Identifying and characterizing nocturnal motion and stages of sleep
Disclosed herein are systems, methods and a computer readable storage medium that determine an individual's movement during sleep. In some embodiments, the identification is based on data from a motion sensor included in wearable device. In some embodiments, the individual wears the wearable device on his wrist during the individual's resting and/or sleep activities. In some embodiments, the system, methods and the computer readable storage medium automatically identify the various (stages) types of sleep during an individual's sleep periods that include periods of REM sleep, deep sleep, and light sleep and that are indicative of the individual's overall well-being and health.
1. A method for identifying rapid eye movement (REM) sleep, the method comprising:
receiving heart rate data associated with a sleeping user at a plurality of time points;
identifying a first time domain characteristic of the heart rate data, wherein identifying the first time domain characteristic of heart rate data comprises analyzing changes in a standard deviation of heart rate in the heart rate data and determining a first time span based on the analyzed changes in the standard deviation of heart rate for the first time domain characteristic;
identifying a second time domain characteristic of the heart rate data, wherein identifying the second time domain characteristic of heart rate data comprises determining a median heart rate for the plurality of time points, analyzing changes in the median heart rate of the heart rate data, and determining a second time span based on the analyzed changes in the median heart rate for the second time domain characteristic; and
determining a presence of REM sleep based on an overlap of the first and second time spans for the identified first and second time domain characteristics of the heart rate data.
2. The method of claim 1 , wherein the changes in the standard deviation of heart rate in the heart rate data comprise an increase in the standard deviation of heart rate, and wherein the changes in the median heart rate of the heart rate data comprise an increase in the median heart rate.
3. The method of claim 1 further comprising:
identifying a frequency domain characteristic of the heart rate data, analyzing changes in the frequency domain characteristic, and determining a third time span based on the analyzed changes in the frequency domain characteristic; and
determining the presence of REM sleep is further based on an overlap of the first, the second, and the third time spans for the identified first and second time domain characteristics and the identified frequency domain characteristic of the heart rate data.
4. The method of claim 3 , wherein the frequency domain characteristic comprises a change in low frequency oscillation.
5. The method of claim 4 , wherein the change in low frequency oscillation comprises an increase in low frequency oscillations.
6. The method of claim 1 , wherein receiving the heart rate data comprises receiving data from an optical sensor.
7. The method of claim 6 , wherein the data from the optical sensor comprises data indicating an amount of light absorbed by tissue of the user.
8. A system for identifying rapid eye movement (REM) sleep, the system comprising:
a memory element configured to store data;
a processor operable to execute instructions associated with the data; and
a module configured to:
receive heart rate data associated with a sleeping user at a plurality of time points;
identify a first time domain characteristic of the heart rate data, wherein identifying the first time domain characteristic of heart rate data comprises analyzing changes in a standard deviation of heart rate in the heart rate data and determining a first time span based on the analyzed changes in the standard deviation of heart rate for the first time domain characteristic;
identify a second time domain characteristic of the heart rate data, wherein identifying the second time domain characteristic of heart rate data comprises determining a median heart rate for the plurality of time points, analyzing changes in the median heart rate of the heart rate data, and determining a second time span based on the analyzed changes in the median heart rate for the second time domain characteristic; and
determine a presence of REM sleep based on an overlap of the first and second time spans for the identified first and second time domain characteristics of the heart rate data.
9. The system of claim 8 , wherein the changes in the standard deviation of heart rate in the heart rate data comprise an increase in the standard deviation of heart rate, and wherein the changes in the median heart rate of the heart rate data comprise an increase in the median heart rate.
10. The system of claim 8 , wherein the module is further configured to identify a frequency domain characteristic of the heart rate data, analyze changes in the frequency domain characteristic, and determine a third time span based on the analyzed changes to the frequency domain characteristic; and determine the presence of REM sleep is further based on an overlap of the first, the second, and the third time spans for the identified first and second time domain characteristics and the identified frequency domain characteristic of the heart rate data.
11. The system of claim 10 , wherein the changes in the standard deviation of heart rate in the heart rate data comprise an increase in the standard deviation of heart rate, and wherein the frequency domain characteristic comprises a change in low frequency oscillation.
12. The system of claim 11 , wherein the change in low frequency oscillation comprises an increase in low frequency oscillations.
13. The system of claim 8 , wherein receiving the heart rate data comprises receiving data from an optical sensor.
14. The system of claim 13 , wherein the data from the optical sensor comprises data indicating an amount of light absorbed by tissue of the user.
15. One or more non-transitory tangible media that includes code for execution and when executed by a processor is operable to perform operations comprising:
receiving heart rate data associated with a sleeping user at a plurality of time points;
identifying a first time domain characteristic of the heart rate data, wherein identifying the first time domain characteristic of heart rate data comprises analyzing changes in a standard deviation of heart rate in the heart rate data and determining a first time span based on the analyzed changes in the standard deviation of heart rate for the first time domain characteristic;
identifying a second time domain characteristic of the heart rate data, wherein identifying the second time domain characteristic of heart rate data comprises determining a median heart rate for the plurality of time points, analyzing changes in the median heart rate of the heart rate data, and determining a second time span based on the analyzed changes in the median heart rate for the second time domain characteristic; and
determining a presence of rapid eye movement (REM) sleep based on an overlap of the first and second time spans for the identified first and second time domain characteristics of the heart rate data.
16. The media of claim 15 , wherein the changes in the standard deviation of heart rate in the heart rate data comprise an increase in the standard deviation of heart rate, and wherein the changes in the median heart rate of the heart rate data comprise an increase in the median heart rate.
17. The media of claim 15 , wherein the operations further comprise identifying a frequency domain characteristic of the heart rate data, analyzing changes in the frequency domain characteristic, and determining a third time span based on the analyzed changes in the frequency domain characteristic; and determining the presence of REM sleep is further based on an overlap of the first, the second, and the third time spans for the identified first and second time domain characteristics and the identified frequency domain characteristic of the heart rate data.
18. The media of claim 17 , wherein the frequency domain characteristic comprises a change in low frequency oscillation.