Systems and methods for measuring hemodynamic parameters with wearable cardiovascular sensing

Patent metadata

Field	Value
Publication number	US-12336793-B1
Application number	US-202418751868-A
Country	US
Kind code	B1
Filing date	Jun 24, 2024
Priority date	Nov 24, 2020
Publication date	Jun 24, 2025
Grant date	Jun 24, 2025

Abstract

Systems and methods for measuring hemodynamic parameters with wearable cardiovascular sensing. An apparatus can include one or more sensors configured to measure an electrocardiogram signal of a user and one or more seismocardiogram (SCG) signals of the user, a memory and a processing system including one or more processors operatively coupled to the memory and the one or more sensors, and configured to receive the electrocardiogram and one or more SCG signals, and generate an assessment of heart health by determining one or more hemodynamic parameters based on the signals. The invention further includes a method for non-invasively monitoring heart health of a user including receiving an electrocardiogram signal from a first sensor of a wearable device, receiving one or more SCG signals from a second sensor of the wearable device, and generating the assessment of the heart health of the user by determining the one or more hemodynamic parameters.

First claim

What is claimed is:

1. An apparatus, comprising:
a set of electrodes configured to measure an electrocardiogram signal of a heart of a user, the set of electrodes attachable to a chest of the user; and
a wearable device configured to be worn on the chest of the user below a suprasternal notch of the user, the wearable device including:
a housing having a first side and a second side, the first side configured to face away from the chest and the second side configured to face toward the chest when the wearable device is attached to the set of electrodes;
a set of connectors disposed on the second side of the housing in spaced relation to one another, each connector of the set of connectors configured to releasably attach to a different electrode of the set of electrodes;
an accelerometer disposed within an interior of the housing between the first side and the second side, the accelerometer configured to measure a set of seismocardiogram signals of the user, each seismocardiogram signal of the set of seismocardiogram signals being a channel associated with a different axis of a set of axes; and
electronics operatively coupled to the set of electrodes and the accelerometer, the electronics including a processor and a memory, the processor configured to execute instructions stored in the memory to:
receive the electrocardiogram signal and the set of seismocardiogram signals of the user;
segment the set of seismocardiogram signals based on the electrocardiogram signal to construct seismocardiogram beat arrays for the set of axes, each seismocardiogram beat array including a plurality of seismocardiogram signal segments associated with a plurality of beats;
input the seismocardiogram beat arrays into a motion-artifact detection algorithm that compares morphologies of seismocardiogram signal segments of consecutive beats of the plurality of beats to compute similarity values of those seismocardiogram signal segments to identify one or more seismocardiogram signal segments that are dissimilar based on the similarity values and outputs indices of motion-contaminated beats, the comparing of the morphologies of the seismocardiogram signal segments of consecutive beats by the motion-artifact detection algorithm configured to identify motion artifacts having amplitudes higher than underlying local mechanical vibrations of the chest of the user associated with a movement of blood within vasculature of the user;
remove, from the seismocardiogram beat arrays, the motion-contaminated beats;
apply signal quality indexing to each seismocardiogram signal of the set of seismocardiogram signals using as a template an ensemble average of the seismocardiogram signal segments of the plurality of beats after removing the motion-contaminated beats to extract high quality beats from the plurality of beats, each of the high quality beats having a quality score in a top predetermined percentage of the plurality of beats;
extract, using a machine learning algorithm, features from the seismocardiogram signal segments of the high quality beats;
determine one or more hemodynamic parameters associated with filling characteristics of the heart using the extracted features; and
output, via a display, the one or more hemodynamic parameters.

2. The apparatus of claim 1 , wherein the processor is further configured to execute instructions stored in the memory to:
determine one or more hemodynamic parameters associated with at least one of pulmonary artery pressure or pulmonary capillary wedge pressure of the heart using the extracted features.

3. The apparatus of claim 1 , wherein the set of axes includes a lateral axis, a head-to-foot axis, and a dorso-ventral axis.

4. The apparatus of claim 1 , wherein the processor is further configured to execute instructions stored in the memory to:
generate an assessment of heart health of the user based on the one or more hemodynamic parameters, wherein the heart health of the user is associated with heart failure.

5. The apparatus of claim 1 , wherein the processor is further configured to execute instructions stored in the memory to:
detect R-peaks in the electrocardiogram signal, and
the processor is configured to segment the set of seismocardiogram signals based on the R-peaks.

6. The apparatus of claim 5 , wherein detecting the R-peaks in the electrocardiogram signal includes implementing at least one of Pan Tompkins or a Phasor Transform to process the electrocardiogram signal to detect the R-peaks.

7. The apparatus of claim 5 , wherein the processor is further configured to segment the set of seismocardiogram signals by delimiting, for each beat of the plurality of beats, a first predetermined time before an associated R-peak of the R-peaks as a start of the beat and a second predetermined time after the associated R-peak of the R-peaks as an end of the beat, such that the start and the end of the beat captures ventricular diastolic timing.

8. The apparatus of claim 1 , wherein the one or more hemodynamic parameters includes a filling pressure of the user or a change in the filling pressure of the user.

9. An apparatus, comprising:
a set of electrodes configured to measure an electrocardiogram signal of a heart of a user, the set of electrodes attachable to a chest of the user; and
a wearable device configured to be worn on the chest of the user below a suprasternal notch of the user, the wearable device including:
a housing having a first side and a second side, the first side configured to face away from the chest and the second side configured to face toward the chest when the wearable device is attached to the set of electrodes; and
a set of connectors disposed on the second side of the housing in spaced relation to one another, each connector of the set of connectors configured to releasably attach to a different electrode of the set of electrodes;
an accelerometer disposed within an interior of the housing between the first side and the second side, the accelerometer configured to measure a set of seismocardiogram signals of the user, each seismocardiogram signal of the set of seismocardiogram signals being a channel associated with a different axis of a set of axes; and
electronics operatively coupled to the set of electrodes and the accelerometer, the electronics including a processor and a memory, the processor configured to execute instructions stored in the memory to:
receive the electrocardiogram signal and the set of seismocardiogram signals of the user;
segment the set of seismocardiogram signals based on the electrocardiogram signal to construct seismocardiogram beat arrays for the set of axes, each seismocardiogram beat array including a plurality of seismocardiogram signal segments associated with a plurality of beats;
implement a motion-artifact detection algorithm that:
compares, for each axis of the set of axes, morphologies of seismocardiogram signal segments of consecutive beats of the plurality of beats to compute similarity values of those seismocardiogram signal segments for each axis;
determines combined similarity values for those seismocardiogram signal segments based on the similarity values for each axis;
rejects one or more seismocardiogram signal segments that are dissimilar based on the combined similarity values; and
outputs the seismocardiogram beat arrays without the one or more seismocardiogram signal segments that are dissimilar, the one or more seismocardiogram signal segments that are dissimilar having motion artifacts with amplitudes higher than underlying local mechanical vibrations of the chest of the user associated with a movement of blood within vasculature of the user;

apply signal quality indexing to each seismocardiogram signal of the set of seism

Assignees

• Georgia Tech Res Inst

Inventors

• Inan Omer T
• Aydemir Varol Burak
• Rehg James
• Shandhi Md Mobashir Hasan

Classifications

• A61B5/7214

  using signal cancellation, e.g. based on input of two identical physiological sensors spaced apart, or based on two signals derived from the same sensor, for different optical wavelengths · CPC title

• A61B5/6801

  specially adapted to be attached to or worn on the body surface · CPC title

• A61B5/02141 Primary

  Details of apparatus construction, e.g. pump units or housings therefor, cuff pressurising systems, arrangements of fluid conduits or circuits (A61B5/02233, A61B5/0235 take precedence) · CPC title

• A61B5/352

  Detecting R peaks, e.g. for synchronising diagnostic apparatus; Estimating R-R interval · CPC title

• A61B2562/06

  Arrangements of multiple sensors of different types · CPC title