Selectable heart sound tracking in heart failure

Abstract

Systems and methods for identifying and tracking heart sound components are disclosed. An exemplary medical device system includes a data receiver that receives heart sound information and a heart sound recognition circuit that generates a representative heart sound segment, such as an entirety average of selected portions of the heart sound segment. The heart sound identification circuit identifies heart sound components from the representative heart sound segments using a time-based tracking algorithm, and assesses a performance index of the time-based heart sound tracking algorithm. Based on the performance index, a decision may be made whether to switch to a spectrum-based tracking algorithm to identify a heart sound component from a representative heart sound segment. A physiological event detector may detect a cardiac event using the identified heart sound components.

Description

[0001] Priority requirements

[0002] This application claims the benefit of U.S. Provisional Application No. 63 / 599,867, filed November 16, 2023, which is incorporated herein by reference in its entirety. Technical Field

[0003] This document generally pertains to medical systems, and more specifically, to systems, devices, and methods for identifying and tracking the heart sounds of subjects. Background Technology

[0004] Heart sounds are typically associated with the mechanical vibrations of the heart and the flow of blood through it. Heart sounds repeat with each cardiac cycle and are separated and categorized based on the activity associated with the vibrations. Historically, heart sounds were assessed by humans, and therefore only the audible portion of the vibrations was used. Devices can now assess the full spectrum of cardiac vibrations, including both audible and sub-audible components; therefore, the term "sound" in this document refers to the full spectrum of vibrations. Typically, heart sounds sensed by a subject can include several components within the cardiac cycle, including the first heart sound (S1), the second heart sound (S2), the third heart sound (S3), or the fourth heart sound (S4). S1 is associated with the vibrations produced by the heart during mitral valve tension. S2 is produced by the closure of the aortic and pulmonary valves and marks the beginning of diastole. S3 is produced by the early diastolic vibrations corresponding to the passive filling of the ventricles during diastole (when blood rushes into the ventricles). S4 is produced by the late diastolic vibrations corresponding to the active filling of the ventricles during atrial contraction, which pushes blood into the ventricles. In healthy subjects, S3 is usually faint, and S4 is rarely heard. However, pathological S3 or S4 may be higher in pitch and louder.

[0005] Heart sounds have been used to assess cardiac systolic and diastolic function. A systolic contraction is the contraction of the heart, or a period of contraction, that forces blood out of the heart, such as into the ventricles, and into the aorta and pulmonary arteries. A diastolic contraction is the relaxation of the heart, or a period of relaxation, during which blood flows back into the heart, such as into the ventricles. Patients with heart disease may have impaired systolic or diastolic function. For example, congestive heart failure (CHF) occurs when the heart is unable to supply enough blood to maintain a healthy physiological state.

[0006] Implantable medical devices (IMDs) have been used to monitor patients with heart disease, such as detecting cardiac events that lead to worsening heart failure (WHF). IMDs can sense physiological signals from patients and deliver electrical stimulation therapy to improve cardiac performance in patients with CHF. Frequent patient monitoring via IMDs can help identify patients at increased risk of developing future heart failure events, ensuring timely treatment, reducing heart failure hospitalizations, improving patient outcomes, and lowering healthcare costs. Summary of the Invention

[0007] Ambulatory medical devices (AMDs), such as implantable medical devices (IMDs), subcutaneous medical devices, wearable medical devices, or other external medical devices, can be used to monitor patients with heart disease. AMDs can sense the electrical or mechanical activity of the heart via sensing electrodes and / or physiological sensors and detect cardiac events such as cardiac arrhythmias or heart-wrenching heart failure (WHF). IMDs may include a pulse generator capable of generating and delivering electrical stimulation therapy to the heart or other excitable tissues (e.g., neural targets) to restore or improve cardiac performance in patients with heart failure (CHF) or to correct cardiac arrhythmias. For example, detection of cardiac arrhythmias can trigger cardiac pacing or an electric shock, or detection of WHF events can trigger electrical stimulation therapy, such as resynchronization therapy (CRT), to correct cardiac asynchrony in patients with heart failure.

[0008] AMD can use heart sounds detected from a patient to detect cardiac events. For example, S1 and / or S2 can be used to detect cardiac arrhythmias such as supraventricular tachycardia or ventricular tachycardia. Fluid buildup in the lungs of patients with CHF may lead to elevated ventricular filling pressure and diastolic dysfunction, resulting in a pathologically louder S3. Strong atrial contractions in CHF patients overcoming abnormally stiff ventricles may produce a strong S4. Therefore, monitoring S3 or S4 may help identify diastolic dysfunction, detect WHF events, or assess a patient's risk of developing future WHF.

[0009] Mobile heart sound detection involves placing a heart sound sensor at or near the heart in a location on the skin, subcutaneous, submuscular, intramuscular, or substernal. The heart sound sensor, such as an accelerometer, may be included within an IMD for implantation or associated with an implantable lead for epicardial or endocardial placement. S1 and S2 heart sounds typically have frequencies in the range of approximately 10 Hz–250 Hz. Due to wide interpersonal variability and minor displacement of the technology device, S2 typically has a higher frequency than S1. For example, most of the power of S1 falls within approximately 10 Hz–50 Hz, while most of S2 falls within approximately 20 Hz–70 Hz. The early diastolic sound S3, produced by the rapid filling of the dilated ventricles, and the late diastolic sound S4, produced by the contraction of the left atrium against the non-compliant left ventricle, when present, typically have lower intensity and lower frequency.

[0010] Time-domain, amplitude-based methods can be used to identify heart sound components (e.g., S1, S2, S3, or S4). These methods involve detecting the peak amplitude or its variation within a heart sound detection window that defines a time interval for detecting the heart sound component of interest. This peak amplitude, such as peak signal power or root mean square (RMS) value, is then identified as the timing location of the heart sound component of interest. However, multiple peaks may often exist within a single heart sound window, and the dominant peak (the one with the largest amplitude) may not always represent the actual heart sound component. Therefore, time-domain, amplitude-based methods may sometimes incorrectly identify spurious peaks as target heart sound components. These methods can also be sensitive to electromagnetic or physiological noise or interference. At low signal-to-noise ratios (SNR), the timing information of heart sound components (such as S1 or S2) may not be accurately determined. This can lead to further errors in the detection of other heart sound components (e.g., S3 or S4) measured with reference to S1 or S2, or in detection based on cardiac time intervals such as pre-ejection period (PEP, the interval between the start of QRS and S1), systolic timing interval (STI, the interval between the start of QRS and S2), left ventricular ejection time (LVET, the interval between S1 and S2), or diastolic timing interval (DTI, the interval between S2 and the start of QRS in the next cardiac cycle)). Detection performance may be degraded when heart sound timing and / or cardiac time intervals are used to detect cardiac events (e.g., WHF events). Other factors, such as inaccurate heart rate estimation (e.g., during atrial arrhythmias such as atrial fibrillation), flipping of implantable medical devices, or misalignment or torsion of the sensing lead system, can also lead to errors in time-domain, amplitude-based heart sound recognition.

[0011] Alternative methods for detecting and tracking heart sound components are based on the frequency or spectral characteristics of the heart sound signal. Measures of complexity, such as Shannon entropy or other entropy measures, can be obtained from the power spectrum of the heart sound signal (or a portion thereof) and are used to identify the heart sound component of interest, as described in U.S. Patent Application No. 63 / 432,649, entitled “ENTROPY BASED HEART SOUND TRACKING” by Goftari et al. This spectral entropy-based approach quantifies the complexity (or inhomogeneity) of the power spectrum amplitude of the heart sound signal or a portion thereof (such as within the cardiac cycle). Spectral entropy-based heart sound recognition combines the advantages of frequency domain spectral analysis and entropy measures, both of which are less susceptible to noise and interference from various sources compared to time-domain, amplitude-based heart sound detection methods. In particular, for example, spectral entropy-based methods can be less affected by occasional high-amplitude time-domain noise (a major cause of false detections). However, spectral entropy-based methods are generally more computationally intensive than time-domain, amplitude-based recognition methods and may require more computational resources and consume more power from implantable devices.

[0012] The inventors have recognized an unmet need for devices and methods for more robust heart sound recognition and tracking while keeping power consumption and computational complexity under control. Heart sound recognition technologies like these can improve the accuracy and efficiency of detecting various cardiac events, such as various cardiac arrhythmias or worsening heart failure. This document discusses systems, devices, and methods for evaluating the quality of various heart sound recognition algorithms and identifying appropriate algorithms for use in detecting cardiac events. An exemplary medical device system includes a data receiver circuitry that receives heart sound information and a heart sound recognition circuitry that generates representative heart sound segments within a cardiac cycle (such as an ensemble average of selected portions of heart sound segments from multiple cardiac cycles). The heart sound recognition circuitry can use a time-based tracking algorithm to identify heart sound components of interest from the representative heart sound segments and evaluate the performance index of the time-based heart sound tracking algorithm based on morphological or timing information of the identified heart sound components. Based on the performance index, the heart sound recognition circuitry can determine whether to switch to a spectrum-based tracking algorithm to further identify heart sound components. The heart sound components thus identified can be used to detect cardiac events, such as worsening heart failure (WHF) events.

[0013] Example 1 is a medical device system comprising: a data receiver circuit configured to receive heart sound information over multiple cardiac cycles; and a heart sound recognition circuit configured to: generate representative heart sound segments within the cardiac cycle using at least a portion of the received heart sound information; identify heart sound components from the representative heart sound segments using a time-based heart sound tracking algorithm; evaluate the performance index of the time-based heart sound tracking algorithm using morphological or timing information of the identified heart sound components; and determine, based on the evaluated performance index, whether to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from the representative heart sound segments.

[0014] In Example 2, the subject of Example 1 may optionally include a time-based heart sound tracking algorithm that can use amplitude and timing information of representative heart sound segments to identify heart sound components.

[0015] In Example 3, any one or more of the topics in Examples 1 to 2 may optionally include a spectrum-based heart sound tracking algorithm that can identify heart sound components using one or more spectral entropy values ​​calculated from representative heart sound segments.

[0016] In Example 4, the subject of any one or more of Examples 1 to 3 may optionally include a portion of the received heart sound information, which may include multiple heart sound segments corresponding to the corresponding heartbeat, wherein the heart sound recognition circuit is configured to generate a representative heart sound segment using the overall average of a subset of at least multiple heart sound segments.

[0017] In Example 5, the subject of Example 4 may optionally include a data receiver circuit that can be configured to receive information about the instantaneous heart rate of heartbeats corresponding to a plurality of heart sounds, wherein a subset of at least a plurality of heart sounds used to generate representative heart sounds corresponds to heartbeats falling within a predetermined heart rate range.

[0018] In Example 6, the subject matter of Example 5 may optionally include, in order to generate representative heart sounds, a heart sound recognition circuit is configured to: sort a plurality of heart sounds according to a specific order of the instantaneous heart rates of heartbeats corresponding to the plurality of heart sounds; identify from the sorted plurality of heart sounds a first group of heart sounds corresponding to at least a first heartbeat and a second group of heart sounds corresponding to a second heartbeat having a different heart rate than the first heartbeat; determine a first morphological similarity measure of the first group of heart sounds and a second morphological similarity measure of the second group of heart sounds; select a group between the first group and the second group based on one or more of the first morphological similarity measure or the second morphological similarity measure; and generate representative heart sounds using the overall average of the heart sounds of the selected group.

[0019] In Example 7, the subject of Example 6 may optionally include a first group and a second group, each having at least a specific minimum number of heart segments.

[0020] In Example 8, the subject matter of any one or more of Examples 6 to 7 may optionally include a first heartbeat corresponding to a first set of heart sounds and a second heartbeat corresponding to a second set of heart sounds, each having a corresponding instantaneous heart rate falling below a heart rate threshold.

[0021] In Example 9, the subject matter of any one or more of Examples 6 to 8 may optionally include a first heartbeat corresponding to a first set of heart sounds and a second heartbeat corresponding to a second set of heart sounds, each having a corresponding instantaneous heart rate that satisfies the requirements of heart rate variability or range.

[0022] In Example 10, the subject of any one or more of Examples 6 to 9 may optionally include multiple heart sounds, which can be sorted in ascending order of the instantaneous heart rate of the heartbeat, wherein a second heartbeat corresponding to a second group of heart sounds has a higher heart rate than a first heartbeat corresponding to a first group of heart sounds, wherein the selection between the first group and the second group includes: selecting the first group if (i) a first morphological similarity measure is higher than a first threshold, or (ii) both the first morphological similarity measure and the second morphological similarity measure are lower than their respective thresholds, and the first morphological similarity measure is greater than the second morphological similarity measure; and selecting the second group if (i) the first morphological similarity measure is lower than the first threshold and the second morphological similarity measure is higher than the second threshold, or (ii) both the first morphological similarity measure and the second morphological similarity measure are lower than their respective thresholds and the second morphological similarity measure is greater than the first morphological similarity measure.

[0023] In Example 11, any one or more of the subjects in Examples 4 to 10 may optionally include a heart sound recognition circuit, which can be configured to evaluate the performance index of a time-based heart sound tracking algorithm using a morphological similarity measure among subsets of at least a plurality of heart sound segments used to generate representative heart sound segments.

[0024] In Example 12, the subject matter of Example 11 may optionally include, wherein evaluating the performance index of a time-based heart sound tracking algorithm includes further using heart sound tracking stability, which indicates the variability in timing of heart sound components correspondingly determined from a subset of at least a plurality of heart sound segments used to generate representative heart sound segments, wherein, in order to determine whether to switch to a spectrum-based heart sound tracking algorithm, the heart sound recognition circuit is configured to: determine not to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments if (i) the morphological similarity metric exceeds a similarity threshold and (ii) the heart sound tracking stability is below a variability threshold; and determine to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments if (i) the morphological similarity metric is not greater than a similarity threshold or (ii) the heart sound tracking stability exceeds a variability threshold.

[0025] In Example 13, the subject matter of Example 12 may optionally include heart sound tracking stability, which can indicate the timing variability of the S2 component, wherein the heart sound recognition circuit is configured to determine an AS index indicating the presence or severity of aortic stenosis (AS) for each of a subset of at least a plurality of heart sound segments used to generate representative heart sound segments, and to use the variability of the determined AS index to determine heart sound tracking stability.

[0026] In Example 14, the subject matter of any one or more of Examples 4 to 13 may optionally include, wherein, in order to evaluate the performance index of the time-based heart sound tracking algorithm, the heart sound recognition circuit is configured to: generate a heart sound timing trend, the heart sound timing trend comprising a time series of timings of heart sound components identified from a subset of at least a plurality of heart sound segments using the time-based heart sound tracking algorithm; generate a heart rate trend, the heart rate trend comprising a time series of instantaneous heart rates of heartbeats corresponding to a subset of at least a plurality of heart sound segments; and evaluate the performance index of the time-based heart sound tracking algorithm using a consistency metric between the heart sound timing trend and the heart rate trend, wherein, in order to determine whether to switch to a spectrum-based heart sound tracking algorithm, the heart sound recognition circuit is configured to: determine not to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments if the consistency metric falls below a threshold; and determine to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments if the consistency metric exceeds the threshold.

[0027] In Example 15, any one or more of the subjects in Examples 1 to 14 may optionally include a physiological event detector, which can be configured to detect cardiac events using identified heart sound components.

[0028] Example 16 is a method for detecting cardiac events, comprising the following steps: receiving heart sound information over multiple cardiac cycles; using at least a portion of the received heart sound information to generate representative heart sound segments within the cardiac cycles; using a time-based heart sound tracking algorithm to identify heart sound components from the representative heart sound segments; evaluating the performance index of the time-based heart sound tracking algorithm using morphological or timing information of the identified heart sound components; determining, based on the evaluated performance index, whether to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from the representative heart sound segments; and using the identified heart sound components to detect cardiac events.

[0029] In Example 17, the subject of Example 16 may optionally include a time-based heart sound tracking algorithm that can identify heart sound components using amplitude and timing information of representative heart sound segments, wherein the spectrum-based heart sound tracking algorithm uses one or more spectral entropy values ​​calculated from representative heart sound segments to identify heart sound components.

[0030] In Example 18, the subject matter of any one or more of Examples 13 to 17 may optionally include a portion of the received heart sound information, which may include multiple heart sound segments corresponding to corresponding heartbeats, wherein generating representative heart sound segments includes: sorting the multiple heart sound segments according to a specific order of the instantaneous heart rates of the heartbeats corresponding to the multiple heart sound segments; identifying from the sorted multiple heart sound segments a first group of heart sound segments corresponding to at least a first heartbeat, and a second group of heart sound segments corresponding to a second heartbeat, the second heartbeat having a different heart rate than the first heartbeat; determining a first morphological similarity measure of the first group of heart sound segments and a second morphological similarity measure of the second group of heart sound segments; selecting a group between the first group and the second group based on one or more of the first morphological similarity measure or the second morphological similarity measure; and generating representative heart sound segments using the overall average of the heart sound segments of the selected group.

[0031] In Example 19, the subject matter of Example 18 may optionally include a first group and a second group, each having at least a specific minimum number of heart sounds, wherein a first heartbeat corresponding to a heart sound in the first group and a second heartbeat corresponding to a heart sound in the second group each have a corresponding instantaneous heart rate that falls below a heart rate threshold and satisfies a heart rate variability or range requirement.

[0032] In Example 20, the subject matter of any one or more of Examples 18 to 19 may optionally include multiple heart sounds, which can be sorted in ascending order according to the instantaneous heart rate of the heartbeats, wherein a second heartbeat corresponding to a second group of heart sounds has a higher heart rate than a first heartbeat corresponding to a first group of heart sounds, wherein the selection between the first group and the second group includes: selecting the first group if (i) a first morphological similarity measure is higher than a first threshold, or (ii) both the first morphological similarity measure and the second morphological similarity measure are lower than their respective thresholds, and the first morphological similarity measure is greater than the second morphological similarity measure; and selecting the second group if (i) the first morphological similarity measure is lower than the first threshold and the second morphological similarity measure is higher than the second threshold, or (ii) both the first morphological similarity measure and the second morphological similarity measure are lower than their respective thresholds and the second morphological similarity measure is greater than the first morphological similarity measure.

[0033] In Example 21, the subject of any one or more of Examples 16 to 20 may optionally include a portion of the received heart sound information, which may include multiple heart sound segments corresponding to the corresponding heartbeats, wherein evaluating the performance index of the time-based heart sound tracking algorithm includes using (i) a morphological similarity measure among subsets of at least a plurality of heart sound segments used to generate representative heart sound segments, and (ii) heart sound tracking stability, which indicates the variability in timing of heart sound components correspondingly determined from the subsets of at least a plurality of heart sound segments used to generate representative heart sound segments, wherein determining whether to switch to a spectrum-based heart sound tracking algorithm includes: if (i) the morphological similarity measure exceeds a similarity threshold and (ii) the heart sound tracking stability is below a variability threshold, then determining not to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments; and if (i) the morphological similarity measure is not greater than a similarity threshold, or (ii) the heart sound tracking stability exceeds a variability threshold, then determining to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments.

[0034] In Example 22, the subject matter of any one or more of Examples 16 to 21 may optionally include a portion of the received heart sound information, which may include multiple heart sound segments corresponding to the corresponding heartbeats, wherein evaluating the performance index of the time-based heart sound tracking algorithm includes: generating a heart sound timing trend, the heart sound timing trend including a time series of timing of heart sound components identified from a subset of at least a plurality of heart sound segments using the time-based heart sound tracking algorithm; generating a heart rate trend, the heart rate trend including a time series of instantaneous heart rates of heartbeats corresponding to a subset of at least a plurality of heart sound segments; and evaluating the performance index of the time-based heart sound tracking algorithm using a consistency metric between the heart sound timing trend and the heart rate trend, wherein determining whether to switch to a spectrum-based heart sound tracking algorithm includes: if the consistency metric falls below a threshold, determining not to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments; and if the consistency metric exceeds a threshold, determining to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments.

[0035] This overview provides some teachings for this application and is not intended to be an exclusive or exhaustive treatment of the subject matter. Further details regarding the subject matter are found in the detailed description and the appended claims. Other aspects of this disclosure will be apparent to those skilled in the art upon reading and understanding the following detailed description and examining the accompanying drawings, each of which should not be construed in a limiting sense. The scope of this disclosure is defined by the appended claims and their legal equivalents. Attached Figure Description

[0036] Various embodiments are illustrated by way of example in the accompanying drawings. Such embodiments are exemplary and are not intended to be exhaustive or exclusive embodiments of the subject matter.

[0037] Figure 1 This section broadly illustrates an example of a patient management system and portions of the environment in which the system can operate.

[0038] Figure 2 This paper presents an example of a heart sound-based physiological event detection system that identifies heart sound components and uses these components to detect target physiological events.

[0039] Figure 3 This section broadly illustrates an example of generating the entire heart sound set using a pre-selected, sorted subset of heart sound segments.

[0040] Figure 4 This is a flowchart illustrating an example method for selecting a pre-selected, sorted set of heart sounds from a candidate group based on morphological similarity between heart sounds.

[0041] Figure 5AA heart sound tracking evaluation circuit is shown, which uses a heart sound morphological similarity metric and heart sound tracking stability to generate a heart sound recognition performance index.

[0042] Figure 5B It shows the use Figure 5A The flowchart shown illustrates an example method where the heart sound tracking evaluation circuit evaluates the performance of a time-based tracker and uses the evaluated performance to determine whether to switch to a spectrum-based tracker to identify the heart sound components of interest.

[0043] Figure 6A A heart sound tracking evaluation circuit is shown, which uses a consistency measure between heart rate trends and heart sound timing trends to generate a heart sound recognition performance index.

[0044] Figure 6B It shows the use of Figure 6A The flowchart shown illustrates an example method where the heart sound tracking evaluation circuit evaluates the performance of a time-based tracker and uses the evaluated performance to determine whether to switch to a spectrum-based tracker to identify the heart sound components of interest.

[0045] Figures 7A and 7B illustrate examples of quantifying heart sound tracking stability based on aortic stenosis indication and using heart sound tracking stability to determine whether to switch to a spectrum-based tracker to identify heart sound components of interest.

[0046] Figures 8A to 8B An example is shown of determining the consistency between heart rate and S2 timing and using that consistency to determine whether to switch to a spectrum-based tracker to identify the heart sound components of interest.

[0047] Figure 9 Examples of detecting worsening heart failure (WHF) are shown in general, at least in part, based on the S3 heart sounds identified and tracked according to the various examples described in this document.

[0048] Figure 10 The diagram broadly illustrates an example of a method for identifying and tracking heart sound components and using these components to detect physiological events.

[0049] Figure 11 A block diagram is shown in general terms of an example machine on which any one or more of the techniques (e.g., methodologies) discussed in this paper can be executed. Detailed Implementation

[0050] This document discloses systems, apparatus, and methods for identifying and tracking heart sound components using one or more different recognition and tracking algorithms. An exemplary medical device system includes a data receiver that receives heart sound information and a heart sound recognition circuit that generates representative heart sound segments (such as the aggregate average of selected portions of heart sound segments from multiple cardiac cycles). The heart sound recognition circuit uses a time-based tracking algorithm to identify heart sound components from the representative heart sound segments and evaluates a performance index of the time-based heart sound tracking algorithm. Based on the performance index, a decision can be made as to whether to switch to a spectrum-based tracking algorithm to further identify heart sound components from the representative heart sound segments. A physiological event detector can use the identified heart sound components to detect cardiac events.

[0051] Figure 1 The example patient management system 100 and portions of the environment in which it may operate are shown in general. The patient management system 100 can perform a range of activities, including remote patient monitoring and diagnosis of disease conditions. These activities can be performed near the patient 101 (such as in the patient's home or office), via a centralized server (such as in a hospital, clinic, or doctor's office), or via remote workstations (such as secure wireless mobile computing devices).

[0052] The patient management system 100 may include one or more mobile medical devices, an external system 105, and a communication link 111, which provides communication between the one or more mobile medical devices and the external system 105. The one or more mobile medical devices may include an implantable medical device (IMD) 102, a wearable medical device (WMD) 103, or one or more other implantable, leadless, subcutaneous, external, wearable, or mobile medical devices configured to monitor, sense, or detect information from the patient 101, determine physiological information about the patient 101, or provide one or more treatments to treat various conditions of the patient 101, such as one or more cardiac or non-cardiac conditions (e.g., dehydration, sleep apnea, etc.).

[0053] In one example, IMD 102 may include one or more conventional cardiac rhythm management devices implanted in a patient's chest, having a lead system including one or more transvenous, subcutaneous, or non-invasive leads or catheters to position one or more electrodes or other sensors (e.g., heart sound sensors) in, on, or around the heart, chest, abdomen, or neck of patient 101. In another example, IMD 102 may include, for example, a monitor implanted subcutaneously in the chest of patient 101, including a housing containing circuitry and, in some examples, one or more sensors, such as temperature sensors.

[0054] IMD 102 may include assessment circuitry configured to detect or determine specific physiological information of patient 101, or to determine one or more conditions, or to provide information or alerts to users such as patient 101 (e.g., a patient), a clinician, or one or more other caregivers or processes. In examples, IMD 102 may be an implantable cardiac monitor (ICM) configured to collect cardiac information from the patient, optionally along with other physiological information. IMD 102 may alternatively or additionally be configured as a treatment device configured to treat one or more medical conditions of patient 101. Treatment may be delivered to patient 101 via a lead system and associated electrodes or using one or more other delivery mechanisms. Treatment may include delivering one or more medications to patient 101, such as using IMD 102 or one or more other mobile medical devices. In some examples, treatment may include cardiac resynchronization therapy to correct asynchrony in patients with heart failure and improve cardiac function in patients with heart failure. In other examples, IMD 102 may include a drug delivery system, such as a drug infusion pump, to deliver drugs to a patient for managing arrhythmias or complications arising from arrhythmias, hypertension, or one or more other physiological conditions. In other examples, IMD 102 may include one or more electrodes configured to stimulate the patient's nervous system or to provide stimulation to the muscles of the patient's airway.

[0055] WMD 103 may include one or more wearable or external medical sensors or devices (e.g., automatic external defibrillator (AED), Holter monitor, patch-based device, smartwatch, smart accessory, wrist or finger-worn medical device, such as finger-based photoplethysmography sensor, etc.).

[0056] In the examples, IMD 102 or WMD 103 may include or be coupled to an implantable or wearable sensor to sense heart sound signals and includes heart sound recognition circuitry to identify one or more heart sound components, such as S1, S2, S3, or S4, based on a spectral entropy time series derived from the sensed heart sound signals. IMD 102 or WMD 103 also includes heart sound-based event detector circuitry that can detect physiological events (e.g., cardiac arrhythmia onset or worsening heart failure (WHF) events) based at least on heart sound metrics of the detected one or more heart sound components. Examples of such heart sound metrics may include the amplitude or timing of a heart sound component within a cardiac cycle relative to a reference point. In some examples, at least a portion of the heart sound recognition circuitry and / or the heart sound-based event detector circuitry may be implemented in and executed by external system 105.

[0057] External system 105 may include dedicated hardware / software systems, such as a programmer, a remote server-based patient management system, or alternatively, a software-defined system primarily running on a standard personal computer. External system 105 may manage patient 101 via IMD 102 or one or more other mobile medical devices connected to external system 105 via communication link 111. In other examples, IMD 102 may be connected to WMD 103, or WMD 103 may be connected to external system 105 via communication link 111. This may include, for example, programming IMD 102 to perform one or more of the following: acquiring physiological data, performing at least one self-diagnostic test (such as for device operating status), analyzing physiological data, or optionally delivering or adjusting treatment for patient 101. Additionally, external system 105 may send or receive information from IMD 102 or WMD 103 via communication link 111. Examples of information may include real-time or stored physiological data from patient 101, diagnostic data such as detection of patient hydration status, hospitalization, response to treatment delivered to patient 101, or device operating status (e.g., battery status, lead impedance, etc.) of IMD 102 or WMD 103. Communication link 111 may be an inductive telemetry link, a capacitive telemetry link, or a radio frequency (RF) telemetry link, or wireless telemetry based on standards such as “Strong” Bluetooth or IEEE 802.11 Wireless Fidelity “Wi-Fi”. Other configurations and combinations of patient data source docking are possible.

[0058] External system 105 may include an external device 106 located near one or more mobile medical devices, and a remote device 108 located relatively far from the one or more mobile medical devices, communicating with external device 106 via communication network 107. Examples of external device 106 may include a medical device programmer. Remote device 108 may be configured to, among other possible functions, evaluate collected patient or patient information and provide alarm notifications. In an example, remote device 108 may include a centralized server acting as a central hub for data storage and analysis of the collected data. The server may be configured as a single, multiple, or distributed computing and processing system. Remote device 108 may receive data from multiple patients. Data may be collected by one or more mobile medical devices, in addition to other data acquisition sensors or devices associated with patient 101. The server may include a memory device to store data in a patient database. The server may include alarm analyzer circuitry to evaluate the collected data to determine whether specific alarm conditions are met. The fulfillment of alarm conditions may trigger the generation of alarm notifications, such as those provided by one or more human-perceptible user interfaces. In some examples, alert conditions may be evaluated alternatively or additionally by one or more mobile medical devices, such as implantable medical devices. For example, alert notifications may include web page updates, telephone or pager calls, emails, SMS messages, text or "instant" messages, as well as messages to patients and simultaneous direct notifications to emergency services and clinicians. Other alert notifications are possible. The server may include alert priority sorting circuitry configured to prioritize alert notifications. For example, alerts for detected physiological events may be prioritized using a similarity metric between physiological data associated with a detected physiological event and physiological data associated with historical alerts.

[0059] Remote device 108 may additionally include one or more locally configured clients or remote clients securely connected to the server via communication network 107. Examples of clients may include personal desktop computers, laptops, mobile devices, or other computing devices. System users, such as clinicians or other qualified medical professionals, can use the clients to securely access stored patient data aggregated in a database on the server and select and prioritize patients and alerts for healthcare provisioning. In addition to generating alert notifications, remote device 108, including the server and interconnected clients, can also execute follow-up protocols by sending follow-up requests to one or more mobile medical devices, or by sending messages or other communications as compliance notifications to patient 101 (e.g., a patient), clinicians, or authorized third parties.

[0060] The communication network 107 can provide wired or wireless interconnection. In this example, the communication network 107 may be based on the Transmission Control Protocol / Internet Protocol (TCP / IP) network communication standard, although other types or combinations of networking implementations are possible. Similarly, other network topologies and arrangements are possible.

[0061] One or more of the external device 106 or remote device 108 can output detected physiological events to a system user, such as a patient or clinician, or to a process, including instances of computer programs, such as those executable in a microprocessor. In examples, the process may include automatically generating recommendations for antiarrhythmic treatment, or recommendations for further diagnostic tests or treatments. In examples, external device 106 or remote device 108 may include corresponding display units for displaying physiological or functional signals, or alarms, warnings, emergency calls, or other forms of alerts to signal the detection of an arrhythmia. In some examples, external system 105 may include an external data processor configured to analyze physiological or functional signals received by one or more mobile medical devices and confirm or deny the detection of an arrhythmia. Computationally intensive algorithms, such as machine learning algorithms, may be implemented in the external data processor to retrospectively process data to detect cardiac arrhythmias.

[0062] One or more portions of a mobile medical device or external system 105 may be implemented using hardware, software, firmware, or a combination thereof. One or more portions of a mobile medical device or external system 105 may be implemented using dedicated circuitry, which may be constructed or configured to perform one or more functions, or may be implemented using general-purpose circuitry, which may be programmed or otherwise configured to perform one or more functions. Such general-purpose circuitry may include a microprocessor or a portion thereof, a microcontroller or a portion thereof, or programmable logic circuitry, memory circuitry, network interfaces, and various components for interconnecting these components. For example, among others, a “comparator” may include an electronic circuit comparator that may be constructed to perform a specific function of comparing two signals, or a comparator may be implemented as part of a general-purpose circuitry that may be driven by code instructing a portion of the general-purpose circuitry to perform a comparison between two signals. A “sensor” may include electronic circuitry configured to receive information and provide an electronic output representing such received information.

[0063] Treatment device 110 can be configured to send or receive information from one or more mobile medical devices or external systems 105 using communication link 111. In this example, one or more mobile medical devices, external device 106, or remote device 108 can be configured to control one or more parameters of treatment device 110. External system 105 can allow programming of one or more mobile medical devices and can receive information about one or more signals acquired by the one or more mobile medical devices, such as those received via communication link 111. External system 105 may include a local external implantable medical device programmer. External system 105 may include a remote patient management system that can, for example, monitor patient status from a remote location or adjust one or more treatments.

[0064] Figure 2 An example of a heart sound-based physiological event detection system 200 is generally illustrated. This system can identify heart sound components of interest to detect physiological events, such as cardiac arrhythmias or worsening heart failure (WHF) events, using information about the identified heart sound components. The heart sound-based physiological event detection system 200 may include one or more of a data receiver circuitry 210, a controller circuitry 220, a user interface 230, and a treatment circuitry 240. At least a portion of the heart sound-based physiological event detection system 200 may be implemented in an IMD 102, a WMD 103, or an external system 105 (such as one or more of an external device 106 or a remote device 108).

[0065] Data receiver circuitry 210 can receive physiological information from a patient. In an example, data receiver circuitry 210 may include sensing amplifier circuitry configured to sense physiological signals from the patient via physiological sensors, such as implantable, wearable, or otherwise mobile sensors or electrodes associated with the patient. The sensors may be incorporated into or otherwise associated with a mobile device such as IMD 102 or WMD 103. In some examples, physiological signals sensed from the patient may be stored in a storage device, such as an electronic medical record (EMR) system. Data receiver circuitry 210 can receive physiological signals from the storage device, such as in response to user commands or triggering events. This is by way of example and not limitation, and as... Figure 2As shown, the data receiver circuit 210 may include one or more of a heart sound sensing circuit 212, a heart sound sensing circuit 214, or a heart rate circuit 216. The heart sound sensing circuit 212 may receive heart sound information, such as heart sound signals sensed from a patient. In this example, the heart sound sensing circuit 212 may be coupled to a heart sound sensor to sense body motion / vibration signals that indicate heart sounds, which are correlated with or indicate heart sounds. The heart sound sensor may take the form of an accelerometer, an acoustic sensor, a microphone, a piezoelectric-based sensor, or other vibration or acoustic sensor. The accelerometer may be a single-axis, dual-axis, or triaxial accelerometer. Examples of accelerometers may include a flexible piezoelectric crystal (e.g., quartz) accelerometer or a capacitive accelerometer manufactured using microelectromechanical systems (MEMS) technology. The heart sound sensor may be included in IMD 102 or WMD 103, or disposed on leads (such as part of a lead system associated with IMD 102 or WMD 103). In the example, an accelerometer (or other sensor type) can sense endocardial or epicardial acceleration (EA) signals from a part of the heart, such as on the endocardial or epicardial surface of one of the left ventricle, right ventricle, left atrium, or right atrium. The EA signal may contain components corresponding to various heart sound components, such as one or more of S1, S2, S3, or S4.

[0066] The cardiac sensing circuit 214 can sense cardiac electrical signals. Examples of cardiac electrical signals may include surface electrocardiography (ECG) sensed from electrodes placed on the body surface, subcutaneous ECG sensed from electrodes placed under the skin, and intracardiac electrogram (EGM) sensed from one or more implantable electrodes. Sensing electrodes may be included in or communicatively coupled to IMD 102 or WMD 103. The heart rate circuit 216 can detect the patient's heart rate, such as from signals sensed by the cardiac sensing circuit 214. Alternatively, heart rate (or pulse rate) may be detected from cardiac mechanical signals. As discussed further below, one or more of the cardiac electrical signals or heart rate information may be used by the controller circuitry to construct representative heart segments from which a spectral entropy time series can be derived.

[0067] In some examples, the data receiver circuit 210 may receive other physiological or functional signals, among others, including, for example, body activity signals, posture signals, thoracic or cardiac impedance signals, arterial pressure signals, pulmonary artery pressure signals, left atrial pressure signals, RV pressure signals, LV coronary artery pressure signals, coronary artery blood temperature signals, blood oxygen saturation signals, heart sound signals, physiological responses to activity, apnea-hypopnea index, one or more respiratory signals (such as respiratory rate signals or tidal volume signals), brain natriuretic peptide (BNP), blood panel, sodium and potassium levels, glucose levels, and other biomarkers and biochemical markers.

[0068] The controller circuit 220 can identify and track heart sound components and detect physiological events based on at least one heart sound component. The controller circuit 220 is implemented as part of a microprocessor circuit, which can be a dedicated processor, such as a digital signal processor, an application-specific integrated circuit (ASIC), a microprocessor, or other type of processor for processing information including bodily activity information. Alternatively, the microprocessor circuit can be a general-purpose processor that can receive and implement a set of instructions to perform the functions, methods, or techniques described herein.

[0069] Controller circuitry 220 may include a circuit group comprising one or more other circuits or sub-circuits, such as heart sound recognition circuitry 222, heart sound tracking and evaluation circuitry 226, and physiological event detector 227. These circuits may perform the functions, methods, or techniques described herein, individually or in combination. In the example, the hardware of the circuit group may be invariably designed to perform a specific operation (e.g., hardwired). In the example, the hardware of the circuit group may include physically connected components (e.g., execution units, transistors, simple circuits, etc.), including computer-readable media that are physically modified (e.g., magnetically grounded, electrically grounded, movable placement of invariant aggregated particles, etc.) to encode instructions for a specific operation. When the physical components are connected, the underlying electrical characteristics of the hardware composition change, for example, from an insulator to a conductor, or vice versa. Instructions enable embedded hardware (e.g., execution units or loading mechanisms) to create members of the circuit group within the hardware via variable connections to perform portions of a specific operation during operation. Thus, the computer-readable medium is communicatively coupled to other components of the circuit group members during device operation. In the example, any physical component may be used in more than one member of more than one circuit group. For example, under operation, the execution unit can be used at one point in time in the first circuit of the first circuit group, and reused at different times by the second circuit in the first circuit group or by the third circuit in the second circuit group.

[0070] Heart sound recognition circuit 222 can preprocess the heart sound signal received from heart sound sensing circuit 212. In an example, heart sound recognition circuit 222 may include a filter or filter bank to remove or attenuate one or more of low-frequency signal baseline drift, high-frequency noise, or other unwanted frequency components. In an example, the heart sound segment may be bandpass filtered to a frequency range of approximately 5Hz–90Hz or approximately 9Hz–90Hz. In an example, the filter may include a second-order or higher-order differentiator configured to calculate the second-order or higher-order derivative of the heart sound signal.

[0071] Heart sound recognition circuit 222 can identify heart sound components of interest (e.g., one or more of S1, S2, S3, or S4) from a preprocessed heart sound signal. Cardiac electrical signals sensed by cardiac sensing circuit 214 and / or heart rate information sensed by heart rate circuit 216 can be used to assist in identifying heart sound components. To improve heart sound signal quality and recognition accuracy, heart sound recognition circuit 222 can use at least a portion of the received heart sound signal to generate a representative heart sound segment within the cardiac cycle. In the example, a representative heart sound segment can be generated using the overall average of a selected set of heart sound segments. The entirety of heart sounds includes heart sound segments within each cardiac cycle. In the example, the entirety of heart sounds circuit 323 can use cardiac electrical signals sensed simultaneously with the heart sound signal (from cardiac sensing circuit 214). The entirety of heart sounds circuit 323 can detect the cardiac cycle as the time interval between two consecutive ventricular excitations (such as R waves or QRS complexes) on an ECG or intracardiac EGM signal. The entirety of heart sounds circuit 323 can align heart sound segments relative to various reference points within multiple cardiac cycles. Reference points may include, for example, ventricular activation timing on ECG or EGM signals over multiple cardiac cycles.

[0072] The heart sound whole circuit 323 can filter heart sound segments across multiple cardiac cycles and select a subset of heart sound segments that meet specific criteria to form the heart sound whole. Selection criteria may include, among others, one or more of the following: heart rate, heart rate variability or range, activity level, respiratory rate, or time of day. In one example, a subset of heart sound segments may be selected to correspond to heart rates below a predetermined or user-set heart rate threshold (e.g., 100 bpm–120 bpm). In another example, a subset of heart sound segments may be selected to correspond to heart rates or cycle lengths falling within a specified range, such as substantially the same cycle length within a margin of + / - 100 milliseconds, or in one example + / - 5 beats per minute (bpm), or in another example substantially the same heart rate within a margin of + / - 10 bpm (heart rate variability or range).

[0073] In addition to or alternative to heart rate and heart rate variability or range criteria, other requirements may be imposed on a selected subset of heart sound segments. For example, a subset of heart sound segments may be selected to correspond to the physical activity level of a subject falling within a specified range, such as substantially the same physical activity level within a specified margin. Alternatively, a subset of heart sound segments may be selected to correspond to the respiratory rate of a subject falling within a specified range, such as substantially the same respiratory rate within a specified margin. In yet another example, heart sound signals may be sensed during a specified time period of day, and the overall heart sound circuit 323 may select a subset of heart sound segments sensed during substantially the same time period of day within a specified margin.

[0074] In some examples, the total heart sound circuit 323 can sort heart sound segments across multiple cardiac cycles based on a specific order of heart rates corresponding to the heartbeats of the heart sound segments. From the “sorting pool” of heart sound segments, at least a first group of heart sound segments corresponding to a first heartbeat, denoted by {HS1}, and a second group of heart sound segments corresponding to a second heartbeat, denoted by {HS2}, can be identified. The second heartbeat has a different heart rate than the first heartbeat. Morphological similarity can be assessed between heart sound segments in each group, resulting in a first morphological similarity measure MS1 between the first group of heart sound segments {HS1} and a second morphological similarity measure MS2 between the second group of heart sound segments {HS2}. Now refer to Figure 3 Figure 300 illustrates an example of creating the entirety of heart sounds using a selected subset of a pre-selected “sorting pool” of heart sound segments {310A, 310B, 310C, ...310N}. In this non-limiting example, the heart sound segments are sorted in ascending order of the instantaneous heart rate of the heartbeats corresponding to the heart sound segments {310A, 310B, 310C, ...310N}. The heart sound segments in the sorting pool are pre-selected such that they meet, among other things, screening criteria based on one or more of heart rate, heart rate variability or range, respiratory rate, or activity level, as described above. Different portions of the sorting pool of heart sound segments can be used to assemble different groups of heart sound segments corresponding to heartbeats with different heart rates. This is an example and not a limitation. Figure 3 This diagram shows a first group 311{HS1} comprising the first K1 heart sounds in the sorting pool and a second group 312{HS2} comprising the last K2 heart sounds in the sorting pool. The first group {HS1} corresponds to the first K1 heartbeats, and the second group of heart sounds {HS2}... HR2The last K2 heartbeats correspond to the last heartbeats with a higher heart rate than the first K1 heartbeats. The number of beats K1 and K2 can be user-specified or programmable numbers within a range, such as between 8 and 16 heartbeats in the non-limiting example. In the example, for each of the first and second groups, at least eight qualifying heartbeats are required to meet the heart rate and / or heart rate variability requirements. The first K1 heartbeats and the last K2 heartbeats each meet the heart rate and heart rate variability requirements, such as the instantaneous heart rate falling below a heart rate threshold (e.g., 120 bpm in the example) in each group, and the heart rate variability falling below a variability threshold in each group. In the example, heart rate variability can be calculated as the difference between the maximum and minimum instantaneous heart rates in a group of heartbeats. The difference in heart rate during the first K1 heartbeats. The difference in heart rate between the last K2 heartbeats Each is required to be less than a threshold. (For example, 10 bpm in the example).

[0075] The overall heart sound circuit 223 may include a heart sound segment group selector 330 to select a group between a first group {HS1} and a second group {HS2}, and generate representative heart sound segments using the overall average of the heart sound segments in the selected group. This selection may be based on a morphological similarity check 320, where one or more of a first morphological similarity measure (MS1) between heart sound segments in the first group {HS1} or a second morphological similarity measure (MS2) between heart sound segments in the second group {HS2} may be determined. The morphological similarity measure represents the similarity of waveforms of heart sound portions or components extracted from heart sound segments, such as those from the first group {HS1} or the second group {HS2}. In the example, the morphological similarity measure may be calculated using a portion of the heart sound waveform within a specified window, such as the S2 detection window. The S2 detection window may be defined relative to a reference point such as ventricular activation timing (e.g., the R wave on an ECG waveform). In the example, morphological similarity can be calculated as the cumulative deviation or root-mean-squared (RMS) value of heart sound waveforms within the same group. In another example, morphological similarity can be calculated as the correlation between morphological features of heart sound waveforms within the same group.

[0076] Now for reference Figure 4This flowchart illustrates an example method 400 for selecting a pre-selected, sorted set of heart sound segments based on morphological similarity between them. Method 400 can be used by the heart sound whole circuit 323 to select between a first group {HS1} and a second group {HS2} based on one or more of a first morphological similarity measure MS1 or a second morphological similarity measure MS2. At step 410, the first group {HS1} can be assembled from a sorted pool of heart sound segments, such as... Figure 3 As shown. At step 420, the morphological similarity measure MS1 of {HS1} is calculated and compared with the threshold MS. TH1 Compare them. If MS1 is greater than MS... TH1 In step 460, the first group {HS1} is considered satisfactory and can be selected as the entire set of heart sounds for calculating the overall average. If MS1 is not greater than MS... TH1 Then, at step 430, a second group {HS2} can be assembled from the sorting pool of heart sounds, such as... Figure 3 As shown. At step 440, the morphological similarity measure MS2 of {HS2} is calculated and compared with the threshold MS. TH2 Comparison. MS TH2 Can be used with MS in an example TH1 Same, or in another example, with MS TH1 Different. If MS2 is greater than MS TH2 In step 470, the second group {HS2} is considered satisfactory and can be selected as the entire heart sound set for calculating the overall mean. If MS2 is not greater than MS... TH2 Then, at step 450, the morphological similarity measure MS2 of the second group is compared with the morphological similarity measure MS1 of the first group, and any group with a larger similarity measure value can be selected as the whole heart sound group. The whole heart sound circuit 323 can align the heart sound segments of the selected group relative to their respective reference points (such as ventricular excitation (e.g., the R wave in ECG)) and calculate the whole average of the heart sound segments as representative heart sound segments.

[0077] Return to reference Figure 2The heart sound recognition circuit 222 can use either a time-based tracker 224 or a spectrum-based tracker 225 to identify and track heart sound components of interest, such as S1, S2, S3, or S4 components. The time-based tracker 224 uses a time-based tracking algorithm to identify the heart sound components of interest from representative heart sound segments (such as those generated by the heart sound whole circuit 223). The time-based tracking algorithm involves, among other things, amplitude and timing information of the heart sound segments. An example of the time-based tracking algorithm used by the time-based tracker 224 is described in U.S. Patent No. 7,853,327 entitled “HEARTSOUND TRACKING SYSTEM AND METHOD” by Patangay et al., which, among other things, mentions using heart sound information from a particular heart sound waveform and heart sound information from at least one other heart sound waveform, using a first heart sound energy indicator and a corresponding first heart sound time indicator to identify heart sound components from a heart sound waveform, the disclosure of which is incorporated herein by reference in its entirety. Spectrum-based tracker 225 uses a spectrum-based tracking algorithm to identify heart sound components of interest from representative heart sound segments, such as those generated by the overall heart sound circuit 223. Among other information, the spectrum-based tracking algorithm uses one or more spectral features that can be derived from the heart sound segments. An example of a spectrum-based tracking algorithm used by spectrum-based tracker 225 is described in U.S. Patent Application No. 63 / 432,649, entitled “ENTROPY BASED HEART SOUND TRACKING” by Goftari et al., which describes the identification of heart sound components from a heart sound waveform using a complexity metric, including entropy (or a variant thereof) obtained from the power spectrum of the heart sound signal, the disclosure of which is incorporated herein by reference in its entirety. Spectrum entropy-based methods quantify the complexity (or inhomogeneity) of the power spectrum amplitude of a heart sound signal or a portion thereof, such as within a cardiac cycle. High spectral entropy indicates a more uniform distribution of signal energy across a wide frequency range in the frequency domain, while low spectral entropy indicates less uniformity in the distribution of signal energy. Wideband signals (such as white noise signals) are characterized by their signal energy propagating almost uniformly over a wide frequency range, and therefore have a large spectral entropy. Conversely, narrowband signals concentrate most of their signal energy in a narrow frequency band and are less uniformly distributed, and therefore have a smaller spectral entropy. Some heart sound components, such as S1 and S2, are narrowband signals. For example, the power of S1 typically falls within approximately 10Hz–50Hz, while the power of S2 typically falls within approximately 20Hz–70Hz. Therefore, S1 and S2 typically have lower spectral entropy values.

[0078] Spectral entropy-based heart sound recognition combines the advantages of frequency domain spectral analysis and Shannon entropy, both of which are more robust to random noise and interference from various sources than time-domain, amplitude-based heart sound detection methods. However, spectral entropy-based methods are generally more computationally intensive than time-based heart sound recognition methods and may require more computational resources and consume more power from implantable devices. To improve heart sound recognition performance while balancing accuracy and efficiency, in this example, heart sound recognition circuit 222 can select between a time-based tracker 224 and a spectrum-based tracker 225 to identify and track the heart sound components of interest. Alternatively, heart sound recognition circuit 222 can perform initial identification of the heart sound components of interest using a less computationally intensive algorithm (such as the time-based tracking algorithm used by the time-based tracker 224) and, based on the performance evaluation of the initial identification, determine whether to switch to a more computationally intensive algorithm (such as the spectral entropy-based tracking algorithm used by the spectrum-based tracker 225) to further identify the heart sound components of interest. Figure 2 As shown, the heart sound tracking evaluation circuit 226 can evaluate the performance of one or more of the time-based heart sound tracking algorithms or the spectrum-based tracking algorithms, at least based on the morphological or timing information of the heart sound components of interest. Figures 5A to 5B and Figures 6A to 6B An embodiment of the heart sound recognition circuit 222 is shown in which the quality of the time-based tracker 224 in recognizing the heart sound components of interest is evaluated. Figure 5A A heart sound tracking evaluation circuit 510 is shown, configured to generate a heart sound recognition performance index using a heart sound morphological similarity measure 512 and a heart sound tracking stability 514. The heart sound morphological similarity measure 512 is a measure of similarity between at least a subset of heart sound segments (such as selected groups of heart sound segments, e.g., a first group {HS1} or a second group {HS2}) used to generate representative heart sound segments. (See above regarding...) Figures 3 to 4 The morphological similarity measure can be calculated using a portion of the heart sound waveform within a specified window, such as an S2 detection window used to assess the S2 morphological similarity measure. In one example, the morphological similarity measure can be calculated as the cumulative deviation or RMS value of heart sound waveforms within the same group, or the correlation between morphological features of heart sound waveforms within the selected group.

[0079] Heart sound tracking stability 514 is a measure of the stability over time of the timing of a heart sound component of interest (e.g., S2) at different heart rates. In the case of identifying and tracking the S2 heart sound, aortic stenosis (AS) is one of the factors affecting the timing stability of S2 in heart sounds. Heart sound tracking stability 514 can be approximated by the AS index, which indicates the presence or severity of AS. The AS index measures the physical properties of the aortic valve, such as the aortic valve surface area or changes in its surface area. The surface area value of the aorta (S... AV ) or S AV The variation can be estimated using the heart sound-based Gorlin formula, as detailed in U.S. Patent Application No. 63 / 523,152, entitled "AORTIC STENOSIS SEVERITY QUANTIFICATION USING HEART SOUNDS" by Goftari et al., the disclosure of which is incorporated herein by reference in its entirety. Specifically, the heart sound-based Gorlin formula reproduced below relates to heart rate (HR), heart sound-based ejection time (ET) (estimated as the interval between the S1 and S2 heart sound components (T... S1-S2 The surface area of ​​the aorta, S), and S2 intensity, such as S2 amplitude or peak power (||S2||). AV With HR, T S1-S2 Each of the square roots of the S2 intensity is inversely proportional. An increase in S2 intensity and / or a prolonged S1-S2 time interval may result in an increase in aortic valve surface area S. AV A decrease in the level of AS indicates the presence or severity of the condition.

[0080]

[0081] Figures 7A and 7B illustrate examples of quantifying heart sound tracking stability based on aortic stenosis (AS) indications (e.g., calculated using a heart sound-based Gorlin formula) and using it to determine whether to switch to a spectrum-based tracker to identify heart sound components of interest. Figure 7A shows a time series of AS indices 710 (on the y-axis) calculated over time (along the x-axis) for consecutive heart sound segments using the heart sound-based Gorlin formula above. Figure 7B shows a heart sound map 720 representing multiple heart sound segments stacked together (along the y-axis, indexed by the number of segments). The heart sound segments are extracted over multiple cardiac cycles with different heart rates and aligned relative to various reference points (the origin of the x-axis) of the R waves of each cardiac cycle. Each heart sound segment (along the x-axis) has a color-coded or grayscale-coded signal amplitude to show how the amplitude changes over time. Figure 7B also shows the S2 component 722 of each of the heart sound segments as identified and tracked by the time-based tracker 224.

[0082] Figure 7A illustrates a sudden “jump” 711 in the AS index at approximately time T (corresponding to approximately 3400–3500 heart sound indexes), indicating an increase in heart sound tracking variability. Figure 7B shows that at approximately the same time T (corresponding to approximately 3400–3500 heart sound indexes), S2 begins to be falsely detected at approximately the timing position of the S1 component, indicating that the time-based tracker 224 is locked onto the S1 component 723 instead of S2, resulting in significant variability in the identified S2 position. The examples in Figures 7A and 7B demonstrate that the variability in the AS index calculated according to the heart sound-based Gorlin formula above (e.g., the sudden jump 711) can be used to detect S2 tracking performance (e.g., false identification of S2).

[0083] Heart sound tracking stability 514 can be determined as the AS index (each calculated as) of a subset of heart sound segments used to generate representative heart sound segments (such as a selected group of heart sound segments (e.g., the first group {HS1} or the second group {HS2}) generated by the total heart sound circuit 223). The variability of ). For example, if according to such Figure 4 If the method 400 selected the first group {HS1}, the AS index can be calculated for each of the K1 heart sounds in that group. Similarly, if the first group {HS1} is selected according to method 400, the AS index can be calculated for each of the K2 heart sounds in that group. Heart sound tracking stability 514 can then be determined as the variability of the AS index for the selected group.

[0084] It should be recognized that using the AS index as described above is a non-limiting example for determining heart sound tracking stability.514 Other measures, such as ejection time (ET) based on heart sounds, are also possible. S1-S2 Alternatively, the S2 intensity ||S2|| can each be used to approximate heart sound tracking stability 514. For example, heart sound tracking stability 514 can be based on T S1-S2 The deviation from the ET threshold, or the deviation from the S2 intensity threshold based on ||S2||, is determined.

[0085] Figure 5B This is a flowchart illustrating example method 520, which can be... Figure 5A The heart sound tracking evaluation circuit 510 is used to evaluate the performance of the time-based tracker 224 and uses the evaluated performance to determine whether to switch to a spectrum-based tracker to identify the heart sound component of interest (e.g., S2). At step 521, the morphological similarity of the selected heart sound segment group (e.g., {HS1} or {HS2}) can be compared with a similarity threshold MS. TH Compare the morphological similarity. If the morphological similarity is higher than the MS threshold... THThen, at step 522, the stability threshold TS can be used as a reference. TH Check the heart sound tracking stability of the selected heart sound segment group. If the heart sound tracking stability is below the variability threshold TS... TH Then, at step 523, the time-based tracker 224 is considered acceptable. If, at step 521, the morphological similarity is below the threshold MS... TH Or if the stability of heart sound tracking is greater than the variability threshold TS TH If time-based trackers are not accepted, then at step 524, spectrum-based trackers 225 can be used to identify and track the heart sound components of interest.

[0086] Figure 6A A heart sound tracking evaluation circuit 610 is illustrated, configured to generate a heart sound recognition performance index using a consistency metric 613 between a heart rate trend 611 and a heart sound timing trend 612. The heart rate trend 611 and the heart sound timing trend 612 can each be calculated using at least a subset of heart sound segments used to generate representative heart sound segments, such as selected groups of heart sound segments (e.g., a first group {HS1} or a second group {HS2}) generated by the heart sound whole circuit 223. The heart rate trend 611 represents the trend of the instantaneous heart rate of the heartbeat corresponding to the selected group of heart sound segments. The heart sound timing trend 612 represents the timing trend of a heart sound component of interest (such as S2) detected from the selected group of heart sound segments relative to a corresponding reference point. The consistency metric 613, indicating the degree of consistency between changes in heart rate and changes in the timing of heart sound components over time, can be calculated using the heart rate trend 611 and the heart sound timing trend 612. In one example, the consistency metric 613 can be calculated using the correlation between the heart rate trend 611 and the heart sound timing trend 612. In another example, linear or nonlinear regression analysis can be used to generate a first regression line or curve for heart rate and a second regression line or curve for heart sound component timing (e.g., S2 timing). The consistency metric 613 can be determined based on the degree of deviation between the first and second regression lines or curves.

[0087] Figure 6B This is a flowchart illustrating example method 620, which can be derived from... Figure 6A The heart sound tracking evaluation circuit 610 is used to evaluate the performance of the time-based tracker 224 and uses the evaluated performance to determine whether to switch to a spectrum-based tracker to identify the heart sound components of interest (e.g., S2). At step 621, a consistency metric can be calculated using the heart rate trend 611 and the heart sound timing trend 612. The consistency metric can be correlated with a consistency threshold C. TH Compare and contrast the checks. If the consistency metric is below the consistency threshold C... THIf, at step 622, the time-based tracker 224 is considered acceptable, then at step 621, the consistency metric is above the threshold C. TH If the time-based tracker 224 is deemed unacceptable, then at step 623, the spectrum-based tracker 225 is used to identify and track the heart sound components of interest.

[0088] Figures 8A to 8B The example shown is an example of determining the consistency between heart rate and S2 timing, and using that consistency to determine whether to switch to a spectrum-based tracker to identify the heart sound components of interest. Figure 8A A phonogram 810 is shown, representing multiple heart sound segments stacked together (indexed by the number of segments along the y-axis). Each heart sound segment (on the x-axis) has a color-coded or grayscale-coded signal amplitude to show how the amplitude changes over time. Figure 8A The diagram also shows the S1 component 811 and S2 component 812 for each of the heart sounds. S2 can be identified and tracked using a time-based tracker 224, and the timing of S2 can be determined relative to a corresponding reference point (such as the R wave of the corresponding cardiac cycle). Figure 8B It shows the same as Figure 8A The instantaneous heart rate 821 corresponding to the heartbeat segment shown, and the S2 timing 822 (relative to the corresponding reference point) determined by the detected S2 component 812. Regression analysis can be performed to determine the heart rate regression line 831 from the instantaneous heart rate 821 and the S2 timing regression line 832 from the S2 timing 822. Figure 8B The heart rate trend and the S2 timing trend are shown to be diverging, which can be quantified as the difference between the corresponding slopes of the heart rate regression line 831 and the S2 timing regression line 832. If the difference exceeds a threshold, the time-based tracker 224 is considered unacceptable, and a decision or suggestion can be made to the user to switch to the spectrum-based tracker 225 to identify and track the S2 component.

[0089] Return to reference Figure 2The physiological event detector 227 can detect physiological events using one or more of the heart sound components detected by the heart sound recognition circuit 222. In some examples, the physiological event detector 227 can generate a heart sound metric using the detected heart sound components, and use the heart sound metric, optionally in conjunction with other physiological information obtained from the patient, to detect physiological events. Examples of heart sound metrics may include, among others, the intensity of the heart sound component (e.g., amplitude or signal energy under curve), or one or more heart sound-based cardiac timing intervals, such as the pre-ejection period (PEP), such as measured between the start of the QRS complex and the S1 heart sound; the systolic timing interval (STI), such as measured between the start of the QRS complex on the ECG and the S2 heart sound; the left-ventricular ejection time (LVET), such as measured as the interval between the S1 and S2 heart sounds; or the diastolic timing interval (DTI), such as measured between the S2 heart sound and the start of the subsequent QRS complex on the ECG. These cardiac timing intervals based on heart sounds can be correlated with cardiac systolic or diastolic function. Heart sound measures may also include the PEP / LVET ratio, STI / DTI ratio, STI / cycle length (CL) ratio, or DTI / CL ratio, or other composite measures.

[0090] In the example, the physiological event detector 227 can track heart sound components over time and generate heart sound trends, such as S1 amplitude trend, S2 amplitude trend, S1 timing trend, S2 timing trend, or trends based on cardiac timing intervals of heart sounds (e.g., PEP trend, STI trend, LVET trend, DEI trend, etc.). Based on heart sound trends, the physiological event detector 227 can generate cardiac function indicators indicative of myocardial contractility, cardiac synchronicity, and cardiac hemodynamics. In the example, the physiological event detector 227 can use one or more of the measures of S1 intensity, S2 intensity, or STI to detect cardiac arrhythmia episodes or to distinguish between different arrhythmias (e.g., atrial tachyarrhythmias, supraventricular tachyarrhythmias, or ventricular tachyarrhythmias). For example, a decrease in S1 intensity can indicate reduced cardiac contractility, and a decrease in S2 intensity can indicate reduced cardiac output; both can be used to detect cardiac arrhythmias and deterioration of cardiac hemodynamics during arrhythmias. In another example, heart sound measures, such as S3 intensity, can be used to detect WHF. An increase in S3 intensity indicates reduced ventricular compliance and deterioration of diastolic function, suggesting the occurrence of WHF. Alternatively, a decrease in S1 intensity or STI can indicate poor cardiac contractility or reduced electromechanical coupling, indicating the occurrence of WHF. (See below for reference.) Figure 9 Further descriptions are based on various examples described in this document (such as those above regarding...). Figure 2 , Figures 5A to 5B and Figures 6A to 6B Examples of time-based and / or spectrum-based tracking algorithms described are used to identify and track heart sound components to detect WHF. The physiological event detector 227 can additionally or alternatively detect medical conditions such as respiratory, renal, and neurological conditions based on heart sound metrics.

[0091] User interface 230 may include an input unit and an output unit. In this example, at least a portion of user interface 230 may be implemented in external system 105. The input unit may receive user input for programming data receiver circuitry 210 and controller circuitry 220, such as parameters for sensing heart sound signals, generating a full set of heart sounds, detecting heart sound components of interest using one or more time-based or spectrum-based trackers, or detecting physiological events, among other things. The input unit may include a keyboard, on-screen keyboard, mouse, trackball, touchpad, touchscreen, or other pointing or navigation device. The output unit may include a display for displaying, among other things, the sensed heart sound signals, representative heart sound segments, spectral entropy time series, heart sound metrics, information about detected physiological events, and any intermediate measurements or calculations. The output unit may also present treatment titration protocols and recommended treatments to the user, such as via the display unit, including changes in parameters during treatment provided by the implanted device, prescriptions for the implanted device, initiation or changes to drug therapy, or other treatment options for the patient. The output unit may include a printer for printing hard copies of information that can be displayed on the display unit. Signals and information can be presented in tables, charts, graphs, or any other type of text, table, or graphic format. The presentation of output information can include audio or other media formats. In the example, the output unit can generate alarms, alerts, emergency calls, or other forms of warnings to signal to the system user about a detected medical event.

[0092] The treatment circuit 240 can be configured to deliver treatment to a patient, such as in response to a detected physiological event. The treatment can be preventative or therapeutic in nature, such as modifying, restoring, or improving a patient's neurological, cardiac, or respiratory function. Examples of treatment may include electrical stimulation therapy delivered to the heart, nerve tissue, or other target tissues; cardioversion therapy; defibrillation therapy; or pharmaceutical therapy involving the delivery of medicines to the patient. In some examples, the treatment circuit 240 can modify existing treatments, such as adjusting stimulation parameters or drug dosage.

[0093] Figure 9 The examples shown are at least in part based on the various examples described in this document (such as those mentioned above). Figure 2 , Figures 5A to 5B and Figures 6A to 6BExamples of detecting worsening heart failure (WHF) are described, using time-based and / or spectrum-based tracking algorithms to identify and track heart sound components (e.g., S2 and S3). From each cardiac cycle, S3 can be detected based on the timing of S2 detected within that cardiac cycle, using an entropy-based method as described above. In the example, S3 can be detected within an S3 detection window that begins at the S2 timing or at a specified offset after the detected S2 (e.g., approximately 50-125 msec after the S2 timing) and lasts for a duration of approximately 125 msec. S3 metrics, such as the root mean square (RMS) of S3, can be calculated and tracked over time, as represented by the daily S3 trend. The WHF detection index can be calculated using the S3 RMS value (optionally along with other sensor information) via a weighted combining unit or other type of sensor fusion engine. In this example, S3 is tracked daily for approximately 90 days until the WHF detection index value reaches and exceeds a predetermined WHF detection threshold. At this point (time 0920), a WHF event is considered detected, and an alarm can be issued to the user. In this example, the WHF detection index may include the ratio of the intensity (e.g., amplitude or RMS value) of the S3 component to the S1 component. U.S. Patent Application No. 15 / 473,783, entitled "SYSTEMS AND METHODS FOR DETECTING WORSENING HEART FAILURE" by Pramodsingh et al., describes techniques for detecting worsening cardiac events such as WHF events using information from multiple sensors, including heart sounds, the disclosure of which is incorporated herein by reference in its entirety.

[0094] Figure 10 An example of a method 1000 for identifying and tracking heart sound components from a heart sound signal and using the identified heart sound components to detect physiological events is generally shown. Method 1000 can be implemented and performed in a mobile medical device, such as an implantable or wearable medical device, or in a remote patient management system. In the example, method 1000 can be implemented and performed by an IMD 102 or WMD 103, an external system 105, or a heart sound-based physiological event detection system 200.

[0095] Method 1000 begins at step 1010 to receive physiological information from the subject, including heart sounds over multiple cardiac cycles. Heart sounds can be detected using sensors associated with or included in a mobile or wearable device. In some examples, endocardial acceleration signals sensed from inside the heart can be used to analyze heart sounds. Other physiological information may also be received, among others, including cardiac electrical signals such as an electrocardiogram (ECG) or electrocardiogram (EGM), heart rate, signals indicating cardiac mechanical activity, including, for example, thoracic or cardiac impedance signals, arterial pressure signals, pulmonary artery pressure signals, left atrial pressure signals, RV pressure signals, LV pressure signals, heart sounds or endocardial acceleration signals, physiological responses to activity, apnea-hypopnea index, and one or more respiratory signals, such as respiratory rate signals or tidal volume signals.

[0096] At step 1020, at least a portion of the received heart sounds can be used to generate a representative heart sound segment. An example of a representative heart sound segment is the average of multiple heart sound segments taken from various cardiac cycles and time-aligned relative to a corresponding reference point (such as the R wave in multiple cardiac cycles of an ECG signal). In some examples, a subset of heart sound segments that meet specified criteria can be selected and used to establish the representative heart sound segment. Selection criteria may include, among others, one or more of heart rate, heart rate variability, activity level, respiratory rate, or time of day. In one example, a subset of heart sound segments may be selected to correspond to heart rates below a predetermined or user-set heart rate threshold (e.g., 100 bpm–120 bpm). In another example, a subset of heart sound segments may be selected to correspond to heart rates or cycle lengths falling within specified ranges, such as substantially the same cycle length within a margin of + / - 100 milliseconds, or in one example + / - 5 beats per minute (bpm), or in another example substantially the same heart rate within a specific heart rate variability margin of + / - 10 bpm.

[0097] In some examples, heart sounds within multiple cardiac cycles can be ordered according to a specific order of the instantaneous heart rate of the heartbeats corresponding to the heart sounds. From the ordered multiple heart sounds, at least the first group of heart sounds corresponding to the first heartbeat is determined by {HS}. HR1} represents, and the second set of heart sounds corresponding to the second heartbeat (with a heart rate different from the first heartbeat), by {HS HR2} indicates that it can be identified. Morphological similarity between heart sounds can be assessed separately for each group within the group, including, for example, the first group of heart sounds {HS}. HR1 The first morphological similarity measure between} and the second group of heart segments {HS} HR2A second morphological similarity measure between {HS}. This can be based on one or more of the first or second morphological similarity measures in the first group {HS}. HR1} and the second group {HS HR2 Choose between them, such as based on the above regarding Figure 3 and Figure 4 The method described.

[0098] At step 1030, a time-based heart sound tracking algorithm (such as implemented in time-based tracker 224) can be used to identify the heart sound component of interest (e.g., S2) from representative heart sound segments. The time-based tracking algorithm uses amplitude and timing information of the heart sound segment among other information. An example of a time-based tracking algorithm to be used is described in U.S. Patent No. 7,853,327 entitled "HEART SOUND TRACKING SYSTEM AND METHOD" by Patangay et al., which describes identifying heart sound components from heart sound waveforms using heart sound information from a particular heart sound waveform and heart sound information from at least one other heart sound waveform, using a first intratone energy indicator and a corresponding first intratone time indicator.

[0099] At step 1040, a performance index of the time-based heart sound tracking algorithm can be evaluated, such as using the heart sound tracking evaluation circuit 226. The evaluation can be based on the morphological or timing information of the identified heart sound components, as described above regarding... Figures 5A to 5B and Figures 6A to 6B As described above. One method, as mentioned above... Figures 5A to 5B The description relates to the use of heart sound morphological similarity measures and heart sound tracking stability. Heart sound morphological similarity measures are a measure of the similarity between at least a subset of heart sound segments (such as selected groups of heart sound segments, e.g., a first group {HS1} or a second group {HS2}) used to generate representative heart sound segments. Heart sound tracking stability is a measure of the stability of the heart sound component of interest (e.g., S2) over time at different heart rates. In the example of identifying the S2 heart sound, the variability of the aortic stenosis (AS) index can be used to quantify heart sound tracking stability, which can be calculated using the heart sound-based Gorlin formula as described above.

[0100] The above about Figures 6A to 6BAnother approach to evaluating the performance of time-based heart sound tracking algorithms is described, involving determining a consistency metric between a heart rate trend and a heart sound timing trend. The heart rate trend represents the trend of the instantaneous heart rate of a heartbeat corresponding to a selected group of heart sound segments. The heart sound timing trend represents the timing trend of a heart sound component of interest (such as S2) detected from the selected group of heart sound segments relative to a corresponding baseline. Both the heart rate trend and the heart sound timing trend can be computed using at least a subset of heart sound segments used to generate representative heart sound segments (such as selected groups of heart sound segments (e.g., a first group {HS1} or a second group {HS2})). The consistency metric indicates the degree of consistency between changes in heart rate and heart sound component timing over time. In one example, the consistency metric can be computed using the correlation between the heart rate trend and the heart sound timing trend. In another example, linear or nonlinear regression analysis can be used to generate a first regression line or curve for heart rate and a second regression line or curve for heart sound component timing (e.g., S2 timing), and the consistency metric can be determined based on the degree of deviation between the first and second regression lines or curves.

[0101] At step 1050, based on the evaluated performance index, a decision or recommendation can be made to the user regarding whether to switch to a spectrum-based heart sound tracking algorithm to identify heart sound components from representative heart sound segments. About Figure 5B An example method is described that uses heart sound morphological similarity metrics and heart sound tracking stability to evaluate the performance of a time-based S2 recognition algorithm and determine whether to switch to a spectrum-based tracker for S2 recognition. (About...) Figure 6B An example method is described to evaluate the performance of a time-based S2 recognition algorithm and determine whether to switch to a spectrum-based tracker to recognize S2 heart sounds by using the consistency between heart rate trends and heart sound component timing (e.g., S2 timing) trends.

[0102] At step 1060, one or more heart sound components may be used to detect cardiac events, such as using a physiological event detector 227. Physiological events may include indicators of myocardial contractility, cardiac synchronicity and cardiac hemodynamics, cardiac arrhythmias, or worsening heart failure (WHF) events. (The above is about...) Figure 9Examples of detecting WHF events are described, at least in part, based on heart sound components (e.g., S2 and S3 components) identified and tracked according to various examples described in this document. In some examples, heart sound measurements can be used in conjunction with other sensor information to detect medical conditions such as respiratory, renal, and neurological conditions based on heart sound measurements generated from heart sound segments. In some examples, recommendations can be generated and provided to the user based on the detection of physiological events. These recommendations may include one or more of the following: further diagnostic tests to be performed, adjustments to one or more parameters used to detect the physiological event, or adjustments to one or more treatment parameters to be delivered. The system user can review and adjudicate the detected physiological event and reprogram one or more detection or treatment parameters, such as using user interface 230. In some examples, treatment can be delivered to the patient in response to the detected physiological event, such as via... Figure 2 The treatment circuit 240 is shown. Examples of treatment may include electrical stimulation therapy delivered to the heart, nerve tissue, other target tissues, cardioversion therapy, defibrillation therapy, or pharmaceutical therapy involving the delivery of medicines to tissues or organs. In some examples, existing treatments or treatment plans may be modified to treat detected arrhythmias, such as modifying patient follow-up schedules or adjusting stimulation parameters or drug dosages.

[0103] Figure 11 A block diagram of an example machine 1100 on which any one or more of the techniques (e.g., methodologies) discussed herein can be generally illustrated. Parts of this description can be applied to the computational framework of various parts of IMD 102, WMD 103, external system 105, or the heart sound-based physiological event detection system 200.

[0104] In alternative embodiments, machine 1100 may operate as a standalone device or may be connected (e.g., networked) to other machines. In a networked deployment, machine 1100 may operate as a server machine, a client machine, or both in a server-client network environment. In the example, machine 1100 may act as a peer machine in a peer-to-peer (P2P) (or other distributed) network environment. Machine 1100 may be a personal computer (PC), tablet PC, set-top box (STB), personal digital assistant (PDA), mobile phone, web appliance, network router, switch, or bridge, or any machine capable of executing instructions (sequentially or otherwise) specifying actions to be taken by that machine. Furthermore, although only a single machine is shown, the term "machine" should also be considered to include any collection of machines, such as cloud computing, software as a service (SaaS), and other computer cluster configurations, that individually or jointly execute a set (or more) of instructions to perform any one or more of the methods discussed herein.

[0105] Examples, as described herein, may include logic or multiple components or mechanisms, or those that can be operated by them. A circuit group is a collection of circuits implemented in a tangible entity, which includes hardware (e.g., simple circuits, gates, logic, etc.). Circuit group members are flexible and can change over time and with the underlying hardware. A circuit group includes members that can perform a specified operation individually or in combination during operation. In the example, the hardware of the circuit group may be designed invariably to perform a specific operation (e.g., hardwired). In the example, the hardware of the circuit group may include physically connected components (e.g., execution units, transistors, simple circuits, etc.) and computer-readable media that are physically modified (e.g., magnetic ground, electrical ground, movable placement of invariant aggregate particles, etc.) to encode instructions for a specific operation. When the physical components are connected, the underlying electrical characteristics of the hardware composition change, for example, from an insulator to a conductor, or vice versa. Instructions enable embedded hardware (e.g., execution units or loading mechanisms) to create members of the circuit group in the hardware via variable connections to perform portions of a specific operation during operation. Thus, the computer-readable media is communicatively coupled to other components of the circuit group members during device operation. In the example, any one of the physical components can be used in more than one member of more than one circuit group. For example, under operation, the execution unit can be used in the first circuit of the first circuit group at one point in time, and reused at different times by the second circuit of the first circuit group or by the third circuit of the second circuit group.

[0106] Machine (e.g., computer system) 1100 may include a hardware processor 1102 (e.g., a central processing unit (CPU), graphics processing unit (GPU), hardware processor core, or any combination thereof), main memory 1104, and static memory 1106, some or all of which may communicate with each other via interconnect (e.g., bus) 1108. Machine 1100 may also include a display unit 1110 (e.g., raster display, vector display, holographic display, etc.), an alphanumeric input device 1112 (e.g., keyboard), and a user interface (UI) navigation device 1114 (e.g., mouse). In this example, display unit 1110, input device 1112, and UI navigation device 1114 may be a touchscreen display. Machine 1100 may additionally include a storage device (e.g., drive unit) 1116, a signal generation device 1118 (e.g., speaker), a network interface device 1120, and one or more sensors 1121, such as a global positioning system (GPS) sensor, a compass, an accelerometer, or other sensors. Machine 1100 may include output controller 1128, such as serial (e.g., universal serial bus, USB), parallel, or other wired or wireless (e.g., infrared, near field communication, NFC, etc.) connection, to communicate with or control one or more peripheral devices (e.g., printer, card reader, etc.).

[0107] Storage device 1116 may include machine-readable medium 1122 thereon storing one or more sets of data structures or instructions 1124 (e.g., software) embodying or utilized by any one or more of the techniques or functions described herein. Instructions 1124 may also reside wholly or at least partially within main memory 1104, static memory 1106, or hardware processor 1102 during execution by machine 1100. In this example, one or any combination of hardware processor 1102, main memory 1104, static memory 1106, or storage device 1116 may constitute a machine-readable medium.

[0108] Although machine-readable medium 1122 is shown as a single medium, the term "machine-readable medium" can include a single medium or multiple media (e.g., a centralized or distributed database and / or associated caches and servers) configured to store one or more instructions 1124.

[0109] The term "machine-readable medium" can include any medium capable of storing, encoding, or carrying instructions executed by machine 1100 and causing machine 1100 to perform any one or more of the techniques of this disclosure, or any medium capable of storing, encoding, or carrying data structures used by or associated with those instructions. Examples of non-limiting machine-readable media can include solid-state memory as well as optical and magnetic media. In examples, mass-capacity machine-readable media includes machine-readable media with a plurality of particles having invariant (e.g., rest) mass. Therefore, mass-capacity machine-readable media are not transient propagating signals. Specific examples of mass-capacity machine-readable media can include: non-volatile memory, such as semiconductor memory devices (e.g., electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EESOM)) and flash memory devices; magnetic disks, such as internal hard disks and removable hard disks; magneto-optical disks; and CD-ROM and DVD-ROM discs.

[0110] Instruction 1124 can be further transmitted or received on communication network 1126 via network interface device 1120 using a transmission medium, utilizing any of a variety of transport protocols (e.g., Frame Relay, Internet Protocol (IP), Transmission Control Protocol (TCP), User Datagram Protocol (UDP), Hypertext Transfer Protocol (HTTP), etc.). Example communication networks may include, among others, local area networks (LANs), wide area networks (WANs), packet data networks (e.g., the Internet), mobile phone networks (e.g., cellular networks), plain old telephone (POTS) networks, and wireless data networks (e.g., the Institute of Electrical and Electronics Engineers (IEEE) 802.11 standard family known as WiFi®, the IEEE 802.16 standard family known as WiMax®), the IEEE 802.15.4 standard family, and peer-to-peer (P2P) networks. In the example, network interface device 1120 may include one or more physical jacks (e.g., Ethernet, coaxial, or telephone jacks) or one or more antennas for connection to communication network 1126. In the example, network interface device 1120 may include multiple antennas for wireless communication using at least one of single-input multiple-output (SIMO), multiple-input multiple-output (MIMO), or multiple-input single-output (MISO) technologies. The term "transmission medium" should be considered to include any intangible medium capable of storing, encoding, or carrying instructions executed by machine 1100, and includes digital or analog communication signals or other intangible media to facilitate communication of such software.

[0111] Various embodiments are illustrated in the figures above. One or more features from one or more of these embodiments may be combined to form other embodiments.

[0112] The method examples described herein can be implemented, at least partially, on a machine or computer. Some examples may include a computer-readable or machine-readable medium encoded with instructions operable to configure an electronic device or system to perform the methods described in the examples above. Implementations of such methods may include code, such as microcode, assembly language code, or high-level language code or similar code. Such code may include computer-readable instructions for performing various methods. The code may form part of a computer program product. Furthermore, the code may be tangibly stored on one or more volatile or non-volatile computer-readable media during execution or at other times.

[0113] The detailed description above is intended to be illustrative and not restrictive. Therefore, the scope of this disclosure should be determined by reference to the appended claims together with the full scope of the equivalents to which such claims are entitled.

Claims

1. A medical device system, comprising: A data receiver circuit is configured to receive heart sound information over multiple cardiac cycles; as well as The heart sound recognition circuit is configured as follows: Use at least a portion of the received heart sound information to generate representative heart sound segments within the cardiac cycle; A time-based heart sound tracking algorithm was used to identify heart sound components from the representative heart sound segments; The performance index of the time-based heart sound tracking algorithm is evaluated using morphological or timing information of the identified heart sound components. and Based on the evaluated performance index, determine whether to switch to a spectrum-based heart sound tracking algorithm to identify the heart sound components from the representative heart sound segments.

2. The medical device system according to claim 1, wherein, The time-based heart sound tracking algorithm uses the amplitude and timing information of the representative heart sound segments to identify the heart sound components.

3. The medical device system according to any one of claims 1 to 2, wherein, The spectrum-based heart sound tracking algorithm uses one or more spectral entropy values ​​calculated from the representative heart sound segments to identify the heart sound components.

4. The medical device system according to any one of claims 1 to 3, wherein, The received heart sound information includes multiple heart sound segments corresponding to the corresponding heartbeats. The heart sound recognition circuit is configured to generate the representative heart sound segment using the overall average of at least a subset of the plurality of heart sound segments.

5. The medical device system according to claim 4, wherein, The data receiver circuit is configured to receive information about the instantaneous heart rate of the heartbeats corresponding to the plurality of heart sound segments. Wherein, a subset of at least the plurality of heart sounds used to generate the representative heart sound segments corresponds to heartbeats falling within a predetermined heart rate range.

6. The medical device system according to claim 5, wherein, In order to generate the representative heart sound segments, the heart sound recognition circuit is configured as follows: The multiple heart sounds are ordered according to a specific order of the instantaneous heart rate of the heartbeats corresponding to the multiple heart sounds; Identify at least a first group of heart sounds corresponding to a first heartbeat and a second group of heart sounds corresponding to a second heartbeat, the second heartbeat having a different heart rate than the first heartbeat, from a sorted plurality of heart sounds. Determine the first morphological similarity measure of the first group of heart sounds and the second morphological similarity measure of the second group of heart sounds; Based on one or more of the first morphological similarity measure or the second morphological similarity measure, select one group between the first group and the second group; and The representative heart sound segment is generated by using the overall average of the heart sound segments of the selected group.

7. The medical device system according to claim 6, wherein, The first group and the second group each have at least a specific minimum number of heart sounds.

8. The medical device system according to any one of claims 6 to 7, wherein, The first heartbeat corresponding to the first set of heart sounds and the second heartbeat corresponding to the second set of heart sounds each have a corresponding instantaneous heart rate that falls below the heart rate threshold.

9. The medical device system according to any one of claims 6 to 8, wherein, The first heartbeat corresponding to the first group of heart sounds and the second heartbeat corresponding to the second group of heart sounds each have a corresponding instantaneous heart rate that meets the requirements of heart rate variability or range.

10. The medical device system according to any one of claims 6 to 9, wherein, The plurality of heart sound segments are ordered in ascending order according to the instantaneous heart rate of the heartbeat. The second heartbeat corresponding to the second set of heart sounds has a higher heart rate than the first heartbeat corresponding to the first set of heart sounds. The selection between the first group and the second group includes: If (i) the first morphological similarity measure is higher than a first threshold, or (ii) both the first and second morphological similarity measures are lower than their respective thresholds, and the first morphological similarity measure is greater than the second morphological similarity measure, then the first group is selected; and If (i) the first morphological similarity measure is lower than the first threshold and the second morphological similarity measure is higher than the second threshold, or (ii) both the first morphological similarity measure and the second morphological similarity measure are lower than their respective thresholds and the second morphological similarity measure is greater than the first morphological similarity measure, then the second group is selected.

11. The medical device system according to any one of claims 4 to 10, wherein, The heart sound recognition circuit is configured to evaluate the performance index of the time-based heart sound tracking algorithm using a morphological similarity measure among subsets of at least the plurality of heart sound segments used to generate the representative heart sound segments.

12. The medical device system according to claim 11, in, Evaluating the performance index of the time-based heart sound tracking algorithm includes further using heart sound tracking stability, which indicates the timing variability of heart sound components determined from subsets of at least the plurality of heart sound segments used to generate the representative heart sound segments. In order to determine whether to switch to the spectrum-based heart sound tracking algorithm, the heart sound recognition circuit is configured as follows: If (i) the morphological similarity measure exceeds a similarity threshold, and (ii) the heart sound tracking stability is below a variability threshold, then it is determined not to switch to the spectrum-based heart sound tracking algorithm to identify the heart sound components from the representative heart sound segments; and If (i) the morphological similarity measure is not greater than the similarity threshold, or (ii) the heart sound tracking stability exceeds the variability threshold, then it is determined to switch to the spectrum-based heart sound tracking algorithm to identify the heart sound components from the representative heart sound segments.

13. The medical device system according to claim 12, wherein, The heart sound tracking stability indicator shows the timing variability of the S2 component. The heart sound recognition circuit is configured to determine an AS index indicating the presence or severity of aortic stenosis (AS) for each of a subset of at least a plurality of heart sound segments used to generate the representative heart sound segments, and to use the variability of the determined AS index to determine the heart sound tracking stability.

14. The medical device system according to any one of claims 4 to 13, in, To evaluate the performance index of the time-based heart sound tracking algorithm, the heart sound recognition circuit is configured as follows: Generate a heart sound timing trend, the heart sound timing trend comprising a time series of timing of heart sound components identified from at least a subset of the plurality of heart sound segments using the time-based heart sound tracking algorithm; Generate a heart rate trend, said heart rate trend comprising a time series of instantaneous heart rates corresponding to subsets of at least the plurality of heart sound segments; and The performance index of the time-based heart sound tracking algorithm is evaluated using a consistency metric between the heart sound timing trend and the heart rate trend. In order to determine whether to switch to the spectrum-based heart sound tracking algorithm, the heart sound recognition circuit is configured as follows: If the consistency metric falls below a threshold, it is determined not to switch to the spectrum-based heart sound tracking algorithm to identify the heart sound components from the representative heart sound segments; and If the consistency metric exceeds the threshold, then it is determined to switch to the spectrum-based heart sound tracking algorithm to identify the heart sound components from the representative heart sound segments.

15. The medical device system according to any one of claims 1 to 14, further comprising a physiological event detector configured to detect cardiac events using identified heart sound components.

Images