System and method to detect significant arrhythmic events through a photoplethysmogram (PPG) and accelerometer

• US 10,524,736 B2
• Filed: 12/06/2013
• Issued: 01/07/2020
• Est. Priority Date: 12/14/2012
• Status: Active Grant

• Current Assignee: Koninklijke Philips N.V.
• Original Assignee: Koninklijke Philips N.V.
• Inventors: Gross, Brian David
• Primary Examiner(s): Layno, Carl H
• Assistant Examiner(s): Xie, Dacheng
• Application Number:
  US14/648,724
• Publication Number:
  US 20150305684A1
• Time in Patent Office: 2,223 Days
• Field of Search: 600300-301
• CPC Class Codes: A61B 2562/0219 Inertial sensors, e.g. accelerometers, gyroscopes, tilt switches
• 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 5/02416 using photoplethysmograph signals, e.g. generated by infrared radiation A61B5/14552 takes precedence
• CPC Class Codes: A61B 5/02427 Details of sensor
• CPC Class Codes: A61B 5/08 Detecting, measuring or recording devices for evaluating the respiratory organs A61B5/0205 takes precedence
• CPC Class Codes: A61B 5/1102 Ballistocardiography
• CPC Class Codes: A61B 5/1118 Determining activity level
• CPC Class Codes: A61B 5/113 occurring during breathing
• CPC Class Codes: A61B 5/721 using a separate sensor to detect motion or using motion information derived from signals other than the physiological signal to be measured
• CPC Class Codes: A61B 5/7221 Determining signal validity, reliability or quality A61B5/7207 and A61B5/7217 take precedence
• CPC Class Codes: A61B 5/7246 using correlation, e.g. template matching or determination of similarity
• CPC Class Codes: A61B 5/7278 Artificial waveform generation or derivation, e.g. synthesising signals from measured signals A61B5/327 takes precedence
• CPC Class Codes: A61B 5/7282 Event detection, e.g. detecting unique waveforms indicative of a medical condition A61B5/4094, A61B5/4818, A61B5/349, A61B5/0823 take precedence
• CPC Class Codes: A61N 1/05 for implantation or insertion into the body, e.g. heart electrode A61N1/06 takes precedence
• CPC Class Codes: A61N 1/36542 controlled by body motion, e.g. acceleration
• CPC Class Codes: A61N 1/36578 controlled by mechanical motion of the heart wall, e.g. measured by an accelerometer or microphone

Abstract

The medical system and method detect arrhythmic events. The medical system includes at least one processor programmed to perform the method. A photoplethysmogram (PPG) signal generated using a PPG probe positioned on or within a patient and a pulse signal generated using an accelerometer positioned on or within the patient received. Features from the PPG signal are extracted to PPG feature vectors, and features are extracted from the pulse signal to pulse feature vectors. The PPG feature vectors are correlated with the pulse feature vectors, and correlated PPG feature vectors and correlated pulse feature vectors are evaluated to detect arrhythmic events.

Claims

1. A medical system for detecting arrhythmic events, said system comprising:

at least one processor programmed to;

receive a photoplethysmogram (PPG) signal generated using a PPG probe positioned on or within a patient;

receive a pulse signal generated using an accelerometer positioned on or within the patient;

extract features from the PPG signal to generate PPG feature vectors;

extract features from the pulse signal to generate pulse feature vectors;

correlate the PPG feature vectors with the pulse feature vectors; and

evaluate correlated PPG feature vectors and correlated pulse feature vectors to detect arrhythmic events;

wherein features extracted from the PPG signal include inter pulse intervals (IPIs) for pairs of adjacent pulses of a last predetermined number of pulses; and

wherein features extracted from the pulse signal include inter pulse intervals (IPIs) for pairs of adjacent pulses of a last predetermined number of pulses.

2. The medical system according to claim 1, wherein the at least one processor is further programmed to:

receive a respiration signal generated using the accelerometer positioned on or within the patient;

extract features from the respiration signal to generate respiration feature vectors;

correlate the PPG feature vectors with the respiration feature vectors; and

evaluate correlated respiration feature vectors to detect arrhythmic events.

3. The medical system according to claim 2, wherein the at least one processor is further programmed to:

extract features from the pulse and respiration signals when both a signal quality index (SQI) of the pulse signal exceeds a predetermined threshold and an SQI of the respiration signal exceeds a predetermined threshold.

4. The medical system according to claim 1, wherein correlating the PPG feature vectors with the pulse feature vectors includes:

aligning inter pulse interval (IPI) trends of the pulse feature vectors with IPI trends of the PPG feature vectors.

5. The medical system according to claim 1, wherein the at least one processor is further programmed to:

receive a position signal, and an activity signal, generated using the accelerometer; and

extract features from the position and activity signals to generate position feature vectors and activity feature vectors, respectively, wherein evaluating the correlated PPG feature vectors and the correlated pulse feature vectors to detect the arrhythmic events uses the position and activity feature vectors.

6. The medical system according to claim 1, wherein features are extracted from the PPG signal in parallel with, and independent of, extraction of features from the pulse signal.

7. The medical system according to claim 1, wherein:

the at least one processor is further programmed to receive an activity signal generated using the accelerometer, wherein the activity signal indicates an activity level; and

evaluating the correlated PPG feature vectors and the correlated pulse feature vectors to detect the arrhythmic events includes at least one of;

detecting onset of atrial fibrillation or flutter (A-Fib) in response to inter pulse interval (IPI) trend going from regular to persistently irregularly irregular and the activity level being below a predetermined threshold;

detecting end of the A-Fib in response to IPI trend going from irregularly irregular to persistently regular and the activity level being below the predetermined threshold; and

detecting tachyarrhythmia with hemodynamic compromise in response to IPI becoming regular, fast, and persistent and the activity level being below the predetermined threshold and PPG signal amplitude dropping.

8. The medical system according to claim 1, further including

a photoplethysmogram (PPG) device generating the PPG signal using the PPG probe positioned on or within the patient;

and

an accelerometer based respiration and pulse (ABRP) device generating the pulse signal using the accelerometer positioned on or within the patient.

9. The medical system according to claim 1, wherein:

the at least one processor is further programmed to receive an activity signal generated using the accelerometer, wherein the activity signal indicates an activity level; and

evaluating the correlated PPG feature vectors and the correlated pulse feature vectors to detect the arrhythmic events includes;

detecting tachyarrhythmia with hemodynamic compromise in response to IPI becoming regular, fast, and persistent and the activity level being below the predetermined threshold and PPG signal amplitude dropping.

10. A medical method, performed by one or more processors, for detecting arrhythmic events, said method comprising:

receiving a photoplethysmogram (PPG) signal generated using a PPG probe positioned on or within a patient;

receiving a pulse signal generated using an accelerometer positioned on or within the patient;

extracting features from the PPG signal to generate PPG feature vectors;

extracting features from the pulse signal to generate pulse feature vectors;

correlating the PPG feature vectors with the pulse feature vectors; and

,

evaluating correlated PPG feature vectors and correlated pulse feature vectors to detect arrhythmic events;

wherein features extracted from the PPG signal include inter pulse intervals (IPIs) for pairs of adjacent pulses of a last predetermined number of pulses;

wherein features extracted from the pulse signal include inter pulse intervals (IPIs) for pairs of adjacent pulses of a last predetermined number of pulses; and

wherein the evaluating of the correlated PPG feature vectors and correlated pulse feature vectors to detect arrhythmic events does not use electrocardiograph (ECG) data.

11. The medical method according to claim 10, further including:

receiving a respiration signal generated using the accelerometer positioned on or within the patient;

extracting features from the respiration signal to generate respiration feature vectors;

correlating the PPG feature vectors with the respiration feature vectors; and

evaluating correlated respiration feature vectors to detect arrhythmic events.

12. The medical method according to claim 11, further including:

detecting that both a signal quality index (SQI) of the pulse signal exceeds a predetermined threshold and an SQI of the respiration signal exceeds a predetermined threshold; and

in response to the detection, extracting features from the pulse and respiration signals.

13. The medical method according to claim 10, wherein correlating the PPG feature vectors with the pulse feature vectors includes:

aligning inter pulse interval (IPI) trends of the ABRP pulse feature vectors with IPI trends of the PPG feature vectors.

14. The medical method according to claim 10, further including:

receiving a position signal, and an activity signal, generated using the accelerometer;

extracting features from the position and activity signals to generate position feature vectors and activity feature vectors, respectively; and

evaluating the correlated PPG feature vectors and the correlated pulse feature vectors to detect the arrhythmic events using the position and activity feature vectors.

15. The medical method according to claim 10, wherein features are extracted from the PPG signal in parallel with, and independent of, extraction of features from the pulse signal.

16. The medical method according to claim 10, wherein:

the method further includes receiving an activity signal generated using the accelerometer, wherein the activity signal indicates an activity level; and

evaluating the correlated PPG feature vectors and the correlated pulse feature vectors to detect the arrhythmic events includes at least one of;

detecting end of atrial fibrillation (A-Fib) in response to inter pulse interval (IPI) trend going from irregularly irregular to persistently regular and the activity level being below the predetermined threshold; and

detecting tachyarrhythmia with hemodynamic compromise in response to IPI becoming regular, fast, and persistent and the activity level being below the predetermined threshold and PPG signal amplitude dropping.

17. The medical method according to claim 10, further comprising:

based on a detected arrhythmic event, prompting a user to check for the detected arrhythmic event.

18. The medical method according to claim 10, further comprising:

based on a detected arrhythmic event, providing an alert on a display device.

19. The medical method according to claim 10, further comprising:

with the PPG probe, generating the PPG signal.

20. A non-transitory computer readable medium carrying software which contains one or more processors to perform the method according to claim 10.

21. A medical system for detecting arrhythmic events, said system comprising:

at least one processor programmed to;

receive a photoplethysmogram (PPG) signal generated using a PPG probe positioned on or within a patient;

receive a pulse signal generated using an accelerometer positioned on or within the patient;

extract features from the PPG signal to generate PPG feature vectors;

extract features from the pulse signal to generate pulse feature vectors;

correlate the PPG feature vectors with the pulse feature vectors; and

evaluate correlated PPG feature vectors and correlated pulse feature vectors to detect arrhythmic events;

wherein features extracted from the PPG signal include inter pulse intervals (IPIs) for pairs of adjacent pulses of a last predetermined number of pulses;

wherein features extracted from the pulse signal include inter pulse intervals (IPIs) for pairs of adjacent pulses of a last predetermined number of pulses; and

wherein the at least one processor is further programmed to;

extract the features from the PPG signal according to a predetermined sampling rate, the predetermined sampling rate set based upon computational resources of the medical system, the computational resources including a processing power and a memory.