CN112971721B

CN112971721B - Device for detecting falling asleep point

Info

Publication number: CN112971721B
Application number: CN202110167853.4A
Authority: CN
Inventors: 郑捷文; 黄鑫; 兰珂; 郝艳丽; 王钊
Original assignee: Beijing Haisi Ruige Technology Co ltd
Current assignee: Beijing Haisi Ruige Technology Co ltd
Priority date: 2021-02-07
Filing date: 2021-02-07
Publication date: 2024-03-08
Anticipated expiration: 2041-02-07
Also published as: CN112971721A

Abstract

The application proposes a device of detection point of falling asleep, it includes: an electrocardiosignal sensor, a triaxial acceleration sensor, an MP matrix calculating unit, a sleep period calculating unit and a sleeping point calculating unit; the electrocardiosignal sensor is used for acquiring an electrocardiosignal of the subject and inputting the electrocardiosignal to the MP matrix calculation unit; the MP matrix calculation unit obtains an MP matrix through the electrocardiosignal, and obtains a state transition point through the MP matrix; the triaxial acceleration sensor is used for acquiring triaxial acceleration signals of the subject, which are synchronous with the electrocardiosignals, and inputting the triaxial acceleration signals into the sleep period calculation unit; the sleep period calculating unit obtains a sleep period through the triaxial acceleration signal; the falling asleep point calculating unit determines a falling asleep point in the state transition points through the sleep period.

Description

Device for detecting falling asleep point

Technical Field

The present application relates to sleep detection technology, and in particular, to a device for detecting a point of falling asleep.

Background

Sleep time accounts for one third of human life, poor sleep quality can lead to physical and mental tiredness during the day, and sleep disorders can lead to depression, diabetes, hypertension and numerous other cardiovascular diseases. The falling sleep 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 gold standard of the current sleep monitoring is Polysomnography (PSG), and the sleeping point of a subject can be accurately detected based on the monitored brain electricity, electrocardio, myoelectricity, eye movement, respiration, blood oxygen signals and the like, so that a basic guarantee is provided for accurately judging the sleeping quality. However, PSG requires the subject to wear the electroencephalogram electrode in the sleep laboratory, which makes the subject feel uncomfortable, affects the subject to sleep to a certain extent, and uses PSG equipment to be expensive, inefficient, limited in site, requires a professional doctor to make judgment, etc., so that the use of PSG is greatly limited.

The patent with application number 201510371905.4 proposes a solution for detecting a sleeping point, monitoring the turnover number and the heart rate of a subject within 15 minutes, setting a threshold value of the turnover number and a threshold value of the heart rate, and comparing the turnover number and the heart rate within 15 minutes with the threshold value to obtain a corresponding sleeping point. The method is too simple and is extremely easy to cause erroneous judgment. The turn-over habit of each tested person is different, the individual difference is larger, the setting of the turn-over frequency threshold value is extremely not universal, and the scientific basis is lacked.

Disclosure of Invention

In view of the above, the present application aims to propose a device for detecting a point of falling asleep.

The device for detecting the falling asleep point comprises: an electrocardiosignal sensor, a triaxial acceleration sensor, an MP matrix calculating unit, a sleep period calculating unit and a sleeping point calculating unit;

the electrocardiosignal sensor is used for acquiring an electrocardiosignal of the subject and inputting the electrocardiosignal to the MP matrix calculation unit; the MP matrix calculation unit obtains an MP matrix through the electrocardiosignal, and obtains a state transition point through the MP matrix;

the triaxial acceleration sensor is used for acquiring triaxial acceleration signals of the subject, which are synchronous with the electrocardiosignals, and inputting the triaxial acceleration signals into the sleep period calculation unit; the sleep period calculating unit obtains a sleep period through the triaxial acceleration signal;

the falling asleep point calculating unit determines a falling asleep point in the state transition points through the sleep period.

Preferably, the MP matrix calculating unit forms a heart rate value sequence by using the electrocardiograph signal of the subject, obtains an MP matrix according to the heart rate value sequence, marks an arc line between waveforms of the electrocardiograph signal by using the MP matrix, counts the number of the arc lines passing through above each position, that is, ICA, and obtains a maximum value of each column element of the MP matrix, and marks the maximum value as an MP value.

Preferably, the MP matrix calculating unit performs sliding window on the electrocardiograph signal with a 1s duration as a first window, to obtain the heart rate value sequence.

Preferably, the MP matrix calculating unit performs sliding window value on the heart rate value sequence with a second window with a duration of 20s to obtain a plurality of heart rate value arrays A1, A2, … Ak … An, where Ak is the heart rate value array obtained by the kth sliding window, k is 1.ltoreq.k.ltoreq.n, and n is a natural number greater than 1.

Preferably, the MP matrix is

xij represents the Euclidean distance between the heart rate value array Ai and Aj, wherein i is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to n.

Preferably, the MP matrix calculating unit obtains a minimum value of each column element except for the diagonal element in the MP matrix, marks an arc between the electrocardiographic signal waveform diagram corresponding to the minimum value and the electrocardiographic signal waveform diagram corresponding to the diagonal element, counts ICA above each position, and records a point of time when ICA is minimum as a state transition point.

Preferably, the MP matrix calculating unit records a point of time at which the MP value is maximum as the state transition point.

Preferably, the sleep period calculating unit calculates an acceleration standard deviation of each second for the triaxial acceleration signal, the acceleration standard deviation is >0.08 and marked as a body movement, the body movement times of each third window are calculated through the third windows with the sliding time length of 20 minutes per second, and the time period with the body movement times of <30 and the interval length of more than 40 minutes is judged as the sleep period.

Preferably, if there is no sleep period, the falling asleep point calculating unit outputs that there is no falling asleep point;

if one or more sleep periods exist, the falling asleep point calculating unit judges whether the time of the ICA minimum value differs from the time of the MP maximum value by more than one sleep period; if at least one of the time of the ICA minimum value and the time of the MP maximum value is within the sleep period and the time of the minimum value and the time of the MP maximum value differ by more than one sleep period, the falling asleep point calculating unit takes the time of the ICA minimum value as the falling asleep point; if the time of the ICA minimum value and the time of the MP maximum value are in the same sleep period or at least one of the times is in the sleep period, and the difference between the time of the minimum value and the time of the MP maximum value is smaller than one sleep period, the sleep point calculating unit takes the time of the MP maximum value as the sleep point; if neither the timing of the ICA minimum nor the timing of the MP maximum is within the sleep period, the sleep-in-point calculation unit takes the timing of the MP maximum as the sleep-in-point.

According to the device for detecting the falling asleep point, the state transition point is obtained through the MP matrix, the sleeping period obtained by combining the triaxial acceleration signals is combined, and the falling asleep point is determined from the state transition point.

Drawings

FIG. 1 is a preliminary determination flow of a state transition point of a device for detecting a point of falling asleep of the present application;

FIG. 2 is a schematic illustration of an arc marking of the apparatus for detecting a point of sleep of the present application;

FIG. 3 is a graph of arc statistics at an electrocardiographic signal location;

FIG. 4 is a flow chart of a point of sleep confirmation;

FIG. 5 is a graph of the electrocardiographic and triaxial acceleration signals monitored simultaneously during sleep;

FIG. 6 is a plot of the results of the point of sleep analysis for subject 1;

FIG. 7 is a plot of the results of a point of sleep analysis for subject 2;

FIG. 8 is a plot of the results of a point of sleep analysis for subject 3;

fig. 9 is a schematic view of the structure of the device for detecting the falling asleep point of the present application.

Detailed Description

The present application will be described in detail with reference to the accompanying drawings.

In the present application, the MP matrix calculating unit, the sleep period calculating unit, and the sleep point calculating unit may be implemented by running corresponding programs on a computer, or may be dedicated computing devices for implementing the functions.

ICA and MP values

Since the heart rate fluctuates within a certain range of a certain value (which varies from individual to individual) when the human body is in the same state. If people are at rest, the heart rate can fluctuate about 80 times/min, and if people are at rest, the heart rate can fluctuate about 60 times/min; however, for the same person, the heart rate may fluctuate within a certain range of different values in different states, such as about 80 beats/min when the person is stationary, about 120 beats/min in a running state, and about 60 beats/min in sleep.

When the state of the human body is unchanged, the heart rate value does not change greatly (fluctuation around a small range of a certain value), the Euclidean distance (Euclidean distance at the moment corresponding to two sliding windows) obtained through the sliding windows is also small, namely the calculated MP value is small, which indicates that the states at the two moments are closest, and the states between the two moments are the same or similar because the heart rate value gradually changes and does not suddenly increase or decrease greatly, namely the states of the time periods of the arc drawn between the minimum column element of the MP matrix and the diagonal line are the same or similar. And at a state transition point such as a point of wakefulness-sleep (point of falling asleep), the state at this point is greatly different from that at other points, so the number of arcs (ICA) above the position at this point is also minimized. Since the state change includes more, such as wake-to-sleep, fall asleep-to-wake, standing-to-walk, etc., further judgment is required by the sleep period.

In addition, when the state of the human body is unchanged, the heart rate value of the human body does not change greatly (fluctuation is small), and the Euclidean distance value obtained through the sliding window is also small, namely the MP value is small; when the state of the human body is changed, the heart rate value is changed greatly, (for example, the heart rate value is changed from about 60 to about 80 from sleep-wake), and the Euclidean distance obtained through the sliding window is large, namely, the MP value is large, so that the state transition point can be judged through the MP maximum value.

Sleep point confirmation process

1. State transition point validation

The state transition point preliminary confirmation flow is shown in fig. 1.

1) Sliding the heart rate data (1 HZ) for a20 s time window seconds by seconds to obtain a series of arrays comprising 20 heart rate values [ A1, A2, A3, A4 … An ], wherein each of A1, A2 … An comprises 20 heart rate values, such as a1= [ A1, A2, A3 … a20];

2) Calculating MP value of heart rate signal: obtaining MP matrix for the acquired array pairwise Euclidean distance;

a1 and A1, A2, A3 … An find Euclidean distance, get MP 1 column element of matrix, [ x11, x12, x13, … x1n ];

a2 is respectively subjected to Euclidean distance with A1, A2 and A3 … An to obtain the 2 nd column element of the MP matrix, [ x12, x22, x23 … x2n ];

and so on:

an and A1, A2 and A3 … An are respectively subjected to Euclidean distance to obtain An nth column element of the MP matrix, [ xn1, xn2, xn3 … xnn ].

The MP matrix is:

an n×n MP matrix is obtained, and since the euclidean distance is 0 for each group, the MP matrix is a symmetric matrix with 0 diagonal elements.

3) For index vectors of MP values, each index pair is connected by an arc: and acquiring the minimum value of each column element of the MP matrix, and drawing an arc line between the electrocardiosignal corresponding to the minimum value and the electrocardiosignal corresponding to the diagonal element, as shown in figure 2. For example: in the MP matrix, in the 1 st column, X15 is the minimum value of the column, which indicates that the state of the human body corresponding to the first electrocardiosignal sliding window and the fifth electrocardiosignal sliding window is the closest, and then an arc line is drawn between the first electrocardiosignal sliding window and the fifth electrocardiosignal sliding window.

In fig. 2, the waveform diagram is an electrocardiographic signal; the following numbers correspond to the positions, such as 1892 and 1270, and indicate that the electrocardiograph signal acquired by the 1892 th sliding window is closest to the electrocardiograph signal acquired by the 1270 th sliding window, i.e. the euclidean distance value is the smallest, and then an arc is drawn between the electrocardiograph signals corresponding to 1892 and 1270.

4) Counting how many arcs pass through above each position: calculating the number of arcs passing through each position and recording the number as ICA;

5) The position with the least number of marked arcs is defined as the state transition point. The number of arcs obtained is normalized by the formula (1) to make the values between 0 and 1, as shown in fig. 3, the dotted line mark is a state transition point.

Delta: normalized values;

x _i i, the number of arcs passing through the position;

an average of the number of arcs;

sigma: standard variance of the number of arcs.

Sleep stage validation

And calculating an acceleration standard deviation according to the triaxial acceleration signal by a sliding window method, calculating the body movement times according to the acceleration standard deviation, and determining the possible sleep period according to the body movement times.

1) The standard deviation of acceleration per second is calculated from the triaxial acceleration signal as the sum of standard deviations per second of acceleration per axis, as shown in formula (2).

d_std＝sum(std(x_1sec)+std(y_1sec)+std(z_1sec)) (2)

2) Calculating the number of body movements: the standard deviation of acceleration is more than 0.08, the standard deviation of acceleration is marked as one body movement, the body movement times of each window are calculated by a sliding window per second method, and the length of the characteristic window is 20min.

3) Sleep period is determined by the number of body movements.

The time period with the body movement times smaller than 30 and the interval length longer than 40 minutes is the sleep period.

Sleeping point confirmation

The point of fall asleep confirmation procedure is shown in fig. 4, where the curve value is minimal at the correct location 2400. Wherein MP maximum: and extracting the maximum value of each column element of the MP matrix as the maximum value of the corresponding position.

1) There is no falling asleep point: no sleep period, i.e. no period of time with body movement times <30, interval length >40 minutes;

2) There is one/more sleep periods:

if one or more sleep periods exist, judging whether the time of the ICA minimum value differs from the time of the MP maximum value by more than one sleep period;

if at least one of the time of the ICA minimum value and the time of the MP maximum value is in the sleep period and the difference between the time of the minimum value and the time of the MP maximum value exceeds one sleep period, taking the time of the ICA minimum value as a sleep point;

if the time of the ICA minimum value and the time of the MP maximum value are in the same sleep period or at least one of the times is in the sleep period, and the difference between the time of the minimum value and the time of the MP maximum value is smaller than one sleep period, taking the time of the MP maximum value as a sleep point;

if neither the instant of ICA minimum nor the instant of MP maximum is within the sleep period, the instant of MP maximum is taken as the point of falling asleep.

Examples

Based on the sleeping point determined by the sleeping doctor according to the PSG signal monitored by the patient, the accuracy of the automatic sleeping point detection algorithm of the patent is verified.

1) Subject 1: women, 163cm in height, 55kg in weight, and 34 years old were subjected to falling asleep point analysis based on the electrocardiographic signals and triaxial acceleration signals monitored by the women, as shown in fig. 6.

2) Subject 2: women, 158cm tall, 50kg heavy, age 28 years, performed a sleep-onset analysis based on their monitored electrocardiographic and triaxial acceleration signals, as shown in fig. 7.

3) Subject 3: man, 175cm height, 73kg weight, 41 years age, performed sleep point analysis based on the electrocardiosignals and triaxial acceleration signals monitored by the man, as shown in FIG. 8.

Wherein the red line: a starting time point of the sleep session; green line: an end time point of the sleep session; black line: a point of sleep determined by a sleeping physician; yellow line: and the falling asleep point is judged by an automatic falling asleep point detection algorithm.

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 pertains. The materials, methods, and examples mentioned in this application are illustrative only and not intended to be limiting.

Although the present invention has been described in connection with specific embodiments thereof, those skilled in the art will appreciate that various substitutions, modifications and changes may be made without departing from the spirit of the invention.

Claims

1. A device for detecting a point of sleep, comprising: an electrocardiosignal sensor, a triaxial acceleration sensor, an MP matrix calculating unit, a sleep period calculating unit and a sleeping point calculating unit;

a falling-to-sleep point calculation unit determining a falling-to-sleep point in the state transition points through the sleep period;

the MP matrix calculation unit forms a heart rate value sequence through the electrocardiosignals of the subject, obtains an MP matrix according to the heart rate value sequence, marks an arc line between waveforms of the electrocardiosignals by using the MP matrix, and counts the number of the arc lines passing through above each position, namely ICA;

the MP matrix calculation unit takes the 1s duration as a first window and performs sliding window on the electrocardiosignals to obtain the heart rate value sequence;

the MP matrix calculation unit performs sliding window value on the heart rate value sequence by using a second window with the duration of 20s to obtain a plurality of heart rate value arrays A1, A2 and … Ak … An, wherein Ak is the heart rate value array obtained by the kth sliding window, k is more than or equal to 1 and less than or equal to n, and n is a natural number greater than 1;

the MP matrix is

xij represents the Euclidean distance between the heart rate value array Ai and Aj, wherein i is more than or equal to 1 and less than or equal to n, and j is more than or equal to 1 and less than or equal to n;

the MP matrix calculation unit obtains the minimum value of each column of elements except diagonal elements in the MP matrix, marks an arc line between an electrocardiosignal waveform diagram corresponding to the minimum value and an electrocardiosignal waveform diagram corresponding to the diagonal elements, counts ICA above each position, and records the minimum moment point of the ICA as a state transition point;

the MP matrix calculating unit obtains the maximum value of each column element of the MP matrix and marks the maximum value as MP value;

the MP matrix calculating unit records the moment point with the maximum MP value as a state transition point.

2. The apparatus for detecting a point of sleep of claim 1, wherein:

and the sleep period calculating unit calculates the acceleration standard deviation of each second of the triaxial acceleration signal, the acceleration standard deviation is marked as a body movement, the body movement times of each third window are calculated through the third windows with the sliding time length of 20 minutes every second, and the time period with the body movement times of <30 and the interval length of more than 40 minutes is judged as the sleep period.

3. The device for detecting a point of sleep according to claim 1 or 2, characterized in that:

if the sleep period does not exist, outputting that the sleep point does not exist by the sleep point calculating unit;