the preset step length can be set in advance, for example, a length of 200 signal sampling points and a step length of 50 signal sampling points for sliding window. The present embodiment does not make specific restrictions on this.

[0091] In step A20, the signal value standard deviation of each secondary sub-heart sound signal is determined, and the variance of all signal value standard deviations is taken as a first variance feature.

[0092] Let the first variance feature of the i-th segment of sub-heart sound signal be σ std,i Then wherein q represents a sliding step length for the sliding window processing of the sub-heart sound signal, and m represents a sliding window length for the sliding window processing of the sub-heart sound signal, std represents a signal value standard deviation, S i,j Sij represents a signal value of a jth signal sampling point of an ith sub-heart sound signal, i,qk:qk+m Sik+m represents a signal value corresponding to each signal sampling point between a qkth signal sampling point and a qk+mth signal sampling point of the ith sub-heart sound signal, and M represents a length of the ith sub-heart sound signal, and in the embodiment, preferably, M is a total number of signal sampling points included in the ith sub-heart sound signal.

[0093] In step A30, an average of all the signal value standard deviations is taken as a first average, and all the signal value standard deviations are sorted to obtain a standard deviation sequence, wherein the standard deviation sequence includes the signal value standard deviations sorted in descending order based on values.

[0094] In step A40, a preset number of signal value standard deviations in the standard deviation sequence are determined.

[0095] In step A50, an average of the preset number of signal value standard deviations is taken as a second average, and a ratio of the first average to the second average is taken as a second variance feature.

[0096] Let the second variance feature of the ith sub-heart sound signal be max Npeaki wherein N is a preset number, and p n,i Sik+m represents a signal value corresponding to each signal sampling point between a qkth signal sampling point and a qk+mth signal sampling point of the ith sub-heart sound signal, and M represents a length of the ith sub-heart sound signal, and in the embodiment, preferably, M is a total number of signal sampling points included in the ith sub-heart sound signal.

[0097] As one of the implementation manners, the signal sampling point position can be a signal sampling point number, for example, the signal sampling points of the sub-heart sound signal are sequentially numbered from 1, and a sampling point number corresponding to each signal sampling point can be obtained, which can be taken as the signal sampling point position of the signal sampling point. As another implementation manner, a time corresponding to each signal sampling point can be taken as the signal sampling point position, and in the embodiment, preferably, the sampling point number is taken as the signal sampling point position of the signal sampling point.

[0098] ​Step A60, taking the variance of all the window starting sampling point positions as a third variance feature;

[0099] Let the third variance feature of the i-th segment of the sub-heart sound signal be Then Wherein,

[0100] It should be noted that the third variance feature can be one or more, and each different value of N corresponds to a third variance feature. N can be any value set in advance, such as 24, 12, 6, etc., and the embodiment does not make specific limitations on this.

[0101] Step A70, taking one or more of the first variance feature, the second variance feature, and the third variance feature as a local variance feature.

[0102] In the embodiment, preferably, the first variance feature, the second variance feature, and the third variance feature are taken as the local variance feature.

[0103] In the embodiment, for the collected heart sound signal, a suitable sliding window with a time window length and a step length is further designed for the sub-heart sound signal, and the variance change feature of the sliding window is calculated, which can effectively extract the peak value generated by the vibration of the heart in the sampling time and the peak value caused by external noise. Extracting the distribution feature of these peak values (the first variance feature, the second variance feature, and the third variance feature in the embodiment) can effectively reflect the strength of the noise in the heart sound signal, and has good sensitivity and robustness.

[0104] In a possible implementation, in the case where the signal feature includes a local peak feature, the step of extracting a signal feature corresponding to the sub-heart sound signal from the sub-heart sound signal includes:

[0105] Step B10, determining a signal peak value of the sub-heart sound signal, and adjusting the signal peak value based on a first preset adjustment coefficient to obtain a first threshold value, wherein the first threshold value is less than the signal peak value;

[0106] Step B20, determining a first target signal value greater than the first threshold value in the signal value of the sub-heart sound signal, and taking the number of sampling points corresponding to the first target signal value as a first sampling point number;

[0107] Let the first sampling point number of the i-th segment of the sub-heart sound signal be C h1,i Then Wherein, h1 is a first preset adjustment coefficient, and max(Si) represents a signal peak value of the i-th segment of the sub-heart sound signal, that is, a maximum signal value, wherein, is an indicator function, which is 1 when (condition) is true, and 0 otherwise.

[0108] It should be noted that the first sampling point quantity can be one or more, and each different value of h1 corresponds to a first sampling point quantity. h1 can be any value set in advance, such as 0.8, 0.6, and the like. The present embodiment does not make a specific limitation on this.

[0109] In step B30, a second threshold value is obtained by adjusting the signal peak value based on a second preset adjustment coefficient, wherein the first threshold value is less than the signal peak value, and the second preset adjustment coefficient is less than the first preset adjustment coefficient.

[0110] In step B40, a second target signal value less than the second threshold value is determined in the signal value of the sub-heart sound signal, and a sampling point quantity of a signal sampling point corresponding to the second target signal value is taken as a second sampling point quantity.

[0111] Let the first sampling point quantity of the i-th segment of the sub-heart sound signal be C h2,i Then Wherein, h1 is the second preset adjustment coefficient.

[0112] It should be noted that the second sampling point quantity can be one or more, and each different value of h2 corresponds to a second sampling point quantity. h2 can be any value set in advance, such as 0.4, 0.1, and the like. The present embodiment does not make a specific limitation on this.

[0113] In step B50, the first sampling point quantity and / or the second sampling point quantity are taken as a local peak value feature.

[0114] In the present embodiment, preferably, the first sampling point quantity and the second sampling point quantity are taken as a local peak value feature.

[0115] In a possible implementation, in the case where the signal feature includes a signal energy feature, the step of performing feature extraction on the sub-heart sound signal to obtain a signal feature corresponding to the sub-heart sound signal includes:

[0116] In step C10, Fourier transformation is performed on the sub-heart sound signal, and the sub-heart sound signal after Fourier transformation is taken as a target signal.

[0117] Specifically, short-time Fourier transformation can be performed on the sub-heart sound signal. It can be understood that after Fourier transformation, an energy value corresponding to F different frequency points at each time t of the sub-heart sound signal is obtained. That is, the target signal includes an energy value corresponding to F different frequency points at each time t.

[0118] Step C20, determining signal energy corresponding to each frequency of the target signal, and determining total signal energy of all the signal energy;

[0119] Let total signal energy of the i-th sub-heart sound signal be total_power_i, then Wherein, T is total time length of the target signal, f represents the f-th frequency after transformation. Pxx[f, t] represents signal energy of the target signal at t (s) time and f (Hz) frequency.

[0120] Step C30, determining first frequency less than first preset frequency among all frequencies included in the target signal, determining signal energy corresponding to each of the first frequency, and taking total energy of all the signal energy corresponding to the first frequency as low frequency energy distribution proportion;

[0121] Let low frequency energy distribution proportion of the i-th sub-heart sound signal be energy_below_i_f1, then energy_below_i_f1 = ∑ f<f1 Pxx[f], wherein, f1 is the first preset frequency, and Pxx[f] represents signal energy of the target signal at all times and f (Hz) frequency.

[0122] It should be noted that the low frequency energy distribution proportion can be one or more, and each different value of f1 corresponds to a low frequency energy distribution proportion. f1 can be any value set in advance, such as 100 Hz, 120 Hz, etc., and the embodiment does not make specific limitation thereto.

[0123] Step C40, determining second frequency greater than second preset frequency among all frequencies included in the target signal, determining signal energy corresponding to each of the second frequency, and taking total energy of all the signal energy corresponding to the second frequency as high frequency energy distribution proportion, wherein the second preset frequency is greater than or equal to the first preset frequency;

[0124] Let high frequency energy distribution proportion of the i-th sub-heart sound signal be energy_above_i_f2, then energy_above_i_f2 = ∑ f>f2 Pxx[f], wherein, f2 is the second preset frequency, and Pxx[f] represents signal energy of the target signal at all times and f (Hz) frequency.

[0125] It should be noted that the high frequency energy distribution proportion can be one or more, and each different value of f2 corresponds to a high frequency energy distribution proportion. f2 can be any value set in advance, such as 250 Hz, 270 Hz, etc., and the embodiment does not make specific limitation thereto.

[0126] Step C50, inputting the target signal into a preset log energy feature extraction model to output log energy;

[0127] The preset log energy feature extraction model can include one or more of the following calculation formulas: log_energy_diff[t]=log_energy[t]-log_energy[t-1] log_energy_std=std(log_energy) log_energy_diff_max=max(log_energy_diff) log_energy_diff_std=std(log_energy_diff)

[0128] wherein std represents standard deviation.

[0129] Therefore, the log energy includes one or more of log_energy[t], log_energy_diff[t], log_energy_std, log_energy_diff_mean, log_energy_diff_max, and log_energy_diff_std, and in this embodiment, preferably, the log energy includes log_energy[t], log_energy_diff[t], log_energy_std, log_energy_diff_mean, log_energy_diff_max, and log_energy_diff_std.

[0130] Step C60, taking one or more of the total signal energy, the low-frequency energy distribution proportion, the high-frequency energy distribution proportion, and the log energy as signal energy features.

[0131] In this embodiment, preferably, the total signal energy, the low-frequency energy distribution proportion, the high-frequency energy distribution proportion, and the log energy are taken as signal energy features.

[0132] Considering that the heart sound signal is essentially a vibration signal generated by heart vibration, the signal characteristics are similar to those of a sound signal. In this embodiment, the spectrum, zero-crossing rate, and log energy spectrum commonly used in sound signal processing can effectively determine whether the signal at a certain time contains the desired sound signal. In heart sound signal processing, these characteristics can effectively determine the occurrence time and intensity of the systolic and diastolic periods, as well as the occurrence time and intensity of other noises. From the time distribution of these characteristics, it can be determined whether the processed heart sound signal meets the physiological heart vibration condition, and thus whether the signal quality meets the requirements, thereby improving the accuracy of heart sound signal quality evaluation.

[0133] In a possible implementation, in the case where the signal feature includes a log energy feature, and in the case where the signal feature includes a zero-crossing rate feature, the step of performing feature extraction on the sub-heart sound signal to obtain the signal feature corresponding to the sub-heart sound signal includes:

[0134] Step D10, determining the zero-crossing rate of each frame of the sub-heart sound signal, wherein the zero-crossing rate includes the ratio of the number of zero-crossing signal sampling points to the total number of signal sampling points of a frame of the heart sound signal;

[0135] Specifically, a predetermined number of signal sampling points can be a frame of signal, such as 50 signal sampling points for a frame of signal. That is, the zero-crossing rate of the kth frame of the ith segment of sub-heart sound signal is Z i [k], then wherein q is the number of signal sampling points included in a frame of signal, and sign is a sign function that returns 1 for positive numbers, -1 for negative numbers, and 0 for 0.

[0136] Step D20, determining the maximum zero-crossing rate among all the zero-crossing rates and determining the minimum zero-crossing rate among all the zero-crossing rates;

[0137] Let the maximum zero-crossing rate of the ith segment of sub-heart sound signal be zero_crossing_rate_i_max, then zero_crossing_rate_i_max = max(Z i ), wherein Z i represents the zero-crossing rate of each frame of signal of the ith segment of sub-heart sound signal.

[0138] Let the minimum zero-crossing rate of the ith segment of sub-heart sound signal be zero_crossing_rate_i_min, then zero_crossing_rate_i_min = min(Z i ), wherein Z i represents the zero-crossing rate of each frame of signal of the ith segment of sub-heart sound signal.

[0139] Step D30, differentiating all the zero-crossing rates to obtain first-order differential zero-crossing rates;

[0140] Step D40, determining the maximum first-order differential zero-crossing rate among all the first-order differential zero-crossing rates, and determining the variance of all the first-order differential zero-crossing rates;

[0141] Let the maximum first-order differential zero-crossing rate of the i-th segment of sub-heart sound signal be zero_crossing_rate_i_diff_max, then zero_crossing_rate_i_diff_max = max(diff(Z i )), where diff represents calculating the difference.

[0142] Let the variance of the first-order differential zero-crossing rate of the i-th segment of sub-heart sound signal be zero_crossing_rate_i_diff_var, then zero_crossing_rate_i_diff_var = var(diff(Z i )), where var represents calculating the variance.

[0143] Step D50, differentiating all the first-order differential zero-crossing rates to obtain second-order differential zero-crossing rates;

[0144] Step D60, determining the maximum second-order differential zero-crossing rate among all the second-order differential zero-crossing rates, and determining the standard deviation of all the second-order differential zero-crossing rates;

[0145] Let the maximum second-order differential zero-crossing rate of the i-th segment of sub-heart sound signal be zero_crossing_rate_i_diff2_max, then zero_crossing_rate_i_diff2_max = max(diff(diff(Z i ))).

[0146] Let the standard deviation of the second-order differential zero-crossing rate of the i-th segment of sub-heart sound signal be zero_crossing_rate_i_diff2_std, then zero_crossing_rate_i_diff2_std = std(diff(diff(Z i ))).

[0147] Step D70, taking one or more of the maximum zero-crossing rate, the minimum zero-crossing rate, the maximum first-order differential zero-crossing rate, the variance, the maximum second-order differential zero-crossing rate, and the standard deviation as the zero-crossing rate features.

[0148] In this embodiment, preferably, the maximum zero-crossing rate, the minimum zero-crossing rate, the maximum first-order differential zero-crossing rate, the variance of the first-order differential zero-crossing rate, the maximum second-order differential zero-crossing rate, and the standard deviation of the second-order differential zero-crossing rate are taken as the zero-crossing rate features.

[0149] The signal feature extraction manner of the embodiments of the present application is not based on a heart beat cycle, and does not need to perform position recognition of the first heart sound and the second heart sound. The position recognition algorithm of the first heart sound and the second heart sound is relatively complex and has a large error. The features involved in the embodiments of the present application can be realized with very small calculation complexity, while the sensitivity and specificity in heart sound signal quality evaluation are guaranteed, so as to facilitate deployment on wearable devices and the like with weak calculation capability, to perform real-time analysis of signal quality, so that the heart sound signal quality evaluation method of the embodiments of the present application can be applied to more electronic devices, and the universality of heart sound signal quality evaluation is improved.

[0150] Embodiment three

[0151] Based on the first and second embodiments of the present application, in another embodiment of the present application, the same or similar contents as the above-mentioned embodiments one and two can be referred to the above introduction, and will not be described in detail hereinafter. On this basis, before the step of extracting signal features from the heart sound signal, the method further comprises:

[0152] Step E10, performing signal preprocessing on the heart sound signal to obtain a preprocessed heart sound signal;

[0153] Step E20, based on the preprocessed heart sound signal, performing: extracting signal features from the heart sound signal.

[0154] The preprocessing manner of signal preprocessing can be set in advance, such as normalization, filtering and noise reduction. Alternatively, the collected signal is subjected to a normalization operation, and then the heart sound signal is filtered by a pre-designed filter, and the filtered data and the acceleration signal are subjected to secondary processing by an adaptive filter to filter out noise generated due to motion artifacts, to obtain a clean high-quality heart sound signal.

[0155] In one possible implementation, the step of performing signal preprocessing on the heart sound signal to obtain a preprocessed heart sound signal comprises

[0156] Step F10, performing signal filtering processing on the heart sound signal to obtain a filtered heart sound signal, wherein the signal filtering comprises zero-phase shift Butterworth filtering;

[0157] In this embodiment, preferably, the heart sound signal is filtered and processed by a Butterworth filter in a zero-phase shift manner, to avoid phase shift of the signal after filtering, and to affect subsequent feature extraction.

[0158] In the acquisition of signals, there will inevitably be various noises, including motion artifacts, power frequency interference, etc. Known noise such as power frequency interference can be filtered out by filtering, thereby improving the signal-to-noise ratio of the acquired signal. The signal quality of the signal can be guaranteed to the greatest extent. If the signal quality is not filtered directly, it is likely that some large signals will be filtered out, and the signal will be classified as a poor quality signal due to the existence of noise, which will affect subsequent analysis.

[0159] Step F20, signal clipping processing is performed on the filtered heart sound signal to obtain a pre-processed heart sound signal, wherein the signal clipping includes clipping the first end heart sound signal of a preset time length, and / or clipping the tail end heart sound signal of a preset time length.

[0160] In the embodiment, preferably, the first end heart sound signal of a preset time length and the tail end heart sound signal of a preset time length are clipped, respectively, to clip the data head and tail, remove the influence of lead establishment time and signal distortion caused by band pass filtering, and obtain available heart sound signals for subsequent data processing.

[0161] Exemplarily, in a specific application scenario, when the quality level is good, the signal quality evaluation model outputs a value 1, when the quality level is medium, the signal quality evaluation model outputs a value 2, and when the quality level is poor, the signal quality evaluation model outputs a value 2. Referring to FIG. 2-3, in the specific application scenario, the signal quality evaluation model outputs a value 1 when the quality level is good, a value 2 when the quality level is medium, and a value 2 when the quality level is poor.

[0162] 1. Collecting electrocardiogram (ECG), pulse (PPG), and heart sound (PCG) physiological signals. First, the data is processed by the data preprocessing module, then the signal quality is evaluated by the signal quality evaluation module, and the signal is classified into three levels: good, medium, and poor. Then, the signal with good quality is processed by the signal segmentation module and the feature engineering module, and finally, the extracted features are sent to the machine learning blood pressure estimation analysis module to obtain the systolic blood pressure (SBP), diastolic blood pressure (DBP), and heart rate (HR).

[0163] 2. The data preprocessing module, signal quality evaluation module, signal segmentation module, feature engineering module, and blood pressure estimation module are included.

[0164] 3、Data preprocessing module includes data filtering and data cropping. Data filtering uses different passband Butterworth filters for ECG, pulse, and heart sound signals, and uses zero-phase shift implementation method to avoid phase shift after three-way signal filtering, which affects subsequent feature extraction. Data cropping synchronously crops the lengths of three signals, and crops the data head and tail to remove the influence of lead establishment time and signal distortion caused by band-pass filtering, and obtain usable, equal-length, and synchronous three-way physiological signals for subsequent data processing.

[0165] 4、Signal quality assessment module, including ECG signal quality assessment, pulse signal quality assessment module, and heart sound signal quality assessment module. The signal quality assessment module uses a five-second length and a two-second step for sliding window, calculates the proposed feature values for the signals in the sliding window, and uses a pre-trained machine learning integrated model for classification prediction. The model can obtain the signal quality distribution of different parts of the whole signal.

[0166] For the ECG signal quality assessment module, multiple signal features suitable for ECG signal quality assessment are extracted, including kurtosis (ssqi), skewness (ksqi), etc.

[0167] For the pulse signal quality assessment module, multiple signal features suitable for pulse signal quality assessment are used, including zero-crossing number (zero_cross), local less than zero minimum value number (L i ), average distance between adjacent minimum values standard deviation of adjacent minimum value distance minimum minimum value ratio (R i ), average position of minimum value, average peak-valley amplitude, standard deviation of peak-valley amplitude, maximum peak-valley amplitude ratio, average peak-valley time difference, standard deviation of peak-valley time difference, valley amplitude coefficient of variation, etc.

[0168] For the heart sound signal quality assessment module, by referring to multiple features commonly used in speech signal processing and combining heart sound signal features, multiple signal features suitable for heart sound signal quality assessment are innovatively proposed, including time domain features and frequency domain features. Time domain features include signal mean, variance, local variance features, and local peak value features. Frequency domain features include total energy, energy distribution ratio, zero-crossing rate features, first and second order differences of zero-crossing rate, logarithmic energy features, first and second order differences of logarithmic energy.

[0169] 5. A signal segmentation module segments the signal to obtain different heartbeats. The signal segmentation is divided into two steps. Firstly, the QRS complex or the trough of the pulse of the electrocardiosignal is identified, and the signal is preliminarily segmented into different heartbeats. Subsequently, the segmented heartbeats are post-processed, the upper and lower threshold values of the length of the heartbeat are set, and the abnormal points are removed (for example, the heartbeat signal with the length of the heartbeat greater than the upper threshold value or the length of the heartbeat less than the lower threshold value is removed), to obtain a signal segmentation result with high accuracy.

[0170] 6. A feature engineering module includes electrocardiosignal features, pulse signal features, heart sound signal features, multi-signal joint time difference features, and personal information features. The electrocardiosignal features include HRV time domain features, HRV frequency domain features, and nonlinear features. The pulse signal features include pulse amplitude features, heart rate features, pulse morphology features, and pulse variability features. The heart sound signal features include the first heart sound duration, the systolic duration, the diastolic duration, the energy feature, and the main frequency component feature. The multi-signal joint time features include the R-wave time point of the electrocardiosignal, the pulse trough, and the first heart sound peak time point. The joint features are obtained by calculating the difference values of the above-mentioned time points, such as the pre-ejection period (PEP), the pulse transmission time, the pulse arrival time, the ratio of the systolic duration of the pulse signal to the systolic duration of the heart sound signal, and the ratio of the diastolic duration of the pulse signal to the diastolic duration of the heart sound signal. The personal information features are obtained by collecting and analyzing the personal data of the collection object, including age, height, weight, and the like.

[0171] In the physiological signal analysis algorithm and disease diagnosis algorithm using electrocardiosignal, pulse signal, and heart sound signal, the high-quality segments analyzed by the algorithm are used to implement subsequent signal analysis and calculation, which can greatly improve the accuracy and robustness of the data analysis algorithm.

[0172] Embodiment Four

[0173] The embodiment of the present application also provides a heart sound signal quality evaluation device, which refers to FIG. 4. The device includes:

[0174] A signal collection module 10 is configured to collect a heart sound signal.

[0175] A feature extraction module 20 is configured to extract features from the heart sound signal to obtain signal features, wherein the signal features include one or more of local variance features, local peak features, signal energy features, and zero-crossing rate features.

[0176] A quality evaluation module 30 is configured to input the signal features into a pre-trained signal quality evaluation model to output signal quality evaluation results.

[0177] The feature extraction module 20 is further configured to:

[0178] performing sliding window processing on the heart sound signal, taking the heart sound signal in the sliding window as a sub-heart sound signal;

[0179] For each of the sub-heart sound signals, performing feature extraction on the sub-heart sound signal to obtain a signal feature corresponding to the sub-heart sound signal.

[0180] In a case where the signal feature includes a local variance feature, the feature extraction module 20 is further configured to:

[0181] performing sliding window processing on the sub-heart sound signal, taking the sub-heart sound signal in the sliding window as a secondary sub-heart sound signal;

[0182] determining a signal value standard deviation of each of the secondary sub-heart sound signals, and taking a variance of all the signal value standard deviations as a first variance feature;

[0183] taking a mean value of all the signal value standard deviations as a second mean value, and sorting all the signal value standard deviations to obtain a standard deviation sequence, wherein the standard deviation sequence includes the signal value standard deviations sorted in descending order based on numerical values;

[0184] determining a first preset number of signal value standard deviations in the standard deviation sequence;

[0185] taking a mean value of the first preset number of signal value standard deviations as a second mean value, and taking a ratio of the first mean value to the second mean value as a second variance feature;

[0186] taking a variance of all the window start sampling point positions as a third variance feature;

[0187] taking one or more of the first variance feature, the second variance feature, and the third variance feature as the local variance feature.

[0188] In a case where the signal feature includes a local peak value feature, the feature extraction module 20 is further configured to:

[0189] determining a signal peak value of the sub-heart sound signal, adjusting the signal peak value based on a first preset adjustment coefficient to obtain a first threshold value, wherein the first threshold value is less than the signal peak value;

[0190] determining a first target signal value greater than the first threshold value in the signal value of the sub-heart sound signal, and taking a sampling point number of a signal sampling point corresponding to the first target signal value as a first sampling point number;

[0191] adjusting the signal peak value based on a second preset adjustment coefficient to obtain a second threshold value, wherein the first threshold value is less than the signal peak value, and the second preset adjustment coefficient is less than the first preset adjustment coefficient.

[0192] determining a second target signal value in the signal values of the sub-heart sound signal that is less than the second threshold value, taking a number of sampling points of a signal sampling point corresponding to the second target signal value as a second number of sampling points;

[0193] taking the first number of sampling points and / or the second number of sampling points as a local peak feature.

[0194] In a case where the signal feature includes a signal energy feature, the feature extraction module 20 is further configured to:

[0195] performing Fourier transformation on the sub-heart sound signal, and taking the sub-heart sound signal after the Fourier transformation as a target signal;

[0196] determining a signal energy corresponding to each frequency of the target signal, and determining a total signal energy of all the signal energies;

[0197] determining a first frequency less than a first preset frequency among all the frequencies included in the target signal, determining a signal energy corresponding to each of the first frequencies, and taking a total energy of all the signal energies corresponding to the first frequencies as a low-frequency energy distribution proportion;

[0198] determining a second frequency greater than a second preset frequency among all the frequencies included in the target signal, determining a signal energy corresponding to each of the second frequencies, and taking a total energy of all the signal energies corresponding to the second frequencies as a high-frequency energy distribution proportion, wherein the second preset frequency is greater than or equal to the first preset frequency;

[0199] inputting the target signal into a preset logarithmic energy feature extraction model to output a logarithmic energy;

[0200] taking one or more of the total signal energy, the low-frequency energy distribution proportion, the high-frequency energy distribution proportion and the logarithmic energy as a signal energy feature.

[0201] In a case where the signal feature includes a logarithmic energy feature, the feature extraction module 20 is further configured to:

[0202] determining a zero-crossing rate of each frame of the sub-heart sound signal, wherein the zero-crossing rate includes a ratio of a number of signal sampling points of zero-crossing to a total number of signal sampling points of a frame of the heart sound signal;

[0203] determining a maximum zero-crossing rate among all the zero-crossing rates, and determining a minimum zero-crossing rate among all the zero-crossings;

[0204] performing differential processing on all the zero-crossing rates to obtain a first-order differential zero-crossing rate;

[0205] determining a maximum first-order differential zero-crossing rate among all the first-order differential zero-crossing rates, and determining a variance of all the first-order differential zero-crossing rates;

[0206] differentially processing all the first-order differential zero-crossing rates to obtain second-order differential zero-crossing rates;

[0207] determining a maximum second-order differential zero-crossing rate among all the second-order differential zero-crossing rates, and determining a standard deviation of all the second-order differential zero-crossing rates;

[0208] taking one or more of the maximum zero-crossing rate, the minimum zero-crossing rate, the maximum first-order differential zero-crossing rate, the variance, the maximum second-order differential zero-crossing rate, and the standard deviation as a zero-crossing rate feature.

[0209] The quality assessment module 30 is further configured to:

[0210] inputting the signal feature corresponding to each of the sub-heart sound signals into a pre-trained signal quality assessment model to output a signal quality assessment result of each of the sub-heart sound signals.

[0211] The device further comprises a preprocessing module configured to:

[0212] performing signal filtering processing on the heart sound signal to obtain a filtered heart sound signal, wherein the signal filtering comprises zero-phase shift Butterworth filtering;

[0213] performing signal clipping processing on the filtered heart sound signal to obtain a preprocessed heart sound signal, wherein the signal clipping comprises clipping a first-end heart sound signal of a preset time length and / or clipping a tail-end heart sound signal of a preset time length;

[0214] The feature extraction module extracts a signal feature from the heart sound signal based on the preprocessed heart sound signal.

[0215] The heart sound signal quality assessment device provided by the application can effectively assess the signal quality of a heart sound signal, and has the same beneficial effects as the heart sound signal quality assessment method provided by the above embodiments, and other technical features of the heart sound signal quality assessment device are the same as those disclosed in the previous embodiment, which will not be repeated here.

[0216] Embodiment Five

[0217] An electronic device is provided, and the electronic device includes at least one processor, and a memory connected to the at least one processor in communication. The memory stores instructions executable by the at least one processor, and the instructions, when executed by the at least one processor, enable the at least one processor to perform the heart sound signal quality evaluation method in Embodiment 1.

[0218] Reference is made below to FIG. 5, which shows a structural schematic diagram of an electronic device suitable for implementing embodiments of the present disclosure. The electronic device in embodiments of the present disclosure can be a portable device or the like. The electronic device shown in FIG. 5 is merely an example, and should not impose any limitation on the functions and use range of embodiments of the present disclosure.

[0219] As shown in FIG. 5, the electronic device can include a processing apparatus 1001 (such as a central processing unit, a graphics processing unit, or the like), which can perform various appropriate actions and processes according to programs stored in a read-only memory (ROM 1002) or loaded from a storage apparatus into a random access memory (RAM 1004). Various programs and data required for operation of the electronic device are also stored in the RAM 1004. The processing apparatus 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface is also connected to the bus 1005.

[0220] Generally, the following systems can be connected to the I / O interface 1006: input apparatus 1007 including, for example, a touch screen, a touch pad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, or the like; output apparatus 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, or the like; storage apparatus 1003 including, for example, a magnetic tape, a hard disk, or the like; and communication apparatus 1009. The communication apparatus 1009 can allow the electronic device to communicate with other devices wirelessly or by wire to exchange data. Although an electronic device having various systems is shown in the figure, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems can alternatively be implemented or provided.

[0221] In particular, according to embodiments of the present disclosure, the processes described above with reference to the flowcharts can be implemented as a computer software program. For example, embodiments of the present disclosure include a computer program product comprising a computer program carried on a computer readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network through the communication apparatus, or installed from the storage apparatus 1003, or installed from the ROM 1002. When the computer program is executed by the processing apparatus 1001, the above-mentioned functions defined in the methods of embodiments of the present disclosure are performed.

[0222] The electronic device provided by the present application can solve the technical problem of how to effectively evaluate the signal quality of the heart sound signal by using the heart sound signal quality evaluation method in the above embodiment. Compared with the prior art, the electronic device provided by the embodiment of the present application has the same beneficial effects as the heart sound signal quality evaluation method provided by the above embodiment, and other technical features in the electronic device are the same as the features disclosed in the previous embodiment method, which will not be repeated here.

[0223] It should be understood that parts of the present disclosure can be realized by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0224] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto, and any person skilled in the art can easily think of changes or replacements within the technical range disclosed by the present application, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

[0225] Embodiment six

[0226] The embodiment of the present application provides a computer readable storage medium having computer readable program instructions stored thereon, and the computer readable program instructions are used to execute the heart sound signal quality evaluation method in the above embodiment one.

[0227] The computer readable storage medium provided by the embodiment of the present application may, for example, be a U disk, but is not limited to an electric, magnetic, optical, electromagnetic, infrared, or semiconductor system, system or device, or any combination thereof. More specific examples of computer readable storage media can include, but are not limited to, an electric connection with one or more conductive wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present embodiment, the computer readable storage medium can be any tangible medium containing or storing a program, which can be used by or in conjunction with an instruction execution system, system or device. The program code contained on the computer readable storage medium can be transmitted by any appropriate medium, including but not limited to: electric wire, optical cable, RF (radio frequency), etc., or any suitable combination thereof.

[0228] The above computer readable storage medium can be contained in the electronic device; or can exist separately without being assembled into the electronic device.

[0229] The computer readable storage medium described above carries one or more programs, when the one or more programs are executed by the electronic device, cause the electronic device to: collect a heart sound signal; perform feature extraction on the heart sound signal to obtain a signal feature, wherein the signal feature includes one or more of a local variance feature, a local peak value feature, a signal energy feature, and a zero-crossing rate feature; input the signal feature into a pre-trained signal quality evaluation model to output a signal quality evaluation result. The computer program code for performing the operations of the present disclosure can be written in one or more programming languages or combinations thereof, including an object-oriented programming language such as Java, Smalltalk, C++, and a conventional procedural programming language such as "C" language or a similar programming language. The program code can be executed entirely on a user computer, partially on a user computer, as an independent software package, partially on a user computer and partially on a remote computer, or entirely on a remote computer or server. In the case involving a remote computer, the remote computer can be connected to the user computer through any kind of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, through the Internet using an Internet service provider).

[0230] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowcharts or block diagrams can represent a module, a program segment, or a portion of code that includes one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions noted in the blocks can occur in different orders than those noted in the accompanying drawings. For example, two blocks that are shown in succession can actually be executed substantially in parallel, and sometimes in reverse order, depending on the involved functions. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented by a dedicated hardware-based system that performs the specified functions or operations, or can be implemented by a combination of dedicated hardware and computer instructions.

[0231] The modules described in the embodiments of the present disclosure can be implemented in a software manner or in a hardware manner. In some cases, the names of the modules do not constitute a limitation on the modules themselves.

[0232] The readable storage medium provided by the application is a computer readable storage medium, and the computer readable storage medium stores computer readable program instructions for executing the heart sound signal quality evaluation method, and can solve the technical problem of how to effectively evaluate the signal quality of the heart sound signal. Compared with the prior art, the computer readable storage medium provided by the embodiment of the application has the same beneficial effects as the heart sound signal quality evaluation method provided by the above-mentioned embodiment one, embodiment two or embodiment three, and will not be repeated here.

[0233] Embodiment seven

[0234] The embodiment of the application further provides a computer program product, comprising a computer program, and the computer program is executed by a processor to realize the steps of the heart sound signal quality evaluation method.

[0235] The computer program product provided by the application can solve the technical problem of how to effectively evaluate the signal quality of the heart sound signal. Compared with the prior art, the computer program product provided by the embodiment of the application has the same beneficial effects as the heart sound signal quality evaluation method provided by the above-mentioned embodiment one, embodiment two or embodiment three, and will not be repeated here.

[0236] The above is only the preferred embodiment of the application, and does not limit the patent scope of the application, and any equivalent structure or equivalent flow transformation using the content of the specification and the drawings, or directly or indirectly applied to other related technical fields, are also included in the patent processing scope of the application.

Claims

1. A method of heart sound signal quality assessment, characterized in that, The heart sound signal quality evaluation method comprises: Collecting a heart sound signal; Extracting features of the heart sound signal to obtain signal features, wherein the signal features comprise one or more of local variance features, local peak value features, signal energy features, and zero-crossing rate features; Inputting the signal features into a pre-trained signal quality evaluation model to output signal quality evaluation results.

2. The heart sound signal quality assessment method of claim 1, wherein, The step of extracting features of the heart sound signal to obtain signal features comprises: Performing sliding window processing on the heart sound signal, and taking the heart sound signal in the sliding window as a sub-heart sound signal; For each sub-heart sound signal, extracting features of the sub-heart sound signal to obtain signal features corresponding to the sub-heart sound signal.

3. The heart sound signal quality assessment method of claim 2, wherein, In the case where the signal features comprise local variance features, the step of extracting features of the sub-heart sound signal to obtain signal features corresponding to the sub-heart sound signal comprises: Performing sliding window processing on the sub-heart sound signal, and taking the sub-heart sound signal in the sliding window as a secondary sub-heart sound signal; Determining a signal value standard deviation of each secondary sub-heart sound signal, and taking the variance of all the signal value standard deviations as a first variance feature; Taking the mean of all the signal value standard deviations as a first mean, and sorting all the signal value standard deviations to obtain a standard deviation sequence, wherein the standard deviation sequence comprises the signal value standard deviations sorted in descending order based on numerical values; Determining a preset number of signal value standard deviations in the standard deviation sequence; Taking the mean of the preset number of signal value standard deviations as a second mean, and taking the ratio of the first mean to the second mean as a second variance feature; Taking the variance of all the window start sampling point positions as a third variance feature; Taking one or more of the first variance feature, the second variance feature, and the third variance feature as a local variance feature.

4. The heart sound signal quality assessment method of claim 2, wherein, In the case where the signal features comprise local peak value features, the step of extracting features of the sub-heart sound signal to obtain signal features corresponding to the sub-heart sound signal comprises: Determining a signal peak value of the sub-heart sound signal, adjusting the signal peak value based on a first preset adjustment coefficient to obtain a first threshold value, wherein the first threshold value is less than the signal peak value; Determining a first target signal value greater than the first threshold value in the signal value of the sub-heart sound signal, and taking the sampling point number of the signal sampling point corresponding to the first target signal value as a first sampling point number; Adjusting the signal peak value based on a second preset adjustment coefficient to obtain a second threshold value, wherein the first threshold value is less than the signal peak value, and the second preset adjustment coefficient is less than the first preset adjustment coefficient; Determining a second target signal value less than the second threshold value in the signal value of the sub-heart sound signal, and taking the sampling point number of the signal sampling point corresponding to the second target signal value as a second sampling point number; Taking the first sampling point number and / or the second sampling point number as a local peak value feature.

5. The heart sound signal quality assessment method of claim 2, wherein, In the case where the signal features comprise signal energy features, the step of extracting features of the sub-heart sound signal to obtain signal features corresponding to the sub-heart sound signal comprises: performing Fourier transform on the sub-heart sound signal, taking the sub-heart sound signal after Fourier transform as a target signal; determining signal energy corresponding to each frequency of the target signal, determining total signal energy of all the signal energy; determining first frequencies less than a first preset frequency among all frequencies included in the target signal, determining signal energy corresponding to each of the first frequencies, and taking total energy of all the signal energy corresponding to the first frequencies as a low-frequency energy distribution proportion; determining second frequencies greater than a second preset frequency among all frequencies included in the target signal, determining signal energy corresponding to each of the second frequencies, and taking total energy of all the signal energy corresponding to the second frequencies as a high-frequency energy distribution proportion, wherein the second preset frequency is greater than or equal to the first preset frequency; inputting the target signal into a preset log energy feature extraction model, and outputting log energy; taking one or more of the total signal energy, the low-frequency energy distribution proportion, the high-frequency energy distribution proportion, and the log energy as a signal energy feature.

6. The heart sound signal quality assessment method of claim 2, wherein, In the case where the signal feature includes the log energy feature, and in the case where the signal feature includes the zero-crossing rate feature, the step of performing feature extraction on the sub-heart sound signal to obtain a signal feature corresponding to the sub-heart sound signal comprises: determining a zero-crossing rate of each frame of the sub-heart sound signal, wherein the zero-crossing rate comprises a ratio of a number of signal sampling points of zero-crossing to a total number of signal sampling points of a frame of the heart sound signal; determining a maximum zero-crossing rate among all the zero-crossing rates, and determining a minimum zero-crossing rate among all the zero-crossings; performing differential processing on all the zero-crossing rates to obtain a first-order differential zero-crossing rate; determining a maximum first-order differential zero-crossing rate among all the first-order differential zero-crossing rates, and determining a variance of all the first-order differential zero-crossing rates; performing differential processing on all the first-order differential zero-crossing rates to obtain a second-order differential zero-crossing rate; determining a maximum second-order differential zero-crossing rate among all the second-order differential zero-crossing rates, and determining a standard deviation of all the second-order differential zero-crossing rates; taking one or more of the maximum zero-crossing rate, the minimum zero-crossing rate, the maximum first-order differential zero-crossing rate, the variance, the maximum second-order differential zero-crossing rate, and the standard deviation as a zero-crossing rate feature.

7. The heart sound signal quality assessment method of claim 2, wherein, The step of inputting the signal feature into a pre-trained signal quality evaluation model and outputting a signal quality evaluation result comprises: inputting the signal feature corresponding to each sub-heart sound signal into the pre-trained signal quality evaluation model, and outputting a signal quality evaluation result of each sub-heart sound signal.

8. The heart sound signal quality assessment method according to any one of claims 1 to 7, characterized in that, Before the step of performing feature extraction on the heart sound signal to obtain a signal feature, the method further comprises: performing signal filtering processing on the heart sound signal to obtain a filtered heart sound signal, wherein the signal filtering comprises zero-phase Butterworth filtering; performing signal clipping processing on the filtered heart sound signal to obtain a pre-processed heart sound signal, wherein the signal clipping comprises clipping a first end heart sound signal of a preset time length and / or clipping a tail end heart sound signal of a preset time length; Based on the pre-processed heart sound signal, the following is performed: feature extraction is performed on the heart sound signal to obtain signal features.

9. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory connected to the at least one processor in communication; wherein The memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the steps of the heart sound signal quality evaluation method according to any one of claims 1 to 8.

10. A readable storage medium, characterized by, The readable storage medium is a computer readable storage medium, and the computer readable storage medium stores a program for implementing a heart sound signal quality evaluation method. The program for implementing the heart sound signal quality evaluation method is executed by a processor to implement the steps of the heart sound signal quality evaluation method according to any one of claims 1 to 8.

Images