Method and system for monitoring sleep

• US 10,154,790 B2
• Filed: 08/11/2008
• Issued: 12/18/2018
• Est. Priority Date: 08/21/2007
• Status: Active Grant

• Current Assignee: ResMed Sensor Technologies Limited (ResMed Incorporated)
• Original Assignee: University College Dublin
• Inventors: Heneghan, Conor, Chua, Eric C.-P., McDarby, Gareth
• Primary Examiner(s): Downey, John R
• Assistant Examiner(s): JIAN, SHIRLEY XUEYING
• Application Number:
  US12/674,344
• Publication Number:
  US 20110124979A1
• Time in Patent Office: 3,781 Days
• CPC Class Codes: A61B 5/024 Detecting, measuring or recording pulse rate or heart rate A61B5/0205, A61B5/021 take precedence
• CPC Class Codes: A61B 5/02405 Determining heart rate variability
• CPC Class Codes: A61B 5/316 Modalities, i.e. specific diagnostic methods
• CPC Class Codes: A61B 5/346 Analysis of electrocardiograms
• CPC Class Codes: A61B 5/4812 Detecting sleep stages or cycles
• CPC Class Codes: A61B 5/4818 Sleep apnoea

Abstract

A method of monitoring sleep comprises simultaneously recording a person'"'"'s electrocardiogram (ECG) and photoplethysmogram (PPG), deriving a plurality of parameters from the recorded data, and providing an output indicative of a sleep characteristic based upon an analysis of the parameters. The ECG and PPG may be recorded using an apparatus which is a combination of a Holter monitor and a pulse oximeter, which is wearable in ambulatory manner.

Claims

1. A method of monitoring sleep stages or level of sleepiness, the method comprising using one or more processors to:

acquire and record data of a person'"'"'s electrocardiogram (ECG) and photoplethysmogram (PPG), the data being acquired and recorded simultaneously,

derive a plurality of parameters from both the recorded ECG and PPG data, wherein the plurality of parameters includes (1) at least one parameter derived from the ECG data independently of the PPG data, and (2) at least one parameter derived from the PPG data independently of the ECG data,

extract features from the plurality of parameters, and

provide an output indicative of a sleep stage or a quantitative level of sleepiness based upon an analysis of at least a subset of the extracted features from the derived plurality of parameters, wherein a trained statistical classifier in a processor of the one or more processors evaluates, on an epoch-by-epoch basis, the extracted features from both (1) the at least one parameter derived from the ECG data and (2) the at least one parameter derived from the PPG data, to generate the output indicative of the sleep stage or the quantitative level of sleepiness.

2. The method of claim 1, wherein the ECG and PPG data is acquired in an ambulatory manner.

3. The method of claim 2, wherein the ECG and PPG data is acquired using a combination of a Holter monitor and a pulse oximeter.

4. The method of claim 1, wherein the plurality of parameters include at least one of RR interval, pulse amplitude and respiratory rate.

5. The method of claim 4, wherein the plurality of parameters are analyzed by extracting the features, wherein the features are suitable for classifying sleep stages or quantitative level of sleepiness, and by feeding a subset of the features into a quantitative classifier to obtain an estimate of the sleep stage or quantitative level of sleepiness.

6. The method of claim 5, wherein the features include at least one of PPG spectral spread, PPG pulse amplitude, respiratory rate variability, detrended fluctuation analysis, time and frequency analysis and multi-scale entropy.

7. The method of claim 5, wherein the output is used to determine sleep stages comprising at least one of wakefulness, rapid-eye-movement sleep, and non-rapid-eye-movement sleep.

8. The method of claim 5, comprising assigning sleep stages or quantitative levels of sleepiness to specific epochs of time of greater than 10 seconds in duration.

9. The method of claim 8, comprising classifying a single epoch according to a classification of surrounding epochs.

10. The method of claim 5, further comprising estimating the output of a psychomotor vigilance test during wakefulness using the output of the quantitative classifier.

11. The method of claim 1, wherein the one or more processors derive an additional parameter from both the recorded ECG data and the recorded PPG data, and wherein the output indicative of a sleep stage or a quantitative level of sleepiness is based upon an analysis of parameters that includes the additional parameter.

12. The method of claim 11, wherein the additional parameter derived from both the recorded PPG data and the recorded ECG data is pulse arrival time.

13. The method of claim 1 wherein the subset of the extracted features from the plurality of parameters includes a feature derived by detrended fluctuation analysis.

14. The method of claim 1 wherein the subset of the extracted features from the plurality of parameters includes a feature derived by multi-scale entropy analysis.

15. The method of claim 1 wherein the trained statistical classifier evaluates a PPG spectral spread feature to generate the output indicative of the sleep stage or the quantitative level of sleepiness.

16. The method of claim 1 wherein the trained statistical classifier evaluates the extracted features from (1) the at least one parameter derived from the ECG data and (2) the at least one parameter derived from the PPG data, to generate the output indicative of the quantitative level of sleepiness.

17. The method of claim 1 wherein the trained statistical classifier evaluates the extracted features from (1) the at least one parameter derived from the ECG data and (2) the at least one parameter derived from the PPG data, to generate the output indicative of the sleep stage.

18. A system for monitoring sleep stage or level of sleepiness, the system comprising:

a sensing apparatus configured to simultaneously acquire electrocardiogram (ECG) and photoplethysmogram (PPG) data from a monitored subject;

a computer executable memory configured to record the acquired ECG and PPG data, and to store computer executable instructions; and

a processor configured to execute the computer executable instructions to derive a plurality of parameters from both the recorded ECG and PPG data, to extract features from the plurality of parameters, and to provide an output indicative of a sleep stage or a quantitative level of sleepiness based upon an analysis of at least a subset of the extracted features from the plurality of parameters, wherein the plurality of parameters includes (1) at least one parameter derived from the ECG data independently of the PPG data, and (2) at least one parameter derived from the PPG data independently of the ECG data, wherein the processor is configured with a trained statistical classifier that evaluates, on an epoch-by-epoch basis, the extracted features from both (1) the at least one parameter derived from the ECG data and (2) the at least one parameter derived from the PPG data, to generate the output indicative of the sleep stage or the quantitative level of sleepiness.

19. The system of claim 18, wherein the sensing apparatus is wearable in ambulatory manner.

20. The system of claim 19, wherein the sensing apparatus comprises a combination of a Holter monitor and a pulse oximeter.

21. The system of claim 18, wherein the processor is configured to derive an additional parameter from both the recorded ECG data and the recorded PPG data, and provide an output indicative of quantitative level of sleepiness or sleep stage based upon an analysis of parameters that includes the additional parameter.

22. The system of claim 21, wherein the additional parameter derived from both the recorded PPG data and the recorded ECG data is pulse arrival time.

23. The system of claim 18 wherein the subset of the extracted features from the plurality of parameters includes a feature derived by detrended fluctuation analysis.

24. The system of claim 18 wherein the subset of the extracted features from the plurality of parameters includes a feature derived by multi-scale entropy analysis.

25. The system of claim 18 wherein the trained statistical classifier is configured to evaluate a PPG spectral spread feature to generate the output indicative of the sleep stage or the quantitative level of sleepiness.

26. The system of claim 18 wherein the trained statistical classifier is configured to evaluate the extracted features from (1) the at least one parameter derived from the ECG data and (2) the at least one parameter derived from the PPG data, to generate the output indicative of the quantitative level of sleepiness.

27. The system of claim 18 wherein the trained statistical classifier is configured to evaluate the extracted features from (1) the at least one parameter derived from the ECG data and (2) the at least one parameter derived from the PPG data, to generate the output indicative of the sleep stage.