CN112971724A - Method for detecting sleeping point - Google Patents
Method for detecting sleeping point Download PDFInfo
- Publication number
- CN112971724A CN112971724A CN202110178505.7A CN202110178505A CN112971724A CN 112971724 A CN112971724 A CN 112971724A CN 202110178505 A CN202110178505 A CN 202110178505A CN 112971724 A CN112971724 A CN 112971724A
- Authority
- CN
- China
- Prior art keywords
- sleep
- period
- point
- standard deviation
- public
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/48—Other medical applications
- A61B5/4806—Sleep evaluation
- A61B5/4812—Detecting sleep stages or cycles
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/08—Detecting, measuring or recording devices for evaluating the respiratory organs
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B5/00—Measuring for diagnostic purposes; Identification of persons
- A61B5/103—Detecting, measuring or recording devices for testing the shape, pattern, colour, size or movement of the body or parts thereof, for diagnostic purposes
- A61B5/11—Measuring movement of the entire body or parts thereof, e.g. head or hand tremor, mobility of a limb
Abstract
According to the sleep-in point detection method, the heart rate, the respiration rate and the three-axis acceleration standard deviation of a testee in a preset time period are calculated through electrocardiosignals, respiration signals and three-axis acceleration signals of the testee in the preset time period; aligning the heart rate and the respiration rate obtained by calculation with the three-axis acceleration standard deviation to obtain the aligned heart rate and respiration rate and the three-axis acceleration standard deviation; performing sliding window on the aligned heart rate, respiration rate and triaxial acceleration standard deviation by using a characteristic window with a preset time length, and calculating the body movement times, the heart rate average value and the respiration rate standard deviation in each characteristic window; judging the time period in which the body movement times are less than the preset times and the duration time exceeds the preset long time as the falling asleep period; calculating a sleep onset candidate point for each of the sleep onset periods; determining a sleep onset point from the sleep onset candidate points.
Description
Technical Field
The application relates to a sleep detection technology, in particular to a method for detecting a sleeping point.
Background
Sleep time accounts for one third of human life, poor sleep quality can lead to physical and mental fatigue during daytime hours, and sleep disorders can lead to depression, diabetes, hypertension, and numerous other cardiovascular disorders. The sleep onset point is used as the starting time point of sleep quality judgment, and plays a very important role in accurately judging the sleep quality.
The current 'gold standard' of sleep monitoring is a Polysomnography (PSG), and the sleep point of a subject can be accurately detected based on monitored electroencephalogram, electrocardio, myoelectricity, eye movement, respiration, blood oxygen signals and the like, so that basic guarantee is provided for accurately judging sleep quality. However, the PSG requires a subject to wear the electroencephalogram electrode in a sleep laboratory, which makes the subject feel uncomfortable and affects the subject to fall asleep to a certain extent, and the use of the PSG is greatly limited due to the fact that the PSG equipment is expensive, low in efficiency, limited in site, required to be judged by a professional doctor, and the like.
The patent with application number 201510371905.4 proposes a solution for detecting the falling asleep point, which monitors the turnover number and heart rate of the subject within 15 minutes, then sets the threshold of the turnover number and the threshold of the heart rate, and compares the turnover number and the heart rate within 15 minutes with the thresholds by comparison to obtain the corresponding falling asleep point. The method is too simple and is easy to cause misjudgment. The turning habits of all tested persons are different, the individual difference is large, the setting of the turning frequency threshold value is extremely not universal, and no scientific basis exists.
Disclosure of Invention
In view of the above problems, the present application aims to provide a method for detecting a falling asleep point.
According to the sleep-in point detection method, the heart rate, the respiration rate and the three-axis acceleration standard deviation of a testee in a preset time period are calculated through electrocardiosignals, respiration signals and three-axis acceleration signals of the testee in the preset time period; aligning the heart rate and the respiration rate obtained by calculation with the three-axis acceleration standard deviation to obtain the aligned heart rate and respiration rate and the three-axis acceleration standard deviation; performing sliding window on the aligned heart rate, respiration rate and triaxial acceleration standard deviation by using a characteristic window with a preset time length, and calculating the body movement times, the heart rate average value and the respiration rate standard deviation in each characteristic window; judging the time period in which the body movement times are less than the preset times and the duration time exceeds the preset long time as the falling asleep period; calculating a sleep onset candidate point for each of the sleep onset periods; determining a sleep onset point from the sleep onset candidate points.
Preferably, the time length of the characteristic window of the predetermined time length is 20 min.
Preferably, the time period in which the number of body movements is less than 30 and the duration time exceeds 40min is judged as the falling asleep period.
Preferably, the alignment of the calculated heart rate, respiration rate and three axis acceleration standard deviation is an alignment within a time window of 1sec width.
Preferably, the aligned heart rate, respiration rate and three axis acceleration standard deviations are sliding windowed in units of 1 sec.
Preferably, for each sleep onset period, the sleep onset candidate point is calculated from the heart rate mean and the respiration rate standard deviation.
Preferably, at each sleep onset period, a time point at which at least one of the heart rate mean value or the respiration rate standard deviation falls to 70% of its maximum value is a sleep onset candidate point.
Preferably, the electrocardio signal, the respiration signal and the triaxial acceleration signal of the preset time period are 18: 00: and 00-10: 00:00 on the next day, namely electrocardiosignals, respiration signals and triaxial acceleration signals.
Preferably, the first and second electrodes are formed of a metal,
the sleep onset period of the sleep onset point generated by the reduction of the average value of the heart rate and the standard deviation of the breathing rate is a public sleep period;
if the public sleep period does not exist, judging the most front sleep candidate point as a sleep point;
if only one public sleep period exists, calculating the time error of a sleep point generated by the reduction of the average value of the heart rate and the sleep point generated by the reduction of the standard deviation of the respiration rate, and if the time error is less than or equal to 10min, judging that the more front sleep candidate point in the public sleep period is the sleep point; if the time error is more than 10min, checking whether a sleep onset period exists before the public sleep period, if so, judging that the most advanced sleep onset candidate point in the previous sleep onset period is a sleep onset point, and if not, judging that the most advanced sleep onset candidate point in the public sleep onset period is a sleep onset point;
if at least two public sleep periods exist, firstly judging whether the public sleep periods exist before 02:00:00 on the next day;
if there is a common sleep period before the next day 02:00: selecting the most front public sleep period with the time error of the sleep candidate point less than or equal to 10min from the public sleep period before 02:00:00 on the next day, recording the time error, judging whether the front of the public sleep period has a sleep period if the time error is 0, if so, keeping the front sleep candidate point in the front sleep period as the sleep point, and if not, keeping the front sleep candidate point in the public sleep period as the sleep point; if the errors of the sleep-in points in the public sleep period before 02:00:00 in the next day are all larger than 10min, recording the time error, and outputting the early sleep-in candidate point in the sleep-in period with the minimum time error as the sleep-in point;
if there was no period of common sleep before the next day 02:00: selecting the most front public sleep period with the time error of the sleep candidate point less than or equal to 10min from the public sleep period after the next day 02:00:00, recording the time error, judging whether the front of the public sleep period has a sleep period if the time error is 0, if so, keeping the front sleep candidate point in the front sleep period as the sleep point, and if not, keeping the front sleep candidate point in the public sleep period as the sleep point; if the time errors of the sleep-in candidate points in the public sleep period after 02:00:00 in the next day are all larger than 10min, recording the time errors, and judging the former sleep-in candidate point in the public sleep period with the minimum time error as the sleep-in point.
According to the method for detecting the sleep onset point, the sleep onset period and the sleep onset candidate point are determined by calculating the average value of the heart rate, the standard deviation of the respiration rate and the body movement times, then the sleep onset point is determined according to the sleep onset candidate point judgment rule, and the algorithm judgment result is compared with the PSG doctor judgment result, so that the method has high accuracy and reliability.
Drawings
Fig. 1 is a flow chart of a method of detecting a point of falling asleep according to the present application;
fig. 2 is a flowchart of confirming a sleep onset point of the sleep onset point detection method of the present application;
FIG. 3 illustrates an example of the original cardiac, respiratory, and three-axis acceleration signals captured at 18:00 on the first day to 10:00 on the second day;
FIG. 4 is a partial enlarged view of the signal of FIG. 3;
fig. 5 is a diagram of the detection result of the falling asleep point of the first example (one falling asleep period);
fig. 6 is a diagram of the detection result of the falling asleep point(s) of the second example;
fig. 7 is a diagram of the detection result of the falling asleep point of the third example (the falling asleep point does not exist).
Detailed Description
The present application will be described in detail below with reference to the accompanying drawings.
The method identifies the point of falling asleep through three characteristics of the mean value of the heart rate, the standard deviation of the respiration rate and the number of movements in the window. The implementation flow is shown in fig. 1.
1. Intercepting 18:00-10:00 (next day) continuously and dynamically monitored physiological signals, comprising: electrocardio, respiration, triaxial acceleration signals and corresponding time stamps. If the wear time does not cover this time period, intercepting the time within this time period provides a corresponding timestamp.
2. And calculating the heart rate, the respiration rate and the triaxial acceleration standard deviation, wherein the calculation result corresponds to each second through interpolation and a window. Because the values of the heart rate, the respiration rate and the acceleration standard deviation calculated according to the intercepted continuous dynamic physiological signals are 1/second, and because the electrocardio signals, the respiration signals and the acceleration sampling rate are different, the heart rate and the respiration rate are corresponding to each second in a window through an interpolation method, namely the sampling rates of the heart rate, the respiration rate and the triaxial acceleration standard deviation are all 1 HZ.
The triaxial acceleration standard deviation is the sum of the standard deviations per second of the acceleration per axis, as shown in the following formula.
d_std＝sum(std(x_1sec)+std(y_1sec)+std(z_1sec))
3. Filtering outliers: the acceleration standard deviation is larger than 0.08 and marked as one-time body movement, the body movement frequency of each window is calculated by a method of sliding the window one second by one second, the length of the characteristic window is 20min, if the body movement frequency of the body position in the window is larger than or equal to 50, the window is marked as 50, the visualization is convenient, and meanwhile, the influence of abnormal values, such as signal noise or abnormal values caused by irregular activities of the human body, is avoided.
4. And calculating the heart rate average value and the respiration rate standard deviation in each characteristic window, and corresponding the heart rate average value and the respiration rate standard deviation calculated by the length of each window to the number of the body movement times calculated in the window.
Window length number of physical movements: number of acceleration standard deviation d _ std >0.08 within window.
5. The possible sleep onset period (also called sleep period) is calculated by the number of body movements.
And judging the time period with the body movement times less than 30 and the interval length more than 40 minutes as a possible sleep period. The step is to search a time period with the body movement times less than 30 and the interval length more than 40 minutes through a step-by-step sliding window.
6. For each calculated sleep period, the possible point of falling asleep is calculated from the heart rate mean and the standard deviation of the respiration rate.
7. Point of sleep confirmation
Calculating sleep monitoring data by the method for calculating the number of physical activity, the average value of heart rate and the standard deviation of respiratory rate, wherein a plurality of falling asleep periods may exist, for example, 1) when a person may get up to a toilet during sleep, the time period is judged as an awakening period, and the front period and the rear period of the time period are judged as falling asleep periods by the method for calculating the sleep period, so that a plurality of falling asleep periods exist; 2) if the human body is early in bed and does not move during lying, the time period may be misjudged as the sleep-in period, then other things are done, and then the user starts to sleep, and a plurality of sleep-in periods can be caused, and meanwhile, the sleep-in point is not in the first sleep-in period; 3) there are multiple periods of arousal and sleep throughout the sleep, i.e., multiple arousals and multiple falls to sleep. Therefore, there is a need to calibrate the possible sleep onset periods by means of the heart rate mean or the standard deviation of the breathing rate, to determine the actual sleep onset period, and to find the sleep onset point during the sleep onset period, to avoid causing erroneous decisions as much as possible. The process of the method for confirming the point of falling asleep is shown in fig. 2.
1) Searching each sleep period respiration rate annotation difference (std _ BR) and heart rate average value (mean _ HR) for possible sleep onset candidate points.
2) And judging whether a sleep point of a public sleep period exists or not.
The common sleep period refers to a point where both the standard deviation of the maximum respiration rate or the average of the maximum heart rate (mean _ HR) decrease to 70% of the standard deviation of the maximum respiration rate or the average of the maximum heart rate in the whole process in one sleep period.
And (4) calculating the bias (error) of the current public period, namely calculating the point that the average value of the heart rate or the standard deviation of the respiration rate in the current public sleep period is reduced to 70% of the standard deviation of the maximum respiration rate or the average value of the maximum heart rate in the whole process.
a) If the difference value is less than or equal to 10min, outputting a sleep-in point which is closer to the front in the public sleep period, namely, a time point when the heart rate average value or the respiration rate standard deviation is firstly reduced to 70% of the maximum heart rate average value or the maximum respiration rate standard deviation is taken as a sleep-in point;
b) if the difference is greater than 10min, it is searched whether there is a sleep period before the common sleep period. If not, retaining the current result, namely, taking the time point when the heart rate average value or the respiration rate standard deviation firstly falls to 70 percent of the maximum heart rate average value or the maximum respiration rate standard deviation as the sleep-in point in the public sleep period; if a sleep period is preceded by the common sleep period, the point in time at which the heart rate mean or the standard deviation of the respiration rate in the preceding sleep period first falls to 70% of its maximum heart rate mean or maximum standard deviation of the respiration rate is the point of falling asleep.
Since most people fall asleep before 02:00:00, whether a public sleep period exists before 02:00:00 is considered preferentially, if so, a reasonable falling asleep point (the falling asleep time point conforms to the range of the sleep rule) can be found by an algorithm, and the public sleep period after 02:00:00 is not considered.
First, there is a public sleep period before 02:00:00
a) If the error of the falling asleep point in the public sleep period is less than or equal to 10min, keeping a more advanced point (marked as a first public sleep period: the error of the point of falling asleep is less than or equal to the first sleep session of 10min), and the error (bias) is recorded.
Error (bias ≠ 0, and is less than or equal to 10 min): the point in time when the heart rate or respiration rate first drops to 70% of the maximum heart rate or maximum respiration rate in the first common sleep period is the point of falling asleep. For example: before 02:00:00, at least 1 sleep onset point error in a public sleep period is less than or equal to 10min and not 0, and selecting a time point when the heart rate average value or the respiration rate standard deviation in the earlier sleep period is firstly reduced to 70% of the maximum heart rate average value or the maximum respiration rate standard deviation as the sleep onset point.
Error-free (bias ═ 0): if the time points of the heart rate average value or the respiration rate standard deviation which are reduced to the maximum heart rate average value or the maximum respiration rate standard deviation of 70% in the first public sleep period coincide, whether a sleep-in period exists before the first public sleep period is checked, if yes, the sleep-in point which is in front of the first sleep-in period (the sleep-in period before the first public sleep period) in the previous sleep-in period is output (the time point of the heart rate average value or the respiration rate standard deviation which is reduced to the maximum heart rate average value or the maximum respiration rate standard deviation of 70% firstly is the sleep-in point); if not, the point of falling asleep for the first common term is output (the point in time at which the heart rate mean or respiration rate standard deviation first falls to 70% of its maximum heart rate mean or maximum respiration rate standard deviation is the point of falling asleep).
b) Before 02:00:00, recording bias when the errors of the sleeping points in the public sleeping period are all more than 10 min. And searching the public sleep period with the minimum bias, and outputting a more advanced point in the public sleep period (the time point when the heart rate average value or the respiration rate standard deviation is firstly reduced to the maximum heart rate average value or the maximum respiration rate standard deviation of 70 percent is the sleep-in point).
(2) after 02:00:00, a public sleep period begins to appear
a) If the error of the sleeping point in the public sleeping period is less than or equal to 10min, keeping a point which is closer to the front (after 02:00:00, the first sleeping period with the sleeping point less than or equal to 10min is recorded as a first public period), and recording bias.
Error (bias ≠ 0, and is less than or equal to 10 min): the point in time at which the heart rate mean or respiration rate standard deviation first falls to 70% of its maximum heart rate mean or maximum respiration rate standard deviation in the first common sleep period is the point of falling asleep. For example: when the error of the falling asleep point in at least 1 public sleep period is less than or equal to 10min and not 0 after 02:00:00, selecting the time point at which the heart rate average value or the respiration rate standard deviation in the earlier sleep period is firstly reduced to the maximum heart rate average value or the maximum respiration rate standard deviation of 70% as the falling asleep point.
Error-free (bias ═ 0): if the time points of the heart rate average value or the respiration rate standard deviation which are reduced to the maximum heart rate average value or the maximum respiration rate standard deviation of 70% in the first public sleep period coincide, whether a sleep-in period exists before the first public sleep period is checked, if yes, the sleep-in point which is in front of the first sleep-in period (the sleep-in period before the first public sleep period) in the previous sleep-in period is output (the time point of the heart rate average value or the respiration rate standard deviation which is reduced to the maximum heart rate average value or the maximum respiration rate standard deviation of 70% firstly is the sleep-in point); if not, the point of falling asleep for the first common term is output (the point in time at which the heart rate mean or respiration rate standard deviation first falls to 70% of its maximum heart rate mean or maximum respiration rate standard deviation is the point of falling asleep).
b) After 02:00:00, the errors of the sleeping points in the public sleeping period are all more than 10min, and the bias is recorded. And searching the public sleep period with the minimum bias, and outputting a more advanced point in the public sleep period (the time point when the heart rate average value or the respiration rate standard deviation is firstly reduced to the maximum heart rate average value or the maximum respiration rate standard deviation of 70 percent is the sleep-in point).
By no common sleep period is meant that either the heart rate mean or the standard deviation of the breathing rate does not occur simultaneously (i.e. only one occurs) during the same sleep period. At this time, the time point at which the heart rate mean or the respiration rate standard deviation falls to 70% of its maximum heart rate mean or maximum respiration rate standard deviation is the point of falling asleep.
This is illustrated in connection with the examples shown in fig. 5-7.
1. Intercepting 18:00-10:00 (day two) physiological signal data comprising: electrocardio signals, respiration signals and triaxial acceleration signals. Raw signal data visualization is shown in fig. 3, and local amplification signals are shown in fig. 4.
2. And calculating the standard deviation of the respiration rate, the heart rate and the acceleration according to the intercepted physiological signals, wherein the calculation result corresponds to each second through interpolation and a window. Because the values of the heart rate, the respiration rate and the acceleration standard deviation calculated according to the intercepted continuous dynamic physiological signals are 1/second, the heart rate, the respiration rate and the standard deviation are corresponding to each second in a window through an interpolation method, namely the sampling rates of the heart rate, the respiration rate and the triaxial acceleration standard deviation are all 1 HZ.
3. Calculating the number of body movements of each window: the acceleration standard deviation is larger than 0.08 and marked as one-time body movement, the body movement frequency of each window is calculated by a method of sliding the window one second by one second, the length of the characteristic window is 20min, if the body movement frequency of the body position in the window is larger than or equal to 50, the window is marked as 50, the visualization is convenient, and meanwhile, the influence of abnormal values, such as signal noise or abnormal values caused by irregular activities of the human body, is avoided.
4. And calculating the heart rate average value and the respiration rate standard deviation in each characteristic window, and corresponding the heart rate average value and the respiration rate standard deviation calculated by the length of each window to the number of the body movement times calculated in the window.
Window length number of physical movements: number of acceleration standard deviation d _ std >0.08 within window.
5. The possible sleep onset period (also called sleep period) is calculated by the number of body movements. And judging the time period with the body movement times less than 30 and the interval length more than 40 minutes as a possible sleep period. The step is to search a time period with the body movement times less than 30 and the interval length more than 40 minutes through a step-by-step sliding window.
6. For each calculated sleep period, the possible point of falling asleep is calculated from the heart rate mean and the standard deviation of the respiration rate. The average value of the heart rate and the standard deviation of the respiration rate in each sleep period are calculated, and the possible sleeping point in the sleep period is determined according to the average value of the heart rate or the standard deviation of the respiration rate.
7. Confirm the point of falling asleep
The point of falling asleep is determined according to the above rules, and the analysis results of each case of the monitored data are represented by four sub-graphs, the first two being the mean heart rate (meanHR) and the standard deviation of respiration rate (stdBR), the third being the number of body movements (numaccs), and the fourth being the three-axis acceleration signal (ACC). The yellow lines indicate the detection results of the automatic detection algorithm for the falling asleep points, the black lines indicate the PSG physician labeling results, and the red and green lines of the third sub-graph indicate the start and end time points of each sleep session.
1) King XX, male, height: 180, body weight: 75kg, age 30, intercepting 18:00-10:00 (the next day) electrocardio, respiration and triaxial acceleration signals of the monitoring data, and carrying out data processing and analysis on the signals to confirm the point of falling asleep, as shown in figure 5.
The physician marks the point of falling asleep as 00:10:00, the point of falling asleep is automatically detected as 00:14:49, and the difference between the points of falling asleep is about 5 minutes, which indicates that the algorithm has higher accuracy and reliability.
2) The monitoring data of the Heya XX, the male, the height of 175cm, the weight of 80kg and the age of 45 are intercepted, the electrocardio, respiration and triaxial acceleration signals of 18:00-10:00 (the second day) are processed and analyzed, and the sleeping point is confirmed, as shown in figure 6.
The physician marks the points of falling asleep at 20:32:30, the points of falling asleep are automatically detected to be 20:19:38, and the difference between the points of falling asleep is about 13 minutes, which indicates that the algorithm has higher accuracy and reliability. And the automatic detection algorithm of the falling asleep point detects two sleep periods and marks the two sleep periods in the second sleep period, the judged falling asleep time point is 20:19:38 and is very close to the marking time point of a doctor, and the marking time point of the algorithm in the second sleep period is more reasonable.
3) King XX, male, height: 170cm, body weight: 65kg, age 38 years old, and 18:00-10:00 (the next day) electrocardio, respiration and triaxial acceleration signals are intercepted from the monitoring data, and the signals are subjected to data processing and analysis to confirm the point of falling asleep, as shown in figure 7.
As can be seen from the analysis of FIG. 7, the number of data body movements (NumACC) in this example is equal to or greater than 50 within the range of the intercepted data segment, and no sleep period exists, that is, no sleep point exists according to the sleep point judgment rule. It may be because the collected monitoring data has problems, such as large signal noise or falling off of the signal collecting sensor, and it may be that the person really does not sleep all night.
Unless defined otherwise, all technical and/or scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The materials, methods, and examples set forth in this application are illustrative only and not intended to be limiting.
Although the present invention has been described in conjunction with specific embodiments thereof, it is evident that many alternatives, modifications, and variations will be apparent to those skilled in the art in light of the teachings of this application and yet remain within the scope of this application.
Claims (9)
1. A method for detecting a point of falling asleep calculates heart rate, respiration rate and triaxial acceleration standard deviation of a testee in a preset time period through electrocardiosignals, respiration signals and triaxial acceleration signals of the testee in the preset time period; aligning the heart rate and the respiration rate obtained by calculation with the three-axis acceleration standard deviation to obtain the aligned heart rate and respiration rate and the three-axis acceleration standard deviation; performing sliding window on the aligned heart rate, respiration rate and triaxial acceleration standard deviation by using a characteristic window with a preset time length, and calculating the body movement times, the heart rate average value and the respiration rate standard deviation in each characteristic window; judging the time period in which the body movement times are less than the preset times and the duration time exceeds the preset long time as the falling asleep period; calculating a sleep onset candidate point for each of the sleep onset periods; determining a sleep onset point from the sleep onset candidate points.
2. The method of claim 1, wherein:
the time length of the characteristic window of the preset time length is 20 min.
3. The method of claim 1, wherein:
and judging the time period with the body movement times less than 30 and the duration time exceeding 40min as the falling asleep period.
4. The method of claim 1, wherein:
the alignment of the calculated heart rate, respiration rate and three-axis acceleration standard deviation is an alignment within a time window of 1sec width.
5. The method of claim 4, wherein:
and performing sliding window on the aligned heart rate, respiration rate and triaxial acceleration standard deviation by taking 1sec as a unit.
6. The method of claim 1, wherein:
for each sleep onset period, sleep onset candidate points are calculated from the heart rate mean and the respiration rate standard deviation.
7. The method of claim 1, wherein:
at each sleep onset period, a time point at which at least one of the heart rate mean value or the respiration rate standard deviation falls to 70% of its maximum value is a sleep onset candidate point.
8. The method of claim 7, wherein:
the electrocardio signals, the respiration signals and the three-axis acceleration signals in the preset time period are 18: 00: and 00-10: 00:00 on the next day, namely electrocardiosignals, respiration signals and triaxial acceleration signals.
9. The method of claim 8, wherein:
the sleep onset period of the sleep onset point generated by the reduction of the average value of the heart rate and the standard deviation of the breathing rate is a public sleep period;
if the public sleep period does not exist, judging the most front sleep candidate point as a sleep point;
if only one public sleep period exists, calculating the time error of a sleep point generated by the reduction of the average value of the heart rate and the sleep point generated by the reduction of the standard deviation of the respiration rate, and if the time error is less than or equal to 10min, judging that the more front sleep candidate point in the public sleep period is the sleep point; if the time error is more than 10min, checking whether a sleep onset period exists before the public sleep period, if so, judging that the most advanced sleep onset candidate point in the previous sleep onset period is a sleep onset point, and if not, judging that the most advanced sleep onset candidate point in the public sleep onset period is a sleep onset point;
if at least two public sleep periods exist, firstly judging whether the public sleep periods exist before 02:00:00 on the next day;
if there is a common sleep period before the next day 02:00: selecting the most front public sleep period with the time error of the sleep candidate point less than or equal to 10min from the public sleep period before 02:00:00 on the next day, recording the time error, judging whether the front of the public sleep period has a sleep period if the time error is 0, if so, keeping the front sleep candidate point in the front sleep period as the sleep point, and if not, keeping the front sleep candidate point in the public sleep period as the sleep point; if the errors of the sleep-in points in the public sleep period before 02:00:00 in the next day are all larger than 10min, recording the time error, and outputting the early sleep-in candidate point in the sleep-in period with the minimum time error as the sleep-in point;
if there was no period of common sleep before the next day 02:00: selecting the most front public sleep period with the time error of the sleep candidate point less than or equal to 10min from the public sleep period after the next day 02:00:00, recording the time error, judging whether the front of the public sleep period has a sleep period if the time error is 0, if so, keeping the front sleep candidate point in the front sleep period as the sleep point, and if not, keeping the front sleep candidate point in the public sleep period as the sleep point; if the time errors of the sleep-in candidate points in the public sleep period after 02:00:00 in the next day are all larger than 10min, recording the time errors, and judging the former sleep-in candidate point in the public sleep period with the minimum time error as the sleep-in point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110178505.7A CN112971724A (en) | 2021-02-07 | 2021-02-07 | Method for detecting sleeping point |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110178505.7A CN112971724A (en) | 2021-02-07 | 2021-02-07 | Method for detecting sleeping point |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112971724A true CN112971724A (en) | 2021-06-18 |
Family
ID=76392801
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110178505.7A Pending CN112971724A (en) | 2021-02-07 | 2021-02-07 | Method for detecting sleeping point |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112971724A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113568505A (en) * | 2021-07-22 | 2021-10-29 | 歌尔科技有限公司 | Method, device and equipment for determining sleep time point and readable storage medium |
| CN113576419A (en) * | 2021-08-20 | 2021-11-02 | 珠海格力电器股份有限公司 | Method and device for determining sleep time of user |
| CN113892907A (en) * | 2021-08-31 | 2022-01-07 | 杭州思立普科技有限公司 | Biological rhythm detection method, device, equipment and medium based on wearable equipment |
| CN114027799A (en) * | 2021-12-13 | 2022-02-11 | 珠海格力电器股份有限公司 | Method and device for determining time point of falling asleep |
-
2021
- 2021-02-07 CN CN202110178505.7A patent/CN112971724A/en active Pending
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113568505A (en) * | 2021-07-22 | 2021-10-29 | 歌尔科技有限公司 | Method, device and equipment for determining sleep time point and readable storage medium |
| CN113576419A (en) * | 2021-08-20 | 2021-11-02 | 珠海格力电器股份有限公司 | Method and device for determining sleep time of user |
| CN113576419B (en) * | 2021-08-20 | 2022-06-14 | 珠海格力电器股份有限公司 | Method and device for determining sleep time of user |
| CN113892907A (en) * | 2021-08-31 | 2022-01-07 | 杭州思立普科技有限公司 | Biological rhythm detection method, device, equipment and medium based on wearable equipment |
| CN114027799A (en) * | 2021-12-13 | 2022-02-11 | 珠海格力电器股份有限公司 | Method and device for determining time point of falling asleep |
| CN114027799B (en) * | 2021-12-13 | 2023-03-14 | 珠海格力电器股份有限公司 | Method and device for determining time point of falling asleep |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112971724A (en) | Method for detecting sleeping point | |
| Shouldice et al. | Detection of obstructive sleep apnea in pediatric subjects using surface lead electrocardiogram features | |
| Chen et al. | An automatic screening approach for obstructive sleep apnea diagnosis based on single-lead electrocardiogram | |
| US9357953B2 (en) | System and method for diagnosing sleep apnea | |
| US8398555B2 (en) | System and method for detecting ventilatory instability | |
| KR101868888B1 (en) | Classification of Sleep/Wakefulness using Nasal Pressure for Patients with Sleep-disordered Breathing | |
| CN107049283B (en) | Sleep apnea detection system based on self-adaptive residual comparison algorithm | |
| US10004452B2 (en) | System and methods for estimating respiratory airflow | |
| CN112806966B (en) | Non-interference type early warning system for sleep apnea | |
| US9498162B2 (en) | Identifying seizures using heart data from two or more windows | |
| US20220167856A1 (en) | Lung function monitoring from heart signals | |
| Rahman et al. | Severity classification of obstructive sleep apnea using only heart rate variability measures with an ensemble classifier | |
| CN116211256B (en) | Non-contact sleep breathing signal acquisition method and device | |
| CN112089413A (en) | Blocking type sleep apnea syndrome screening system | |
| Jaworski et al. | Detection of sleep and wake states based on the combined use of actigraphy and ballistocardiography | |
| CN112971731B (en) | Sleep monitoring system based on snore recognition | |
| US11291406B2 (en) | System for determining a set of at least one cardio-respiratory descriptor of an individual during sleep | |
| JP2012090913A (en) | Sleep measuring device | |
| Lin et al. | Achieving accurate automatic sleep apnea/hypopnea syndrome assessment using nasal pressure signal | |
| Kagawa et al. | Non-contact screening system for sleep apnea-hypopnea syndrome using the time-varying baseline of radar amplitudes | |
| CN114668407B (en) | Sleep apnea prescreening system based on BCG signal | |
| CN112971721B (en) | Device for detecting falling asleep point | |
| CN111481189A (en) | Sleep evaluation method and device | |
| CN112971720B (en) | Method for detecting point of falling asleep | |
| TWI810619B (en) | Sleep state judging system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |