Heart sound normalization
Through frequency-dependent normalization processing, the frequency-dependent attenuation problem caused by incorrect placement of sound sensors in traditional electronic medical systems is solved, and the accurate capture and processing of heart sounds is achieved, ensuring the correct representation of heart sounds.
Patent Information
- Application Number
- CN202411020431.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2019-12-02
- Filing Date
- 2020-12-02
- Publication Date
- 2024-12-03
AI Technical Summary
When traditional electronic medical systems capture heart sounds, incorrect representation and processing of heart sounds due to incorrect placement of sound sensors and frequency-dependent attenuation problems.
Through frequency-dependent normalization processing, the attenuation amount between the capture and control of the first heart sound, the modification amount is determined according to different frequency ranges, and the different parts of the second heart sound are modified to eliminate the attenuation and realize the normalization of the heart sound. .
Effectively eliminates the heart sound attenuation caused by sound sensors, ensuring accurate capture and processing of heart sounds, and avoiding errors caused by frequency-dependent attenuation.
Smart Images

Figure CN119055268A_ABST
Abstract
Description
[0001] This application is a divisional application of the Chinese invention patent application with application date of December 2, 2020, application number 202011389937.4, and invention name “Heart Sound Normalization”. Technical Field
[0002] The present disclosure relates generally to the field of healthcare systems and, more particularly, but not exclusively, to systems and methods for processing heart sounds. Background Art
[0003] In healthcare scenarios, the heart sounds produced by a subject's heart have been used to diagnose the subject's healthcare problems. Historically, caregivers (e.g., physicians) would use a stethoscope to listen to the subject's heart sounds and infer the subject's health status.
[0004] Conventional electronic medical systems have been developed to capture the heart sounds of a subject instead of a caregiver. In particular, conventional electronic medical systems can capture the heart sounds and store representations of the heart sounds. In some of these conventional electronic medical systems, a sound sensor can be placed on the subject to capture the sounds. Summary of the invention
[0005] Examples of systems and methods for processing captured heart sounds by frequency-dependent normalization are disclosed herein. Based on an amount of attenuation of a first heart sound, the second heart sound can be normalized by modifying portions of the second heart sound by an amount determined based on the frequency of the second portion. Thus, the systems and methods disclosed herein can result in different amounts of modification of different portions of the second heart sound based on different frequencies of the portions.
[0006] In certain embodiments, disclosed herein is one or more computer-readable media having instructions stored thereon, wherein the instructions, in response to execution by a device, cause the device to determine an attenuation amount between capture of a first heart sound and control of the first heart sound, capture of the first heart sound captured during a test cycle, modify a first portion of capture of a second heart sound by a first amount, wherein capture of the second heart sound is captured during the test cycle, wherein the first portion corresponds to a first frequency range, and wherein the first amount is determined based on the attenuation amount, and modify a second portion of capture of the second heart sound by a second amount, wherein the second portion corresponds to a second frequency range, and wherein the second amount is determined based on the attenuation amount.
[0007] In certain embodiments, the device disclosed herein for capturing heart sounds of a subject comprises: one or more sound sensors to sense the heart sounds of the subject, and to generate an electronic representation of the heart sounds. In addition in these embodiments, the device disclosed herein comprises: one or more processors to identify the capture of a first heart sound from the electronic representation, compare the capture of the first heart sound with the control of the first heart sound to determine the attenuation between the capture of the first heart sound and the control of the first heart sound, determine a first modification amount and a second modification amount based on the attenuation amount, identify the capture of a second heart sound from the electronic representation, modify a first portion of the capture of the second heart sound by the first modification amount, wherein the first portion of the capture corresponds to a first frequency range, and modify a second portion of the capture of the second heart sound by the second modification amount, wherein the second portion of the capture corresponds to a second frequency range.
[0008] In certain embodiments, a method for normalizing heart sounds includes comparing the capture of a first heart sound of a test cycle with a control of the first heart sound, determining an attenuation amount between the capture of the first heart sound and the control of the first heart sound based on the comparison of the capture of the first heart sound with the control of the first heart sound, modifying a first portion of the capture of a second heart sound of the test cycle by a first amount, wherein the first portion corresponds to a first frequency range, and wherein the first amount is determined based on the attenuation amount, and modifying a second portion of the capture of the second heart sound of the test cycle by a second amount, wherein the second portion corresponds to a second frequency range, and wherein the second amount is determined based on the attenuation amount. BRIEF DESCRIPTION OF THE DRAWINGS
[0009] The present disclosure will be best understood from the following detailed description when read in conjunction with the accompanying drawings. It is emphasized that, in accordance with standard practice in the industry, various features are not necessarily drawn to scale and are for illustration purposes only. Where scale is explicitly or implicitly shown, it is merely provided as an illustrative example. In other embodiments, the sizes of various features may be arbitrarily increased or reduced for clarity.
[0010] Figure 1 An example process for processing heart sounds according to embodiments herein is shown.
[0011] Figure 2 An exemplary electrical signal according to an embodiment of the present invention is shown, which can be received as Figure 1 The input of the process.
[0012] Figure 3 An example frequency domain representation of an S1 heart sound according to embodiments herein is shown.
[0013] Figure 4 An example frequency domain representation of an S2 heart sound according to embodiments herein is shown.
[0014] Figure 5 An example frequency domain representation of an S3 heart sound according to embodiments herein is shown.
[0015] Figure 6 An example frequency domain representation of an S4 heart sound according to embodiments herein is shown.
[0016] Figure 7 An example process for determining attenuation according to embodiments herein is shown.
[0017] Figure 8 An example process for modifying the capture of heart sounds according to embodiments herein is shown.
[0018] Fig. 9 An example of a modified execution of the capture of negatively correlated heart sounds according to embodiments herein is shown.
[0019] Fig.10 It shows that the embodiment according to this article can be implemented Figure 1 An example system of the process.
[0020] Fig.11 An example process for initialization and normalization by a heart sound capture system according to embodiments herein is shown. DETAILED DESCRIPTION
[0021] The following disclosure provides many different embodiments or examples for realizing the different features of the present disclosure. The following describes specific examples of components and arrangements to simplify the present disclosure. Of course, these are merely examples and are not intended to be limiting. In addition, the present disclosure may repeat reference numerals and / or letters in various examples or in some cases across different drawings. This repetition is for the purpose of simplicity and clarity, and does not itself indicate a specific relationship between the various embodiments and / or configurations discussed. Different embodiments may have different advantages, and any embodiment does not require a specific advantage.
[0022] Disclosed herein are systems and methods for processing heart sounds by frequency-dependent normalization. Healthcare systems that capture heart sounds typically utilize one or more sound sensors placed on the skin of a subject. However, the placement of the sound sensors often inadvertently results in attenuation of the heart sounds captured by the healthcare system. For example, incorrect placement of the sound sensors and / or incorrect attachment of the sound sensors to the skin of the subject may result in attenuation of the heart sounds. Additionally, based on the frequencies of the different portions of the heart sounds, different portions of the heart sounds may be attenuated by different amounts. Failure to account for the attenuation of heart sounds and the different amounts of attenuation of different portions of heart sounds based on different frequencies may result in incorrect representation of heart sounds and / or processing of incorrect representations of heart sounds.
[0023] The systems and methods disclosed herein can take into account the attenuation of heart sounds and different attenuation amounts for different parts of heart sounds based on different frequencies. In particular, the systems and methods disclosed herein can compare the capture of a first heart sound with the control for the first heart sound to determine the attenuation amount between the capture of the first heart sound and the control for the first heart sound, wherein the first heart sound should have a uniform amplitude throughout the measurement. Each measurement may include one or more cardiac cycles. The systems and methods can utilize the attenuation amount to modify the capture of other heart sounds, wherein different parts of other heart sounds can be modified by different amounts based on different frequencies of different parts.
[0024] Figure 1 An example method 100 for processing heart sounds according to embodiments herein is shown. In particular, the method 100 may include normalizing one or more captures of a subject's heart sounds using a frequency-based approach.
[0025] The method 100 may begin by receiving a representation of one or more cardiac cycles as input 102. For example, the input 102 may include an electrical signal representing heart sounds included in the cardiac cycle. The electrical signal may be generated by a sound sensor that senses the heart sounds generated by the cardiac cycle and generates an electrical signal representing the heart sounds. For example, the amplitude of the electrical signal may indicate the amplitude of the sound. Furthermore, the electrical signal may be a time domain representation of the heart sound, wherein the electrical signal indicates the amplitude of the heart sound over time.
[0026] Figure 2 An example electrical signal 200 is shown that may be received as an input 102 of method 100 in accordance with an embodiment of the present invention. In particular, electrical signal 200 shows a single cardiac cycle captured from a subject. A cardiac cycle includes a single heartbeat, and electrical signal 200 shows heart sounds produced by the subject's heart during a single heartbeat. In the illustrated example, a single heartbeat produces an S1 heart sound, an S2 heart sound, an S3 heart sound, and an S4 heart sound. In other cases, the S3 heart sound and / or the S4 heart sound may be omitted from the heart sounds produced from the subject's heart during a heartbeat. Although a single cardiac cycle is shown in the illustrated example, it should be understood that in other cases, multiple cardiac cycles may be received as input 102 of method 100. A cardiac cycle and / or multiple cardiac cycles may be captured during a test cycle of the cardiac sound capture system.
[0027] Method 100 may include performing decomposition of the electrical signal at 104. The decomposition of the electrical signal may include identifying different heart sounds from the electrical signal. In particular, an S1 heart sound, an S2 heart sound, an S3 heart sound, and / or an S4 heart sound represented by the electrical signal may be identified. Each heart sound may be identified based on an amplitude of the heart sound represented by the electrical signal, a frequency of the heart sound represented by the electrical signal, a duration of the heart sound represented by the electrical signal, an order of the heart sounds represented by the electrical signal, or some combination thereof.
[0028] Figure 2 Heart sounds identified within method 100 according to embodiments herein are shown. For example, electrical signal 200 may include representations of one or more heart sounds that occur discretely in time. A cardiac cycle may include an S1 heart sound, an S2 heart sound, an S3 heart sound, an S4 heart sound, or some combination thereof. In the example shown, electrical signal 200 includes representations of S1 heart sound 202, S2 heart sound 204, S3 heart sound 206, and S4 heart sound 208. In the decomposition of electrical signal 200, S1 heart sound 202, S2 heart sound 204, S3 heart sound 206, and S4 heart sound 208 may be identified, as shown in FIG. Figure 2 208. In other cases, the representation of S3 heart sound 206 and / or S4 heart sound 208 may be omitted, such as when the subject's heartbeat does not produce S3 heart sound 206 and / or S4 heart sound 208 (which may be normal for the subject). In these cases, any combination of S1 heart sound 202, S2 heart sound 204, S3 heart sound 206, and / or S4 heart sound 208 may be identified, and the heart sounds omitted from electrical signal 200 may not be identified.
[0029] In some embodiments, the electrical signal 200 can be divided into different parts for processing during electrical signal decomposition. In particular, the electrical signal 200 can be divided into different parts, where each part corresponds to a corresponding, identified heart sound. For example, in the illustrated example, the first part can include the S1 heart sound 202, the second part can include the S2 heart sound 204, the third part can include the S3 heart sound 206, and the fourth part can include the S4 heart sound 208. In other embodiments, the electrical signal 200 can be divided into different parts at a later point during the method 100.
[0030] The method 100 may further include performing a spectral analysis of the electrical signal in 106. Specifically, the spectral analysis may be performed for each heart sound identified in the decomposition of the electrical signal performed in 104. The representation of the heart sound is converted into a frequency domain representation of the heart sound. The representation of the heart sound may be converted into a frequency domain representation by applying a Fourier transform to each representation of the heart sound. For example, the Fourier transform applied to each representation may include a fast Fourier transform, a discrete Fourier transform, a continuous Fourier transform, or some combination thereof. For example, in the illustrated example, the S1 heart sound 202, the S2 heart sound 204, the S3 heart sound 206, and the S4 heart sound 208 may be converted into a frequency domain representation of each. Different portions of the electrical signal 200 corresponding to different heart sounds may be converted separately.
[0031] Figure 3 An example frequency domain representation 300 of an S1 heart sound according to an embodiment of the present invention is shown. For example, the frequency domain representation 300 may be a frequency domain representation of the S1 heart sound 202. The frequency domain representation 300 may be generated by performing a spectral analysis on a portion of the electrical signal 200 corresponding to the S1 heart sound 202 in 106. As can be seen, the frequency domain representation 300 shows the amplitudes of different portions of the S1 heart sound 202 corresponding to different frequencies. For example, the frequency domain representation 300 shows a first peak 302 at a first frequency and a second peak 304 at a second frequency. In other embodiments, the peaks may occur at different frequencies and / or may have different amplitudes in the frequency domain representation 300.
[0032] Figure 4 An example frequency domain representation 400 of an S2 heart sound according to an embodiment of the present invention is shown. For example, the frequency domain representation 400 may be a frequency domain representation of the S2 heart sound 204. The frequency domain representation 400 may be generated by performing a spectral analysis on a portion of the electrical signal 200 corresponding to the S2 heart sound 204 in 106. As can be seen, the frequency domain representation 400 shows the amplitudes of different portions of the S2 heart sound 204 corresponding to different frequencies. For example, the frequency domain representation 400 shows a first peak 402 at a first frequency, a second peak 404 at a second frequency, and a third peak 406 at a third frequency. In other embodiments, the peaks may appear at different frequencies and / or may have different amplitudes in the frequency domain representation 400.
[0033] Figure 5An example frequency domain representation 500 of an S3 heart sound according to an embodiment of the present invention is shown. For example, the frequency domain representation 500 can be a frequency domain representation of the S3 heart sound 206. The frequency domain representation 500 can be generated by performing a spectral analysis on a portion of the electrical signal 200 corresponding to the S3 heart sound 206 in 106. It can be seen that the frequency domain representation 500 shows that different portions of the S3 heart sound 206 correspond to amplitudes of different frequencies. For example, the frequency domain representation 500 shows a peak 502 at a certain frequency. In other embodiments, the peaks can appear at different frequencies in the frequency domain representation 500 and / or can have different amplitudes.
[0034] Figure 6 An example frequency domain representation 600 of an S4 heart sound according to an embodiment of the present invention is shown. For example, the frequency domain representation 600 may be a frequency domain representation of the S4 heart sound 208. The frequency domain representation 600 may be generated by performing a spectral analysis on a portion of the electrical signal 200 corresponding to the S4 heart sound 208 in 106. As can be seen, the frequency domain representation 600 shows the amplitudes of different portions of the S4 heart sound 208 corresponding to different frequencies. For example, the frequency domain representation 600 shows a peak 602 at a certain frequency. In other embodiments, the peaks may appear at different frequencies in the frequency domain representation 600 and / or may have different amplitudes.
[0035] Method 100 may further include normalizing the heart sounds at 108. Normalizing the heart sounds may include determining an attenuation of one of the heart sounds and modifying one or more heart sounds using the attenuation. In other embodiments, normalizing the heart sounds may include determining an attenuation of a portion of the heart sounds and modifying one or more heart sounds using the attenuation. Modification of the one or more heart sounds may include: amplifying the one or more heart sounds; attenuating the one or more heart sounds; or amplifying a portion of the one or more heart sounds; and attenuating another portion of the one or more heart sounds. The attenuation may be caused by incorrect application of the sound sensor (e.g., incorrect placement of the sound sensor and / or the sound sensor is not airtightly placed on the subject's skin) and / or other forms of defects (which in some cases may result in audio coupling defects between the subject's body and the sensor used to measure the heart sounds).
[0036] The one or more heart sounds used to determine the attenuation amount may be predetermined or may be indicated by the user. For example, in some embodiments, one or more captures of the S1 heart sound or one or more captures of the S2 heart sound may be used to determine the attenuation amount or attenuation amount, where in most cases the amplitudes of the S1 heart sound and the S2 heart sound may be consistent over multiple cardiac cycles. In some embodiments, the user may be presented with a choice between utilizing one or more captures of the S1 heart sound or one or more captures of the S2 heart sound, and the capture of the S1 heart sound or the capture of the S2 heart sound may be indicated.
[0037] In some embodiments, the heart sound used to determine the attenuation amount can be selected based on the heart sound or heart sounds used to determine the attenuation amount, which heart sound or heart sound is expected to be negatively correlated with the heart sound or negative sound to be modified. For example, one or more heart sounds used to determine the attenuation amount can be expected to remain constant, while one or more heart sounds to be modified can change, or one or more heart sounds used to determine the attenuation amount can be expected to change in a direction opposite to the direction of the heart sound or heart sound to be changed if the heart sound or heart sound to be changed changes. Selecting one or more heart sounds to be negatively correlated with one or more heart sounds to be modified can allow coupling factors to be eliminated, and one or more heart sounds can be modified without showing coupling effects. The capture of the S1 heart sound, S2 heart sound, S3 heart sound, and / or S4 heart sound can be modified based on the determined attenuation amount or attenuation amount.
[0038] Figure 7 An example process 700 for determining attenuation according to an embodiment of the present invention is shown. Figure 1 ) to perform process 700 as part of the normalized heart sound in the image processing apparatus. Process 700 may include comparing a capture 702 of the S1 heart sound with a control 704 for the S1 heart sound to determine an amount of attenuation. In particular, a frequency domain representation of the capture 702 of the S1 heart sound may be compared with the control 704 for the S1 heart sound to determine an amount of attenuation. In the example shown, the S1 heart sound may be a predetermined heart sound for determining an amount of attenuation, or may be an indication that the S1 heart sound should be used to determine an amount of attenuation. In other examples, other heart sounds may be predetermined heart sounds or indications that other heart sounds should be used to determine an amount of attenuation, for example, the S2 heart sound may be used in other examples.
[0039] Capture 702 of the S1 heart sound may be generated by decomposition in 104 and spectral analysis in 106. Figure 2 ) extracts the capture 702 of the S1 heart sound. In particular, in the illustrated example, the capture 702 of the S1 heart sound may be a frequency domain representation 300 ( Figure 3 ).
[0040] Control 704 may be another frequency domain representation of the S1 heart sound. Control 704 may have been captured during the initialization cycle, may have been captured during a previous test cycle, or may be generated by averaging multiple captures captured during the initialization cycle and / or previous test cycles. Control 704 may be captured in the presence of an authorized user (e.g., a health care provider) to ensure that the frequency domain representation is properly captured by the heart sound capture system. In some embodiments, controls for other heart sounds may be captured during the initialization cycle, may be captured during a previous test cycle, or may be generated by averaging multiple captures captured during the initialization cycle and / or previous test cycles. For example, in addition to controls for the S1 heart sound, controls for S2 heart sounds, S3 heart sounds, and / or S4 heart sounds may also be captured. Controls for other heart sounds may be generated using heart sounds from the same one or more cardiac cycles used to generate the S1 heart sound.
[0041] The process 700 may include comparing the capture 702 of the S1 heart sound with the control 704 for the S1 heart sound to determine the attenuation amount. For example, one or more peak values, average amplitudes, RMS values, or other measurements of amplitude of the control 704 for the S1 heart sound may be compared with corresponding measurements of the capture 702 of the S1 heart sound. In the example shown, a first peak 706 of the control 704 occurring at a first frequency is compared with a first peak 708 of the capture 702 occurring at the first frequency to determine the attenuation amount 714 at the first frequency. In addition, a second peak 710 of the control 704 occurring at a second frequency is compared with a second peak 712 of the capture 702 occurring at the second frequency to determine the attenuation amount 716 at the second frequency. It can be seen that the attenuation amount 714 at the first frequency is greater than the attenuation amount 716 at the second frequency, wherein the first frequency is lower than the second frequency. In other embodiments, one or more attenuations may be determined in the process 700, wherein each attenuation amount may correspond to a corresponding frequency range. The determined one or more attenuations may be used to determine a frequency-dependent modification amount for modifying the capture of the heart sound. In the illustrated example, the capture 702 of the S1 heart sound is attenuated to define an attenuation amount 714 compared to the control 704. In other examples, the control 704 may be attenuated to define an attenuation amount 714 compared to the capture 702 of the S1 heart sound.
[0042] Figure 8An example process 800 for modifying a capture of heart sounds according to an embodiment of the present invention is shown. A portion of the process 800 may be performed as part of normalization of heart sounds at 108. The process 800 may include modifying a capture 802 of an S4 heart sound to produce a normalized representation 804 of the S4 heart sound. In particular, in the example shown, a frequency domain representation of the capture 802 of the S4 heart sound may be amplified to produce a normalized representation 804 of the S4 heart sound. In other cases, a frequency domain representation of the capture 802 of the S4 heart sound may be attenuated to produce a normalized representation 804 of the S4 heart sound.
[0043] The capture 802 of the S4 heart sound may have been generated by the decomposition in 104 and the spectral analysis in 106. The capture 802 of the S4 heart sound may have been generated from the electrical signal 200 ( Figure 2 ) is extracted. In particular, in the example shown, the capture 802 of the S4 heart sound can be a frequency domain representation 600 ( Figure 6 ). Although modifications to the capture 802 of heart sounds are shown in this example, it should be understood that the example shown is merely an example and the modification process described can be applied to other captures and other heart sounds.
[0044] The capture 802 of the S4 heart sound may be modified to produce a normalized representation 804 of the S4 heart sound. The amount of modification to the capture 802 of the S4 heart sound to produce the normalized representation 804 of the S4 heart sound may be determined based on an amount of attenuation of another heart sound. For example, the capture 702 of the S1 heart sound may be determined based on the attenuation of the other heart sound. Figure 7 ) and control 704 for S1 heart sound ( Figure 7 ) between the attenuation amount. In addition, whether the modification is an amplification of capture 802 or an attenuation of capture 802 can be determined based on whether capture 702 is attenuated relative to control 704 or whether control 704 is attenuated relative to capture 702. For example, if control 704 is captured in a state that results in an attenuated signal, and the attenuation source has decreased or no longer exists when capture 702 is performed, the modification will be the attenuation of capture 802 to maintain a consistent trend. Similarly, if control 704 is captured in a state without attenuation, and if the attenuation source increases when capture 702 is performed, the modification will be an amplification of capture 802 to maintain a consistent trend.
[0045] The modification amount may be frequency dependent, where the modification amount varies based on different frequencies within the capture 802 of the S4 heart sound. In some embodiments, an equation may be applied to the attenuation amount to determine the modification amount for each frequency within the capture 802 of the S4 heart sound. In other embodiments, the attenuation amount may be determined in process 700 ( Figure 7 ) in which an attenuation amount is determined for each frequency within capture 802 of the S4 heart sound, wherein the modification amount may be used to determine the modification amount for each corresponding frequency.
[0046] As can be seen, first peak 806 of capture 802 is modified by first amount of modification 808 to produce first peak 810 of normalized representation 804. Additionally, second peak 812 of capture 802 is modified by second modification amount 814 to produce second peak 816 of normalized representation 804. In the example shown, the attenuation amount 716 ( Figure 7 ) and attenuation 714( Figure 7 ) determines a first modification amount 808 and a second modification amount 814. Based on the frequency difference being modified, the first modification amount 808 can be greater than the second modification amount 814. Specifically, based on the first modification amount 808 modifying a first frequency that is lower than a second frequency modified by the second modification amount 814, the first modification amount 808 can be greater than the second modification amount 814. The amount of modification applied to the capture 802 can have a specific modification amount for each frequency range, where each frequency range can include a single frequency or a frequency range.
[0047] Process 700 may include performing inverse spectral analysis in 710. In particular, inverse spectral analysis may be performed on the normalized representation of the heart sounds generated by the normalization of the heart sounds in 108. The inverse spectral analysis may include performing an inverse Fourier transform on the normalized representation of the heart sounds. The inverse spectral analysis may generate a time domain representation of the normalized representation of the heart sounds. The time domain representation may be a representation of the heart sounds without attenuating the heart sounds that may result in an incorrect analysis of the heart sounds. Therefore, the time domain representation generated using the normalized representation of the heart sounds may be used to analyze the heart sounds of the subject without having to worry about the attenuation of the heart sounds resulting in an incorrect heart sound analysis result.
[0048] Figure 8 An example of the performance of the inverse spectral analysis in 710 is further shown. In particular, a normalized representation 804 of the S4 heart sound is shown, wherein an inverse spectral analysis is performed to produce a time domain representation 818 of the S4 heart sound. The time domain representation 818 can be generated by applying an inverse Fourier transform to the normalized representation 804 of the S4 heart sound. The time domain representation 818 of the S4 heart sound can be used to analyze the captured S4 heart sound without having to worry about attenuation caused by improper application of the sound sensor. Although the inverse spectral analysis is shown to be performed on the capture of the S4 heart sound, it should be understood that the inverse spectral analysis can be performed on the capture of any heart sound, including the S1 heart sound, the S2 heart sound, the S3 heart sound, and the S4 heart sound.
[0049] In some embodiments, modification of the capture of the heart sounds may include determining whether the capture of the heart sounds to be modified is expected to be negatively correlated with the capture of the heart sounds used to determine the attenuation amount. Fig. 9 An example of a modified execution of the capture of negatively correlated heart sounds according to embodiments herein is shown.
[0050] Whether the heart sound being modified and the heart sound being used to determine the attenuation amount are positively or negatively correlated can be determined based on the relationship between each heart sound and the corresponding control. For example, if both heart sounds are approximately (within 5%) equal to the corresponding control, or both heart sounds differ from the corresponding control in the same direction and by approximately (within 5%) the same amount, then the modified heart sound and the heart sound used to determine the attenuation amount can be determined to be positively correlated. If the heart sound used to determine the attenuation amount remains approximately (within 5%) equal to the corresponding control, while the modified heart sound differs from the corresponding control, or the heart sound used to determine the attenuation amount differs from the corresponding control in one direction, while the modified heart sound differs from the corresponding control in the opposite direction, then the modified heart sound and the heart sound used to determine the attenuation amount can be determined to be negatively correlated.
[0051] In the illustrated embodiment, a capture 902 of the S1 heart sound is compared to a control 904 of the S1 heart sound to determine an amount of attenuation. Specifically, a first peak 906 of the control 904 is compared to a first peak 908 of the capture 902. The comparison determines that the first peak 908 of the capture 902 is attenuated below the first peak 906 of the control 904, wherein a first amount of attenuation 914 of the capture 902 is determined based on the comparison. A second peak 910 of the control 904 is compared to a second peak 912 of the capture 902. The comparison determines that the second peak 912 of the capture is attenuated below the second peak 910 of the control 904, wherein a second amount of attenuation 916 of the capture 902 is determined based on the comparison.
[0052] In addition, the capture 918 of the S3 heart sound is compared to the control 920 for the S3 heart sound. The comparison determines that the peak 922 of the capture 918 has increased compared to the peak 924 of the control 920. In particular, the peak 922 of the capture 918 may have an increase 926 relative to the peak 924 of the control 920. Therefore, it can be determined that the capture 918 of the S3 heart sound (i.e., the capture to be modified) has been changed from the control 920 by increasing, while the capture 902 of the S1 heart sound (i.e., the capture used to determine the attenuation amount) has been changed from the control 904 by decreasing, thereby changing the capture 902 in the opposite direction compared to the direction of change of the capture 918. Therefore, in the example shown, the capture 918 of the S3 heart sound and the capture 902 of the S1 heart sound are determined to be negatively correlated.
[0053] In the case where the capture to be modified is determined to be negatively correlated with the capture used to determine the attenuation amount, it can be determined that the change in the capture to be modified is due to a change in coupling, and the capture to be modified can be modified to eliminate the coupling effect. In the case where it is determined that the capture to be modified is not negatively correlated with the capture used to determine the attenuation amount, the change in the capture to be modified can be determined to be due to the change in the capture to be modified alone, rather than due to a change caused by coupling.
[0054] Fig.10 It shows that the embodiment according to this article can be implemented Figure 1 An example system 1000 of method 100 of the present invention is provided. For example, system 1000 can capture heart sounds via method 100 and normalize the captured heart sounds.
[0055] The system 1000 may include one or more sound sensors 1002. The sound sensor 1002 may be placed on a subject and may capture the subject's heart sounds. In particular, the sound sensor 1002 may detect the subject's heart sounds and generate an electrical signal representing the detected heart sounds. In some embodiments, the sound sensor 1002 may include an electronic stethoscope, a piezoelectric sensor, an accelerometer, a microphone, any other sensor that may detect sound and / or vibration and generate an electrical signal representing the detected sound and / or vibration, or some combination thereof.
[0056] The system 1000 may further include a device 1004. The device 1004 may be coupled to the sound sensor 1002 and may receive the captured heart sounds from the sound sensor 1002.
[0057] Device 1004 may include a storage device 1006 in which the captured heart sounds may be stored after being received and processed. In some embodiments, storage device 1006 may further include one or more instructions that, when executed by device 1004, cause the device to perform the operations described herein. In other embodiments, system 1000 may include one or more other computer-readable media having instructions stored thereon.
[0058] The device 1004 may further include a normalization circuit 1008. The normalization circuit 1008 may include a processor 1010. The normalization circuit 1008 may perform the method 100 to normalize the captured heart sounds received from the sound sensor 1002. In particular, the normalization circuit 1008 may retrieve the captured heart sounds from the storage device 1006 and normalize the heart sounds for further processing. In some embodiments, the normalization circuit 1008 may generate a control (e.g., the control 704 ( Figure 7 )), and stores the control in storage device 1006 for normalization of captured heart sounds. One or more captured heart sounds used to generate the control may have been captured during an initialization cycle of system 1000. Normalization circuit 1008 may retrieve the control from storage device 1006 for normalization of heart sounds.
[0059] Device 1004 may perform further processing of the normalized heart sounds. In other embodiments, device 1004 may be coupled to another device (such as a server) and may provide the normalized heart sounds to the other device. The other device may perform further processing on the heart sounds.
[0060] Although in the illustrated embodiment system 1000 includes a sound sensor 1002 coupled to device 1004, it should be understood that the illustrated embodiment is an example embodiment and other embodiments may have other arrangements and / or additional elements. For example, in other embodiments, sound sensor 1002 may be implemented within device 1004.
[0061] Although the captured heart sounds are disclosed as being stored on the storage device 1006 of the system 1000 in the illustrated embodiment, in other embodiments, the captured heart sounds may be stored on one or more separate elements. For example, in other embodiments, the captured heart sounds may be stored on a separate computing device or on one or more servers (e.g., a cloud). The normalization circuit 1008 may obtain the captured heart sounds from the one or more separate elements and perform normalization on the obtained heart sounds. In some embodiments, the device 1004 may include a computing device or one or more servers (e.g., a cloud) located away from the sound sensor 1002, and may receive the captured heart sounds via a network for communication between another device coupled to the sound sensor 1002 and the device 1004.
[0062] Fig.11 An example method 1100 for initialization and normalization by a heart sound capture system according to embodiments herein is shown. For example, the method 1100 may be performed by the system 1000.
[0063] The method 1100 may be initiated in an initialization cycle 1102. The initialization cycle 1102 may be performed in the presence of a care provider (e.g., a physician) to ensure that the system's sound sensors are properly placed on the subject during the initialization cycle 1102. In some embodiments, the system implementing the method 1100 may require authorization by a user who is a care provider to enter the initialization cycle 1102, such as embodiments where controls are generated based on the heart sounds captured in the initialization cycle 1102.
[0064] Initialization cycle 1102 may be initiated by capturing one or more heart sounds at 1104. For example, one or more sound sensors (e.g., sound sensor 1002 ( Fig.10 )) can capture the heart sounds of the subject, and the device (e.g., device 1004 ( Fig.10)) The captured heart sounds may be stored. In some embodiments, the capture of one or more heart sounds in 1104 may be omitted. In these embodiments, method 1100 may begin at 1112.
[0065] At 1106, a control may be generated from the one or more heart sounds captured at 1104. In some embodiments, a single heart sound from the one or more heart sounds may be stored as a control. In other embodiments, certain portions of the one or more heart sounds may be averaged to produce a representation of the heart sound that may be stored as a control. At 1108, the control may be stored in a storage device (such as storage device 1006 ( Fig.10 ))middle.
[0066] The method 1100 may proceed with a test cycle 1110. The test cycle 1110 may be initiated by capturing one or more heart sounds at 1112. For example, one or more sound sensors may capture the heart sounds of the subject, and the device may store the captured heart sounds for analysis. In some embodiments, the one or more heart sounds captured at 1112 may be used to generate a control at 1106.
[0067] In 1114, the heart sounds captured in 1112 may be normalized. In particular, the heart sounds may be normalized by performing method 100 ( Figure 1 ) to normalize the heart sounds. The control generated in 1106 can be used to normalize the heart sounds. The normalized heart sounds can be stored for further processing.
[0068] Example Implementation
[0069] The following examples are provided by way of example.
[0070] Example 1 may include one or more computer-readable media having instructions stored thereon, wherein the instructions, in response to execution by a device, cause the device to determine an attenuation amount between capture of a first heart sound and control of the first heart sound, capture of the first heart sound captured during a test cycle, modify a first portion of capture of a second heart sound by a first amount, wherein capture of the second heart sound was captured during the test cycle, wherein the first portion corresponds to a first frequency range, and wherein the first amount is determined based on the attenuation amount, and modify a second portion of capture of the second heart sound by a second amount, wherein the second portion corresponds to a second frequency range, and wherein the second amount is determined based on the attenuation amount.
[0071] Example 2 may include one or more computer-readable media of Example 1, wherein the instructions, in response to execution by a device, further cause the device to convert the capture of the first heart sound into a frequency domain representation of the capture of the first heart sound, and compare the frequency domain representation of the capture of the first heart sound with a control of the first heart sound to determine the attenuation amount.
[0072] Example 3 may include one or more computer-readable media of Example 1, wherein the instructions, in response to execution of the device, further cause the device to convert the capture of the second heart sound into a frequency domain representation of the capture of the second heart sound, wherein modifying a first portion of the capture of the second heart sound includes identifying a first portion of the captured frequency domain representation of the capture of the second heart sound, wherein the first portion corresponds to a first frequency range, and modifying the first portion of the captured frequency domain representation of the capture of the second heart sound by a first amount, and modifying a second portion of the captured frequency domain representation of the capture of the second heart sound includes identifying a second portion of the captured frequency domain representation of the capture of the second heart sound, wherein the second portion corresponds to the second frequency range, and modifying the second portion of the captured frequency domain representation of the capture of the second heart sound by a second amount, and the instructions, in response to execution of the device, further cause the device to convert the frequency domain representation of the capture of the second heart sound into a time domain representation of the capture of the second heart sound.
[0073] Example 4 may include one or more computer-readable media of Example 1, wherein determining the attenuation amount includes determining an attenuation amount of a capture of a first heart sound corresponding to the first frequency range, wherein the first amount is determined based on the attenuation of the capture of the first heart sound corresponding to the first frequency range, and determining an attenuation amount of the capture of the first heart sound corresponding to the second frequency range, wherein the second amount is determined based on the attenuation of the capture of the first heart sound corresponding to the second frequency range.
[0074] Example 5 may include one or more computer-readable media of Example 1, wherein the instructions, in response to execution by a device, further cause the device to receive an indication to use an S1 heart sound or an S2 heart sound as the first heart sound, wherein based on the indication, the first heart sound is an S1 heart sound or an S2 heart sound.
[0075] Example 6 may include one or more computer-readable media of Example 1, wherein the capture of the first heart sound is a first capture of the first heart sound, wherein the instructions, in response to execution by the device, further cause the device to capture a second capture of the first heart sound in an initialization cycle, and wherein the initialization cycle occurs before the test cycle, and the second capture of the first heart sound is stored as a control of the first heart sound.
[0076] Example 7 may include one or more computer-readable media of Example 1, wherein the instructions, in response to execution by a device, further cause the device to capture multiple captures of the first heart sound in multiple test cycles, wherein the multiple test cycles occur before the test cycle, generate an average representation of the first heart sound by averaging the multiple captures of the first heart sound, and store the average representation of the first heart sound as a control of the first heart sound.
[0077] Example 8 may include the one or more computer-readable media of Example 1, wherein the capture of the first heart sound is captured within a cardiac cycle, and wherein the capture of the second heart sound is captured within the cardiac cycle.
[0078] Example 9 may include a system for capturing heart sounds of a subject, comprising: one or more sound sensors to sense the heart sounds of the subject, and generate an electronic representation of the heart sounds, and one or more processors to identify capture of a first heart sound from the electronic representation, compare the capture of the first heart sound with control of the first heart sound to determine an attenuation amount between the capture of the first heart sound and the control of the first heart sound, determine a first modification amount and a second modification amount based on the attenuation amount, identify capture of a second heart sound from the electronic representation, modify a first portion of the capture of the second heart sound by the first modification amount, wherein the captured first portion corresponds to a first frequency range, and modify a second portion of the capture of the second heart sound by a second modification amount, wherein the captured second portion corresponds to a second frequency range.
[0079] Example 10 may include the system of Example 9, wherein identifying capture of the first heart sound comprises: identifying a cardiac cycle from the electronic representation, and identifying capture of the first heart sound within the cardiac cycle, and identifying capture of the second heart sound comprises identifying capture of the second heart sound within the cardiac cycle.
[0080] Example 11 may include the system of Example 9, wherein comparing the capture of the first heart sound to a control includes converting the capture of the first heart sound to a frequency domain representation of the capture of the first heart sound, and comparing the frequency domain representation of the capture of the first heart sound to the control of the first heart sound.
[0081] Example 12 may include the system of Example 9, wherein determining a first modification amount and a second modification amount includes determining the first modification amount and the second modification amount using an attenuation amount in a frequency-dependent equation, wherein the first modification amount and the second modification amount are different based on a first modification amount for a first frequency range and a second modification amount for a second frequency range.
[0082] Example 13 may include the system of Example 9, wherein the one or more processors are further used to convert the capture of the second heart sound into a frequency domain representation of the capture of the second heart sound, modifying a first portion of the capture of the second heart sound includes identifying a first portion of the frequency domain representation, wherein the first portion of the frequency domain representation corresponds to the first frequency range, and modifying the first portion of the frequency domain representation by a first modification amount, modifying a second portion of the capture of the second heart sound includes identifying a second portion of the frequency domain representation, wherein the second portion of the frequency domain representation corresponds to the second frequency range, and modifying the second portion of the frequency domain representation by a second modification amount, and the one or more processors are further used to convert the frequency domain representation into a time domain representation of the capture of the second heart sound.
[0083] Example 14 may include the system of Example 9, wherein the heart sound of the subject is a first heart sound of the subject, the capture of the first heart sound is a first capture of the first heart sound, the one or more sound sensors are used to sense a second heart sound of the subject during an initialization cycle, generate an electronic representation of the second heart sound, and the one or more processors are used to identify the second capture of the first heart sound from the electronic representation of the second heart sound, and store the second capture of the first heart sound as a control of the first heart sound.
[0084] Example 15 may include the system of Example 9, wherein the one or more processors are further used to receive an indication to use the S1 heart sound or the S2 heart sound as the first heart sound, wherein identifying capture of the first heart sound includes identifying capture of the S1 heart sound or the S2 heart sound from the electronic representation based on the indication.
[0085] Example 16 may include a method for normalizing heart sounds, comprising: comparing the capture of a first heart sound of a test cycle with the control of the first heart sound, and determining the attenuation amount between the capture of the first heart sound and the control of the first heart sound based on the comparison of the capture of the first heart sound with the control of the first heart sound; modifying a first portion of the capture of a second heart sound of the test cycle by a first amount, wherein the first portion corresponds to a first frequency range, and wherein the first amount is determined based on the attenuation amount, and modifying a second portion of the capture of the second heart sound of the test cycle by a second amount, wherein the second portion corresponds to a second frequency range, and wherein the second amount is determined based on the attenuation amount.
[0086] Example 17 may include the method of Example 16, further including: converting the capture of the second heart sound into a frequency domain representation of the capture of the second heart sound, wherein modifying a first portion of the capture of the second heart sound includes identifying a first portion of the captured frequency domain representation of the capture of the second heart sound, wherein the first portion corresponds to a first frequency range, and modifying the first portion of the captured frequency domain representation of the capture of the second heart sound by a first amount, and modifying a second portion of the capture of the second heart sound includes identifying a second portion of the captured frequency domain representation of the capture of the second heart sound, wherein the second portion corresponds to the second frequency range, and modifying the second portion of the captured frequency domain representation of the capture of the second heart sound by a second amount, and the method also includes converting the captured frequency domain representation of the second heart sound into a captured time domain representation of the second heart sound.
[0087] Example 18 may include the method of Example 16, wherein determining the attenuation amount includes determining the attenuation amount between capture of the first heart sound and control of a first heart sound corresponding to a first frequency range, wherein the first amount is determined based on the attenuation between capture of the first heart sound and control of the first heart sound corresponding to the first frequency range, and determining the attenuation amount between capture of the first heart sound and control of the first heart sound corresponding to a second frequency range, wherein the second amount is determined based on the attenuation between capture of the first heart sound and control of the first heart sound corresponding to the second frequency range.
[0088] Example 19 may include the method of Example 16, wherein the capture of the first heart sound is a first capture of the first heart sound, wherein the method further includes capturing a second capture of the first heart sound in an initialization cycle and storing the second capture as the control, and wherein the initialization cycle occurs before the test cycle.
[0089] Example 20 may include the method of Example 16, further comprising capturing multiple captures of the first heart sound in multiple test cycles, wherein the multiple test cycles occur before the test cycle, and generating an average representation of the first heart sound by averaging the multiple captures of the first heart sound, wherein the control of the first heart sound is the average representation of the first heart sound.
[0090] The foregoing summarizes the features of one or more embodiments of the subject matter disclosed herein. These embodiments are provided to enable a person of ordinary skill in the art (PHOSITA) to better understand the various aspects of the present disclosure. Without describing in detail, certain easily understood terms and underlying technologies and / or standards may be referenced. It is expected that PHOSITA will have or have access to background knowledge or information in those technologies and standards sufficient to practice the teachings of this specification.
[0091] PHOSITA will appreciate that they can easily use the present disclosure as a basis for designing or modifying other processes, structures or variations to achieve the same purposes and / or achieve the same advantages as the embodiments described herein. PHOSITA will also appreciate that such equivalent constructions do not depart from the spirit and scope of the present disclosure, and that they can make various changes, substitutions and modifications without departing from the spirit and scope of the present disclosure.
[0092] In some cases, the teachings of this specification may be encoded into one or more tangible, non-transitory computer-readable media having executable instructions stored thereon that, when executed, instruct a programmable device (e.g., a processor or DSP) to perform the methods or functions disclosed herein. Where the teachings herein are at least partially embodied in a hardware device (e.g., an ASIC, an IP block, or an SoC), the non-transitory medium may include a hardware device that is programmed with logic to perform the methods or functions disclosed herein. The teachings may also be practiced in the form of register transfer level (RTL) or other hardware description languages (e.g., VHDL or Verilog), which may be used to program a manufacturing process to produce the disclosed hardware elements.
[0093] In an example embodiment, at least some parts of the processing activities outlined herein may also be implemented in software. In some embodiments, one or more of these features may be implemented in hardware provided outside the element of the disclosed accompanying drawings, or may be merged in any appropriate manner to realize the expected function. Various components may include software (or reciprocating software) that may be coordinated to realize the operation outlined herein. In other embodiments, these elements may include any suitable algorithm, hardware, software, component, module, interface or object that promotes its operation.
[0094] Any appropriately configured processor component can execute any type of instruction associated with data to implement the operations detailed herein. Any processor disclosed herein can convert an element or article (e.g., data) from one state or thing to another state or thing. In another example, some activities outlined herein can be implemented with fixed logic or programmable logic (e.g., software and / or computer instructions executed by a processor), and the element identified herein can be a certain type of programmable processor, programmable digital logic (e.g., FPGA, erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM)), including digital logic, software, code, electronic instructions, flash memory, optical disk, CD-ROM, DVD ROM, magnetic card or optical card ASIC, other types of machine-readable media suitable for storing electronic instructions, or any suitable combination thereof. In operation, the processor can store information in any suitable type of non-temporary storage medium (e.g., random access memory (RAM), read-only memory (ROM), FPGA, EPROM, electrically erasable programmable ROM (EEPROM), etc.), software, hardware, or any other suitable component, device, element or object, as appropriate and according to specific needs. Furthermore, information being tracked, sent, received, or stored in the processor may be provided in any database, register, table, cache, queue, control list, or storage structure, all of which may be referenced in any suitable time frame, depending on specific needs and implementations. Any storage item discussed herein should be interpreted as being included in the broad sense of "memory." Similarly, any potential processing elements, modules, and machines described herein should be interpreted as being included within the broad terms "microprocessor" or "processor." Furthermore, in various embodiments, the processor, memory, network card, bus, storage device, related peripheral devices, and other hardware elements described herein may be implemented by processors, memories, and other related devices that are configured by software or firmware to emulate or virtualize the functionality of those hardware elements.
[0095] The computer program logic that implements all or part of the functions described herein is embodied in various forms, including but not limited to source code form, computer executable form, hardware description form, and various intermediate forms (e.g., mask works, or forms generated by assemblers, compilers, linkers, or locators). In an example, the source code includes a series of computer program instructions implemented in various programming languages (e.g., object code, assembly language) or high-level languages (e.g., OpenCL, RTL, Verilog, VHDL, Fortran, C, C++, JAVA, or HTML for various operating systems or operating environments). The source code can define and use various data structures and communication messages. The source code can be in a computer executable form (e.g., via an interpreter), or the source code can be converted (e.g., via a translator, assembler, or compiler) into a computer executable form.
Claims
1. One or more computer-readable media having instructions stored thereon, wherein: The instructions, in response to execution by a device, cause the device to: determining an attenuation amount between capture of a first heart sound and control of said first heart sound, the capture of the first heart sound captured during a test cycle; modifying a first portion of a capture of a second heart sound by a first amount, wherein the capture of the second heart sound is captured during a test cycle, wherein the first portion corresponds to a first frequency range, and wherein the first amount is determined based on the attenuation amount; and A second portion of the capture of the second heart sound is modified by a second amount, wherein the second portion corresponds to a second frequency range, and wherein the second amount is determined based on the attenuation amount.
2. The one or more computer-readable media of claim 1, wherein the instructions, in response to execution by a device, further cause the device to: converting the capture of the first heart sound into a frequency domain representation of the capture of the first heart sound; and The captured frequency domain representation of the first heart sound is compared to a control of the first heart sound to determine the amount of attenuation.
3. The one or more computer-readable media of claim 1, wherein the instructions, in response to execution by a device, further cause the device to convert the capture of the second heart sound into a frequency domain representation of the capture of the second heart sound, wherein: Modifying the first part of the capture of the second heart sound includes: identifying a first portion of the captured frequency domain representation of the second heart sound, wherein the first portion corresponds to a first frequency range; and modifying a first portion of the captured frequency domain representation of the second heart sound by a first amount; and Modifying the second part of the capture of the second heart sound includes: identifying a captured second portion of the captured frequency domain representation of the second heart sound, wherein the second portion corresponds to the second frequency range; and modifying a second portion of the captured frequency domain representation of the second heart sound by a second amount; and The instructions, in response to execution by the device, further cause the device to convert the captured frequency domain representation of the second heart sound into a captured time domain representation of the second heart sound.
4. The one or more computer-readable media of claim 1 , wherein determining the attenuation amount comprises: determining an attenuation of capture of a first heart sound corresponding to the first frequency range, wherein determining the first amount based on an attenuation of a capture of a first heart sound corresponding to the first frequency range; and An amount of attenuation of the capture of the first heart sound corresponding to the second frequency range is determined, wherein the second amount is determined based on the attenuation of the capture of the first heart sound corresponding to the second frequency range.
5. The one or more computer-readable media of claim 1, wherein the instructions, in response to execution by a device, further cause the device to receive an indication to use an S1 heart sound or an S2 heart sound as the first heart sound, wherein based on the indication, the first heart sound is the S1 heart sound or the S2 heart sound.
6. The one or more computer-readable media of claim 1, wherein the capturing of the first heart sound is a first capturing of the first heart sound, wherein the instructions, in response to execution by a device, further cause the device to: capturing a second capture of the first heart sound during an initialization period, and wherein the initialization period occurs before the test period; and A second capture of the first heart sound is stored as a control for the first heart sound.
7. The one or more computer-readable media of claim 1, wherein the instructions, in response to execution by a device, further cause the device to: capturing a plurality of captures of the first heart sound in a plurality of test cycles, wherein the plurality of test cycles occur prior to the test cycle; generating an average representation of the first heart sound by averaging multiple captures of the first heart sound; and The average representation of the first heart sound is stored as a control for the first heart sound.
8. The one or more computer-readable media of claim 1, wherein the capture of the first heart sound is captured within a cardiac cycle, and wherein the capture of the second heart sound is captured within the cardiac cycle.
9. A system for capturing heart sounds of a subject, comprising: One or more sound sensors to: sensing heart sounds of the subject; and generating an electronic representation of said heart sounds; and One or more processors to: identifying capture of a first heart sound from the electronic representation; comparing the capture of the first heart sound to the control of the first heart sound to determine an amount of attenuation between the capture of the first heart sound and the control of the first heart sound; determining a first modification amount and a second modification amount based on the attenuation amount; identifying capture of a second heart sound from the electronic representation; modifying a captured first portion of the second heart sound by the first modification amount, wherein the captured first portion corresponds to a first frequency range; and A captured second portion of the second heart sound is modified by the second modification amount, wherein the captured second portion corresponds to a second frequency range.
10. The system of claim 9, wherein: Identifying the capture of the first heart sound includes: identifying a cardiac cycle from the electronic representation; and identifying capture of the first heart sound within the cardiac cycle; and Identifying capture of the second heart sound includes identifying capture of the second heart sound within the cardiac cycle.
11. The system of claim 9, wherein comparing the capture of the first heart sound to a control comprises: converting the capture of the first heart sound into a frequency domain representation of the capture of the first heart sound; and The captured frequency domain representation of the first heart sound is compared to a control of the first heart sound.
12. The system of claim 9, wherein determining a first modifier and a second modifier comprises determining the first modifier and the second modifier using an attenuation in a frequency-dependent equation, wherein the first modifier and the second modifier are different based on a first modifier for a first frequency range and a second modifier for a second frequency range.
13. The system of claim 9, wherein: The one or more processors are further configured to convert the capture of the second heart sound into a frequency domain representation of the capture of the second heart sound; Modifying the first part of the capture of the second heart sound includes: identifying a first portion of the frequency domain representation, wherein the first portion of the frequency domain representation corresponds to the first frequency range; and modifying a first portion of the frequency domain representation by a first modification amount; Modifying the second part of the capture of the second heart sound includes: identifying a second portion of the frequency domain representation, wherein the second portion of the frequency domain representation corresponds to the second frequency range; and modifying a second portion of the frequency domain representation by a second modification amount; and The one or more processors are further configured to convert the frequency domain representation to a time domain representation of the capture of the second heart sound.
14. The system of claim 9, wherein: the subject's heart sound is a first heart sound of the subject; The capturing of the first heart sound is a first capturing of the first heart sound; The one or more sound sensors are used to: sensing a second heart sound of the subject during an initialization cycle; generating an electronic representation of the second heart sound; and The one or more processors are to: identifying a second capture of the first heart sound from the electronic representation of the second heart sound; and A second capture of the first heart sound is stored as a control for the first heart sound.
15. The system of claim 9, wherein the one or more processors are further configured to receive an indication to use an S1 heart sound or an S2 heart sound as the first heart sound, wherein identifying capture of the first heart sound comprises identifying capture of the S1 heart sound or the S2 heart sound from the electronic representation based on the indication.
16. A method for normalizing heart sounds, comprising: comparing the capture of a first heart sound of the test cycle to a control of the first heart sound; determining an amount of attenuation of the capture of the first heart sound based on a comparison of the capture of the first heart sound and the control of the first heart sound; modifying a first portion of a capture of a second heart sound of the test cycle by a first amount, wherein the first portion corresponds to a first frequency range, and wherein the first amount is determined based on an amount of attenuation; and A second portion of the capture of a second heart sound of the test cycle is modified by a second amount, wherein the second portion corresponds to a second frequency range, and wherein the second amount is determined based on the amount of attenuation.
17. The method of claim 16, further comprising converting the capture of the second heart sound into a frequency domain representation of the capture of the second heart sound, wherein: Modifying the first part of the capture of the second heart sound includes: identifying a first portion of the captured frequency domain representation of the second heart sound, wherein the first portion corresponds to a first frequency range; and modifying a first portion of the captured frequency domain representation of the second heart sound by a first amount; and Modifying the second part of the capture of the second heart sound includes: identifying a captured second portion of the captured frequency domain representation of the second heart sound, wherein the second portion corresponds to the second frequency range; and modifying a second portion of the captured frequency domain representation of the second heart sound by a second amount; and The method also includes converting the captured frequency domain representation of the second heart sound into a captured time domain representation of the second heart sound.
18. The method of claim 16, wherein determining the amount of attenuation comprises: determining an amount of attenuation between capture of the first heart sound and control of the first heart sound corresponding to a first frequency range, wherein the first amount is determined based on the attenuation between capture of the first heart sound and control of the first heart sound corresponding to the first frequency range; and An attenuation amount between capture of the first heart sound and control of the first heart sound corresponding to a second frequency range is determined, wherein a second amount is determined based on the attenuation between capture of the first heart sound and control of the first heart sound corresponding to the second frequency range.
19. The method of claim 16, wherein the capture of the first heart sound is a first capture of the first heart sound, wherein the method further comprises capturing a second capture of the first heart sound in an initialization cycle and storing the second capture as the control, and wherein the initialization cycle occurs before the test cycle.
20. The method of claim 16, further comprising: capturing a plurality of captures of the first heart sound in a plurality of test cycles, wherein the plurality of test cycles occur prior to the test cycle; and An average representation of the first heart sound is generated by averaging a plurality of captures of the first heart sound, wherein the control of the first heart sound is the average representation of the first heart sound.
