CN113951826B

CN113951826B - Method, system and equipment for evaluating sleep condition

Info

Publication number: CN113951826B
Application number: CN202111237975.2A
Authority: CN
Inventors: 邓研辉; 戴涛; 王启帆; 向江梅; 徐现红; 金键
Original assignee: Sealand Technology Chengdu Ltd
Current assignee: Sealand Technology Chengdu Ltd
Priority date: 2021-10-25
Filing date: 2021-10-25
Publication date: 2023-09-05
Anticipated expiration: 2041-10-25
Also published as: CN113951826A

Abstract

The invention discloses a method for assessing sleep states, which is characterized by comprising the following steps: s1, acquiring skin conductivity through a skin conductivity sensor; s2, calculating to obtain a skin electric change signal according to the skin conductivity in the step S1; s3, judging a sleep state according to the skin electric change signal obtained in the step S2; the signal of the skin electric change in the step S2 may be any one or several of the following: a first order differential signal of skin conductivity value, a second order differential signal of skin conductivity, a percentage change in skin conductivity per unit time. The invention judges the sleep stage and the deep sleep stage by using the skin conductivity signal, monitors the skin conductivity signal and the heart rate signal or the pulse wave signal in a matching way, separates and extracts other physiological parameter ranges in the deep sleep state, further perfects and supplements the sleep data, and solves the problem of lower accuracy of monitoring the shallow sleep by using the skin conductivity signal.

Description

Method, system and equipment for evaluating sleep condition

Technical Field

The invention relates to the field of sleep monitoring and detection, in particular to a method, a system and equipment for assessing sleep conditions.

Background

Sleep is an indispensable vital activity in daily life and is closely related to basic operation of human body, including normal growth, physical recovery, emotion adjustment, memory arrangement, etc. Adequate sleep is the basis for maintaining mental and physiological health of the human body, and insufficient sleep is harmful to the human body and increases the risk of disease invasion. Therefore, the daily sleep quality is related to the living aspects of people, and the situation of the sleep quality of the people is timely obtained and the people actively take measures to adjust the sleep quality is a focus of attention. The monitoring and analysis of sleeping in the prior art are mainly based on brain waves, but brain electrical equipment is not portable and expensive, and is difficult to be widely used in ordinary daily household life. In order to solve the problem of inconvenient electroencephalogram monitoring, monitoring of human body conditions based on the dermatography technology is gradually accepted by researchers. The skin electric technique refers to the skin electric technique, in which the skin resistance of a human body changes with the skill of sweat glands of the skin, sweat gland secretion increases under the condition of emotional stress, fear or anxiety of the human body, and sweat on the skin surface increases, thereby causing the skin electric activity to change.

In the prior art, a method and a system for evaluating sleep quality based on skin electricity are also proposed by using skin electricity detection technology to monitor sleep, for example, CN106491120a, and the proposed method includes obtaining skin electricity activity data during sleep of a human body, extracting a preset skin electricity activity index from the skin electricity activity data, obtaining a sleep score according to the skin electricity activity index, evaluating sleep quality according to a preset sleep evaluation system and the sleep score, and returning an evaluation result.

CN201811287167.5 discloses a sleeping monitoring bracelet based on skin electricity, which comprises a bracelet body, this internal being equipped with of bracelet: and the acquisition module is used for: the device comprises a skin electric sensor and a plurality of human body physiological index sensors, wherein the module is used for collecting human body sleep data; the processing module is used for: the module carries out comprehensive treatment on sleep data; and an output module: the module outputs sleep data to an external device; and (3) an evaluation and adjustment module: the module evaluates sleep quality based on the sleep data and determines whether to output an adjustment signal for triggering an external sleep quality adjustment device based on the sleep quality.

However, there are large individual differences in the data such as body temperature and heart rate during sleep, which also results in poor final accuracy.

Disclosure of Invention

The invention aims at: aiming at the problems, the method, the system and the equipment for evaluating the sleep state are provided, and the problems that the monitoring accuracy of the dermatology technology is low and the accuracy of analyzing the sleep stage is low are solved.

The technical scheme adopted by the invention is as follows:

a method for assessing sleep state, comprising the steps of:

s1, acquiring skin conductivity through a skin conductivity sensor;

s2, calculating to obtain a skin electric change signal according to the skin conductivity in the step S1;

s3, judging a sleep state according to the skin electric change signal obtained in the step S2;

the signal of the skin electric change in the step S2 may be any one or several of the following: a first order differential signal of skin conductivity value, a second order differential signal of skin conductivity, a percentage change in skin conductivity per unit time.

Further, the step S3 judges the sleep state: when the change value of the skin electric change signal is smaller than the first threshold value and the duration time is longer than the first time parameter, judging the state of the time period to be a sleep state, and otherwise, judging the state to be an awake state.

Further, when the step S3 determines that the sleep state is sleep, the method further includes a step S4 of determining a sleep stage, where the step S4 includes: and when the value of the skin electric change signal is smaller than the first threshold and larger than the second threshold and the duration time is larger than the second time parameter, judging the deep sleep stage I.

Furthermore, the step S1 further obtains a heart rate signal and/or a pulse wave signal, separates the heart rate variability and/or the pulse wave variability from the heart rate signal and/or the pulse wave variability, and extracts the step S4 to determine that the physiological parameter range of the deep sleep stage i is a deep sleep parameter.

Furthermore, the sleep stage which does not meet the deep sleep stage I is set as a sleep stage to be identified, the deep sleep parameter is matched with the physiological parameter of the sleep stage to be identified, and the sleep stage which is separated to meet the deep sleep parameter is set as a deep sleep stage II.

Further, the method also comprises the step of S5 judging the sleep quality, wherein the time period in the sleep state judged in the step of S3 is the total sleep time length, and the sum of the sleep time periods of the deep sleep I and the deep sleep II is the total deep sleep time length.

Further, the physiological parameters include LF, HF and LF/HF. LF, HF and LF/HF are frequency domain features in heart rate variability or pulse variability.

Further, the physiological parameters further include SD1, SD2 and SD2/SD1.

The poincare analysis of heart rate variability is a graphical representation of the time dependence of heart rate variability, i.e. derived from an electrocardiogram, of the heart beat of a time series of continuous cycles. This method of visualization provides information about heart rate, especially short-term and long-term variations in behavior summaries and details to each beat, and it is considered an autonomous regulation of heart rate. SD1 and SD2 are two descriptors of poincare plots of heart rate variability, where the distribution of SD1 across the plot is across the marker line and the distribution of SD2 along the marker line, these descriptions relate to the calculation of the projection sum variance, also including the standard deviation of the projection. Thus, SD1 is calculated, the poincare plot number is projected onto a line perpendicular to the marker line, the resulting point is represented by the coordinates of that line, and the square root of the variance is calculated, if the point is substituted for the point projected onto the parallel and marker line, SD2 is obtained in the same procedure. Variability from beat to beat in the RR interval described by SD1 is a measure of short-term HRV, and SD2 is related to long-term HRV.

A system for assessing sleep state, comprising a signal acquisition unit, a signal analysis unit and an output unit;

the signal acquisition unit is used for acquiring skin conductivity signals;

the signal analysis unit calculates a skin conductivity signal to obtain a skin electricity change signal, and analyzes a sleep state according to the skin electricity change signal;

the output unit outputs the obtained sleep state.

Further, the signal acquisition unit also acquires a heart rate signal.

An apparatus for assessing sleep state comprising a finger cuff and a wristband, the finger cuff and wristband being connected by a wire; the finger stall is provided with a pulse wave sensor and a display, and the finger stall is provided with a skin conductivity sensor which is in direct contact with skin.

In summary, due to the adoption of the technical scheme, the beneficial effects of the invention are as follows:

1. the invention judges the sleep stage and the deep sleep stage by using the skin conductivity signal, monitors the skin conductivity signal and the heart rate signal or the pulse wave signal in a matching way, separates and extracts other physiological parameter ranges in the deep sleep state, further perfects and supplements the sleep data, and solves the problem of lower accuracy of monitoring the shallow sleep by using the skin conductivity signal.

2. According to the invention, the acceleration sensor is arranged, and the acceleration sensor is used for correcting the monitored skin conductivity signal, so that the influence of unintentional actions on sleep state monitoring in a sleep state is solved, and the accuracy of sleep state judgment is improved.

3. In the prior art, electroencephalogram or electrocardiograph is adopted for sleep analysis, a signal acquisition unit needs to be worn at a designated position, professional assistance is usually needed, and otherwise, the accuracy and stability of signal acquisition are low. The skin conductivity signal acquisition unit is arranged on the wrist part, so that the wearing mode is simpler, the comfort is better, and the falling off is not easy to cause.

Drawings

Fig. 1 is a front view of a third embodiment;

FIG. 2 is a schematic block diagram of the present invention;

FIG. 3 is a flow chart of skin conductivity signal determination of sleep state;

FIG. 4 is a flow chart of determining sleep state by pulse wave signals;

fig. 5 is a graph of the skin electrical change signal, PRV parameters, and sleep monitoring change profiles.

Description of the drawings:

1-bracelet, 2-dactylotheca, 3-display, 4-wire.

Detailed Description

All of the features disclosed in this specification, or all of the steps in a method or process disclosed, may be combined in any combination, except for mutually exclusive features and/or steps.

Any feature disclosed in this specification may be replaced by alternative features serving the same or equivalent purpose, unless expressly stated otherwise. That is, each feature is one example only of a generic series of equivalent or similar features, unless expressly stated otherwise.

Example 1

The embodiment provides a system for assessing sleep conditions, which comprises a signal acquisition unit, a signal analysis unit and an output unit;

the signal acquisition unit comprises a skin electric sensor for acquiring skin conductivity signals;

the signal analysis unit comprises a processor, and the processor obtains a skin electric variation signal by calculating a skin conductivity signal and analyzes a sleep state according to the skin electric variation signal;

the output unit outputs the analyzed sleep state for the display 3.

Example two

The present embodiment provides a device for assessing sleep conditions, which includes a bracelet 1, a finger cuff 2 and 4 impedance electrodes (in other embodiments, a technician may use other numbers of impedance electrodes according to actual needs), and the finger cuff 2 is connected with the bracelet 1 through a lead 4. A skin conductivity sensor for collecting skin conductivity signals is arranged on the fingerstall 2, and the skin conductivity sensor is in direct contact with skin; a processor and a display 3 are provided on the bracelet 1.

The impedance electrode is stuck on the chest and abdomen of a tested person to collect heart rate signals, the finger stall 2 is worn on the index finger and the middle finger of the tested person to collect skin conductivity signals, then the heart rate signals and the skin conductivity signals are transmitted to the processor for comprehensive treatment, and the display 3 outputs the treated result.

In order to reduce the influence of unintentional actions in sleep on the monitoring result, the finger stall 2 is further provided with a first acceleration sensor and an impedance signal amplifier for acquiring finger movement signals.

In order to improve the accuracy of sleep monitoring, a drift offset compensator and a filter are further arranged in the bracelet 1 and are used for correcting and denoising the base line of the heart rate signal and the skin conductivity signal.

Example III

As shown in fig. 1 and 2, the present embodiment provides a device for assessing sleep conditions, which includes a bracelet 1 and a finger cuff 2, wherein the finger cuff 2 is connected with the bracelet 1 through a wire 4. A pulse wave sensor for collecting pulse wave signals is arranged on the bracelet 1, a skin conductivity sensor for collecting skin conductivity signals is arranged on the fingerstall 2, and the skin conductivity sensor is in direct contact with skin.

In order to reduce the influence of unintentional actions in sleep on a monitoring result, the fingerstall 2 is further provided with a first acceleration sensor and an impedance signal amplifier, the bracelet 1 is provided with a second acceleration sensor, and the first acceleration sensor and the second acceleration sensor respectively collect movement signals of fingers and wrists.

In order to improve the accuracy of sleep monitoring, a drift offset compensator and a filter are further arranged in the bracelet 1 and are used for correcting and denoising the base line of the pulse wave signal and the skin conductivity signal.

The bracelet 1 is provided with a sensor and a display 3, the processor is used for comprehensively processing the acquired sleep data, and the display 3 outputs the processed result.

Example IV

According to the apparatus for assessing sleep conditions provided in embodiment three, the present embodiment provides a method for assessing sleep conditions, comprising the steps of:

s1, wearing the device provided in the second embodiment, wherein the bracelet 1 is worn on a wrist, and the fingerstall 2 is wound on a middle finger and a ring finger and is used for acquiring a skin conductivity signal and a first acceleration signal;

s2, the processor carries out correction conversion analysis on the skin conductivity obtained in the step S1, and calculates to obtain corrected skin electricity change signals, specifically;

as shown in fig. 3, the skin conductivity signal is corrected according to the first acceleration signal processor, specifically, the first acceleration and the skin conductivity signal are aligned according to time sequence; when the skin conductivity signal and the acceleration signal are suddenly changed, the skin conductivity signal at the moment is replaced by the skin conductivity signal in the previous unit time. The purpose of this is to address the motion artifact of the skin conductivity signal caused by finger motion. The processor then obtains a skin electrical change correction signal, which is a first order differential signal or a second order differential signal of skin conductivity percent change per unit time.

when the dermatologic change correction signal is smaller than a threshold value SC1 and the duration is longer than T1, the time is determined that the subject is in a sleep state, and further judgment is made; otherwise, the state is a waking state;

s4, when the skin electric change correction signal is larger than a second threshold value SC2 and smaller than a threshold value SC1 and the duration time is longer than T2, the section is considered to be the deep sleep I;

s5, judging sleep quality:

the sleeping time in the sleeping state is judged to be the total sleeping time in the step S3, the sleeping time in the deep sleeping I is judged to be the deep sleeping time in the step S4, and the evaluation of the sleeping quality can be quantitatively analyzed by the ratio of the total deep sleeping time to the shallow sleeping time or can be obtained by the ratio of the total deep sleeping time to the sleeping time.

Finally, the display 3 outputs the skin electric change signal and the sleep quality after the calibration of the processor.

Example five

Based on the fourth embodiment, the monitoring of the pulse wave signal (heart rate signal may be adopted in other embodiments) of the tested person is added, and the skin conductivity signal is combined to evaluate the sleep quality together, specifically:

s1, wearing the device provided in the first embodiment, wherein the bracelet 1 is worn on a wrist, and the fingerstall 2 is wound on a middle finger and a ring finger to obtain a pulse wave signal (PPG), a skin conductivity signal, a first acceleration signal and a second acceleration signal. The method comprises the steps of carrying out a first treatment on the surface of the

S2, the processor carries out correction conversion analysis on the skin conductivity signal and the pulse wave signal obtained in the step S1, and calculates to obtain a skin electric variation correction signal and a pulse rate variability correction signal (PRV), specifically;

as shown in fig. 3, the skin conductivity signal is corrected according to the first acceleration signal processor, and then the processor obtains a corrected skin electrical change signal, which in this embodiment is a skin conductivity change percentage per unit time (in other embodiments, it may be a first-order differential signal of skin conductivity values or a second-order differential signal of skin conductivity).

A process flow of the processor to process the time series signal of the pulse signal is shown. Correcting the pulse wave signal according to a second acceleration signal processor in the bracelet 1 to obtain a corrected pulse wave correction signal with motion artifacts removed, and separating pulse wave variability; and simultaneously outputs a body movement signal of the subject.

when the dermatologic change signal is smaller than a first threshold (SC 1) and the duration is longer than T1, the subject is considered to be in a sleep state and further judgment is made; otherwise, the state is a waking state;

s4, deep sleep judgment:

as shown in fig. 4, when the skin electric variation signal is greater than the second threshold (SC 2) and less than the first threshold (SC 1) and the duration is longer than T2, the segment is considered to be deep sleep i; otherwise, the period of time is counted as 'sleep to be identified';

the feature range of pulse wave variability in the corresponding time of deep sleep I is extracted as a deep sleep parameter, and in the embodiment, the deep sleep parameter can be a low-frequency high-frequency ratio (LF/HF), the low-frequency LF range is 0.04-0.15, and the high-frequency HF range is 0.15-0.4. (also the SD1, SD2 and SD2/SD1 ratios in Poincare analysis in other embodiments).

Then, the processor extracts pulse wave variability parameters and deep sleep parameters in the sleep to be identified time to carry out matching analysis, and outputs a time period conforming to the sleep parameter range as deep sleep II; and the period output that does not fall within this characteristic range is light sleep.

S5, judging sleep quality:

the time length in the sleep state is judged to be the total sleep time length in the step S3, the sum of the sleep time lengths of the deep sleep I and the deep sleep II which are judged according to the pulse wave signals in the step S4 is the total deep sleep time length, and the evaluation of the sleep quality can be quantitatively analyzed by the ratio of the total deep sleep time length to the shallow sleep time length or can be obtained by the ratio of the total deep sleep time length to the sleep time length.

FIG. 5 is a graph of monitored skin electrical change signals, PRV characteristic parameter data, and chest sleep monitoring patterns tested between 00:00 and 7:30. In the embodiment, the skin conductivity change signal is a skin conductivity change percentage in unit time, and the first threshold value is set to be 5%, and the first time parameter is set to be 5 minutes; the second threshold is 1% and the second time parameter is 10 minutes.

As shown in fig. 5, in the early morning about 0:00, the skin electric change signal of the tested person gradually falls below the SC1 threshold for a duration of more than 5 minutes, so that the tested person is judged to enter a sleep state.

And in the morning between 2:28 and 2:55, the skin electric change signal of the tested person is between SC1 and SC2, and the duration exceeds 10 minutes, so that the sleeping state of the period is judged to be deep sleep I, and the PRV parameter range in the period is separated to be sleeping parameters.

Then, PRV parameters in a sleep state are screened out in the ranges of 00:15 to 1:15, 03:30 to 04:15 and 5: PPV parameters in the time periods of 30-06:00 are in the range of sleep parameters, so that the time periods are judged to be deep sleep II. In the morning, the skin electric change signals of the tested person do not exceed SC2, but the PRV parameters monitored at the moment do not fall in the range of sleep parameters but the skin electric change signals are smaller than SC1, so that the tested person is judged to be in a light sleep stage.

And in the time periods of 3:00-3:05, 6:40 and 7:20, the skin electric change signal of the tested person exceeds SC1, and the tested person is judged to enter the awake state from the sleep state.

The invention is not limited to the specific embodiments described above. The invention extends to any novel one, or any novel combination, of the features disclosed in this specification, as well as to any novel one, or any novel combination, of the steps of the method or process disclosed.

Claims

1. A method for assessing sleep state, comprising the steps of:

s1, acquiring skin conductivity through a skin conductivity sensor;

step S1 also acquires a heartbeat signal and/or a pulse wave signal, and separates the heartbeat signal and/or the pulse wave signal from the heartbeat signal and/or the pulse wave signal to obtain heart rate variability and/or pulse wave variability;

the signal of the skin electric change in the step S2 may be any one or several of the following: a first order differential signal of skin conductivity values, a second order differential signal of skin conductivity, a percentage change in skin conductivity per unit time;

step S3 is to judge the sleep state: when the change value of the skin electricity change signal is smaller than a first threshold value and the duration time is longer than a first time parameter, judging that the state of the time period is a sleep state, and otherwise, judging that the state is an awake state; when the step S3 judges that the sleep state is sleep, the method further includes a step S4 of judging a sleep stage, and the step S4 includes: when the value of the skin electric change signal is smaller than the first threshold and larger than the second threshold and the duration time is larger than the second time parameter, the deep sleep stage I is judged, the physiological parameter range of the deep sleep stage I is extracted to be the deep sleep parameter, the sleep stage which does not meet the deep sleep stage I is set to be the sleep to be identified, the deep sleep parameter is matched with the physiological parameter of the sleep to be identified, and the sleep stage which accords with the deep sleep parameter is separated to be the deep sleep stage II.

2. The method for assessing sleep state of claim 1 wherein the physiological parameters include LF, HF and LF/HF ratios.

3. The method for assessing a sleep state of claim 1 wherein the physiological parameter further comprises SD1, SD2 and SD2/SD1 ratios.

4. A system for assessing sleep state, comprising a signal acquisition unit, a signal analysis unit and an output unit; the signal acquisition unit is used for acquiring skin conductivity signals, heartbeat signals and/or pulse wave signals; the signal analysis unit calculates a skin conductivity signal to obtain a skin electricity change signal, and analyzes a sleep state according to the skin electricity change signal; the output unit outputs the obtained sleep state; wherein analyzing the sleep state comprises: when the change value of the skin electricity change signal is smaller than a first threshold value and the duration time is longer than a first time parameter, judging that the state of the time period is a sleep state, and otherwise, judging that the state is an awake state; when the value of the skin electric change signal is smaller than the first threshold and larger than the second threshold and the duration time is longer than the second time parameter, judging that the skin electric change signal is in the deep sleep stage I; and obtaining heart rate variability and/or pulse wave variability according to the heart beat signals and/or the pulse wave signals, extracting the physiological parameter range of the deep sleep stage I as a deep sleep parameter, setting the sleep stage which does not meet the deep sleep stage I as 'sleep to be identified', matching the deep sleep parameter with the physiological parameter of the 'sleep to be identified', and separating the sleep stage which meets the deep sleep parameter as a deep sleep stage II.

5. An apparatus for assessing sleep state comprising a finger cuff and a wristband, the finger cuff and wristband being connected by a wire; a skin conductance sensor and a first acceleration sensor which are in direct contact with skin are arranged on the fingerstall; the bracelet is provided with a pulse wave sensor, a second acceleration sensor, a processor and a display; the processor obtains a skin electric variation signal according to the skin electric conductivity signal, and analyzes a sleep state according to the skin electric variation signal, wherein when the variation value of the skin electric variation signal is smaller than a first threshold value and the duration time is longer than a first time parameter, the processor judges the time period as a sleep state, otherwise, the processor is in a waking state, and when the sleep state is judged to be sleep, when the value of the skin electric variation signal is smaller than the first threshold value and larger than a second threshold value and the duration time is longer than a second time parameter, the processor judges the time period as a deep sleep stage I; the processor obtains pulse wave variability according to pulse wave signals, extracts the physiological parameter range of the deep sleep stage I as a deep sleep parameter, sets the sleep stage which does not meet the deep sleep stage I as 'sleep to be identified', matches the deep sleep parameter with the physiological parameter of 'sleep to be identified', and separates the sleep stage which meets the deep sleep parameter as a deep sleep stage II.