Wearable biometric measurement device

• US 10,660,536 B2
• Filed: 09/15/2015
• Issued: 05/26/2020
• Est. Priority Date: 09/15/2015
• Status: Active Grant

• Current Assignee: ZEPP
• Original Assignee: Huami, Inc.
• Inventors: Wang, Fei, Chen, Ting
• Primary Examiner(s): Flory, Christopher A
• Assistant Examiner(s): Huh, Vynn V
• Application Number:
  US14/854,569
• Publication Number:
  US 20170071483A1
• Time in Patent Office: 1,715 Days
• 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 5/01 Measuring temperature of body parts ; Diagnostic temperature sensing, e.g. for malignant or inflamed tissue clinical contact thermometers G01K13/20
• CPC Class Codes: A61B 5/0245 by using sensing means generating electric signals, i.e. ECG signals
• CPC Class Codes: A61B 5/14551 for measuring blood gases
• CPC Class Codes: A61B 5/332 Portable devices specially adapted therefor
• CPC Class Codes: A61B 5/349 Detecting specific parameters of the electrocardiograph cycle
• CPC Class Codes: A61B 5/4818 Sleep apnoea
• CPC Class Codes: A61B 5/681 Wristwatch-type devices
• CPC Class Codes: A61B 5/7207 of noise induced by motion artifacts
• CPC Class Codes: A61B 5/726 using Wavelet transforms
• 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 5/742 using visual displays A61B5/339 takes precedence
• CPC Class Codes: G16H 40/63 for local operation

Abstract

A wearable biometric measurement device includes a ring body having interior and exterior surfaces, an electrocardiogram component including first and second electrodes wherein a single lead electrocardiogram is formed upon contact with the first electrode and the second electrode, a pulse oximeter component including a light emitter and a photodetector wherein the photodetector receives one or more signals transmitted by the light emitter, and a communication component. A system for measuring and processing biometric data includes a wearable biometric measurement device, a monitor device, and a server. A method for measuring biometric data using a wearable biometric measurement device includes pre-processing biometric signals, feature extraction on the signals, and event prediction with respect to the signals.

Claims

1. A method for analysis and prediction of disease risk factors or disease diagnoses of a user of a wearable device based on biometric data generated at the wearable device and communicated to a cloud server using a monitor device intermediate to the wearable device and the cloud server, the method comprising:

generating, at the wearable device, at least one electrocardiograph (ECG) signal and at least one pulse oximeter (PO) signal associated with the user of the wearable device, wherein the wearable device is defined by a ring body having an interior surface and an exterior surface, wherein the at least one ECG signal is generated responsive to the user contacting a portion of the exterior surface using a first finger of the user while the interior surface is in communication with a second finger of the user;

outputting the at least one ECG signal and the at least one PO signal to a display of the wearable device;

transmitting the at least one ECG signal and the at least one PO signal from the wearable device to the monitor device using a short range wireless connection established between the wearable device and the monitor device;

responsive to the monitor device receiving the at least one ECG signal and the at least one PO signal using the short range wireless connection, processing the at least one ECG signal and the at least one PO signal at the monitor device by;

performing, at the monitor device, pre-processing on the at least one ECG signal and the at least one PO signal to denoise the at least one ECG signal and to denoise the at least one PO signal; and

combining, at the monitor device, the pre-processed at least one ECG signal and the pre-processed at least one PO signal into a combined biometric signal;

outputting the pre-processed at least one ECG signal, the pre-processed at least one PO signal, and the combined biometric signal to a display of the monitor device;

transmitting the combined biometric signal from the monitor device to the cloud server; and

responsive to the cloud server receiving the combined biometric signal from the monitor device, analyzing the combined biometric signal at the cloud server to determine at least one predicted event for the user of the wearable device by;

performing, at the cloud server, feature extraction against the combined biometric signal to determine one or more extracted wave features of the combined biometric signal; and

determining, at the cloud server, the at least one predicted event associated with the user based on the one or more extracted wave features and based on one or more disease models corresponding to the disease risk factors or the disease diagnoses of the user of the wearable device; and

storing, at the cloud server, data indicative of the at least one predicted event.

2. The method of claim 1, wherein performing the pre-processing on the at least one ECG signal and the at least one PO signal to denoise the at least one ECG signal and to denoise the at least one PO signal comprises:

removing a baseline wander from a signal, wherein the signal is at least one of the at least one ECG signal or the at least one PO signal;

applying a band-pass filter to the signal;

detecting a periodicity of the signal; and

applying an adaptive motion noise reduction filter to the signal.

3. The method of claim 1, wherein performing the feature extraction against the combined biometric signal to determine the one or more extracted wave features of the combined biometric signal comprises:

identifying QRS complex features of the combined biometric signal; and

identifying each of;

wavelet transformation features of the combined biometric signal;

wave magnitude features relative to the QRS complex features and related wave features; and

periodicity interval features relative to the QRS complex features and related wave features,

wherein the one or more extracted wave features comprise the at least one of the wavelet transformation features, the wave magnitude features, and the periodicity interval features identified.

4. The method of claim 1, wherein determining the at least one predicted event associated with the user based on the one or more extracted wave features and based on one or more disease models corresponding to the disease risk factors or the disease diagnoses of the user of the wearable device comprises:

analyzing, at the cloud server, the combined biometric signal and the one or more extracted wave features to present an analyzed signal;

determining, at the cloud server, a first result by comparing the analyzed signal to the one or more disease models;

determining, at the cloud server, a second result by comparing the analyzed signal to previously stored data, wherein the previously stored data is stored at the cloud server and includes at least one of historical ECG signal data, historical PO signal data, and historical combined biometric signal data; and

determining, at the cloud server, the at least one predicted event based on the first result and the second result.

5. The method of claim 1, further comprising:

extracting, at the wearable device, the at least one ECG signal responsive to the user contacting the portion of the exterior surface of the ring body using the first finger of the user.

6. The method of claim 1, further comprising:

determining, by a multi-modal decision fusion performed at the cloud server, an identity of the user.

7. An apparatus for analysis and prediction of disease risk factors or disease diagnoses of a user of a wearable device based on biometric data generated at the wearable device and communicated to a cloud server using the apparatus, wherein the apparatus is intermediate to the wearable device and the cloud server, the apparatus comprising:

a display;

a non-transitory memory; and

a processor configured to execute instructions stored in the non-transitory memory to;

receive, using a short range wireless connection established between the wearable device and the apparatus, at least one electrocardiograph (ECG) signal and at least one pulse oximeter (PO) signal associated with a user from the wearable device, wherein the wearable device is defined by a ring body having an interior surface and an exterior surface, wherein the at least one ECG signal is generated responsive to the user contacting a portion of the exterior surface using a first finger of the user while the interior surface is in communication with a second finger of the user, wherein the at least one ECG signal and the at least one PO signal are output to a wearable device display;

perform pre-processing on the at least one ECG signal and the at least one PO signal to denoise the at least one ECG signal and to denoise the at least one PO signal;

combine the pre-processed at least one ECG signal and the pre-processed at least one PO signal into a combined biometric signal;

output the pre-processed at least one ECG signal, the pre-processed at least one PO signal, and the combined biometric signal to the display; and

transmit the combined biometric signal to the cloud server, wherein the cloud server analyzes the combined biometric signal to determine at least one predicted event for the user of the wearable device and stores data indicative of the at least one predicted event by;

performing feature extraction against the combined biometric signal to determine one or more extracted wave features of the combined biometric signal;

determining the at least one predicted event associated with the user based on the one or more extracted wave features and based on one or more disease models corresponding to the disease risk factors or the disease diagnoses of the user of the wearable device; and

storing data indicative of the at least one predicted event.

8. The apparatus of claim 7, wherein the instructions to perform the pre-processing on the at least one ECG signal and the at least one PO signal to denoise the at least one ECG signal and to denoise the at least one PO signal comprise instructions to:

remove a baseline wander from a signal, wherein the signal is at least one of the at least one ECG signal or the at least one PO signal;

apply a band-pass filter to the signal;

detect a periodicity of the signal; and

apply an adaptive motion noise reduction filter to the signal.

9. The apparatus of claim 7, wherein the cloud server performs the feature extraction against the combined biometric signal to determine the one or more extracted wave features of the combined biometric signal by:

identifying QRS complex features of the combined biometric signal; and

identifying each of;

wavelet transformation features of the combined biometric signal;

wave magnitude features relative to the QRS complex features and related wave features; and

periodicity interval features relative to the QRS complex features and related wave features,

wherein the one or more extracted wave features comprise the at least one of the wavelet transformation features, the wave magnitude features, and the periodicity interval features identified.

10. The apparatus of claim 7, wherein the cloud server the at least one predicted event associated with the user based on the one or more extracted wave features and based on one or more disease models corresponding to the disease risk factors or the disease diagnoses of the user of the wearable device by:

analyzing the combined biometric signal and the one or more extracted features to present an analyzed signal;

determining a first result by comparing the analyzed signal to the one or more disease models;

determining a second result by comparing the analyzed signal to previously stored data, wherein the previously stored data is stored at the cloud server and includes at least one of historical ECG signal data, historical PO signal data, and historical combined biometric signal data; and

determining the at least one predicted event based on the first result and the second result.

11. The apparatus of claim 7, wherein the cloud server is further configured to determine, by a multi-modal decision fusion, an identity of the user.

12. The apparatus of claim 7, wherein the apparatus comprises a monitor device configured to receive the at least one ECG signal and the at least one PO signal from the wearable device.

13. An apparatus for analysis and prediction of disease risk factors or disease diagnoses of a user of the apparatus, comprising:

a wearable ring body having an interior surface and an exterior surface;

a display;

a non-transitory memory;

a processor configured to execute instructions stored in the non-transitory memory to;

identify at least one electrocardiograph (ECG) signal and at least one pulse oximeter (PO) signal associated with a user while the interior surface of the wearable ring body is in communication with a first finger of the user;

output the at least one ECG signal and the at least one PO signal to the display;

perform pre-processing on the at least one ECG signal and the at least one PO signal to denoise the at least one ECG signal and to denoise the at least one PO signal;

combine the pre-processed at least one ECG signal and the pre-processed at least one PO signal into a combined biometric signal;

transmit the pre-processed at least one ECG signal, the pre-processed at least one PO signal, and the combined biometric signal to a monitor device, wherein the monitor device outputs the pre-processed at least one ECG signal, the pre-processed at least one PO signal, and the combined biometric signal at a monitor device display;

perform feature extraction against the combined biometric signal to determine one or more extracted wave features of the combined biometric signal;

determine at least one predicted event associated with the user based on the one or more extracted wave features and based on one or more disease models corresponding to the disease risk factors or the disease diagnoses of the user; and

output data indicative of the at least one predicted event to the display; and

at least one biometric sensor coupled to a portion of the exterior surface of the wearable ring body to produce the at least one ECG signal associated with the user responsive to the user contacting the portion of the exterior surface of the wearable ring body using a second finger of the user.

14. The apparatus of claim 13, wherein the instructions to perform the pre-processing on the at least one ECG signal and the at least one PO signal to denoise the at least one ECG signal and to denoise the at least one PO signal comprise instructions to:

remove a baseline wander from a signal, wherein the signal is at least one of the at least one ECG signal or the at least one PO signal;

apply a band-pass filter to the signal;

detect a periodicity of the signal; and

apply an adaptive motion noise reduction filter to the signal.

15. The apparatus of claim 13, wherein the instructions to perform the feature extraction against the combined biometric signal to determine the one or more extracted features comprise instructions to:

identify QRS complex features of the combined biometric signal; and

identify each of;

wavelet transformation features of the combined biometric signal;

wave magnitude features relative to the QRS complex features and related wave features; and

periodicity interval features relative to the QRS complex features and related wave features,

wherein the one or more extracted wave features comprise the at least one of the wavelet transformation features, the wave magnitude features, and the periodicity interval features identified.

16. The apparatus of claim 13, wherein the instructions to determine the at least one predicted event associated with the user based on the one or more extracted wave features and based on one or more disease models corresponding to the disease risk factors or the disease diagnoses of the user comprise instructions to:

analyze the combined biometric signal and the one or more extracted features to present an analyzed signal;

determine a first result by comparing the analyzed signal to the one or more disease models;

determine a second result by comparing the analyzed signal to previously stored data, wherein the previously stored data includes at least one of historical ECG signal data, historical PO signal data, and historical combined biometric signal data; and

determine the at least one predicted event based on the first result and the second result.

17. The apparatus of claim 13, wherein the processor is further configured to execute instructions stored in the non-transitory memory to:

extract the at least one ECG signal responsive to the user contacting the portion of the exterior surface of the wearable ring body using the second finger of the user.

18. The apparatus of claim 13, wherein the processor is further configured to execute instructions stored in the non-transitory memory to:

determine, by a multi-modal decision fusion, an identity of the user.