Method and system for measuring beat parameters

• US 10,524,734 B2
• Filed: 10/08/2015
• Issued: 01/07/2020
• Est. Priority Date: 10/08/2014
• Status: Active Grant

• Current Assignee: MAD Apparel Incorporated
• Original Assignee: MAD Apparel Incorporated
• Inventors: Korzinov, Lev, Gordhandas, Ankit, Wiebe, Christopher
• Primary Examiner(s): Eiseman, Lynsey C
• Assistant Examiner(s): Hough, Jessandra F
• Application Number:
  US14/878,831
• Publication Number:
  US 20160100803A1
• Time in Patent Office: 1,552 Days
• CPC Class Codes: A61B 2562/04 Arrangements of multiple sensors of the same type
• CPC Class Codes: A61B 5/0002 Remote monitoring of patients using telemetry, e.g. transmission of vital signals via a communication network A61B5/07 takes precedence; transmission systems for measured values G08C; transmission H04B; transmission of digital information H04L; wireless communication networks H04W
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/318 Heart-related electrical modalities, e.g. electrocardiography [ECG]
• CPC Class Codes: A61B 5/366 Detecting abnormal QRS complex, e.g. widening
• CPC Class Codes: A61B 5/389 Electromyography [EMG]
• CPC Class Codes: A61B 5/6804 Garments; Clothes
• CPC Class Codes: A61B 5/7203 for noise prevention, reduction or removal
• CPC Class Codes: A61B 5/7207 of noise induced by motion artifacts
• CPC Class Codes: A61B 5/7221 Determining signal validity, reliability or quality A61B5/7207 and A61B5/7217 take precedence
• CPC Class Codes: A61B 5/7275 Determining trends in physiological measurement data; Predicting development of a medical condition based on physiological measurements, e.g. determining a risk factor
• CPC Class Codes: A61B 5/746 Alarms related to a physiological condition, e.g. details of setting alarm thresholds or avoiding false alarms alarms per se G08B
• CPC Class Codes: A61B 5/7475 User input or interface means, e.g. keyboard, pointing device, joystick
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G16H 40/63 for local operation

Abstract

A method for communicating beat parameters to a user includes: providing an electrode module comprising a first and a second set of electrodes, associated with a first and a second sensor channel, respectively; receiving a first and a second dataset based on a first and a second set of bioelectrical signals detected from the first and the second sensor channel, respectively; receiving a supplemental dataset based on supplemental bioelectrical signals detected from a supplemental sensor module; generating a noise-mitigated power spectrum upon: generating a combined dataset based upon combining the first and second datasets, calculating 1) a heart power spectrum based on the combined data set, and 2) a supplemental power spectrum based on the supplemental dataset, and generating a noise-mitigated power spectrum based on processing the heart power spectrum with the supplemental power spectrum; and rendering information derived from a beat parameter analysis to the user.

Claims

1. A method for measuring heart beat parameters of a user with a frequency domain-based approach, the method comprising:

receiving, at a processing subsystem in communication with an electrode module coupled to a garment worn by the user and comprising a first electrode pair associated with a first sensor channel and a second electrode pair associated with a second sensor channel,

     1) a first dataset based on a first set of signals detected from the first sensor channel, and

     2) a second dataset based on a second set of signals detected from the second sensor channel, wherein the first dataset and the second dataset comprise a local noise component and a heart signal component, the garment configured to abut a first body region located inferior to a L1 lumbar vertebrae region of a user;

receiving, at the processing subsystem, an electromyography (EMG) dataset based on a set of EMG signals detected from an EMG sensor module coupled to the garment;

generating noise-mitigated power spectrum by;

calculating a combined EMG power spectrum based on combining a plurality of EMG power spectra calculated from the EMG dataset,

generating a combined dataset based upon non-linearly combining the first and second datasets, wherein the combined dataset comprises a dampened local noise component and an accentuated heart signal component,

after generating the combined dataset, calculating a heart power spectrum based on a sample of the combined data set, the sample having a time interval including signal components for a plurality of heart beats, wherein calculating the heart power spectrum comprises generating a power spectral density from a non-linear combination of multiple heart power spectra including the heart power spectrum, and

after calculating the heart power spectrum, generating the noise-mitigated power spectrum at least in part by dividing the heart power spectrum by the combined EMG power spectrum, wherein generating the noise-mitigated power spectrum further comprises smoothing power spike features in the noise-mitigated power spectrum upon applying a moving average operation to the noise-mitigated power spectrum;

generating a beat parameter analysis based upon the noise-mitigated power spectrum in the frequency domain;

and

rendering information derived from the beat parameter analysis on a user interface of an electronic device associated with the user.

2. The method of claim 1, wherein generating the noise-mitigated power spectrum comprises combining a power spectrum component corresponding to a harmonic frequency of the heart signal with a power spectrum component corresponding to a fundamental frequency of the heart signal.

3. The method of claim 1, wherein generating the noise-mitigated power spectrum comprises combining a power spectrum component corresponding to a harmonic frequency of the combined EMG power spectrum with a power spectrum component corresponding to a fundamental frequency of the combined EMG power spectrum.

4. The method of claim 1, further comprising identifying a frequency parameter with highest power in the noise-mitigated power spectrum, and wherein generating the beat parameter analysis comprises generating the beat parameter analysis based upon the frequency parameter with the highest power.

5. The method of claim 1, wherein generating the beat parameter analysis comprises:

determining a set of heart rate components, wherein the set of heart rate components is associated with a set of time intervals in a one-to-one manner; and

calculating a heart rate variability based on the set of heart rate components.

12. The method of claim 1, wherein generating the combined dataset comprises:

identifying a noise removed subspace substantially orthogonal to the heart signal component based on a correlation between the first dataset and the second dataset; and

extracting the noise removed subspace from the combined dataset.

13. The method of claim 1, wherein generating the beat parameter analysis comprises predicting a confidence parameter indicating an accuracy level of the beat parameter analysis, based on running a predictive model trained on features predictive of the accuracy level, and wherein rendering information derived from the beat parameter analysis comprises:

comparing the confidence parameter to a threshold condition; and

in response to the confidence parameter satisfying the threshold condition, rendering information derived from the beat parameter analysis.

21. The method of claim 1, wherein applying the exponential moving average operation comprises applying a function with the form:

EMAk=γ

*EMAk−

1+1−

γ

*x[k]

where EMA_k is an exponential moving average at a time period k, coefficient γ

is the degree of weighting decrease, and x[k] represents signal value during time period k.

22. The method of claim 1, wherein generating the combined dataset based upon non-linearly combining the first and second datasets comprises applying a combination function with the following form (where x is a first signal from the first dataset and y is a second signal from the second dataset):

fx,y=x*y1+|x−

y|.

6. A system for measuring heart beat parameters of a user with a frequency domain-based approach, the system comprising:

a garment configured to abut a first body region located inferior to a L1 lumbar vertebrae region of the user;

a heart signal sensor module coupled to the garment and comprising;

a first electrode pair and a second electrode pair oriented such that a first vector between electrodes of the first electrode pair and a second vector between electrodes of the second electrode pair are substantially crossed,

a first sensor channel associated with the first electrode pair, and configured to generate a first dataset based on a first set of detected signals, and

a second sensor channel associated with the second electrode pair, and configured to generate a second dataset based on a second set of detected signals, wherein the first dataset and the second dataset comprise a local noise component and a heart signal component;

an electromyography (EMG) sensor module coupled to the garment, and configured to generate an EMG dataset based on a set of EMG signals detected by the EMG sensor module; and

a control module comprising;

a communication subsystem in communication with the heart signal sensor module and the EMG sensor module; and

a processing subsystem in communication with the communication subsystem, and configured to;

calculate a combined EMG power spectrum based on combining a plurality of EMG power spectra calculated from the EMG dataset;

generate a combined dataset based upon non-linearly combining the first and second datasets, wherein the combined dataset comprises a dampened local noise component and an accentuated heart signal component,

after generating the combined dataset, calculate a heart power spectrum based on a sample of the combined data set, the sample having a time interval including signal components for a plurality of heart beats, wherein calculating the heart power spectrum comprises generating a power spectral density from a non-linear combination of multiple heart power spectra including the heart power spectrum, and,

after calculating the heart power spectrum, generate a noise-mitigated power spectrum at least in part by dividing the heart power spectrum by the combined EMG power spectrum, wherein generating the noise-mitigated power spectrum further comprises smoothing power spike features in the noise-mitigated power spectrum upon applying a moving average operation to the noise-mitigated power spectrum,

generate an analysis of a beat parameter based upon the noise-mitigated power spectrum in the frequency domain,

and

generate information to be rendered on a user interface of an electronic device associated with the user, wherein the information is derived from the beat parameter analysis.

7. The system of claim 6, wherein the EMG sensor module interfaces with the user at a second body region located substantially proximal to at least one of a group of muscles comprising:

a gluteus maximus muscle, a vastus lateralis muscle, a bicep femoris muscle, and a triceps surae muscle.

8. The system of claim 6, wherein the processing subsystem is further configured to identify a frequency parameter with highest power in the noise-mitigated power spectrum, and wherein the beat parameter analysis is generated based on the frequency parameter with the highest power.

9. The system of claim 6, wherein generating the noise-mitigated power spectrum comprises combining a power spectrum component corresponding to a harmonic frequency of the heart signal with a power spectrum component corresponding to a fundamental frequency of the heart signal.

10. The system of claim 6, wherein generating the noise-mitigated power spectrum comprises combining a power spectrum component corresponding to a harmonic frequency of the combined EMG power spectrum with a power spectrum component corresponding to the fundamental frequency of the combined EMG power spectrum.

11. The system of claim 6, wherein generating the beat parameter analysis comprises:

determining a set of heart rate components, wherein the set of heart rate components is associated with a set of time intervals in a one-to-one manner; and

calculating a heart rate variability based on the set of heart rate components.

14. A method for measuring heart beat parameters of a user with a frequency domain-based approach, the method comprising:

receiving, at a processing subsystem in communication with an electrode module coupled to a garment worn by the user and comprising a first electrode pair associated with a first sensor channel and a second electrode pair associated with a second sensor channel,

     1) a first dataset based on a first set of signals detected from the first sensor channel, and

     2) a second dataset based on a second set of signals detected from the second sensor channel, wherein the first dataset and the second dataset comprise a local noise component and a heart signal component, the garment configured to abut a first body region located inferior to a L1 lumbar vertebrae region of a user;

receiving, at the processing subsystem, an electromyography (EMG) dataset based on a set of EMG signals detected from an EMG sensor module coupled to the garment;

generating a noise-mitigated power spectrum by;

calculating a combined EMG power spectrum based on combining a plurality of EMG power spectra calculated from the EMG dataset;

generating a combined dataset based upon non-linearly combining the first and second datasets, wherein the combined dataset comprises a dampened local noise component and an accentuated heart signal component;

after generating the combined dataset, calculating a heart power spectrum based on a sample of the combined data set, the sample having a time interval including signal components for a plurality of heart beats, wherein calculating the heart power spectrum comprises generating a power spectral density from a non-linear combination of multiple heart power spectra including the heart power spectrum, and;

after calculating the heart power spectrum, generating the noise-mitigated power spectrum by;

identifying a frequency range of the combined EMG power spectrum; and

filtering a portion of the heart power spectrum corresponding to the identified frequency range to produce the noise-mitigated power spectrum, wherein generating the noise-mitigated power spectrum further comprises smoothing power spike features in the noise-mitigated power spectrum upon applying a moving average operation to the noise-mitigated power spectrum;

generating a beat parameter analysis based upon the noise-mitigated power spectrum in the frequency domain;

and

rendering information derived from the beat parameter analysis on a user interface of an electronic device associated with the user.

15. The method of claim 14, wherein the identified frequency range of the combined EMG power spectrum corresponds to a local power maximum of the combined EMG power spectrum.

16. The method of claim 14, wherein generating the noise-mitigated power spectrum comprises combining a power spectrum component corresponding to a harmonic frequency of the heart signal with a power spectrum component corresponding to a fundamental frequency of the heart signal.

17. The method of claim 14, wherein generating the noise-mitigated power spectrum comprises combining a power spectrum component corresponding to a harmonic frequency of the combined EMG power spectrum with a power spectrum component corresponding to a fundamental frequency of the combined EMG power spectrum.

18. The method of claim 14, wherein rendering information derived from the beat parameter analysis comprises sending a notification indicative of a cardiovascular health condition of the user.

19. The method of claim 14, wherein generating the noise-mitigated power spectrum further comprises filtering signals associated with an estimate heart rate less than 54 beats per minute.

20. The method of claim 14, wherein generating the noise-mitigated power spectrum further comprises filtering signals associated with a sampling rate less than a lower sampling rate threshold or greater than a higher sampling rate threshold.