Heart sound signal processing method, device and wearable device

Abstract

The present disclosure relates to a heart sound signal processing method, device and wearable device, the method comprising: in the case that the signal quality of a first heart sound signal does not meet a preset requirement, continuously segmenting the first heart sound signal into a plurality of first sub-signals and continuously segmenting a second heart sound signal into a plurality of second sub-signals; determining a weight coefficient of the frequency domain signal corresponding to each frequency value of the first sub-signals and the second sub-signals according to the phase difference of the frequency domain signal corresponding to each frequency value; determining a fused frequency domain signal corresponding to each frequency value based on each window, according to the frequency domain signal corresponding to each frequency value of the first sub-signals and the second sub-signals and the weight coefficient of the frequency domain signal corresponding to each frequency value; determining a time domain signal of each sampling point based on each window, according to the fused frequency domain signal corresponding to each frequency value; and generating a reconstructed heart sound signal according to the time domain signal of each sampling point of each window.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to a heart state detection technology, and more particularly, to a heart sound signal processing method, device and wearable device. BACKGROUND

[0002] As a sound signal reflecting mechanical activity of heart valves, the heart sound signal has important value in non-invasive heart function monitoring and is widely used in many fields such as heart murmur detection and heart valve disease screening. In recent years, with the development of wearable devices and remote medical technology, heart sound signal sensors are being integrated into wearable devices, such as smart watches, to facilitate users to monitor their own heart function in daily life. In the prior art, a wearable device usually integrates a heart sound sensor. When a user places the wearable device worn on the chest, the heart sound sensor can collect heart sound signals. This scheme can provide a signal with sufficient signal-to-noise ratio under ideal wearing conditions, meeting the subsequent heart sound analysis requirements.

[0003] However, in actual use of the wearable device, in the case of a shift in the wearing position or a change in the body posture, the user often cannot correctly keep the wearable device placed flat on the chest, resulting in a significant decrease in the signal quality received by the heart sound sensor and a deviation in the analysis result based on the heart sound signal. SUMMARY

[0004] An object of the present application is to provide a new technical solution for a heart sound signal processing method.

[0005] According to a first aspect of the present application, a heart sound signal processing method is provided, comprising:

[0006] obtaining a first heart sound signal collected by a first heart sound sensor and a second heart sound signal collected by a second heart sound sensor, wherein the first heart sound sensor is arranged on a chassis of a wearable device;

[0007] in a case where the signal quality of the first heart sound signal does not meet a preset requirement, continuously dividing the first heart sound signal into a plurality of first sub-signals and continuously dividing the second heart sound signal into a plurality of second sub-signals based on a sliding window;

[0008] determining a phase difference between the first sub-signals and the second sub-signals corresponding to each window in a frequency domain signal corresponding to each frequency value;

[0009] determining a weight coefficient of the first sub-signals and the second sub-signals in the frequency domain signal corresponding to each frequency value according to the phase difference between the first sub-signals and the second sub-signals in the frequency domain signal corresponding to each frequency value;

[0010] determine, based on the first partial signal and the second partial signal, the frequency-domain signal corresponding to each frequency value according to the frequency-domain signal corresponding to each frequency value, a weight coefficient of the first partial signal and the second partial signal;

[0011] determine, based on the first partial signal and the second partial signal, the frequency-domain signal corresponding to each frequency value according to the frequency-domain signal corresponding to each frequency value, a weight coefficient of the first partial signal and the second partial signal;

[0012] generate the reconstructed heart sound signal according to the time-domain signal of each sampling point of each window.

[0013] Optionally, the two adjacent first partial signals overlap and the two adjacent second partial signals overlap, and the method further comprises:

[0014] obtain the time-domain signal of the sampling point with overlap based on the two adjacent windows;

[0015] perform weighted average processing on the two time-domain signals corresponding to each sampling point with overlap to obtain the time-domain signal of the corresponding sampling point.

[0016] Optionally, the second heart sound sensor is arranged on the side of the wearable device.

[0017] Optionally, the method further comprises:

[0018] determine the average energy value of the first heart sound signal according to the time-domain signal of each sampling point of the first heart sound signal, and determine the average energy value of the second heart sound signal according to the time-domain signal of each sampling point of the second heart sound signal;

[0019] determine that the signal quality of the first heart sound signal does not meet the preset requirement in a case where the ratio of the average energy value of the first heart sound signal to the average energy value of the second heart sound signal is less than a preset threshold.

[0020] Optionally, the method further comprises:

[0021] determine that the signal quality of the first heart sound signal meets the preset requirement in a case where the ratio of the average energy value of the first heart sound signal to the average energy value of the second heart sound signal is greater than or equal to the preset threshold, and only use the first heart sound signal to perform subsequent analysis processing.

[0022] Optionally, before the first heart sound signal is continuously divided into a plurality of first partial signals and the second heart sound signal is continuously divided into a plurality of second partial signals based on the sliding window in a case where the signal quality of the first heart sound signal does not meet the preset requirement, the method further comprises:

[0023] determine delay information of the second heart sound signal relative to the first heart sound signal according to the cross-correlation function;

[0024] perform time alignment processing on the first heart sound signal and the second heart sound signal according to the delay information, to obtain aligned first and second heart sound signals.

[0025] Optionally, the determining of the phase difference between the first and second partial signals corresponding to each window in the frequency domain signal corresponding to each frequency value comprises:

[0026] performing fast Fourier transform on the first and second partial signals corresponding to each window, to obtain the frequency domain signal of the first and second partial signals corresponding to each window;

[0027] determining the phase difference between the first and second partial signals corresponding to each window in the frequency domain signal corresponding to each frequency value according to the frequency domain signal of the first and second partial signals corresponding to each window;

[0028] The determining of the time domain signal of each sampling point according to the fused frequency domain signal corresponding to each frequency value based on each window comprises:

[0029] performing inverse fast Fourier transform on the fused frequency domain signal corresponding to each frequency value based on each window, to obtain the time domain signal of each sampling point.

[0030] According to a second aspect of the present application, a heart sound signal processing device is provided, comprising:

[0031] a signal acquisition module configured to acquire a first heart sound signal collected by a first heart sound sensor and a second heart sound signal collected by a second heart sound sensor, wherein the first heart sound sensor is arranged on a chassis of a wearable device;

[0032] a signal segmentation module configured to, in a case where the signal quality of the first heart sound signal does not meet a preset requirement, continuously segment the first heart sound signal into a plurality of first partial signals and continuously segment the second heart sound signal into a plurality of second partial signals based on a sliding window;

[0033] a phase difference determination module configured to determine the phase difference between the first and second partial signals corresponding to each window in the frequency domain signal corresponding to each frequency value;

[0034] a weight coefficient determination module configured to determine the weight coefficient of the frequency domain signal of the first and second partial signals corresponding to each frequency value according to the phase difference between the frequency domain signal of the first and second partial signals corresponding to each frequency value;

[0035] a fusion module configured to determine a fused frequency domain signal corresponding to each frequency value based on the first partial signal and the second partial signal and a weight coefficient of the first partial signal and the second partial signal corresponding to each frequency value according to the first partial signal and the second partial signal based on each window;

[0036] a time domain signal determination module configured to determine a time domain signal of each sampling point based on the fused frequency domain signal corresponding to each frequency value according to each window;

[0037] a signal reconstruction module configured to generate a reconstructed heart sound signal based on the time domain signal of each sampling point of each window.

[0038] According to a third aspect of the present disclosure, a heart sound signal processing apparatus is provided, comprising a memory and a processor, the memory storing a computer program for controlling the processor to operate to perform the method according to any one of the first aspect.

[0039] According to a fourth aspect of the present disclosure, a wearable device is provided, comprising a first heart sound sensor, a second heart sound sensor and a heart sound signal processing apparatus according to the second aspect or the third aspect.

[0040] The present disclosure provides a heart sound signal processing method, additionally provided with a heart sound sensor, i.e. a second heart sound sensor, to improve the spatial collection coverage of the heart sound sensor, make up for the deficiency of single acoustic detection dimension, and in the case that the signal quality of the first heart sound signal does not meet the preset requirement, fuse the first heart sound signal and the second heart sound signal to obtain a fused heart sound signal, thereby improving the accuracy of the heart sound signal and providing an accurate data basis for subsequent analysis.

[0041] The features of the embodiments of the present disclosure, and the advantages thereof, will become more apparent from the following detailed description of exemplary embodiments of the present disclosure with reference to the drawings. BRIEF DESCRIPTION OF DRAWINGS

[0042] The accompanying drawings, which are incorporated in and constitute a part of the specification, illustrate embodiments of the present disclosure and, together with the description, serve to explain the principles of the embodiments of the present disclosure.

[0043] Figure 1 is a flowchart of a heart sound signal processing method according to an embodiment of the present disclosure.

[0044] Figure 2 is a perspective view of the setting positions of a first heart sound sensor and a second heart sound sensor according to an embodiment of the present disclosure.

[0045] Figure 3is a top view of the setting positions of the first heart sound sensor and the second heart sound sensor according to an embodiment of the present application.

[0046] Figure 4 is a flowchart of a heart sound signal processing method according to an embodiment of the present application.

[0047] Figure 5 is a principle block diagram of a heart sound signal processing device according to an embodiment of the present application.

[0048] Figure 6 is a structural schematic diagram of a heart sound signal processing device according to an embodiment of the present application.

[0049] Figure 7 is a structural schematic diagram of a wearable device according to an embodiment of the present application. DETAILED DESCRIPTION

[0050] Various exemplary embodiments of the present specification will now be described in detail with reference to the accompanying drawings.

[0051] The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the description of embodiments of the present specification and its applications or uses.

[0052] It should be noted that similar reference numbers and letters refer to similar items in the following drawings, and thus, once an item is defined in one drawing, it need not be discussed further in subsequent drawings.

[0053] To solve the above technical problems, the present disclosure provides a heart sound signal processing method, an additional heart sound sensor, i.e., a second heart sound sensor, is provided to improve the spatial collection coverage capability of the heart sound sensor, to make up for the deficiency of single acoustic detection dimension, and in the case that the signal quality of the first heart sound signal does not meet the preset requirements, the fusion processing of the first heart sound signal and the second heart sound signal is performed to obtain the fused heart sound signal, thereby improving the accuracy of the heart sound signal and providing an accurate data basis for subsequent analysis.

[0054] In an embodiment of the present application, a heart sound signal processing method is provided. According to Figure 1 As shown in the figure, the heart sound signal processing method of the present embodiment includes the following steps S110-S170.

[0055] In step S110, a first heart sound signal collected by a first heart sound sensor and a second heart sound signal collected by a second heart sound sensor are obtained, wherein the first heart sound sensor is arranged on the chassis of a wearable device.

[0056] The propagation of the heart sound signal in the body tissue has directionality, and the first heart sound sensor is arranged on the chassis of the wearable device, which is more conducive to collecting the heart sound signal and ensuring the signal collection quality, such asFigure 2 and Figure 3 as shown.

[0057] When the user places the wearable device worn on the chest, the signal quality of the first heart sound signal collected by the first heart sound sensor is good. When the user places the wearable device on the chest at an angle or the user cannot continuously and stably maintain the posture of placing the wearable device on the chest, the quality of the first heart sound signal collected by the first heart sound sensor is poor, which is specifically manifested in signal amplitude attenuation, loss of high-frequency components, and intensified interference of environmental noise and electromyographic noise, further leading to deviation of the analysis result based on the first heart sound signal. To solve this problem, in the embodiment, an additional heart sound sensor is provided to improve the spatial collection coverage of the heart sound sensor and make up for the deficiency of single acoustic detection dimension.

[0058] The additional heart sound sensor, i.e., the second heart sound sensor, can be arranged on the side of the wearable device, as shown in Figure 2 and Figure 3 In this way, the first heart sound sensor can capture the normal component of the heart sound signal, and the second heart sound sensor can pick up the tangential component of the heart sound signal, and the two are fused to obtain the complete heart sound signal. The normal direction here refers to the direction perpendicular to the chest plane of the user, and the tangential direction refers to the direction parallel to the chest plane.

[0059] The first heart sound signal and the second heart sound signal are both time domain signals.

[0060] In step S120, in the case where the signal quality of the first heart sound signal does not meet the preset requirement, the first heart sound signal is continuously segmented into a plurality of first sub-signals and the second heart sound signal is continuously segmented into a plurality of second sub-signals based on a sliding window.

[0061] The length of the sliding window can be set according to requirements, that is, the number of sampling points of the heart sound signal included in the sliding window can be set according to requirements, for example, the sliding window includes 100 sampling points, the sliding window includes 500 sampling points, or the sliding window includes 1000 sampling points.

[0062] The first heart sound signal is continuously segmented into a plurality of first sub-signals, and one window corresponds to one first sub-signal. The second heart sound signal is continuously segmented into a plurality of second sub-signals, and one window corresponds to one second sub-signal.

[0063] In step S130, the phase difference between the first sub-signal and the second sub-signal corresponding to each window in the frequency domain signal corresponding to each frequency value is determined.

[0064] In some embodiments, step S130 specifically includes S131-S132.

[0065] Step S131, the first and second sub-signals corresponding to each window are subjected to fast Fourier transform to obtain the frequency domain signals of the first and second sub-signals corresponding to each window at each frequency value.

[0066] The first and second sub-signals corresponding to each window are time domain signals. Based on fast Fourier transform, the time domain signals can be converted into frequency domain signals.

[0067] The first sub-signal is subjected to fast Fourier transform based on the following calculation formula to obtain the frequency domain signal of the first sub-signal at the kth frequency value,

[0068]

[0069] wherein, is the frequency domain signal of the first sub-signal at the kth frequency value, is the time domain signal amplitude of the first sub-signal at the nth sampling point, the total number of sampling points and the total number of frequency values are both N, The calculation formula corresponding to the fast Fourier transform is that the total number of sampling points and the total number of frequency values are both N, which is to ensure that the fused frequency domain signal can be converted back to the time domain signal by inverse fast Fourier transform.

[0070] The second sub-signal is subjected to fast Fourier transform based on the following calculation formula to obtain the frequency domain signal of the second sub-signal at the kth frequency value,

[0071]

[0072] wherein, is the frequency domain signal of the second sub-signal at the kth frequency value, is the time domain signal amplitude of the second sub-signal at the nth sampling point, the total number of sampling points and the total number of frequency values are both N, The calculation formula corresponding to the fast Fourier transform is that the total number of sampling points and the total number of frequency values are both N, which is to ensure that the fused frequency domain signal can be converted back to the time domain signal by inverse fast Fourier transform.

[0073] Step S132, according to the frequency domain signals of the first and second sub-signals corresponding to each window at each frequency value, the phase difference of the frequency domain signals of the first and second sub-signals corresponding to each window at each frequency value is determined.

[0074] Based on the first and second sub-signals, each frequency value corresponds to a phase difference.

[0075] The phase difference of the frequency domain signals of the first and second sub-signals at the kth frequency value is determined based on the following calculation formula,

[0076]

[0077] wherein, is a phase difference of the frequency domain signal corresponding to the kth frequency value of the first partial signal and the second partial signal, is the frequency domain signal corresponding to the kth frequency value of the first partial signal, is the frequency domain signal corresponding to the kth frequency value of the second partial signal, is a taking complex phase.

[0078] The value of is .

[0079] Step S140, according to the phase difference of the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal, determine the weight coefficient of the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal.

[0080] In some embodiments, step S140 specifically includes steps S141-S142.

[0081] Step S141, based on the cosine function, the phase difference of the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal is converted into a coefficient.

[0082] Based on the following calculation formula, the phase difference of the frequency domain signal corresponding to the kth frequency value of the first partial signal and the second partial signal is converted into a coefficient,

[0083]

[0084] wherein, is a coefficient converted from the phase difference of the frequency domain signal corresponding to the kth frequency value of the first partial signal and the second partial signal, is the phase difference of the frequency domain signal corresponding to the kth frequency value of the first partial signal and the second partial signal.

[0085] Step S142, according to the converted coefficient, determine the first weight coefficient of the frequency domain signal corresponding to each frequency value of the first partial signal and the second weight coefficient of the frequency domain signal corresponding to each frequency value of the second partial signal.

[0086] Based on the following calculation formula, determine the first weight coefficient of the frequency domain signal corresponding to the kth frequency value of the first partial signal and the second weight coefficient of the frequency domain signal corresponding to the kth frequency value of the second partial signal,

[0087]

[0088] wherein, is the first weight coefficient of the frequency domain signal corresponding to the kth frequency value of the first partial signal, is a second weight coefficient of the second partial signal corresponding to the frequency domain signal of the kth frequency value.

[0089] At step S150, based on each window, a fused frequency domain signal corresponding to each frequency value is determined according to the frequency domain signals of the first partial signal and the second partial signal corresponding to each frequency value, and the weight coefficients of the frequency domain signals of the first partial signal and the second partial signal corresponding to each frequency value.

[0090] Based on each window, a frequency domain signal can be fused for each frequency value. Specifically, based on each frequency value, the frequency domain signal corresponding to the first partial signal is multiplied by the corresponding first weight coefficient to obtain a first frequency domain partial signal, the frequency domain signal corresponding to the second partial signal is multiplied by the corresponding second weight coefficient to obtain a second frequency domain partial signal, and the first frequency domain partial signal and the second frequency domain partial signal are added to obtain a fused frequency domain signal corresponding to the frequency value.

[0091] Based on each window, a fused frequency domain signal corresponding to the kth frequency value is determined based on the following calculation formula,

[0092]

[0093] wherein, is a fused frequency domain signal corresponding to the kth frequency value, is a first weight coefficient of the first partial signal corresponding to the frequency domain signal of the kth frequency value, is a second weight coefficient of the second partial signal corresponding to the frequency domain signal of the kth frequency value, is the frequency domain signal of the first partial signal corresponding to the kth frequency value, is the frequency domain signal of the second partial signal corresponding to the kth frequency value.

[0094] When the phase of the frequency domain signal of the first partial signal corresponding to a certain frequency value and the phase of the frequency domain signal of the second partial signal corresponding to the frequency value are consistent, it is determined that the frequency domain signal of the first partial signal corresponding to the frequency value and the frequency domain signal of the second partial signal corresponding to the frequency value are both valid signals, the phase difference is 0, and the converted coefficient is 1, so that the fused frequency domain signal is enhanced at the frequency value. When the phase of the frequency domain signal of the first partial signal corresponding to a certain frequency value and the phase of the frequency domain signal of the second partial signal corresponding to the frequency value are opposite, it is determined that at least one of the frequency domain signal of the first partial signal corresponding to the frequency value and the frequency domain signal of the second partial signal corresponding to the frequency value is a noise signal, the phase difference is -1, and the converted coefficient is 0, so that the fused frequency domain signal is suppressed at the frequency value.

[0095] Step S160, based on each window, determining the time domain signal of each sampling point according to the fused frequency domain signal corresponding to each frequency value.

[0096] Based on each window, performing inverse fast Fourier transform on the fused frequency domain signal corresponding to each frequency value to obtain the time domain signal of each sampling point.

[0097] Based on the following calculation formula, for each window, determining the time domain signal of the nth sampling point according to the fused frequency domain signal corresponding to each frequency value,

[0098]

[0099] wherein, is the time domain signal of the nth sampling point, is the fused frequency domain signal corresponding to the kth frequency value, the total number of sampling points and the total number of frequency values are both N, corresponding to the calculation formula of inverse fast Fourier transform.

[0100] Step S170, generating the reconstructed heart sound signal according to the time domain signal of each sampling point of each window.

[0101] In some embodiments, there is an overlap between the two adjacent first partial signals and there is an overlap between the two adjacent second partial signals. The overlap rate is set according to requirements, for example, 25%, 50%. The two adjacent first partial signals are set to have an overlap, and the two adjacent second partial signals are set to have an overlap, in order to ensure the continuity and accuracy of the subsequent determination of the frequency domain signal.

[0102] In this embodiment, the method further includes: based on the two adjacent windows, obtaining the time domain signal of the sampling points with an overlap; and performing weighted average processing on the two time domain signals corresponding to each sampling point with an overlap to obtain the time domain signal of the corresponding sampling point.

[0103] In some embodiments, the method further includes: determining the average energy value of the first heart sound signal according to the time domain signal of each sampling point of the first heart sound signal, and determining the average energy value of the second heart sound signal according to the time domain signal of each sampling point of the second heart sound signal; and determining that the signal quality of the first heart sound signal does not meet the preset requirement in a case where the ratio of the average energy value of the first heart sound signal to the average energy value of the second heart sound signal is less than a preset threshold.

[0104] The first heart sound signal and the second heart sound signal are time-aligned, and the first heart sound signal and the second heart sound signal include an equal number of sampling points.

[0105] The average energy value of the first heart sound signal is determined based on the following calculation formula,

[0106]

[0107] wherein, is an average energy value of the first heart sound signal, is a time domain signal amplitude of the first heart sound signal at the i th sampling point, and M is a number of sampling points included in the first heart sound signal.

[0108] An average energy value of the second heart sound signal is determined based on the following calculation formula,

[0109]

[0110] wherein, is an average energy value of the second heart sound signal, is a time domain signal amplitude of the second heart sound signal at the i th sampling point, and M is a number of sampling points included in the second heart sound signal.

[0111] In some embodiments, the method further comprises: in a case where a ratio of the average energy value of the first heart sound signal and the average energy value of the second heart sound signal is greater than or equal to a preset threshold value, determining that the signal quality of the first heart sound signal meets a preset requirement, and only using the first heart sound signal to perform subsequent analysis processing, without needing the second heart sound signal.

[0112] In some embodiments, before step S120, the method further comprises: determining delay information of the second heart sound signal relative to the first heart sound signal according to the cross-correlation function; and performing time alignment processing on the first heart sound signal and the second heart sound signal according to the delay information, to obtain the aligned first heart sound signal and the second heart sound signal. The alignment of the first heart sound signal and the second heart sound signal in time provides a more accurate data basis for the signal processing involved in the subsequent steps.

[0113] The cross-correlation function can refer to the following calculation formula,

[0114]

[0115] wherein, is a correlation corresponding to a delay of m sampling points of the second heart sound signal relative to the first heart sound signal, is a time domain signal amplitude of the first heart sound signal at the i th sampling point, is a time domain signal amplitude of the second heart sound signal at the i+m th sampling point, and M is a number of sampling points included in the first heart sound signal and the second heart sound signal. m is a quantity to be solved.

[0116] The heart sound signal processing method provided by the present application is described below with reference to a specific embodiment.

[0117] According to Figure 4As shown, the heart sound signal processing method of the embodiment includes steps S401-S412.

[0118] In step S401, a first heart sound signal collected by a first heart sound sensor and a second heart sound signal collected by a second heart sound sensor are acquired. The first heart sound sensor is arranged on a bottom plate of a wearable device, and the second heart sound sensor is arranged on a side of the wearable device.

[0119] In step S402, delay information of the second heart sound signal relative to the first heart sound signal is determined according to a cross-correlation function, and the first heart sound signal and the second heart sound signal are time-aligned according to the delay information to obtain aligned first and second heart sound signals.

[0120] In step S403, an average energy value of the first heart sound signal is determined according to a time domain signal of each sampling point of the first heart sound signal, and an average energy value of the second heart sound signal is determined according to a time domain signal of each sampling point of the second heart sound signal.

[0121] In step S404, it is determined whether a ratio of the average energy value of the first heart sound signal and the average energy value of the second heart sound signal is less than a preset threshold value, to obtain a determination result.

[0122] In a case where the determination result is yes, step S405 is performed, in which the first heart sound signal is continuously segmented into a plurality of first sub-signals and the second heart sound signal is continuously segmented into a plurality of second sub-signals based on a sliding window, and there is an overlap between adjacent two first sub-signals and there is an overlap between adjacent two second sub-signals.

[0123] In a case where the determination result is no, step S406 is performed, in which it is determined that a signal quality of the first heart sound signal meets a preset requirement, and only the first heart sound signal is used to perform subsequent analysis processing.

[0124] After step S405, step S407 is performed, in which a phase difference between a frequency domain signal of each frequency value corresponding to each window of the first sub-signal and the second sub-signal is determined.

[0125] In step S408, a weight coefficient of the frequency domain signal of each frequency value corresponding to the first sub-signal and the second sub-signal is determined according to the phase difference between the frequency domain signal of each frequency value corresponding to the first sub-signal and the second sub-signal.

[0126] In step S409, a fused frequency domain signal corresponding to each frequency value is determined according to the frequency domain signal of each frequency value corresponding to the first sub-signal and the second sub-signal and the weight coefficient of the frequency domain signal of each frequency value corresponding to the first sub-signal and the second sub-signal based on each window.

[0127] Step S410, based on each window, determining the time domain signal of each sampling point according to the fused frequency domain signal corresponding to the frequency value.

[0128] Step S411, based on the adjacent two windows, obtaining the time domain signal of the sampling point with overlap; performing weighted average processing on the two time domain signals corresponding to each sampling point with overlap to obtain the time domain signal of the corresponding sampling point.

[0129] Step S412, generating the reconstructed heart sound signal according to the time domain signals of all sampling points.

[0130] One embodiment of the present application provides a heart sound signal processing device. Figure 5 As shown in the figure, the heart sound signal processing device comprises a signal acquisition module 510, a signal segmentation module 520, a phase difference determination module 530, a weight coefficient determination module 540, a fusion module 550, a time domain signal determination module 560 and a signal reconstruction module 570.

[0131] The signal acquisition module 510 is used for acquiring a first heart sound signal collected by a first heart sound sensor and a second heart sound signal collected by a second heart sound sensor, wherein the first heart sound sensor is arranged on a chassis of a wearable device.

[0132] The signal segmentation module 520 is used for, in the case that the signal quality of the first heart sound signal does not meet a preset requirement, continuously segmenting the first heart sound signal into a plurality of first sub-signals and continuously segmenting the second heart sound signal into a plurality of second sub-signals based on a sliding window.

[0133] The phase difference determination module 530 is used for determining the phase difference of the frequency domain signal corresponding to each frequency value between the first sub-signal and the second sub-signal corresponding to each window.

[0134] The weight coefficient determination module 540 is used for determining the weight coefficient of the frequency domain signal corresponding to each frequency value between the first sub-signal and the second sub-signal according to the phase difference of the frequency domain signal corresponding to each frequency value between the first sub-signal and the second sub-signal.

[0135] The fusion module 550 is used for, based on each window, determining the fused frequency domain signal corresponding to each frequency value according to the frequency domain signal corresponding to each frequency value between the first sub-signal and the second sub-signal and the weight coefficient of the frequency domain signal corresponding to each frequency value between the first sub-signal and the second sub-signal.

[0136] The time domain signal determination module 560 is used for, based on each window, determining the time domain signal of each sampling point according to the fused frequency domain signal corresponding to each frequency value.

[0137] The signal reconstruction module 570 is used for generating a reconstructed heart sound signal according to the time domain signal of each sampling point of each window.

[0138] In some embodiments, there is overlap between the two adjacent first partial signals and there is overlap between the two adjacent second partial signals. The apparatus further comprises a weighted average processing module. The weighted average processing module is configured to obtain, based on the two adjacent windows, a time domain signal of a sampling point with overlap; and perform weighted average processing on the two time domain signals corresponding to each sampling point with overlap to obtain a time domain signal of the corresponding sampling point.

[0139] In some embodiments, the second heart sound sensor is disposed on a side of the wearable device.

[0140] In some embodiments, the apparatus further comprises an average energy value determination module and a signal quality determination module. The average energy value determination module is configured to determine an average energy value of the first heart sound signal based on the time domain signal of each sampling point of the first heart sound signal, and determine an average energy value of the second heart sound signal based on the time domain signal of each sampling point of the second heart sound signal. The signal quality determination module is configured to determine that the signal quality of the first heart sound signal does not meet a preset requirement if a ratio of the average energy value of the first heart sound signal to the average energy value of the second heart sound signal is less than a preset threshold.

[0141] In some embodiments, the signal quality determination module is configured to determine that the signal quality of the first heart sound signal meets a preset requirement if the ratio of the average energy value of the first heart sound signal to the average energy value of the second heart sound signal is greater than or equal to the preset threshold, and perform subsequent analysis processing only using the first heart sound signal.

[0142] In some embodiments, the apparatus further comprises an alignment module. The alignment module is configured to determine delay information of the second heart sound signal relative to the first heart sound signal based on the cross-correlation function; and perform time alignment processing on the first heart sound signal and the second heart sound signal based on the delay information to obtain an aligned first heart sound signal and an aligned second heart sound signal.

[0143] In some embodiments, the phase difference determination module 530 is configured to perform fast Fourier transform on the first partial signal and the second partial signal corresponding to each window to obtain a frequency domain signal of the first partial signal and the second partial signal corresponding to each frequency value corresponding to each window; and determine a phase difference of the frequency domain signal of the first partial signal and the second partial signal corresponding to each frequency value corresponding to each window based on the frequency domain signal of the first partial signal and the second partial signal corresponding to each frequency value corresponding to each window. The time domain signal determination module 560 is configured to perform inverse fast Fourier transform on the fused frequency domain signal corresponding to each frequency value based on each window to obtain a time domain signal of each sampling point.

[0144] One embodiment of the present application provides a heart sound signal processing apparatus. According to the present application,Figure 6 As shown, the heart health state detection apparatus comprises a memory 620 and a processor 610. The memory 620 stores a computer program for controlling the processor 610 to operate to perform the heart sound signal processing method according to any of the above embodiments.

[0145] The processor 610 is configured to execute computer instructions, which can be written in an instruction set of an architecture such as x86, Arm, RISC, MIPS, SSE, etc. The memory 620 includes, for example, a ROM (Read-Only Memory), a RAM (Random Access Memory), a nonvolatile memory such as a hard disk, etc., without limitation.

[0146] One embodiment of the present application provides a wearable device. According to the embodiment, the wearable device comprises a first heart sound sensor, a second heart sound sensor, and the heart sound signal processing apparatus according to any of the above embodiments. Figure 7 As shown, the wearable device comprises a first heart sound sensor, a second heart sound sensor, and the heart sound signal processing apparatus according to any of the above embodiments.

[0147] The wearable device can be any of the following: a smart watch, a smart bracelet.

[0148] Each of the embodiments in the present specification is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other. Each of the embodiments focuses on the difference from other embodiments.

[0149] The above describes specific embodiments of the present specification. Other embodiments are within the scope of the appended claims. In some cases, the acts or steps recited in the claims can be performed in a different order than the order in which they are recited and still achieve desirable results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0150] The embodiments of the present specification can be a system, a method, and / or a computer program product. The computer program product can include a computer readable storage medium having computer instructions loaded thereon for causing a processor to implement various aspects of the embodiments of the present specification.

[0151] Computer readable storage media can be tangible storage media which can retain and store instructions for use by an instruction execution device. Computer readable storage media can be, for example, but is not limited to, an electronic storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples (a non-exhaustive list) of computer readable storage media include the following: a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), a static random access memory (SRAM), a portable compact disc read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanically encoded device such as punch-cards or raised structures in a groove having instructions recorded thereon, and any suitable combination of the foregoing. A computer readable storage medium, as used herein, is not to be construed as being transitory signals per se, such as radio waves or other freely propagating electromagnetic waves, electromagnetic waves propagating through a waveguide or other transmission media (e.g., light pulses passing through a fiber-optic cable), or electrical signals transmitted through a wire.

[0152] Computer instructions described herein can be downloaded to respective computing / processing devices from a computer readable storage medium or to an external computer or external storage device via a network, for example, the Internet, a local area network, a wide area network and / or a wireless network. The network can comprise copper transmission cables, optical transmission fibers, wireless transmission, routers, firewalls, switches, gateway computers and / or edge servers. A network adapter card or network interface in each computing / processing device receives computer instructions from the network and forwards the computer instructions for storage in a computer readable storage medium within the respective computing / processing device.

[0153] The computer program product of the second aspect can include a computer readable storage medium. The computer readable storage medium can include instructions. The instructions can include one or both of: instructions for causing a computer to implement a method as described above; and instructions for causing a computer to operate based on a system as described above. The computer readable storage medium can include one or more of: a magnetic disk; a magnetic disk drive; a magnetic tape; a magneto-optical drive; a solid state drive; a semiconductor drive; a flash drive; an optical drive; a holographic drive; a holographic medium; a memory stick; a floppy disk; a flexible disk; a hard disk; a hard disk drive; a holographic disk; a holographic disk drive; a RAM drive; a ROM drive; a flash drive; an optical drive; a solid state drive; a solid state drive; a DVD; a DVD drive; a DVD-ROM; a DVD-RW; a DVD+RW; a Blu-Ray disk; a Blu-Ray disk drive; a memory stick; a memory card; an electrical connection via one or more busses; an other appropriate device.

[0154] Embodiments of the present description have been described above, the description is exemplary only, and is not exhaustive or limited to the disclosed embodiments. Many modifications and variations of the described embodiments are possible in light of the above teachings. The choice of words in this document is intended to convey the best of the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other ordinary skilled persons in the art to understand the embodiments disclosed herein.

Claims

1. A heart sound signal processing method, characterized by, The method comprises: obtaining a first heart sound signal collected by a first heart sound sensor and a second heart sound signal collected by a second heart sound sensor, wherein the first heart sound sensor is arranged on a chassis of a wearable device; in a case where a signal quality of the first heart sound signal does not meet a preset requirement, continuously segmenting the first heart sound signal into a plurality of first sub-signals and continuously segmenting the second heart sound signal into a plurality of second sub-signals based on a sliding window; determining a phase difference between the first sub-signals and the second sub-signals corresponding to each window in a frequency domain signal corresponding to each frequency value; determining a weight coefficient of the first sub-signals and the second sub-signals in the frequency domain signal corresponding to each frequency value according to the phase difference between the first sub-signals and the second sub-signals in the frequency domain signal corresponding to each frequency value; based on each window, determining a fused frequency domain signal corresponding to each frequency value according to the first sub-signals and the second sub-signals in the frequency domain signal corresponding to each frequency value and the weight coefficient of the first sub-signals and the second sub-signals in the frequency domain signal corresponding to each frequency value; based on each window, determining a time domain signal of each sampling point according to the fused frequency domain signal corresponding to each frequency value; generating a reconstructed heart sound signal according to the time domain signal of each sampling point of each window.

2. The method of claim 1, wherein, There is an overlap between adjacent two first sub-signals and there is an overlap between adjacent two second sub-signals, and the method further comprises: based on adjacent two windows, obtaining a time domain signal of a sampling point with an overlap; performing weighted average processing on two time domain signals corresponding to each sampling point with an overlap to obtain a time domain signal of the corresponding sampling point.

3. The method of claim 1, wherein, The second heart sound sensor is arranged on a side of the wearable device.

4. The method of claim 1, wherein, The method further comprises: determining an average energy value of the first heart sound signal according to the time domain signal of each sampling point of the first heart sound signal and determining an average energy value of the second heart sound signal according to the time domain signal of each sampling point of the second heart sound signal; in a case where a ratio of the average energy value of the first heart sound signal to the average energy value of the second heart sound signal is less than a preset threshold, determining that the signal quality of the first heart sound signal does not meet the preset requirement.

5. The method of claim 4, wherein, The method further comprises: in a case where the ratio of the average energy value of the first heart sound signal to the average energy value of the second heart sound signal is greater than or equal to the preset threshold, determining that the signal quality of the first heart sound signal meets the preset requirement and only using the first heart sound signal to perform subsequent analysis processing.

6. The method of claim 1, wherein, Before the first heart sound signal is continuously segmented into a plurality of first sub-signals and the second heart sound signal is continuously segmented into a plurality of second sub-signals based on the sliding window in a case where the signal quality of the first heart sound signal does not meet the preset requirement, the method further comprises: determining delay information of the second heart sound signal relative to the first heart sound signal according to a cross-correlation function; performing time alignment processing on the first heart sound signal and the second heart sound signal according to the delay information to obtain aligned first heart sound signal and second heart sound signal.

7. The method according to any one of claims 1 to 6, characterized in that, The determination of the phase difference between the first sub-signals and the second sub-signals corresponding to each window in the frequency domain signal corresponding to each frequency value comprises: performing fast Fourier transform on the first partial signal and the second partial signal corresponding to each window to obtain a frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal corresponding to each window; determining a phase difference of the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal corresponding to each window according to the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal corresponding to each window; the method further includes: performing inverse fast Fourier transform on the fused frequency domain signal corresponding to each frequency value based on each window to obtain the time domain signal of each sampling point.

8. A heart sound signal processing apparatus, characterized by comprising: The method further includes: a signal acquisition module configured to acquire a first heart sound signal collected by a first heart sound sensor and a second heart sound signal collected by a second heart sound sensor, wherein the first heart sound sensor is arranged on a chassis of a wearable device; a signal segmentation module configured to, in a case where a signal quality of the first heart sound signal does not meet a preset requirement, continuously segment the first heart sound signal into a plurality of first partial signals and continuously segment the second heart sound signal into a plurality of second partial signals based on a sliding window; a phase difference determination module configured to determine a phase difference of a frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal corresponding to each window; a weight coefficient determination module configured to determine a weight coefficient of the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal according to the phase difference of the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal; a fusion module configured to, based on each window, determine a fused frequency domain signal corresponding to each frequency value according to the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal and the weight coefficient of the frequency domain signal corresponding to each frequency value of the first partial signal and the second partial signal; a time domain signal determination module configured to, based on each window, determine a time domain signal of each sampling point according to the fused frequency domain signal corresponding to each frequency value; a signal reconstruction module configured to generate a reconstructed heart sound signal according to the time domain signal of each sampling point of each window.

9. A heart sound signal processing apparatus, characterized by comprising: The device includes a memory and a processor, the memory stores a computer program, and the computer program is used to control the processor to perform operations to execute the method according to any one of claims 1 to 7.

10. A wearable device, comprising: The device includes a first heart sound sensor, a second heart sound sensor, and a heart sound signal processing apparatus according to claim 8 or 9.

Images