Cardiac condition detection

• US 10,123,741 B2
• Filed: 11/30/2016
• Issued: 11/13/2018
• Est. Priority Date: 11/30/2016
• Status: Active Grant

• Current Assignee: ZEPP
• Original Assignee: Huami, Inc.
• Inventors: Wang, Yuanxiang, Wang, Yuchen, Wang, Fei
• Primary Examiner(s): Schaetzle, Kennedy
• Application Number:
  US15/364,743
• Publication Number:
  US 20180146922A1
• Time in Patent Office: 713 Days
• Field of Search: None
• CPC Class Codes: A41D 1/002 adapted to accommodate electronic equipment patients' garments for medical monitoring A41D13/1281; luminous ornaments A41D27/085
• CPC Class Codes: A61B 2562/0209 Special features of electrodes classified in A61B5/24, A61B5/25, A61B5/283, A61B5/291, A61B5/296, A61B5/053
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• CPC Class Codes: A61B 5/0006 ECG or EEG signals
• CPC Class Codes: A61B 5/0022 Monitoring a patient using a global network, e.g. telephone networks, internet
• CPC Class Codes: A61B 5/0205 Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
• CPC Class Codes: A61B 5/11 Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb for measuring pulse A61B5/02 ;A61B5/1038 takes precedence; motion detection to correct for motion artifacts in physiological signals A61B5/721
• CPC Class Codes: A61B 5/1102 Ballistocardiography
• CPC Class Codes: A61B 5/1116 Determining posture transitions
• CPC Class Codes: A61B 5/1117 Fall detection
• CPC Class Codes: A61B 5/282 Holders for multiple electrodes for introduction into the body A61B5/287
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/332 Portable devices specially adapted therefor
• CPC Class Codes: A61B 5/339 Displays specially adapted therefor
• CPC Class Codes: A61B 5/35 by template matching
• CPC Class Codes: A61B 5/361 Detecting fibrillation
• CPC Class Codes: A61B 5/363 Detecting tachycardia or bradycardia
• CPC Class Codes: A61B 5/6804 Garments; Clothes
• CPC Class Codes: A61B 5/7264 Classification of physiological signals or data, e.g. using neural networks, statistical classifiers, expert systems or fuzzy systems neural networks per se G06N3/00; expert systems per se G06N5/00
• 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 5/747 : in case of emergency, i.e. alerting emergency services
• CPC Class Codes: A61N 1/025 : Digital circuitry features of electrotherapy devices, e.g. memory, clocks, processors
• CPC Class Codes: A61N 1/046 : Specially adapted for shock therapy, e.g. defibrillation
• CPC Class Codes: A61N 1/0484 : Garment electrodes worn by the patient
• CPC Class Codes: A61N 1/3987 : characterised by the timing or triggering of the shock
• CPC Class Codes: G16H 50/20 : for computer-aided diagnosis, e.g. based on medical expert systems

Abstract

Systems and methods for determining a cardiac condition of the human based on acceleration measurements received from one or more accelerometers of a garment worn by a human, electrocardiogram measurements received from electrodes of a garment worn by the human, and a machine learning model previously determined based on training data. An alert message that indicates the determined cardiac condition may be transmitted or displayed to the human wearing the garment or another person who will assist the human. For example, a posture pattern may be determined based on the acceleration measurements. The cardiac condition may be determined based in part on the posture pattern.

Claims

1. A system, comprising:

a garment comprising a plurality of electrodes and one or more accelerometers; and

a processing apparatus that is configured to;

receive acceleration measurements from the one or more accelerometers;

receive electrocardiogram measurements from the plurality of electrodes;

determine a posture pattern of a human based on the acceleration measurements;

determine a cardiac condition of the human based on the posture pattern and the electrocardiogram measurements by inputting features extracted from the electrocardiogram measurements and the posture pattern to a machine learning model previously determined based on training data; and

transmit or display an alert message that indicates the determined cardiac condition.

2. The system of claim 1, in which the one or more accelerometers comprise:

a first accelerometer that is positioned, as a component of the garment, such that the first accelerometer is capable of being fastened to a chest of the human.

3. The system of claim 2, in which the one or more accelerometers comprise:

a second accelerometer that is positioned, as a component of the garment, such that the second accelerometer is capable of being fastened to a waist of the human.

4. The system of claim 1, in which the processing apparatus is attached to the garment.

5. The system of claim 1, in which the garment comprises:

one or more motion sensor modules that include the one or more accelerometers;

an electrocardiogram sensor module that is connected via conductors to the plurality of electrodes and detects the electrocardiogram measurements from the electrodes; and

conductors that connect the processing apparatus to the electrocardiogram sensor module and to the one or more motion sensor modules.

6. The system of claim 1, in which the processing apparatus is configured to receive the acceleration measurements and the electrocardiogram measurements via wireless communications from one or more sensor modules connected to the one or more accelerometers and the plurality of electrodes.

7. The system of claim 1, comprising a defibrillator, and in which the processing apparatus is configured to:

activate the defibrillator responsive to the determination of the cardiac condition.

8. The system of claim 1, in which the cardiac condition is ventricular fibrillation or supraventricular tachycardia.

9. The system of claim 1, in which the machine learning module is a first machine learning model, and in which the processing apparatus is configured to:

input features extracted from the acceleration measurements to a second machine learning module that has been determined based on training data; and

determine the posture pattern based on output of the second machine learning module.

10. The system of claim 9, in which the posture pattern is selected from a set of posture patterns that comprises sitting, standing, walking, running, lying, and fallen.

11. The system of claim 9, in which the processing apparatus is configured to:

determine a feature based on an estimate of an entropy of a normalized power spectral density of a sequence of the acceleration measurements; and

determine the posture pattern based in part on the feature.

12. The system of claim 9, in which the processing apparatus is configured to:

determine a feature based on an estimate of a power spectral density within a frequency band for a sequence of the acceleration measurements; and

determine the posture pattern based in part on the feature.

13. The system of claim 1, in which the processing apparatus is configured to:

determine a feature based on an estimate of an entropy of a normalized power spectral density of a sequence of the acceleration measurements; and

determine the cardiac condition based in part on the feature.

14. The system of claim 1, in which the processing apparatus is configured to:

determine a feature based on an estimate of a power spectral density within a frequency band for a sequence of the acceleration measurements; and

determine the cardiac condition based in part on the feature.

15. The system of claim 1, in which the processing apparatus is configured to:

determine a feature based on an estimate of an entropy of a normalized power spectral density of a sequence of the electrocardiogram measurements; and

determine the cardiac condition based in part on the feature.

16. The system of claim 1, in which the processing apparatus is configured to:

determine a feature based on an estimate of a power spectral density within a frequency band for a sequence of the electrocardiogram measurements; and

determine the cardiac condition based in part on the feature.

17. The system of claim 1, in which the processing apparatus is configured to collect training data with the garment, and in which some of the training data has been gathered from the human using the garment.

18. The system of claim 1, in which the processing apparatus is configured to:

determine a profile of the human including a plurality of features based on the electrocardiogram measurements;

compare the profile to a registered template associated with an identity;

confirm the identity of the human based on comparison of the profile to the registered template; and

transmit or display information about the confirmed identity along with the alert message that indicates the determined cardiac condition.

19. A method comprising:

receiving acceleration measurements from one or more accelerometers of a garment worn by a human;

receiving electrocardiogram measurements from electrodes of a garment worn by the human;

determining a posture pattern of the human based on the acceleration measurements;

determining a cardiac condition of the human based on the posture pattern and the electrocardiogram measurements by inputting features extracted from the electrocardiogram measurements and the posture pattern to a machine learning model previously determined based on training data; and

transmitting or displaying an alert message that indicates the determined cardiac condition.

20. A system comprising:

a processor;

a network interface; and

a memory storing instructions executable by the processor that upon execution by the processor cause the processor to perform operations comprising;

receiving acceleration measurements from one or more accelerometers of a garment worn by a human;

receiving electrocardiogram measurements from electrodes of a garment worn by the human;

determining a posture pattern of the human based on the acceleration measurements;

determining a cardiac condition of the human based on the posture pattern and the electrocardiogram measurements by inputting features extracted from the electrocardiogram measurements and the posture pattern to a machine learning model previously determined based on training data; and

transmitting, via the network interface, an alert message that indicates the determined cardiac condition.