Multi-sensor MEMS system and machine learning based analysis method for hypertrophic cardiomyopathy prediction

Abstract

Exemplary methods are disclosed that can be used in the diagnosis of hypertrophic cardiomyopathy (HCM) using a biophysical sensor system configured to non-invasively and simultaneously acquire electrocardiogram, oscillogram, photoplethysmographic, and / or phonocardiogram signals (collectively referred to herein as biophysical signals) from at least the thoracic region of a subject. The acquired biophysical signals can be assessed for one or more conditions or indicators of hypertrophic cardiomyopathy, and simultaneously with other cardiac diseases, conditions, or indicators of any thereof.

Description

[Technical field]

[0001] Related Applications This PCT international patent application claims priority to and the benefit of U.S. Provisional Patent Application No. 63 / 302,109, entitled "MEMS-SENSOR SYSTEM AND MACHINE-LEARNED ANALYSIS METHOD FOR HYPERTROPHIC CARDIOMYOPATHY ESTIMATION," filed on January 23, 2022, which is incorporated by reference in its entirety herein. [Background technology]

[0002] Hypertrophic cardiomyopathy (HCM) is a disease, typically of autosomal dominant origin, that increases the size of muscle cells in the heart wall (cardiomyocytes), causing one or more areas of the heart wall to become abnormally thickened (enlarged). It can result in diastolic and / or systolic dysfunction with clinical signs of heart failure or valvular heart disease. Approximately 100,000 people in the United States have been diagnosed with HCM, but it is estimated that more than 500,000 people in the United States may have the condition

[16] . HCM can cause sudden cardiac death in approximately 1% of the affected population per year. Its symptoms include, for example, chest pain, fainting dizziness (syncope), sensation of heart palpitations, and shortness of breath. Current diagnostic regimens for HCM include genetic testing, echocardiography, electrocardiograms (ECG / EKG), and cardiac magnetic resonance imaging (MRI). Some of these tests are costly and often require specialized equipment in dedicated rooms at a hospital or clinical center, as well as experienced technicians to perform them. These are often separately scheduled to occur days or weeks after the initial visit with the healthcare provider.

[0003] There is interest in early diagnosis and treatment of HCM, for example, to reduce or prevent the expansion of cardiomyocytes into a further hypertrophied state. Such early diagnosis and / or treatment can result in improved health and overall outcomes for patients with this condition. Early diagnosis and treatment of HCM can save lives and reduce healthcare costs by avoiding or mitigating more costly interventions and treatments.

[0004] There are also advantages in that the presence, absence, and / or severity of HCM can be systematically screened or assessed using non-invasive techniques without the use of radiation, drugs, and / or stress, regardless of what stage of the disease, more quickly and cost-effectively than the present methods allow, and that assessing (e.g., predicting and / or detecting) the presence, absence, severity, and (in some cases) localization of various diseases, pathologies or conditions in mammalian or non-mammalian organisms can be accomplished safely, at lower cost, and / or in less time than current methods and systems provide.

[0005] The methods and systems described herein address this need and may be used for a wide variety of clinical and even research needs in a wide variety of settings, from hospitals to emergency rooms, laboratories, battlefields, remote locations, at the point of care with the patient's primary care physician or other caregiver, and even at home. Summary of the Invention

[0006] Exemplary methods are disclosed that can be used to diagnose hypertrophic cardiomyopathy (HCM) using a biophysical sensor system (e.g., a multi-sensor system) configured to non-invasively and simultaneously acquire, among other things, electrocardiogram, oscillogram, photoplethysmographic, and / or phonocardiogram signals (collectively referred to herein as biophysical signals) from a subject. One or more of these signals may be collected from the patient's chest region. The acquired biophysical signals may be assessed for one or more conditions or indicators of hypertrophic cardiomyopathy and simultaneously with other diseases, conditions, or indicators of any of them. The biophysical sensor system may include MEMS-based accelerometers, transducers, or sensors for acquiring oscillographic and / or phonocardiographic signals as well as other related signals such as ballistocardiographic signals. The biophysical sensor system may include surface electrode-based acquisition circuitry or modules for directly acquiring electrocardiogram or cardiac signals. The biophysical sensor system may include a photoplethysmographic sensor or module for directly acquiring photoplethysmographic or other hemodynamic signals. The biophysical sensor system may use wired or wireless communication or may be an independent or integrated sensor. The biophysical sensor system may be operatively connected to and operate as part of a clinical evaluation system that includes an analysis system with an analysis engine configured to perform machine learning based analysis to provide one or more inferential metrics associated with the presence, absence, and / or severity of hypertrophic cardiomyopathy or conditions that may not be as detectable or understandable in other ways.The clinical evaluation system may, in some embodiments, include an additional analysis engine configured to perform additional machine learning based analysis to assess for other physiological conditions of the patient, including the presence or absence of other diseases, medical conditions, or signs of any of them, including diseases and conditions such as (i) heart failure (e.g., left or right heart failure, heart failure with preserved ejection fraction (HFpEF), heart failure with reduced ejection fraction (HFrEF)), (ii) coronary artery disease (CAD), (iii) various forms of pulmonary hypertension (PH), including but not limited to pulmonary arterial hypertension (PAH), (iv) abnormalities in left ventricular ejection (LVEF), and various other diseases or conditions. Examples of indicators of particular forms of disease that may be evaluated include elevated or abnormal left ventricular end diastolic pressure (LVEDP) as an indication of heart failure, or elevated or abnormal mean pulmonary artery pressure (mPAP) as an indication of pulmonary hypertension.

[0007] In one aspect, a method is disclosed for non-invasively estimating the presence, absence, and / or severity of hypertrophic cardiomyopathy in a mammalian subject, the method comprising: obtaining, by one or more processors, one or more biophysical signals of the patient from one or more sensors; determining, by the one or more processors utilizing at least a portion of the one or more signals, one or more values ​​associated with one or more features and / or machine learning based analyses; and determining, by the one or more processors, an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy using the plurality of values ​​associated with the plurality of features or machine learning based analyses, wherein the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy (e.g., an HCM score) is output for use in diagnosing and / or directly treating hypertrophic cardiomyopathy.

[0008] In another aspect, a method is disclosed for non-invasively estimating the presence, absence, and / or severity of hypertrophic cardiomyopathy in a mammalian subject, comprising: obtaining, by one or more processors, one or more vibrocardiogram signals (SCG signals) and / or phonocardiogram signals (PCG signals) from a multi-sensor device placed or worn by a patient; determining, by the one or more processors utilizing at least a portion of the one or more vibrocardiogram signals and / or phonocardiogram signals, a plurality of values ​​associated with a plurality of features or a machine learning based analysis; and determining, by the one or more processors, an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy using the plurality of values ​​associated with the plurality of features or the machine learning based analysis, wherein the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy (e.g., an HCM score) is output for use in diagnosing hypertrophic cardiomyopathy or for direct treatment of hypertrophic cardiomyopathy.

[0009] In another aspect, a method of non-invasively estimating the presence, absence, and / or severity of hypertrophic cardiomyopathy in a mammalian subject includes obtaining, by one or more processors, a first biophysical signal dataset (PPG signal) associated with a first photoplethysmographic signal and a second photoplethysmographic signal, the first biophysical dataset being acquired over a plurality of cardiac cycles of the subject; and obtaining, by the one or more processors, a second biophysical signal dataset (cardiac signal) associated with the cardiac signals, the second biophysical dataset being acquired over a plurality of cardiac cycles of the subject. a first biophysical signal and a second biophysical signal, the first biophysical signal being acquired simultaneously with the first and second biophysical signals; determining, by the one or more processors utilizing at least a portion of the first and second biophysical signals, a plurality of values ​​associated with the plurality of features or machine learning based analysis; and determining, by the one or more processors, an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy using the plurality of values ​​associated with the plurality of features or machine learning based analysis, wherein the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy (e.g., an HCM score) is output for use in diagnosing hypertrophic cardiomyopathy and / or for direct treatment of hypertrophic cardiomyopathy.

[0010] In some embodiments, the multiple feature or machine learning based analysis is configured to quantify deviations of the VD wave trajectory from a three-dimensional modeled VD wave trajectory (e.g., to evaluate high frequency and low amplitude patterns in the VD trajectory).

[0011] In some embodiments, the multiple feature or machine learning based analysis is configured to quantify beat-to-beat variability of the cardiac signal.

[0012] In some embodiments, a multi-feature or machine learning based analysis is configured to quantify variability in registered landmarks in cardiac, PPG, and / or SCG signals via point-of-care analysis and histogram analysis.

[0013] In some embodiments, multiple feature or machine learning based analyses are configured to quantify dynamic characteristics of cardiac, PPG, and / or SCG signals (e.g., Lyapunov exponent, correlation dimension, entropy, mutual information, correlation, and / or nonlinear filtering).

[0014] In some embodiments, multiple feature or machine learning based analyses are configured to quantify cardiac, PPG, and / or SCG signal properties (e.g., waveform amplitude, duration, heart rate, morphology, PPG, VPG, APG signal characteristics such as peak amplitude, peak-to-peak distance, point-to-point angle, and various ratios).

[0015] In some embodiments, the multiple feature or machine learning based analysis is configured to quantify dominant frequency components of the cardiac, PPG, and / or SCG signals using wavelet analysis.

[0016] In some embodiments, the multiple feature or machine learning based analysis is configured to quantify the power spectrum and frequency content of the cardiac, PPG, and / or SCG signals using power spectrum and coherence analysis (cross-spectral analysis).

[0017] In some embodiments, the multiple feature or machine learning based analysis is configured to quantify properties of cardiac, PPG, and / or SCG signals (e.g., atrial depolarization, ventricular depolarization, and ventricular repolarization) over loop regions in 3D phase space, their projections, and loop vectors.

[0018] In some embodiments, a multiple feature or machine learning based analysis is configured to approximate the respiratory waveform using either (i) the PPG and cardiac signals or (ii) the SCG signal to assess (1) heart rate variability, (2) respiratory rate, (3) dissimilarity features representative of the distance between the respiratory and modulation signals, and (4) squared coherence representative of the correlation between the modulation and respiratory rate signals, and the approximated respiratory waveform is used to generate delineated inhalation and exhalation portions of the SCG signal that are used for analysis, thereby for HCM assessment.

[0019] In some embodiments, the multiple feature or machine learning based analysis is configured to quantify physiological aspects of the cardiac and / or SCG signals.

[0020] In some embodiments, a plurality of feature or machine learning based analysis is configured to quantify characteristic variations in cardiac, SCG, and / or PCG signals associated with inspiration versus expiration in response to a Valsalva maneuver to identify patients with HCM.

[0021] In some embodiments, a plurality of feature or machine learning based analyses are configured to quantify characteristic variations in cardiac, SCG, and / or PCG signals associated with inspiration versus expiration for the Valsalva maneuver to identify a subset of patients with HCM who have obstructive HCM (HOCM).

[0022] In some embodiments, a multi-feature or machine learning based analysis is configured to approximate left ventricular ejection time (LVET) using one or more SCG and / or PCG signals.

[0023] In some embodiments, the plurality of features or machine learning based analysis is configured to quantify propagation characteristics (e.g., wave velocity, trajectory, trajectory frequency, flatness) of ventricular depolarization (VD) waves and / or ventricular repolarization (VR) waves in three-dimensional space.

[0024] In some embodiments, multiple features or machine learning based analyses are evaluated (i) in the inspiration region of one or more vibratory and / or phonocardiogram signals, and / or (ii) in the expiration region of one or more vibratory and / or phonocardiogram signals.

[0025] In another aspect, a device (e.g., an SCG / PCG measurement device) is disclosed, comprising: a sensor body configured to be worn or placed externally on a subject's chest region to acquire biophysical signals from the chest region, including cardiac signals; and two or more MEMS-based biophysical sensors (e.g., accelerometers: single or multi-axis), including a first MEMS-based sensor and a second MEMS-based sensor, wherein the two or more MEMS-based sensors are located within the sensor body and connected to electrodes configured to be placed on the subject, and wherein the first MEMS-based sensor and the second MEMS-based sensor generate first and second biophysical signals that are provided to an analysis system configured to evaluate a plurality of features or machine learning based analysis during operation to generate an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy (e.g., an HCM score) (e.g., for use in diagnosing hypertrophic cardiomyopathy, versus direct treatment of hypertrophic cardiomyopathy).

[0026] In another aspect, an apparatus (e.g., an SCG / PCG measurement device) is disclosed, comprising: a sensor body configured to be worn or placed externally on a subject's chest region to acquire biophysical signals from the chest region, including cardiac signals; and two or more MEMS-based accelerometers (single-axis or multi-axis), including a first MEMS-based accelerometer and a second MEMS-based accelerometer, the two or more MEMS-based accelerometers being located within the sensor body and connected to electrodes configured to be placed on the subject, the first MEMS-based accelerometer and the second MEMS-based accelerometer generating first and second oscillogram signals that are provided to an analysis system configured to evaluate a plurality of features or machine learning based analyses during operation to generate an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy (e.g., an HCM score) (e.g., for use in diagnosing hypertrophic cardiomyopathy, versus directly treating hypertrophic cardiomyopathy).

[0027] In some embodiments, the device further includes a plurality of surface electrodes configured to be positioned on a surface of the subject's chest region and to provide a plurality of cardiac signals of the subject's heart, the plurality of cardiac signals being provided to an analysis system for evaluating a plurality of features or machine learning based analysis to generate an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy.

[0028] In some embodiments, the device further includes a plurality of photoplethysmographic sensors positioned on the subject and configured to provide one or more photoplethysmographic signals, wherein the one or more photoplethysmographic signals are provided to an analysis system for evaluating a plurality of features or a machine learning based analysis to generate an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy.

[0029] In some embodiments, the first MEMS-based accelerometer is configured to be positioned adjacent to the apex region of the heart.

[0030] In some embodiments, the second MEMS-based accelerometer is configured to be positioned adjacent to the base region of the heart.

[0031] In another aspect, a system is disclosed comprising any of the apparatus and analysis systems discussed above, wherein the analysis system is implemented in a cloud-based processing and networking infrastructure.

[0032] In another aspect, a system (e.g., a cloud platform or a local computing platform) is disclosed that includes one or more processors and one or more memories each storing instructions, where execution of the instructions by the one or more processors causes the one or more processors to perform any one of the methods discussed above.

[0033] In another aspect, a non-transitory computer-readable medium is disclosed that includes instructions stored thereon, the execution of the instructions by one or more processors causing the one or more processors to perform any one of the methods discussed above.

[0034] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate the embodiments and, together with the description, serve to explain the principles of the methods and systems.

[0035] Embodiments of the invention may be better understood from the following detailed description when read in conjunction with the accompanying drawings, which are intended for illustrative purposes only and demonstrate novel and non-obvious aspects of the invention. The drawings include the following figures: [Brief description of the drawings]

[0036] [Figure 1]FIG. 1 is a schematic diagram of a clinical assessment system configured to non-invasively and simultaneously acquire electrocardiogram, oscillogram, photoplethysmographic, and / or phonocardiogram signals and evaluate the acquired biophysical signals in a machine learning based analysis to generate one or more estimated metrics associated with the presence, absence, and / or severity of hypertrophic cardiomyopathy. [Diagram 2] 1 shows diagrams of exemplary HCM physiological effects and exemplary machine learning based analyses. [Diagram 3] 1 shows a diagram of an exemplary biophysical signal acquired for assessment of hypertrophic cardiomyopathy, in accordance with an illustrative embodiment; [Figure 4A] 1 shows a diagram of an exemplary SCG / PCG measurement device, in accordance with an illustrative embodiment; [Figure 4B] 1 shows a diagram of another exemplary SCG / PCG measurement device, in accordance with an illustrative embodiment; [Figure 4C] 1 shows a diagram of yet another exemplary SCG / PCG measurement device, in accordance with an illustrative embodiment; [Figure 4D] 1 is a diagram of an exemplary measurement system configured as a wearable MEMS sensor device, in accordance with an illustrative embodiment. [Figure 4E] 4E shows an image of the example fabricated wearable MEMS sensor device of FIG. 4D configured as a wearable chest sensor device, in accordance with an illustrative embodiment. [Figure 5A] FIG. 1 shows a schematic diagram of an exemplary clinical evaluation system configured to use machine learning-based analysis (among other analyses) to generate one or more metrics associated with a patient's physiological status, including, for example, the presence, absence, and / or severity of HCM, another condition, or a symptom and / or the severity of any thereof, according to an illustrative embodiment. [Figure 5B] 5B shows a schematic diagram of the operation of the exemplary clinical evaluation system of FIG. 5A, in accordance with an illustrative embodiment. [Figure 6A] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. [Figure 6B] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. [Figure 6C] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. [Figure 6D] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. [Figure 6E] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. [Figure 6F] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. [Figure 6G] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. [Figure 6H] 13 shows experimental results of various developed ML features for determining HCM physiological effects, according to an illustrative embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0037] Each and every feature described in this specification, and each and every combination of two or more of such features, is included within the scope of the present invention, provided that the features included in such combination are not mutually inconsistent.

[0038] definition As used herein, the terms "subject" and "patient" are used interchangeably to generally refer to people who undergo the analyses performed by the exemplary systems and methods.

[0039] The term "biophysical signal" as used herein includes, but is not limited to, one or more cardiac signals, neural signals, oscillogram signals, ballistocardiographic signals, and / or photoplethysmographic signals, phonocardiogram signals, and / or oscillogram signals, but it also more broadly encompasses any physiological signal from which information may be obtained. By way of example and not intended to be limiting, biophysical signals may be classified into types or categories that may include, for example, electrical (e.g., certain cardiac and nervous system related signals that may be observed, identified, and / or quantified by techniques such as measurements of voltage / potential (e.g., biopotential), impedance, resistivity, conductivity, current, etc. in various domains such as time and / or frequency), magnetic, electromagnetic, optical (e.g., signals that may be observed, identified, and / or quantified by techniques such as reflectance, interferometry, spectroscopy, absorbance, transmittance, visual observation, photoplethysmography, and the like), acoustic, chemical, mechanical (e.g., signals related to fluid flow, pressure, motion, vibration, displacement, strain), thermal, and electrochemical (e.g., signals that may be correlated to the presence of certain analytes such as glucose). In some cases, biophysical signals may be described in the context of a physiological system (e.g., respiratory system, circulatory system (circulatory system, pulmonary system), nervous system, lymphatic system, endocrine system, digestive system, excretory, muscular system, skeletal system, renal / urinary / excretory system, immune system, integumentary / excretory system, and reproductive system), one or more organ systems (e.g., signals that may be unique when the heart and lungs work together), or in the context of tissues (e.g., muscle, fat, nerve, connective tissue, bone), cells, organs, molecules (e.g., water, proteins, fats, carbohydrates, gases, free radicals, organic ions, minerals, acids, and other compounds), elements, and subatomic components thereof. Unless otherwise specified, the term "biophysical signal acquisition" generally refers to any passive or active means of acquiring biophysical signals from a physiological system, such as a mammalian or non-mammalian organism. Passive and active biophysical signal acquisition generally refers to the observation of natural or induced electric, magnetic, optical, and / or acoustic radiation of body tissues.Non-limiting examples of passive and active biophysical signal acquisition means include, for example, voltage / potential, current, magnetic, optical, acoustic, and other non-active ways of observing natural radiation in body tissues, and in some cases inducing such radiation. Non-limiting examples of passive and active biophysical signal acquisition means include, for example, ultrasound, radio waves, microwaves, infrared and / or visible light (e.g., for use in pulse oximetry or photoplethysmography), visible light, ultraviolet light, and other ways of actively interrogating body tissues that do not include ionizing energy or radiation (e.g., X-rays). Active biophysical signal acquisition may include excitation emission spectroscopy (e.g., including excitation emission fluorescence). Active biophysical signal acquisition may also include transmitting ionizing energy or radiation (e.g., X-rays) (also referred to as "ionizing biophysical signals") to body tissues. Passive and active biophysical signal acquisition means may be performed in conjunction with invasive procedures (e.g., via surgery or invasive radiological intervention protocols) or non-invasive (e.g., via imaging, ablation, cardiac contraction modulation (e.g., via a pacemaker), catheter placement, etc.).

[0040] The term "cardiac signal" as used herein refers to one or more signals directly or indirectly associated with the structure, function, and / or activity of the cardiovascular system, including, for example, the electrical / electrochemical conduction aspects of that signal that cause contraction of the cardiac muscle. Cardiac signals, in some embodiments, may include biopotential or electrocardiographic signals, such as electrocardiograms (ECGs), cardiac waveforms, and photoplethysmographic waveforms, or those acquired via signal capture or recording instruments described later herein, or other modalities. In some embodiments, cardiac signals are acquired as orthogonal voltage gradient (OVG) signals.

[0041] The term "photoplethysmographic signal" as used herein refers to one or more signals or waveforms obtained from an optical sensor corresponding to measured changes in light absorption by oxygenated and deoxygenated hemoglobin, such as light having wavelengths in the red and infrared spectrum. Photoplethysmographic signal(s), in some embodiments, includes raw signal(s) obtained via a pulse oximeter or photoplethysmogram (PPG). In some embodiments, photoplethysmographic signal(s) are obtained from commercially available, custom, and / or proprietary equipment or circuitry configured to obtain such signal waveforms for purposes of monitoring health and / or diagnosing disease or abnormal conditions. The photoplethysmographic signal(s) typically include red photoplethysmographic signals (e.g., electromagnetic signals within the visible light spectrum having wavelengths most predominantly between about 625 and 740 nanometers) and infrared photoplethysmographic signals (e.g., electromagnetic signals extending up to about 1 mm from the nominal red edge of the visible spectrum), although other spectra, such as near-infrared, blue, and green, may be used in different combinations depending on the type and / or mode of PPG being employed.

[0042] As used herein, the term "ballistocardiographic signal" refers to a signal or group of signals that generally reflect blood flow through the entire body, which may be observed through vibration, sound, movement, or orientation, for example, using accelerometers or transducers, such as electromechanical systems-based (MEMS) accelerometers. In other embodiments, ballistocardiographic signals may be acquired by external devices, such as bed- or surface-based devices, that measure phenomena such as changes in body weight as blood moves back and forth longitudinally between the head and feet. In such embodiments, the amount of blood at each location may change dynamically, reflected in the weight measured at each location on the bed as well as the rate at which that weight changes.

[0043] As used herein, the term "seismocardiogram signal" refers to a signal or group of signals that generally reflect recorded body vibrations, sounds, or orientations as recorded by a sensor mounted or positioned near the heart, e.g., a MEMS sensor such as a MEMS accelerometer. The term "seismocardiogram signal" is used interchangeably with "seismocardiographic signal."

[0044] As used herein, the term "phonocardiogram signal" refers to a signal or group of signals that generally reflect recorded body sounds, vibrations, and acoustic emissions, for example, as recorded by a microphone or accelerometer sensor mounted or positioned near a subject's heart or about the subject's chest region. "Phonocardiogram signal" is used interchangeably with "phonocardiographic signal."

[0045] Exemplary System FIG. 1 is a schematic diagram of a clinical evaluation system 100 configured to non-invasively and simultaneously acquire electrocardiogram (heart) signals, one or more oscillogram (SCG) signals, photoplethysmographic (PPG) signals, and / or phonocardiogram (PCG) signals, and evaluate the acquired biophysical signals in a machine learning-based analysis to generate one or more estimated metrics associated with the presence, absence, and / or severity of hypertrophic cardiomyopathy. System 100 includes a measurement system 102 and an analysis system 104 via a network 106. In the example shown in FIG. 1, analysis system 104 is implemented in a cloud-based infrastructure. Analysis system 104 includes (i) a data store 108 for receiving one or more data files 114 associated with measurements performed by measurement system 102, and (ii) an analysis engine including an analysis feature analysis module 110 (denoted as “features” 110) and a machine learning classifier module 112 for evaluating the one or more data files 114 in the machine learning-based analysis. Output 116 of analysis system 104 may be provided as a patient report, for example via a healthcare portal, or as output to a wearable device, or to a medical device for treatment of a disease, condition, or any symptom thereof. A healthcare provider, e.g., a physician, can review the report and interpret it to provide a diagnosis of the disease or generate a treatment plan.

[0046] Machine learning based analysis refers to analysis or features that include or are derived from machine learning or artificial intelligence analysis. Machine learning based analysis, in some embodiments, includes evaluation of features from a library of features to downselect or train clinical data, for example, via ElasticNet machine learning classifier models [9], RandomForestClassifier machine learning classifier models

[10] , and extreme gradient boosting (XGB) classifier models

[11] , to features that are statistically significant in estimating a metric associated with the presence, absence, and / or severity of hypertrophic cardiomyopathy. An example of a training system (e.g., for increasing LVEDP estimates) for configuring an analysis system is described in U.S. Provisional Patent Application No. 63 / 235,960, filed August 23, 2021, entitled "METHOD AND SYSTEM TO ASSESS HEART FAILURE," which is incorporated herein by reference in its entirety. Another example of a training system that may be used to configure an analysis engine is described in

[19] , which is incorporated herein by reference in its entirety.

[0047] HCM Physiological Effects and Indicators Figure 2 shows an illustration of an exemplary HCM physiological effect and an exemplary machine learning-based analysis. Hypertrophic cardiomyopathy (HCM) is a genetic disorder characterized by left ventricular hypertrophy unexplained by secondary causes and a nondilated left ventricle, typically with preserved or elevated ejection fraction. It is commonly asymmetric, and the most severe hypertrophy involves the basal interventricular septum. Left ventricular outflow tract obstruction is present at rest in approximately one-third of patients and can be induced in another third

[21] . Pathological features of the disease include cardiomyocyte hypertrophy and disorganization, interstitial and replacement fibrosis, small vessel abnormalities, and electrical remodeling of cardiomyocytes that may form a substrate for ventricular arrhythmias. Hypertrophy is also frequently associated with left ventricular diastolic dysfunction

[22] . In addition, a high prevalence of sinus node dysfunction (66%) and His-Purkinje (HV) conduction (30%) was observed in HCM patients with electrophysiological abnormalities

[23] . The most commonly induced supraventricular arrhythmias are atrial reentrant tachycardia and atrial fibrillation (10% and 11% of patients, respectively).

[0048] In HCM patients, the heart wall(s) (e.g., 202a, 202b) become enlarged (hypertrophy) due to enlarged cardiomyocytes (hypertrophy), resulting in an increase in size, mass, and stiffness (or loss of compliance) of the heart wall. HCM hearts can exhibit different topologies. In some cases, the thickening can be asymmetric, for example involving only the septum between the left and right ventricles (202a), or there can be a concentric increase in thickness around the periphery of the left ventricular wall (202b).

[0049] Dilation Effect (202'). When the septal wall (204) (shown as interventricular septum 204) is dilated such that it encroaches on the left ventricular outflow tract (see 206), it can create a physiological scenario similar to an aortic valve blockage that blocks blood flow from the left ventricle (208) into the aorta (201). This blockage or obstruction of flow (203') is observed as turbulence through the narrowed left ventricular outflow tract (LVOT), causing a "buzzer" (obstruction murmur) or sound and vibrations similar to all the physiological effects of aortic valve stenosis (narrowing the outlet of the left ventricle of the heart).

[0050] In addition to this flow or obstruction blockage (203'), impingement of the left ventricular chamber (208) by the septal wall (204) can also affect the function of the mitral valve (210), which can cause mitral regurgitation (205') (a condition in which the mitral valve, which comprises two cusps or flaps and is located between the left atrium (212) and the left ventricle (208) of the heart, does not close tightly, allowing blood to flow backwards within the left atrium (212). As a result of this systolic anterior motion (SAM) of the mitral valve (210), the anterior mitral leaflet can be pulled towards the septum (214), rendering the mitral valve (210a) ineffective. Leakage (205') across the mitral valve (e.g., 210a) can cause cardiac enlargement and clinical signs of heart failure, similar to aortic stenosis. Mitral regurgitation (205') with associated leakage across the mitral valve often causes turbulence, heart murmurs ("mitral regurgitation murmurs"), and vibrations.

[0051] Altered tissue stiffness (207'). The ventricular muscle (202b) is thickened and stiff, with poor pumping properties, which can result in a physiology similar to heart failure with preserved ejection fraction (HFpEF), where the heart can contract well and completely, but cannot relax properly. HfpEF is characterized by abnormalities in diastolic function, where increased stiffness of the left ventricular wall (202b) during diastole causes decreased left ventricular relaxation, resulting in increased pressure and / or impaired filling. In other cases, patients with HCM have both impaired diastolic function and reduced systolic function. All of these effects (loss of cardiac tissue compliance (207'), obstruction / impaired flow (203'), valve leakage (205')) can impair systolic and / or diastolic function, cause pulmonary congestion, and may cause shortness of breath and fainting (a temporary loss of consciousness usually associated with insufficient blood flow to the brain), among other symptoms.

[0052] The reduced left ventricular relaxation during diastole is essentially a difficulty filling (209') of the left ventricle (208) (not a difficulty in ventricular contraction). This may be modeled or thought of as a muscle-constrained heart that is hindered in its ability to relax sufficiently to allow blood to flow in, thus resulting in increased left atrial pressure (representing pulmonary venous pressure), increased left ventricular pressure (LVP) (pressure within the left ventricle), and increased diastolic pressure (pressure that blood exerts on the arterial walls as the heart relaxes between beats). In order to obtain proper filling, the heart may increase pulmonary venous pressure (211'), which may cause congestion (accumulation of fluid in the lungs) and shortness of breath, among other effects. The effect of the heart muscles being able to squeeze properly but not being able to relax properly to fill for the next heartbeat may also cause the body to compensate by increasing arterial and venous blood pressure, which may also cause shortness of breath and other conditions.

[0053] All of these effects (loss of compliant cardiac tissue, blockage / flow obstruction, valve leakage, loss of cardiac tissue compliance) can affect the cardiac voltage distribution and pattern compared to a normal heart. Due to changes in cardiac geometry (asymmetric or symmetric), the cardiac behavior changes over time and can be indicated by changes in the left ventricular ejection time (LVET) (213'). LVET measures the duration of blood flow through the aortic valve and has a normal value of 0.35±0.08 seconds.

[0054] Additionally, all of these effects (loss of cardiac tissue compliance, blockage / flow obstruction, valve leakage) can affect the observable components of cardiac frequency. For example, cardiac frequency components (e.g., as observed through power spectral density or wavelet frequency components, among others) can vary with respiratory frequency. During inspiration, there is more cardiac filling compared to expiration, as transthoracic pressure during inspiration can increase ventricular filling pressure. Transthoracic pressure can be viewed as a hemodynamic pressure in combination with the negative intrathoracic pressure from inspiration.

[0055] Additionally, there may be observable phasic components in the sounds due to loss of cardiac tissue compliance, blockage / flow obstruction, and valve leakage. With deep inspiration, the heart may fill more and cause less obstruction of the left ventricular outflow, whereas with expiration, there may be less filling and more obstruction of the outflow tract.

[17] Heart murmurs may be heard decreasing in amplitude during inspiration and increasing in amplitude with expiration. Similarly, vibrations detected as vibrogram and / or phonocardiogram signals may decrease in amplitude during inspiration and increase in amplitude with expiration. Similarly, vibrations detected as vibrogram and / or phonocardiogram signals may increase in amplitude through a physiological maneuver known as the Valsalva maneuver. The Valsalva maneuver is a breathing protocol that may be performed by a forceful attempt of expiration against a closed airway, usually performed by closing the mouth and pinching the nose while expelling air as if inflating a balloon.

[0056] Indeed, as noted above, the physical and HCM physiological effects discussed above can generate vibrations, electrical patterns, and frequency components indicative of or unique to HCM and its developing conditions. These physical manifestations can be evaluated by the signature or machine learning-based analyses described herein to estimate metrics associated with the presence, absence, and / or severity of hypertrophic cardiomyopathy.

[0057] In addition, the presence of ventricular tachyarrhythmias may be an indication of the presence of hypertrophic cardiomyopathy. Ventricular tachycardia (V-tach or VT) may be characterized as a fast heart rate arising from the lower chambers of the heart. Programmed ventricular stimulation (PVS) induced non-sustained ventricular tachycardia (VT) in 14% of patients and sustained ventricular arrhythmias in 43% of patients. The sustained ventricular arrhythmias were polymorphic VT in 73% of patients, monomorphic VT in 24% of patients, and ventricular fibrillation in two patients (3%).

[0058] Machine learning analysis module 1 , analysis system 104 includes an analysis engine with analysis feature analysis module 110 configured to calculate features or parameters to generate, via a classifier (e.g., a machine learning classifier), one or more estimated metrics associated with the presence, absence, and / or severity of hypertrophic cardiomyopathy. Analysis system 104 may also include additional analysis engines configured to perform additional machine learning based analyses to assess other physiological conditions of the patient, including the presence, absence, and / or severity of other diseases, medical conditions, or any signs thereof, such as (i) heart failure (e.g., left or right heart failure, heart failure with preserved ejection fraction (HfpEF), heart failure with reduced ejection fraction (HfrEF)), (ii) coronary artery disease (CAD), (iii) various forms of pulmonary hypertension (PH), including but not limited to pulmonary arterial hypertension (PAH), (iv) abnormalities in left ventricular ejection fraction (LVEF). Examples of indicators of particular forms of disease that may be assessed include the presence or absence of elevated or abnormal left ventricular end diastolic pressure (LVEDP) as an indication of heart failure, or elevated or abnormal mean pulmonary artery pressure (mPAP) as an indication of pulmonary hypertension.

[0059] Table 1 shows a list of classes of features and corresponding descriptions that may be used to calculate estimated metrics associated with the presence, absence, and / or severity of hypertrophic cardiomyopathy. The features listed below are also shown in FIG. [Table 1-1] [Table 1-2] [Table 1-3] TIFF2025503130000005.tif24170 [Table 1-4] [Table 1-5] [Table 1-6] [Table 1-7]

[0060] Detailed descriptions of some of the analyses in Table 1 that may be applied to SCG signals and for determining the presence, absence, and / or severity of HCM can be found in, among others, [1]-

[13] and

[20] -

[56] , each of which is incorporated herein by reference in its entirety.

[0061] Other feature modules that may be used are described in U.S. Pat. Nos. 9,289,150, 9,655,536, 9,968,275, 8,923,958, 9,408,543, 9,955,883, 9,737,229, 10,039,468, 9,597,021, 9,968,265, 9,910,964, 10,672,518, 10,566,091, 10,570, 10,770, 10,820, 10,990, 10,102, 10,106, 10,108, 10,110, 10,112, 10,114, 10,116, 10,118 ... No. 10,566,092, U.S. Patent No. 10,542,897, U.S. Patent No. 10,362,950, U.S. Patent No. 10,292,596, U.S. Patent No. 10,806,349, U.S. Patent Application Publication No. 2020 / 0335217, U.S. Patent Application Publication No. 2020 / 0229724, U.S. Patent Application Publication No. 2019 / 0214137, U.S. Patent Application Publication No. 2018 / 0249960, U.S. Patent Application Publication No. 2019 / 0200893, U.S. Patent Application Publication No. 2019 / 0384757, U.S. Patent Application Publication No. 2020 US Patent Application Publication No. 2019 / 0211713, US Patent Application Publication No. 2019 / 0365265, US Patent Application Publication No. 2020 / 0205739, US Patent Application Publication No. 2020 / 0205745, US Patent Application Publication No. 2019 / 0026430, US Patent Application Publication No. 2019 / 0026431, WO 2017 / 033164, WO 2017 / 221221, WO 2019 / 130272, WO 2018 / 158749, WO 2019 / 077414, WO 2019 / 130273 No. 16 / 831,380, U.S. Patent Application No. 17 / 132869, International Application No. PCT / IB2020 / 052889, and International Application No. PCT / IB2020 / 052890, each of which is incorporated herein by reference in its entirety.

[0062] The analysis and related modules (e.g., 110) may evaluate features provided in association with SCG or phonocardiogram signals in addition to PPG and cardiac signals, and may extend various analyses (such as frequency, dynamics, cycle variability as described above) to assess changes in the value of the features between the inspiration and expiration portions of the signal.

[0063] A module (eg, 110) may include an analysis for estimation of LVET or a parameter that correlates with elevated LVET.

[0064] MEMS Accelerometer Systems 1, the measurement system 102 includes a chest-worn or mounted SCG / PCG measurement device 118 configured with two or more MEMS accelerometers, transducers, or sensors to provide at least two or more SCG / PCG measurements at the apical and basal regions of the heart. The SCG / PCG device 118 may be connected to a biophysical signal capture system 120 configured to measure cardiac signals via a set of surface electrodes 122 (shown as 122a-122f) and photoplethysmographic signals via a PPG sensor device 124. Exemplary wireless communication operations and circuitry are described in

[14] , which is incorporated herein by reference in its entirety.

[0065] The SCG / PCG measurement device 118 includes a first MEMS accelerometer, transducer, or sensor 126 and a second MEMS accelerometer, transducer, or sensor 128. The two or more MEMS accelerometers, transducers, or sensors can provide oscillogram signals to the analysis system, and phase difference and differential evaluation of the signals can be performed by the analysis engine. For example, the first MEMS accelerometer, transducer, or sensor 126 can be configured to be positioned proximal to the apex 130, and the second MEMS accelerometer, transducer, or sensor 128 is configured to be positioned near or at the base region 132 of the left ventricle. In the example shown in FIG. 1, the device 118 and associated sensors 126 and 128 are non-invasively placed on the subject's skin in the illustratively shown areas. A cross-section of the heart is shown in FIG. 1 merely to provide an illustrative placement of the sensors relative to certain cardiac structures in certain embodiments.

[0066] Exemplary biophysical signals for assessing hypertrophic cardiomyopathy FIG. 3 shows a diagram of exemplary biophysical signals acquired for assessment of hypertrophic cardiomyopathy, according to an illustrative embodiment. In the example shown in FIG. 3, exemplary oscillogram, phonocardiogram, cardiac signals (shown as "ECG"), and photoplethysmographic signals are shown acquired simultaneously over one cardiac cycle. The diagram further illustrates examples of aortic pressure, atrial pressure, ventricular volume, and ventricular pressure in the same cardiac cycle. Registration / landmark points may be determined within a single signal (e.g., a peak of a cardiac waveform, oscillogram, and / or phonocardiogram signal) or a signal whose synchrony and phase may be assessed in relation to other acquired biophysical signals (e.g., two photoplethysmographic waveforms, a crossing point between two oscillogram waveforms, etc.). See, for example, U.S. Patent Publication No. 2020 / 0397324, entitled "Method and System to Assess Disease Using Dynamical Analysis of Cardiac and Photoplethysmographic Signals," which is incorporated herein by reference in its entirety.

[0067] Exemplary SCG / PCG Measurement System #1 4A shows a diagram of an exemplary SCG / PCG measurement device 118, according to an illustrative embodiment. In the example shown in FIG. 4A, the SCG / PCG measurement device 118 includes two or more MEMS accelerometers, transducers, or sensors 126, 128 (shown as 126a, 128a). In this example, the MEMS accelerometers, transducers, or sensors 126a, 128a include a three-axis accelerometer 302 (shown as 302a, 302b), such as a small, low-power ADXL335 three-axis accelerometer manufactured by Analog Devices of Wilmington MA, integrated into a small 4mm×4mm×1.45mm package IC. The three-axis accelerometers 302a, 302b are coupled to an amplifier and filter 303 to provide a vibrogram signal 304 to a cable 306 that connects to the measurement system 102. In this example, cardiac signals 308 are acquired at each of six surface electrodes 312 and carried over six conductors via cable 306 to the measurement system 102. A reference electrode is also used. In this example, a photoplethysmographic signal 310 is generated at the PPG lead snap 314 when the photoplethysmogram sensor is carried over three conductors (two signal and GND) through cable 306 to the measurement system 102. The measurement system 102 includes conversion circuitry for converting and digitizing the acquired biophysical signals. In some embodiments, the surface electrodes 312 and / or the PPG lead snap 314 may include an accelerometer, for example, as described in

[14] (A4L BCG Application), which is incorporated herein by reference in its entirety.

[0068] TIFF2025503130000010.tif97170

[0069] In some embodiments, the MEMS sensor includes an acoustic sensor or transducer, such as an acoustic respiration sensor (e.g., model. RAS-45 manufactured by Masimo, Corp., Irvine, Calif.). Another example of an acoustic-based sensor is a sensor used in digital stethoscope systems, such as an ECG+ digital stethoscope (e.g., the DUO ECG+ digital stethoscope manufactured by Eko Devices, Inc., Oakland, Calif.).

[0070] In some embodiments, the MEMS sensor includes an accelerometer-based sensor, such as a 9DoF inertial sensing including an accelerometer, a gyroscope, a magnetometer, and a pressure sensor (e.g., Shimmer3 IMU manufactured by Shimmer, Cambridge, MA). Another example of an acceleration-based sensor is an IMU sensor with a 9-axis motion sensor including an acceleration, a gyro, and a magnetometer (Movesense IMU manufactured by Movesense, Vantaa Finland). One example of an accelerometer-based sensor is an IMU tracking sensor with a 9-axis motion sensor including an acceleration, a gyro, and a magnetometer (ICM-20948 manufactured by TDK, InvenSense, San Jose, CA). One example of an accelerometer-based sensor is an IMU tracking sensor with a 9-axis motion sensor including an acceleration, a gyro, and a magnetometer (Sense Connect Detect model no. SCD110 manufactured by Bosch Connected Devices and Solutions GmbH, Germany).

[0071] These MEMS sensors (acoustic, accelerometer, IMU) may be used alone or in any combination with one or more other sensors, including but not limited to the two current sensor types described herein or other sensor types, to gather information to generate diagnostic tools for any indication of HCM, HF, PH, CAD, or combinations thereof or related conditions or indications in their various forms. Additionally, these MEMS sensors may be used to generate diagnostic tools for any indication described herein other than HCM, HF, PH, and CAD in their various forms.

[0072] Figure 4B illustrates the example SCG / PCG measurement device 118 (designated as 118a) of Figure 4A configured as a wireless measurement module that includes a wireless transceiver module 402. The wireless transceiver module 402 includes front-end conversion circuitry 403, a microcontroller 404, and wireless transceiver circuitry 406. The wireless transceiver circuitry 406 can interface via a wireless connection to an interface device.

[0073] The cardiac signal 308 and the photoplethysmographic signal 310 may be acquired using circuitry and computing hardware, software, firmware, middleware, etc. in a biophysical signal capture system as described in U.S. Pat. No. 10,542,898, entitled "Method and Apparatus for Wide-Band Phase Gradient Signal Acquisition," or U.S. Patent Application Publication No. 2018 / 0249960, entitled "Method and Apparatus for Wide-Band Phase Gradient Signal Acquisition," each of which is incorporated by reference in its entirety herein.

[0074] As discussed herein, other configurations and topologies may be used, such as MEMS microphones or acoustic transducers, among others.

[0075] Exemplary SCG / PCG Measurement System #2 4C shows a diagram of another example SCG / PCG measuring device 118 (shown as 118b) according to an illustrative embodiment. In the example shown in FIG. 4C, the SCG / PCG measuring device 118b includes a phonocardiogram device 408 (including an acoustic transducer) and a MEMS accelerometer in a wireless module (shown as 403 in FIG. 4D).

[0076] The phonocardiogram device 408 includes one or more microphones 410 and microphone front-end circuitry 412. The microphone front-end circuitry 412 includes transducer, filter, and amplifier circuitry for converting and digitizing acquired biophysical signals. The microphone 410 and microphone front-end 412 are configured to capture signatures of the heart and nearby structures and noises. In some embodiments, the phonocardiogram device 408 is configured to acquire acoustic signals having a rate of at least 8 kHz with 12-bit resolution. Other sampling rates and resolutions may be used. Other examples of the phonocardiogram device 408 include an acoustic respiratory sensor (model. RAS-45 manufactured by Masimo, Corp., Irvine, CA) or a digital stethoscope system such as an ECG+ Digital Stethoscope (e.g., the DUO ECG+ Digital Stethoscope manufactured by Eko Devices, Inc., Oakland, CA).

[0077] 4C, the SCG / PCG measurement device 118b includes a controller 414, a wireless transceiver 416, and an energy storage 418. The SCG / PCG measurement device 118b is configured as a wireless measurement module, e.g., as a wearable device as described herein, with a wireless transceiver module 416 that communicates with a base station 402 (shown as 402a) that includes a wireless transceiver 406. In this configuration, the base station 402a includes front-end conversion circuitry 403 (for acquisition of cardiac signals 308 and PPG signals 310), a microcontroller 404, and wireless transceiver circuitry 406.

[0078] TIFF2025503130000011.tif173170

[0079] These MEMS sensors (acoustic, accelerometer, IMU) may be used alone or in any combination with one or more other sensors, including but not limited to other sensor types described herein, to gather information to generate diagnostic tools for any indication of HCM, HF, PH, CAD, or combinations thereof or related conditions or indications in their various forms. Additionally, these MEMS sensors may be used to generate diagnostic tools for any indication described herein other than HCM, HF, PH, and CAD in their various forms.

[0080] The cardiac signal 308 and the photoplethysmographic signal 310 may be acquired using circuitry and computing hardware, software, firmware, middleware, etc. in a biophysical signal capture system as described in U.S. Pat. No. 10,542,898, entitled "Method and Apparatus for Wide-Band Phase Gradient Signal Acquisition," or U.S. Patent Application Publication No. 2018 / 0249960, entitled "Method and Apparatus for Wide-Band Phase Gradient Signal Acquisition," each of which is incorporated by reference in its entirety herein.

[0081] During signal acquisition, the patient may be patted (e.g., tapped firmly) to provide a spike in the measurement that can be used to synchronize the measurements between the two acquisition systems 402 and 118b.

[0082] Exemplary Wearable MEMs Sensor 4D is a diagram of an example measurement system 102 (designated as 403) configured as a wearable MEMS sensor device 403. The wearable MEMS sensor device 403 includes a housing 420 configured to house one or more electronic boards 422 including a MEMS accelerometer 302 (designated as "accelerometer" 302), a phonocardiogram device 408 (designated as "acoustic sensor" 408), a controller 414, a wireless transceiver 416 (integrated with 414), and energy storage 418, for example, as described with respect to FIG.

[0083] In the example shown in FIG. 4D, the housing 420 has an outer wall 424 and an inner wall 426. The inner wall 426 has a tapered region 427 that defines and focuses sound at an entrance region 428 of the housing 420 (see FIG. 431) to the phonocardiogram device 408. The phonocardiogram device 408 is located on a portion of the electronic board 422 that faces the entrance region 428. The entrance region 428 forms a volume defined by the housing surface and the body to direct and maintain acoustic energy within the volume. The housing 420 further includes an elastomeric member 430 that can function as a gasket type interface to adhere and maintain contact (434) with the body (shown as 432). FIG. 436 (see also FIG. 4E) is an image of an exemplary fabricated wearable MEMS sensor device 403.

[0084] FIG. 4E shows an image of the exemplary fabricated wearable MEMS sensor device 403 of FIG. 4D configured as a wearable chest sensor device. In FIG. 4E, FIG. 438 shows another example of an exemplary fabricated wearable MEMS sensor device 403 having a front plate 440 disposed on the entrance area 428. FIG. 438 shows an exemplary placement of the wearable MEMS sensor device 403 on a person. As shown in FIG. 438, the wearable MEMS sensor device 403 is placed with the entrance area 428 directed toward the person, and may be placed over the heart (as shown) and at various locations on the body as described herein. FIG. 438 also shows surface electrodes 312. The electrodes 312 are each an ECG pad 442 coupled to a clip connector 444.

[0085] Diagram 440 shows a wearable MEMS sensor device 403 arranged on a wirelessly rechargeable station. An example arrangement is described in U.S. Patent No. 10,542,898, which is incorporated herein by reference in its entirety.

[0086] Exemplary HCM Treatments Following generation of an estimate of a metric associated with the presence or absence of hypertrophic cardiomyopathy, the generated estimate may be provided and used for patient reporting, for example, in a healthcare portal or as output to a wearable device, or as output to a medical device for treatment of a disease, condition, or any symptom thereof.

[0087] Treatment of HCM may include drug therapy or surgery. Pharmacological treatment may include administration of beta-blockers (e.g., metoprolol, propranolol, or atenolol), calcium channel blockers (e.g., verapamil or diltiazem), cardiac rhythm medications (e.g., amiodarone or disopyramide), mabacamen, among others. Surgical intervention may include septal myectomy to remove a portion of the thickened overgrown septal wall, apical myectomy to remove thickened myocardium from near the tip of the heart, septal ablation to destroy a portion of the thickened myocardium, or an implantable cardioverter defibrillator (ICD) to continuously monitor the heartbeat.

[0088] While the present disclosure is directed to the practical assessment of biophysical signals, such as raw or pre-processed photoplethysmographic, biopotential / cardiac, oscillogram, phonocardiogram, etc. signals, in the diagnosis and treatment of cardiac-related pathologies and conditions, such assessment may be applied to the diagnosis, treatment, and tracking / monitoring of any pathology or condition in which a biophysical signal involves any relevant system of the living body, including, but not limited to, surgical, minimally invasive, lifestyle, nutritional, and / or pharmacological treatments, etc. The assessment may be used in the control of medical equipment or wearable devices, or in monitoring applications.

[0089] An exemplary biophysical sensor system may be implemented as a modular medical assessment system

[18] , which is incorporated herein by reference in its entirety.

[0090] Exemplary Clinical Evaluation System 5A illustrates an exemplary clinical evaluation system 500 (also referred to as a clinical and diagnostic system), according to one embodiment, that implements the modules of FIG. 1 to non-invasively perform machine learning based analysis to generate one or more metrics associated with a patient's or subject's HCM-related condition via a classifier (e.g., a machine-learned classifier). Indeed, the feature modules (e.g., of FIGS. 1, 5-14) may generally be considered part of a system (e.g., clinical evaluation system 500) in which any number and / or types of features may be utilized for a disease state, medical condition, any indication, or combination thereof, of interest, e.g., using different embodiments having different configurations of feature modules. This is further illustrated in Figure 5A, where the clinical evaluation system 500 is of a modular design in that disease-specific add-on modules 502 (e.g., for assessing HCM, elevated LVEDP or mPAP, CAD, PH / PAH, abnormal LVEF, hFpEF, and others described herein) can be integrated, singly or in multiple cases, with a single platform (i.e., base system 504) to achieve full operation of the system 500. The modularity allows the clinical evaluation system 500 to be designed to assess for the presence of multiple different diseases, such as disease-specific algorithms being developed, utilizing the same synchronously acquired biophysical signals and data sets and base platform, thereby reducing testing and certification time and costs.

[0091] In various embodiments, different versions of the clinical evaluation system 500 may implement the assessment system 103 (FIG. 1) by including different feature calculation modules that may be configured to indicate a given disease state(s), medical condition(s), or condition(s) of interest. In another embodiment, the clinical evaluation system 500 may include more than one assessment engine 103, which may be selectively utilized to generate different scores specific to the classifier 112 of that engine 103. In this way, the modules of FIGS. 1 and 5 in a more general sense may be considered as one configuration of a modular system, where different and / or multiple engines 103 with different and / or multiple corresponding classifiers 112 may be used depending on the desired configuration of the modules. Thus, any number of embodiments of the modules of FIG. 1 may be present.

[0092] 5A, system 500 can use machine-learned disease-specific algorithms to analyze one or more biophysical signal datasets (e.g., 110) to assess, by way of example, the likelihood of pathology or abnormal conditions, such as elevated LVEDP. System 500 includes hardware and software components designed to work together in combination to facilitate analysis and presentation of an estimated score using the algorithms, and to enable a physician to use the score to assess, for example, the presence, absence, and / or severity of a disease state, medical condition, or any symptom thereof.

[0093] The base system 504 can provide a foundation of functionality and instructions that each add-on module 502 (including disease-specific algorithms) then interfaces with to assess disease states or symptom conditions. The base system 504 includes a base analysis engine or analyzer 506, a web services data transfer API 508 (shown as "DTAPI" 508), a reporting database 510, a web portal services module 513, and a data repository 111 (shown as 111a), as shown in the example of FIG. 5A.

[0094] The data repository 111a, which may be cloud-based, stores data from the signal capture system 102 (shown as 102b). The biophysical signal capture system 102b, in some embodiments, is a reusable device designed as a single unit with a 7-channel lead set and a photoplethysmogram (PPG) sensor securely attached (i.e., non-removable). The signal capture system 102b, along with its hardware, firmware, and software, provides a user interface, collects patient-specific metadata entered therein (e.g., name, sex, date of birth, medical record number, height, weight, etc.), and synchronously acquires the patient's electrical and hemodynamic signals. The signal capture system 102b may securely transmit the metadata and signal data as a single data package directly to the cloud-based data repository. The data repository 111a, in some embodiments, is a secure cloud-based database configured to accept, store, and enable retrieval of patient-specific data packages by the analysis engine 506 or the analyzer 514.

[0095] The base analysis engine or analyzer 506 is a secure cloud-based processing tool that may perform quality assessment of the acquired signals (performed via “SQA” module 516), the results of which may be communicated to a physician or patient at the point of care. The base analysis engine or analyzer 506 may also perform pre-processing (illustrated via pre-processing module 518) of the acquired biophysical signals (e.g., 110 (see FIG. 1 )). The web portal 513 is a secure web-based portal configured to provide healthcare providers with access to patient reports. An exemplary output of the web portal 513 is shown by visualization 536. The report database (RD) 512 is a secure database that may securely interface and communicate with other systems, such as hospital or physician-hosted, remotely hosted, or remote electronic health record systems (e.g., Epic, Cerner, Allscrips, CureMD, Kareo, etc.), such that the output score(s) (e.g., 118) and related information are integrated and stored in the patient's general health record. In some embodiments, the web portal 513 is accessed by a call center to provide output clinical information via telephone. The database 512 may be accessed by other systems capable of generating reports to be delivered via mail, courier, facsimile, hand delivery, etc.

[0096] The add-on module 502 includes a second portion 514 (also referred to herein as an analysis engine (AE) or analyzer 514, and denoted as an "AE add-on module" 514) that operates in conjunction with a base analysis engine (AE) or analyzer 506. The analysis engine (AE) or analyzer 514 may include the main functional loops of a given disease-specific algorithm, such as a feature computation module 520, a classifier model 524 (denoted as an "ensemble" module 524), and an outlier assessment and rejection module 524 (denoted as an "outlier detection" module 524). In certain modular configurations, the analysis engines or analyzers (e.g., 506 and 514) may be implemented within a single analysis engine module.

[0097] The main functional loop may include instructions for (i) validating the execution environment to ensure that all necessary environmental variable values ​​are present, and (ii) executing an analysis pipeline that analyzes new signal capture data files containing acquired biophysical signals to calculate a patient score using disease-specific algorithms. To execute the analysis pipeline, the AE add-on module 514 may include and execute instructions for the various feature modules 110 and classifier modules 112 as described in connection with FIG. 1 to determine an output score (e.g., 116) of a metric associated with the patient's physiological condition. The analysis pipeline within the AE add-on module 514 may calculate features or parameters (denoted as "feature calculation" 520) and identify whether the calculated features are outliers by providing an outlier detection return for outlier vs. non-outlier signal level responses based on the features (denoted as "outlier detection" 522). The outliers may be assessed with respect to the training dataset used to establish the classifier (of module 112). The AE add-on module 514 may use the calculated values ​​of the features and the classifier model to generate an output score (e.g., 116) for the patient (e.g., via the classifier module 524). In the example of an assessment algorithm for estimation of HCM, the output score (e.g., 116) is the HCM score.

[0098] The clinical evaluation system 500 can manage data within and across components using a web service DTAPI 508 (which may also be referred to as HCPP web services in some embodiments). The DTAPI 508 may be used to retrieve acquired biophysical data sets from the data repository 111a and store signal quality analysis results in the data repository 111a. The DTAPI 508 may also be invoked to retrieve and provide stored biophysical data files to an analysis engine or analyzer (e.g., 506, 514), and the results of the analysis engine's analysis of the patient signals may be forwarded to a report database 510 using the DTAPI 508. The DTAPI 508 may also be used to retrieve a given patient data set upon request by a healthcare professional to a web portal module 513, which may present reports to the healthcare practitioner for review and interpretation in a secure web accessible interface.

[0099] The clinical evaluation system 500 includes one or more feature libraries 526 that store the machine learning based analyses, for example as features. The feature libraries 526 may be part of the add-on module 502 (shown in FIG. 5A ) or the base system 504 (not shown) and, in some embodiments, are accessed by the AE add-on module 514.

[0100] Exemplary Operation of a Modular Clinical Evaluation System FIG. 5B shows a schematic diagram of the operation and workflow of the analysis engine or analyzer (eg, 506 and 514) of the clinical evaluation system 500 of FIG. 5A, according to an illustrative embodiment.

[0101] Signal Quality Assessment / Rejection (530). Referring to FIG. 5B, the base analysis engine or analyzer 506, via the SQA module 516, assesses the quality of the biophysical signal data set acquired while the analysis pipeline is running (530). The result of the assessment (e.g., pass / fail) is immediately returned to the user interface of the signal capture system for reading by the user. Acquired signal data that meets the signal quality requirements may be deemed acceptable (i.e., "pass") and may be further processed and analyzed by the AE add-on module 514 for the presence of metrics associated with disease states or conditions (e.g., HCM, elevated LVEDP or mPAP, CAD, PH / PAH, abnormal LVEF, and / or hFpEF). Acquired signals that are deemed unacceptable are rejected (e.g., "fail") and a notification is immediately sent to the user to inform the user to immediately obtain additional signals from the patient (see FIG. 2).

[0102] The base analysis engine or analyzer 506 performs two sets of assessments for signal quality: one for the electrical signal and one for the hemodynamic signal. The electrical signal assessment (530) verifies that the electrical signal is of sufficient length, that there is a lack of high frequency noise (e.g., above 170 Hz), and that there is no power line noise from the environment. The hemodynamic signal assessment (530) verifies that the percentage of outliers in the hemodynamic dataset is below a predefined threshold, and that the percentage and maximum duration of the signal in the hemodynamic dataset that is rail or saturated are below a predefined threshold.

[0103] Feature Value Calculation (532). The AE add-on module 514 performs feature extraction and calculations to calculate feature output values. In the example of the LVEDP algorithm, the AE add-on module 514, in some embodiments, determines feature outputs that belong to different feature families (e.g., generated in module 110).

[0104] Additional descriptions of various features upon which HCM algorithms and associated machine learning analyses may be based are provided in [1], [2], [3], [4], [5], [6], [7], [8],

[12] ,

[13] , and

[19] , each of which is incorporated herein by reference in its entirety.

[0105] Classifier Output Calculation (534). The AE add-on module 514 then uses the feature outputs calculated in the classifier model (e.g., a machine-learned classifier model) to generate a set of model scores. The AE add-on module 514 may combine the set of model scores, for example, in some embodiments in an ensemble of constituent models that average the outputs of the classifier models.

[0106] In some embodiments, the classifier model may include a model developed based on ML techniques described in U.S. Patent Application Publication No. 20190026430, entitled "Discovering Novel Features to Use in Machine Learning Techniques, such as Machine Learning Techniques for Diagnosing Medical Conditions," or U.S. Patent Application Publication No. 20190026431, entitled "Discovering Genomes to Use in Machine Learning Techniques," each of which is incorporated by reference in its entirety. Another example of a training system that may be used to configure the analysis engine is described in

[19] , which is incorporated by reference in its entirety.

[0107] In examples, the machine-learned classifier models may include those described herein, among others, ElasticNet machine-learned classifier models, RandomForest machine-learned classifier models, and extreme gradient boosting (XGB) classifier models. In some embodiments, patient metadata information such as age, sex, and BMI value may be used. The output of the ensemble estimation may be a continuous score.

[0108] Physician Portal Visualization (536). The patient report may include visualization 536 of the acquired patient data and signals, and the results of the disease analysis. The analysis is presented in multiple views within the report in some embodiments. A health care provider, e.g., a physician, can review the report and interpret it to provide a diagnosis of the disease or generate a treatment plan.

[0109] The healthcare portal may list a report for a given patient if the acquired signal dataset for that patient meets the signal quality criteria. The report may indicate disease-specific results (e.g., HCM) that are available if a signal analysis could be performed. The patient's estimated score for the disease-specific analysis may be interpreted against established thresholds.

[0110] The report may be presented, for example, in a healthcare portal for use by a physician or healthcare provider in diagnosing symptoms of HCM, which may in some embodiments include a likelihood or severity score for the presence of a disease, medical condition, or any symptom thereof.

[0111] Outlier assessment and rejection detection (538). Following the AE add-on module 514 calculating the feature value outputs (in process 532) and prior to applying them to the classifier model (in process 534), the AE add-on module 514 is configured in some embodiments to perform an outlier analysis of the feature value outputs (shown in process 538). The outlier analysis assessment process 538 in some embodiments performs a machine-learned outlier detection module (ODM) to identify and exclude abnormally acquired biophysical signals by referencing feature values ​​generated from validation and training data to identify and exclude abnormal feature output values. The outlier detection module assesses for outliers that are present in sparse clusters in isolated regions that are outside the distribution from the rest of the observations. Process 538 can reduce the risk that outlier signals are inappropriately applied to the classifier model and would otherwise generate an inaccurate assessment as seen by the patient or healthcare provider. The accuracy of the outlier module is validated using a held-out validation set where ODM is able to identify all labeled outliers in the test set with an acceptable outlier detection rate (ODR) generalization.

[0112] Experimental Results and Examples A study was conducted to evaluate ML features for the evaluation of phonocardiograms, for example, to determine the presence or absence of HCM. Figures 6A-6G show experimental results of various developed ML features.

[0113] In Figure 6A, a phonocardiogram (PCG) continuous wavelet transform (CWT) scalogram 602 is shown for a high intensity systolic murmur generated from the DigiScope Phonocardiogram Dataset provided by Physionet. A similar signal may be obtained as an acoustic signal 604, for example, as obtained using the systems of Figures 1 and 4A-4E. From Figure 6A, it can be observed that the entire systolic murmur shown in scalogram 602 is high pitched, i.e., has more energy concentrated at high frequencies.

[0114] FIG. 6B shows a scalogram 602 (shown as 602a, 602b) with a wavelet transform 606 (shown as 606a, 606b) applied to an input signal 604 (shown as 604a). In FIG. 6B, a Morlet wavelet transform was applied. The transform 606a has a lower ω o and transform 606b applies a higher ω o Wavelets may be organized by shape as well as by mother wavelet and transform type.

[0115] FIG. 6C illustrates the results of an exemplary operation of ensemble averaging the input signal 602 over both time and CWT scalograms. In operation, an energy scalogram is first calculated from the raw PCG. The operation then determines the maximum energy scale (e.g., around 100 Hz) to identify the "S1" sound using turning points and expected pulse rate. The "S1" criteria (see, e.g., FIGS. 6D-6E) are then used as a trigger point for the ensemble average derivative. In the results shown in FIG. 6C, ω o = 5.5 and performed the Morlet transformation.

[0116] In some embodiments, a 1-D continuous wavelet transform is applied, such as the Morlet (also called Gabor) wavelet as the mother wavelet, as shown above. Other wavelets with equal variance in time and frequency may also be used, such as Gaussian, Mexican hat, spline, and Mayer wavelets. The wavelet may have a resolution of, for example, 48 audio per octave. The Morlet wavelet is a wavelet constructed by multiplying a complex exponential (carrier) by a Gaussian window (envelope), as shown in Equation 1. ψ(t)=exp(iωt)exp(-t 2 / 2σ 2 ) (Formula 1)

[0117] In Equation 1, ω is the wavelet center frequency and σ=n / 2πf is the width of the Gaussian window with n (number of cycles) that controls the time-frequency resolution tradeoff.

[0118] The coherence waveform may be determined as a measure of correlation between time series signals, such as between two observations of a phonocardiogram signal dataset, or between a phonocardiogram signal and a cardiac and / or photoplethysmographic signal dataset(s). Wavelet coherence may be determined, for example, by Equation 2:

number

[0119] In Equation 2, the cross spectrum C xy is a measure of the distribution of the power of two signals x and y in the time-frequency domain given by Equation 3.

number

[0120] In Equation 3, the superscript * denotes complex conjugate and S is a smoothing operator in time and scale. In some embodiments, a coherence spectrum operator is performed, for example, on 32 tones per octave, to find the coherence spectrum of paired channels (e.g., between channels X and Y, channels x and X, and channels Y and Z). The coherence spectrum may be generated between (i) the phonocardiogram signal and (ii) the cardiac signal and / or the photoplethysmographic signal.

[0121] The high power spectral image or data may be generated from the scalogram, for example, by generating a binarized spectral image of the high power spectral content of the waveform signal of interest, e.g., the spectral content of noise. FIG. 6G illustrates an example method of generating a binarized spectral image 614 of the spectral image or spectral data. The method may include performing a wavelet transform to generate a spectral image or spectral data of a given waveform region of interest to which a threshold operator is applied to generate a binarized spectral image or data corresponding to the high spectral power characteristics of the waveform region. The calculations may be performed over multiple cardiac cycles to extract statistical characteristics of the time range, frequency range, time centroid, surface area, eccentricity, circularity, range, orientation, and / or power centroid of the high spectral energy characteristics of the waveform region as wavelet-based features or parameters. FIG. 6G illustrates an example of a binarized spectral image 614 generated from the spectral image or spectral data of the high power spectral content of the waveform signal of interest from the scalogram of the input signal.

[0122] An example of a method for generating such a binarized spectral image is provided in U.S. patent application Ser. No. 17 / 891,259, entitled "METHOD AND SYSTEM TO ASSESS DISEASE USING WAVELET ANALYSIS OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS," filed Aug. 18, 2022, which is incorporated by reference in its entirety into this specification.

[0123] Table 2 shows an exemplary set of extractable high spectral energy features of waveform regions of interest from the generated binarized spectral image or data. [Table 2] TIFF2025503130000015.tif70170

[0124] In Figures 6D and 6E, it can be observed that the ensemble average power described with respect to Figure 6C can easily allow for the identification of prominent noise (II-IV) in clean, well-behaved signals. Figure 6D shows the ensemble average power of a noise-free acoustic signal. Figure 6E shows the ensemble average power of a set of acoustic signals with various stages of holosystole (shown as "I / VI early systole" 608, "II / VI holosystole" 610, and "III / VI holosystole" 612). It can be seen that "pitch" appears to be as correlated with amplitude as with frequency content. It can also be observed that a much higher percentage of grade "I" noise appears to be early systole rather than holosystole.

[0125] Figure 6F shows an exemplary parameterization of the analysis of the ensemble average power analysis of Figures 6D and 6E, for each feature described, for example, in Table 1 and / or Table 2. The parameterized features may be used in a classification output of the presence, absence, and / or severity of HCM, among other diseases and conditions described herein.

[0126] In some embodiments, the analysis may be performed taking into account the patient's respiratory status. Examples of respiratory estimation may be found in U.S. patent application Ser. No. 17 / 891,224, entitled "METHOD AND SYSTEM TO ASSESS DISEASE USING ESTIMATED RESPIRATION PARAMETERS FROM CARDIAC AND PHOTOPLETHYS-MOGRAPHIC SIGNALS."

[0127] 6H shows an example scalogram with detected noise therein, which can be evaluated for noise using wavelet feature analysis as described herein.

[0128] Conclusion. Although the methods and systems have been described with reference to certain specific embodiments and specific examples, the scope is not intended to be limited to the embodiments shown, as the embodiments herein are intended in all respects to be illustrative and not restrictive. The clinical evaluation systems and methods discussed herein may be used to make, or assist physicians or other health care providers in making, non-invasive diagnoses or determinations of the presence or / or severity of other diseases and / or conditions described herein, such as coronary artery disease (CAD), pulmonary hypertension, and other conditions, using similar or other developmental approaches. Furthermore, the exemplary clinical evaluation systems and methods may be used in the diagnosis and treatment of other cardiac-related and neurological-related conditions and conditions, and such assessments may be applied to the diagnosis and treatment (including surgical, minimally invasive, and / or pharmacological treatments) of any disease or condition involving biophysical signals in any relevant system of the living body. An example in the cardiac context is the diagnosis of CAD and other diseases and conditions disclosed herein and their treatment by any number of therapies, alone or in combination, such as placing stents in the coronary arteries, performing atherectomy, angioplasty, prescribing medications, and / or prescribing exercise, nutritional and other lifestyle changes. Other cardiac-related pathologies or conditions that may be diagnosed include, for example, arrhythmias, congestive heart failure, valvular insufficiency, pulmonary hypertension (e.g., pulmonary arterial hypertension, pulmonary hypertension due to left heart disease, pulmonary hypertension due to lung disease, pulmonary hypertension due to chronic thrombosis, and pulmonary hypertension due to other diseases such as blood or other disorders), and other cardiac-related pathologies, conditions and / or diseases.Non-limiting examples of neurological related diseases, conditions, or conditions that may be diagnosed include, for example, epilepsy, schizophrenia, Parkinson's disease, Alzheimer's disease (and all other forms of dementia), autism spectrum (including Asperger's syndrome), attention deficit hyperactivity disorder, Huntington's disease, muscular dystrophy, depression, bipolar disorder, brain / spinal cord tumors (malignant and benign), mobility disorders, cognitive impairment, speech disorders, various psychiatric diseases, brain / spinal cord / nerve injuries, chronic traumatic encephalopathy, cluster headaches, migraines, neurological disorders (various forms thereof), and the like. including peripheral neurological disorders), phantom limb pain, chronic fatigue syndrome, acute and / or chronic pain (including back pain, post-spine surgery pain syndrome, etc.), dyskinesia, anxiety, conditions caused by infection or foreign agents (e.g., Lyme disease, encephalitis, rabies), narcolepsy and other sleep disorders, post traumatic stress disorder, neurological conditions / effects related to stroke, aneurysms, hemorrhagic injuries, etc., tinnitus and other hearing-related diseases / conditions, and vision-related diseases / conditions.

[0129] In addition, the clinical evaluation system described herein may be configured to analyze biophysical signals such as electrocardiogram (ECG), electroencephalogram (EEG), gamma synchrony in signals, respiratory function signals, photoplethysmographic, phonocardiogram, pulse oximetry signals, perfusion data signals, quasi-periodic biological signals, fetal ECG signals, blood pressure signals, cardiac magnetic field signals, heart rate signals, among others. As described herein, the clinical evaluation system may use a single type of biophysical signal for HCM estimation, or may use multiple types of signals for HCM estimation. In addition, it is contemplated that a three-sensor system may be used in which the third sensor is MEMS-based, (a) the third sensor may be non-MEMS-based, (b) may be a single type of sensor (PPG, ECG, MEMS, etc.), (c) may be a dual-sensor system (using three or more of two types of sensors), and (d) may involve more than three sensors. Any combination of known sensor types, contact or non-contact (e.g., non-contact thermometer) sensors may be used.

[0130] Further examples of processes that may be used in the exemplary methods and systems disclosed herein include U.S. Pat. Nos. 9,289,150, 9,655,536, 9,968,275, 8,923,958, 9,408,543, 9,955,883, 9,737,229, 10,039,468, 9,597,021, and 9,968,265. , U.S. Patent No. 9,910,964, U.S. Patent No. 10,672,518, U.S. Patent No. 10,566,091, U.S. Patent No. 10,566,092, U.S. Patent No. 10,542,897, U.S. Patent No. 10,362,950, U.S. Patent No. 10,292,596, U.S. Patent No. 10,806,349, U.S. Patent No. 11,133,109, U.S. Patent No. 11,141,114, U.S. Patent No. 11,160,509, U.S. Patent No. 11,147,516, U.S. Patent Application Publication No. 2020 / 0335217, U.S. Patent Application Publication No. 2020 / 0229724, U.S. Patent Application Publication No. 2019 / 0214137, U.S. Patent Application Publication No. 2018 / 0249960, U.S. Patent Application Publication No. 2019 / 0200893, U.S. Patent Application Publication No. 2019 / 0384757, U.S. Patent Application Publication No. 2020 / 0211713, U.S. Patent Application Publication No. 2019 / 0365265, U.S. Patent Application Publication No. 202 US Patent Application Publication No. 2020 / 0205739, US Patent Application Publication No. 2020 / 0205745, US Patent Application Publication No. 2019 / 0026430, US Patent Application Publication No. 2019 / 0026431, WO 2017 / 033164, WO 2017 / 221221, WO 2019 / 130272, WO 2018 / 158749, WO 2019 / 077414, WO 2019 / 130273, WO 2019 / 2440 43, WO 2020 / 136569, WO 2019 / 234587, WO 2020 / 136570, WO 2020 / 136571, U.S. Design Patent No. D810947, U.S. Design Patent No. D855064, U.S. Design Patent No. D895661, U.S. Design Patent No. D843382, U.S. Design Patent No. D880501, U.S. Design Patent No. D858532, U.S. Patent Application No. 16 / 232586, U.S. Patent Application No. 16 / 83 No. 1,264, U.S. Patent Application No. 16 / 429593, U.S. Patent Application No. 16 / 725402, U.S. Patent Application No. 16 / 831,380, U.S. Patent Application No. 16 / 725430, U.S. Patent Application No. 16 / 725416, U.S. Patent Application No. 17 / 132869, International Application No. PCT / IB2020 / 052889, and International Application No. PCT / IB2020 / 052890, each of which is incorporated herein by reference in its entirety.

[0131] Each of the following patents, applications, and publications listed below and throughout the specification are hereby incorporated by reference in their entirety. List of References [1]U.S.Provisional Patent Application No.63 / 236,072,filed August 23,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING VISUAL FEATURES OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS,” having attorney docket No.10321-49pv1. [2]U.S.Provisional Patent Application No.63 / 235,963,filed August 23,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING POWER SPECTRAL ANALYSIS OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS,” having attorney docket No.10321-50pv1. [3]U.S.Provisional Patent Application No.63 / 235,968,filed August 23,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING ESTIMATED RESPIRATION PARAMETERS FROM CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS,” having attorney docket no.10321-51pv1. [4]U.S.Provisional Patent Application No.63 / 235,968,filed August 23,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING WAVELET ANALYSIS OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS,” having attorney docket no.10321-52pv1. [5]U.S.Patent Application No.17 / 558,702,filed December 22,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING CYCLE VARIABILITY ANALYSIS OF BIOPHYSICAL SIGNALS,” having attorney docket no.10321-053us1. [6]U.S.Provisional Patent Application No.63 / 235,971,filed August 23,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING ANALYSIS OF PHOTOPLETHYSMOGRAPHIC SIGNALS,” having attorney docket no.10321-054pv1. [7]U.S.Provisional Patent Application No.63 / 236,193,filed August 23,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING MORPHOLOGICAL ANALYSIS OF CARDIAC SIGNALS,” having attorney docket no.10321-055pv1. [8]U.S.Provisional Patent Application No.63 / 235,974,filed August 23,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING DEPLORIZATION WAVE PROPAGATION DEVIATIONS,” having attorney docket no.10321-056pv1. [9]H.Zou and T.Hastie,“Regularization and variable selection via the elastic net,,” Journal of the Roy.Statist.Soc.,ser.B,vol.67,no.2,pp.301-320,2005.

[10] Breiman,“Random Forests,” Machine Learning,45(1),5-32,2001.

[11] Chen,Tianqi; Guestrin,Carlos(2016).“XGBoost:A Scalable Tree Boosting System” In Krishnapuram,Balaji; Shah,Mohak; Smola,Alexander J.; Aggarwal,Charu C.; Shen,Dou; Rastogi,Rajeev(eds.).Proceedings of the 2 2n d ACM SIGKDD International Conference on Knowledge Discovery and Data Mining,San Francisco,CA,USA,August 13-17,2016.ACM.pp.785-794.

[12] U.S.Patent Application No.16 / 831,264,filed April 30,2020,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING DYNAMICAL ANALYSIS OF BIOPHYSICAL SIGNALS,” having attorney docket no.10321-040us1.

[13] U.S.Patent Application No.16 / 831,380,filed April 30,2020,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING DYNAMICAL ANALYSIS OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS,” having attorney docket no.10321-041us1.

[14] U.S.Application No.17486609,filed September 27,2021,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING MULTI-SENSOR SIGNALS,” having attorney docket no.10321-047us1.

[15] Fabio Leitao,Eurico Moreira,Filipe Alves,Mario Lourenco,Olga Azevedo,Joao Gaspar,and Luis A.Rocha,“High-Resolution Seismocardiogram Acquisition and Analysis System,” Journal of Sensors 2018,18,3441; doi:10.3390 / s18103441.

[16] Martin S Maron,Jennifer L Hellawell,Jaime C Lucove,Ramin Farzaneh-Far,Iacopo Olivotto,“Occurrence of Clinically Diagnosed Hypertrophic Cardiomyopathy in the United States,” Am J Cardiol 2016 May 15;117(10):1651-1654.

[17] European heart journal cardiovascular imaging,2018-10-01,Vol.19(10),p.1126-1133

[18] U.S.Provisional Application No.63234772,filed August 19,2021,entitled “MEDICAL EVALUATION SYSTEMS AND METHODS USING ADD-ON MODULES,” having attorney docket no.10321-060pv1.

[19] U.S.Provisional Application No.63235960,filed August 23,2021,entitled “METHOD AND SYSTEM TO NON-INVASIVELY ASSESS ELEVATED LEFT VENTRICULAR END-DIASTOLIC PRESSURE,” having attorney docket no.10321-048pv1.

[20] https: / / www.umms.org / ummc / health-services / heart-vascular / services / hypertrophic-cardiomyopathy / types-symptoms-causes

[21] Raj MA,Ranka S,Goyal A.,“Hypertrophic Obstructive Cardiomyopathy,”[Updated 2021 Nov 4].In:StatPearls[Internet].Treasure Island(FL):StatPearls Publishing; 2022 Jan-.Available from:https: / / www.ncbi.nlm.nih.gov / books / NBK430820 /

[21] Maron BJ,Wolfson JK,Roberts WC.Relation between extent of cardiac muscle cell disorganization and left ventricular wall thickness in hypertrophic cardiomyopathy.Am J Cardiol.1992 Sep 15;70(7):785-90.

[22] Marian AJ,Braunwald E.Hypertrophic Cardiomyopathy:Genetics,Pathogenesis,Clinical Manifestations,Diagnosis,and Therapy.Circ Res.2017;121(7):749-770.doi:10.1161 / CIRCRESAHA.117.311059.

[23] Electrophysiologic abnormalities in patients with hypertrophic cardiomyopathy.A consecutive analysis in 155 patients,L Fananapazir,C M Tracy,M B Leon,J B Winkler,R O Cannon, 3r d,R O Bonow,B J Maron,and S E Epstein.

[24] Dubin,Dale.Rapid Interpretation of EK’s:An Interactive Course.Sixth edition.Tampa,Fla.:Cover Pub.Co.,2000.

[25] The electrocardiogram in the diagnosis and management of patients with hypertrophic cardiomyopathy,Finocchiaro,Gherardo,et al.,Heart Rhythm,Volume 17,Issue 1,142 - 151.

[26] Savelieva,Irina,et al.“Comparative reproducibility of QT,QT peak,and T peak-T end intervals and dispersion in normal subjects,patients with myocardial infarction,and patients with hypertrophic cardiomyopathy” Pacing and clinical electrophysiology 21.11(1998):2376-2381.

[27] Dinshaw,Leon,et al.“The T-peak-to-T-end interval:a novel ECG marker for ventricular arrhythmia and appropriate ICD therapy in patients with hypertrophic cardiomyopathy,” Clinical Research in Cardiology 107.2(2018):130-137.

[28] Finocchiaro,Gherardo,et al.“The electrocardiogram in the diagnosis and management of patients with hypertrophic cardiomyopathy” Heart Rhythm 17.1(2020):142-151.

[29] Jothiramalingam,R.,Jude,A.,Patan,R.et al.Machine learning-based left ventricular hypertrophy detection using multi-lead ECG signal.Neural Comput & Applic 33,4445-4455(2021).https: / / doi.org / 10.1007 / s00521-020-05238-2.

[30] A.Englund,K.Hnatkova,P.M.Elliot,W.J.McKenna,M.Malik,Wavelet decomposition analysis of the signal averaged electrocardiogram used for risk stratification of patients with hypertrophic cardiomyopathy,European Heart Journal,Volume 19,Issue 9,1 September 1998,Pages 1383-1390,

[31] Edhouse J,Thakur RK,Khalil JM.ABC of clinical electrocardiography.Conditions affecting the left side of the heart.BMJ.2002 May 25;324(7348):1264-7.

[32] Harrigan RA,Jones K.ABC of clinical electrocardiography.Conditions affecting the right side of the heart.BMJ.2002 May 18;324(7347):1201-4.

[33] Green,E.M.,van Mourik,R.,Wolfus,C.et al.Machine learning detection of obstructive hypertrophic cardiomyopathy using a wearable biosensor.npj Digit.Med.2,57(2019).

[34] Eleid MF,Konecny T,Orban M,Sengupta PP,Somers VK,Parish JM,Mookadam F,Brady PA,Sullivan BL,Khandheria BK,Ommen SR,Tajik AJ.,“High prevalence of abnormal nocturnal oximetry in patients with hypertrophic cardiomyopathy,” J Am Coll Cardiol.2009 Nov 3;54(19):1805-9.

[35] Andres Ricardo Perez-Riera,“Electro / Vectorcardiogram in Left Ventricular Hypertrophy / Enlargement-Part 1,” retrieved on 13 January 2022 from http: / / cardiolatina.com / wp-content / uploads / 2020 / 07 / ECG-VCG-in-Left-Ventricular-Hypertrophy-Enlargements-LVH-Part-1.pdf.

[36] Erick Jimenez,Amr El-Bokl,Daniel Cortez,“Vectorcardiography as a prognostic tool in hypertrophic cardiomyopathy,” Journal of Electrocardiology,Volume 68,2021,Pages 80-84,ISSN 0022-0736.

[37] Perez-Riera AR,de Lucca AA,Barbosa-Barros R,et al.,“Value of electro-vectorcardiogram in hypertrophic cardiomyopathy,” Ann.Noninvasive Electrocardiol.2013;18(4):311-326.

[38] Green,E.M.,van Mourik,R.,Wolfus,C.et al.,“Machine learning detection of obstructive hypertrophic cardiomyopathy using a wearable biosensor,” npj Digit.Med.2,57(2019).

[39] Houston BA,Stevens GR,“Hypertrophic cardiomyopathy:a review,” Clin Med Insights Cardiol.2015;8(Suppl 1):53-65.Published 2015 Jan 26.doi:10.4137 / CMC.S15717.

[40] Takahashi K,Ono K,Arai H,et al.,“Detection of Pathologic Heart Murmurs Using a Piezoelectric Sensor,” Sensors(Basel).2021;21(4):1376.Published 2021 Feb 16.doi:10.3390 / s21041376.

[41] Folino,A.F.,Buja,G.,Bauce,B.,Thiene,G.,Volta,S.D.And Nava,A.,“Heart Rate Variability in Arrhythmogenic Right Ventricular Cardiomyopathy Correlation with Clinical and Prognostic Features,” Pacing and Clinical Electrophysiology,25:1285-1292(2002).

[42] F.Laurent et al.,“Heart Rate Variability in Idiopathic Dilated Cardiomyopathy:Characteristics and Prognostic Value,” Journal of the American College of Cardiology,1997; 30(4):1109-1014.

[43] Tavakolian,Kouhyar,“Systolic time intervals and new measurement methods,” Cardiovascular engineering and technology 7.2(2016):118-125.

[44] Reant P,Dijos M,Donal E,Mignot A,Ritter P,Bordachar P,Dos Santos P,Leclercq C,Roudaut R,Habib G,Lafitte S,“Systolic time intervals as simple echocardiographic parameters of left ventricular systolic performance:correlation with ejection fraction and longitudinal two-dimensional strain,” Eur J Echocardiogr.2010 Dec;11(10):834-44.

[45] Claria,Francesc,et al.,“Heart rate variability analysis based on time-frequency representation and entropies in hypertrophic cardiomyopathy patients,” Physiological measurement 29.3(2008):401.

[46] U.S.Patent Application No.17 / 891,526,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING VISUAL FEATURES OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS.”

[47] U.S.Patent Application No.17 / 891,229,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING POWER SPECTRAL ANALYSIS OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS.”

[48] U.S.Patent Application No.17 / 891,380,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING DEPLORIZATION WAVE PROPAGATION DEVIATIONS.”

[49] U.S.Patent Application No.17 / 558,702,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING CYCLE VARIABILITY ANALYSIS OF BIOPHYSICAL SIGNALS.”

[50] U.S.Patent Application No.11,291,379.

[51] U.S.Publication No.2020 / 0397322.

[52] U.S.Patent Application No.17 / 891,278,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING ANALYSIS OF PHOTOPLETHYSMOGRAPHIC SIGNALS.”

[53] U.S.Patent Application No.17 / 891,259,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING WAVELET ANALYSIS OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS.”

[54] U.S.Patent Application No.17 / 891,526,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING VISUAL FEATURES OF CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS.”

[55] U.S.Patent Application No.17 / 891,224,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING ESTIMATED RESPIRATION PARAMETERS FROM CARDIAC AND PHOTOPLETHYSMOGRAPHIC SIGNALS.”

[56] U.S.Patent Application No.17 / 891,533,entitled “METHOD AND SYSTEM TO ASSESS DISEASE USING MORPHOLOGICAL ANALYSIS OF CARDIAC SIGNALS.”

Claims

1. 1. A method for non-invasively predicting the presence, absence, and / or severity of hypertrophic cardiomyopathy in a mammalian subject, comprising: obtaining, by one or more processors, one or more biophysical signals of the patient from one or more sensors; determining, by the one or more processors utilizing at least a portion of the one or more biophysical signals, one or more values ​​associated with one or more features and / or machine learning based analyses; determining, by the one or more processors, an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy using the one or more values ​​associated with the one or more features or machine learning based analysis; and outputting, by the one or more processors, the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy, wherein the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy is output for use in diagnosing hypertrophic cardiomyopathy and / or for direct treatment of the hypertrophic cardiomyopathy.

2. 1. A method for non-invasively predicting the presence, absence, and / or severity of hypertrophic cardiomyopathy in a mammalian subject, comprising: obtaining, by one or more processors, one or more vibrocardiogram signals (SCG signals) and / or phonocardiogram signals (PCG signals) from a multi-sensor device placed on or worn by a patient; determining, by the one or more processors utilizing at least a portion of the one or more vibrocardiogram and / or phonocardiogram signals, one or more values ​​associated with one or more features or machine learning based analyses; determining, by the one or more processors, an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy using the one or more values ​​associated with the one or more features or machine learning based analysis; and outputting, by the one or more processors, the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy, wherein the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy is output for use in diagnosing hypertrophic cardiomyopathy and / or for direct treatment of the hypertrophic cardiomyopathy.

3. 1. A method for non-invasively predicting the presence, absence, and / or severity of hypertrophic cardiomyopathy in a mammalian subject, comprising: obtaining, by one or more processors, a first biophysical signal dataset associated with a first photoplethysmographic signal and a second photoplethysmographic signal, the first biophysical signal dataset being acquired over a plurality of cardiac cycles of the subject; acquiring, by the one or more processors, a second biophysical signal dataset associated with the cardiac signal, the second biophysical signal dataset being acquired contemporaneously with the first biophysical signal dataset over the plurality of cardiac cycles; determining, by the one or more processors utilizing at least a portion of the first biophysical signal dataset and the second biophysical signal dataset, one or more values ​​associated with one or more features or machine learning based analyses; determining, by the one or more processors, an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy using the one or more values ​​associated with the one or more features or machine learning-based analysis; The method, wherein the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy is output for use in diagnosing hypertrophic cardiomyopathy and / or for directing treatment of said hypertrophic cardiomyopathy.

4. 4. The method of claim 1, wherein the one or more features or machine learning based analysis is configured to quantify deviations of a VD wave trajectory from a three-dimensional modeled VD wave trajectory.

5. The method of any one of claims 1 to 3, wherein the one or more features or machine learning based analysis is configured to quantify beat-to-beat variability of the cardiac signal.

6. 4. The method of claim 1, wherein the one or more feature or machine learning based analyses are configured to quantify variability in registered landmarks in cardiac, PPG, SCG, and / or PCG signals via point-of-care analysis and histogram analysis.

7. 4. The method of claim 1, wherein the one or more features or machine learning based analyses are configured to quantify dynamic characteristics of cardiac, PPG, SCG, and / or PCG signals.

8. The method of any one of claims 1 to 3, wherein the one or more features or machine learning based analyses are configured to quantify properties of cardiac, PPG, and / or SCG signals.

9. 4. The method of claim 1, wherein the one or more feature or machine learning based analyses are configured to quantify dominant frequency components of cardiac, PPG, SCG, and / or PCG signals using wavelet analysis.

10. 4. The method of claim 1, wherein the one or more feature or machine learning based analyses are configured to quantify power spectrum and frequency content of cardiac, PPG, SCG, and / or PCG signals using power spectrum and coherence analysis.

11. 4. The method of claim 1, wherein the one or more feature or machine learning based analyses are configured to quantify properties of the cardiac, PPG, SCG, and / or PCG signals over loop areas, their projections, and loop vectors in 3D phase space.

12. 4. The method of claim 1, wherein the one or more features or machine learning-based analysis is configured to use either (i) a PPG signal and a cardiac signal, or (ii) an SCG signal and / or a PCG signal to approximate a respiratory waveform to assess one of (1) heart rate variability, (2) respiratory rate, (3) a discrepancy feature representing the distance between the respiratory and modulation signals, and (4) squared coherence representing the correlation between the modulation and respiratory rate signals, and the approximated respiratory waveform is used to generate delineated inspiratory and expiratory portions of the SCG signal and / or the PCG signal used for the analysis, thereby used for HCM assessment.

13. 4. The method of claim 1, wherein the one or more features or machine learning based analysis is configured to quantify physiological aspects of cardiac, SCG, and / or PCG signals.

14. 4. The method of claim 1, wherein the one or more features or machine learning-based analysis is configured to quantify characteristic variations in cardiac, SCG, and / or PCG signals associated with inspiration versus expiration for a Valsalva maneuver to identify patients with HCM.

15. 4. The method of claim 1, wherein the one or more features or machine learning-based analysis is configured to quantify characteristic variations in cardiac, SCG, and / or PCG signals associated with inspiration versus expiration for a Valsalva maneuver to identify a subset of patients with obstructive HCM (OHCM).

16. 4. The method of claim 1, wherein the one or more features or machine learning based analysis is configured to approximate left ventricular ejection time using one or more SCG and / or PCG signals.

17. 4. The method of claim 1, wherein the one or more features or machine learning-based analysis is configured to quantify propagation characteristics of ventricular depolarization (VD) waves and / or ventricular repolarization (VR) waves in three-dimensional space.

18. 18. The method of any one of claims 1-2 and 4-17, wherein the one or more features or machine learning based analysis are evaluated (i) in the inspiration region of one or more vibrocardiogram and / or phonocardiogram signals, and / or (ii) in the expiration region of the one or more vibrocardiogram and / or phonocardiogram signals.

19. 1. An apparatus comprising: a sensor body configured to be worn or placed externally on the subject's chest region to acquire biophysical signals from the subject's chest region, including cardiac signals; two or more MEMS-based biophysical sensors including a first MEMS-based sensor and a second MEMS-based sensor, the two or more MEMS-based sensors being located within the sensor body and connected to electrodes configured to be placed on the subject; The device, wherein the first MEMS-based sensor and the second MEMS-based sensor, during operation, generate first and second biophysical signals that are provided to an analysis system configured to evaluate a plurality of features or machine learning-based analysis to generate an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy.

20. 1. An apparatus comprising: a sensor body configured to be worn or placed externally on the subject's chest region to acquire biophysical signals from the subject's chest region, including cardiac signals; two or more MEMS-based sensors including a first MEMS-based sensor and a second MEMS-based sensor, the two or more MEMS-based sensors being connected to electrodes located within the sensor body and configured to be disposed on the object; The device, wherein the first MEMS-based sensor and the second MEMS-based sensor, during operation, generate a first vibrogram signal and / or a first acoustic signal and / or a second vibrogram signal and / or a second acoustic signal that are provided to an analysis system configured to evaluate a plurality of features or machine learning-based analysis to generate an estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy.

21. The method of claim 20, further comprising: a plurality of surface electrodes disposed on a surface of the chest region of the subject and configured to provide a plurality of cardiac signals of the subject's heart; 21. The apparatus of claim 19 or 20, wherein the plurality of cardiac signals is provided to the analysis system for evaluating the plurality of features or machine learning based analysis to generate the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy.

22. a plurality of photoplethysmographic sensors disposed on the subject and configured to provide one or more photoplethysmographic signals; 22. The apparatus of any one of claims 19 to 21, wherein the one or more photoplethysmographic signals are provided to the analysis system for evaluating the plurality of features or a machine learning based analysis to generate the estimate of the presence, absence, and / or severity of hypertrophic cardiomyopathy.

23. 22. The apparatus of any one of claims 19 to 21, wherein the first MEMS-based sensor as an accelerometer or an acoustic sensor is configured to be non-invasively placed on the thoracic region of the subject proximal to an apical region of the subject's heart.

24. 24. The apparatus of any one of claims 19 to 23, wherein the second MEMS-based sensor as an accelerometer or an acoustic sensor is configured to be non-invasively placed on the chest region of the subject proximal to a base region of the subject's heart.

25. 1. A system comprising: An apparatus according to any one of claims 19 to 23; the analysis system, wherein the analysis system is implemented in a cloud-based processing and networking infrastructure.

26. 1. A system comprising: one or more processors; and one or more memories each having instructions stored therein, wherein execution of the instructions by said one or more processors causes said one or more processors to perform any one of the steps of the method according to claims 1 to 18.

27. ​​A program that causes one or more processors to execute any one of the steps of the methods described in claims 1 to 18.

Images