CN115137312B

CN115137312B - Sleep quality evaluation method and device and wearable device

Info

Publication number: CN115137312B
Application number: CN202211050538.4A
Authority: CN
Inventors: 王宁利; 刘勇; 王嘉琪; 陈君亮; 赵子贺; 许文隽; 张弛; 高硕�; 康梦田
Original assignee: Beijing Tongren Hospital
Current assignee: Beijing Tongren Hospital
Priority date: 2022-08-31
Filing date: 2022-08-31
Publication date: 2023-01-24
Anticipated expiration: 2042-08-31
Also published as: CN115137312A

Abstract

The application relates to the technical field of sleep quality assessment, and provides a sleep quality assessment method and device and wearable equipment. The method comprises the following steps: acquiring wrist movement data, pulse blood oxygen saturation data, melatonin data and sympathetic nerve excitation data at the current time, wherein the wrist movement data comprises at least two kinds of movement subdata; generating cross ratio data based on the at least two motion sub-data and the pulse oximetry data; inputting the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the cross ratio data into a pre-trained sleep quality evaluation model, generating and storing sleep state information at the current time; the method comprises the steps of obtaining a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle, and generating a sleep quality evaluation result of the target sleep cycle based on the sleep state information. According to the embodiment of the application, the accuracy of sleep quality evaluation can be greatly improved.

Description

Sleep quality evaluation method and device and wearable device

Technical Field

The application relates to the technical field of sleep quality assessment, in particular to a sleep quality assessment method and device and wearable equipment.

Background

Sleep quality monitoring is an important component of human intelligent health monitoring, and with the rapid development of intelligent sensing and communication technologies, a plurality of commercially available portable sleep monitoring systems are available.

The existing sleep quality monitoring is mainly realized by an intelligent bracelet, a body movement recorder in the bracelet records the wrist movement condition of a user during sleep by utilizing a triaxial acceleration sensor, and meanwhile, the heart rate change condition is monitored by utilizing a photoelectric sensor, and the sleep quality is evaluated based on wrist movement information and heart rate change information.

The accuracy of the sleep evaluation in the prior art is low because the evaluation of the sleep quality also involves melatonin, pulse blood oxygen saturation, sympathetic nerve excitation and other factors.

Disclosure of Invention

In view of this, embodiments of the present application provide a sleep quality assessment method, a sleep quality assessment device, and a wearable device, so as to solve the problem in the prior art that the accuracy of sleep assessment is low.

In a first aspect of the embodiments of the present application, a sleep quality assessment method is provided, including:

acquiring wrist movement data, pulse blood oxygen saturation data, melatonin data and sympathetic nerve excitation data at the current time;

generating cross ratio data based on the at least two kinds of motion sub-data and the pulse blood oxygen saturation data;

inputting the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the cross ratio data into a pre-trained sleep quality evaluation model, generating and storing sleep state information at the current time;

the method comprises the steps of obtaining a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle, and generating a sleep quality evaluation result of the target sleep cycle based on the plurality of pieces of sleep state information, wherein the sleep cycle is a complete sleep cycle or an incomplete sleep cycle.

In a second aspect of the embodiments of the present application, there is provided a sleep quality assessment apparatus, including:

the acquisition module is used for acquiring wrist movement data, pulse blood oxygen saturation data, melatonin data and sympathetic nerve excitation data at the current time;

a first generation module, configured to generate cross ratio data based on the at least two types of motion sub-data and the pulse oximetry data;

the second generation module is used for inputting the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitability data and the cross ratio data into a pre-trained sleep quality evaluation model, generating and storing the sleep state information at the current time;

the third generation module is configured to acquire a plurality of pieces of sleep state information corresponding to a target sleep cycle, and generate a sleep quality evaluation result of the target sleep cycle based on the plurality of pieces of sleep state information, where the sleep cycle is a complete sleep cycle or an incomplete sleep cycle.

In a third aspect of the embodiments of the present application, there is provided a wearable device including a wrist movement detecting means, a pulse oximetry detecting means, a melatonin detecting means and a sympathetic nerve excitation detecting means in contact with skin, and a central processing device, wherein,

the wrist movement detection device is used for detecting wrist movement related signals of a wrist and sending the wrist movement related signals to the central processing equipment;

the pulse blood oxygen saturation detection device is used for detecting a pulse blood oxygen saturation signal and sending the pulse blood oxygen saturation signal to the central processing equipment;

the melatonin detection device is used for detecting a first oxidation-reduction current generated by an electrooxidation reaction through the melatonin in sweat, processing the first oxidation-reduction current to obtain melatonin data and sending the melatonin data to the central processing equipment;

the sympathetic nerve excitation degree detection device is used for detecting an abnormal electric signal of the skin, generating a sympathetic nerve excitation degree signal and sending the sympathetic nerve excitation degree signal to the central processing equipment;

the central processing device is used for realizing the steps of the method when being executed.

In a fourth aspect of the embodiments of the present application, a computer-readable storage medium is provided, in which a computer program is stored, and the computer program realizes the steps of the above method when being executed by a processor.

Advantageous effects

Compared with the prior art, the embodiment of the application has the beneficial effects that at least: the sleep state information is judged by comprehensively using wrist movement data, the pulse blood oxygen saturation, the melatonin, the pulse blood oxygen saturation and the sympathetic nerve excitation degree, and the sleep quality evaluation is carried out by combining the sleep state information corresponding to the sleep process or the sleep cycle, so that the accuracy of the sleep quality evaluation is greatly increased.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for the embodiments or the prior art descriptions will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

Fig. 1 is a schematic diagram of a corresponding scenario of a sleep quality assessment method according to an embodiment of the present application;

fig. 2 is a flow diagram of some embodiments of a sleep quality assessment method provided in accordance with an embodiment of the present application;

FIG. 3 is a flow diagram of further embodiments of another sleep quality assessment method provided in accordance with embodiments of the present application;

fig. 4 is a schematic diagram of a simple structure of a sleep quality assessment apparatus according to an embodiment of the present application;

FIG. 5 is a schematic diagram of a central processing facility provided in accordance with an embodiment of the present application;

fig. 6 is a schematic view of a bracelet provided according to an embodiment of the application.

Detailed Description

Embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While certain embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided for a more thorough and complete understanding of the present application. It should be understood that the drawings and embodiments of the present application are for illustration purposes only and are not intended to limit the scope of the present application.

It should be noted that, for the convenience of description, only the parts related to the present application are shown in the drawings. The embodiments and features of the embodiments in the present application may be combined with each other without conflict.

It should be noted that the terms "first", "second", and the like in the present application are only used for distinguishing different systems, devices, modules or units, and are not used for limiting the order or interdependence of the functions executed by the systems, devices, modules or units.

It is noted that references to "a" or "an" modification in this application are intended to be illustrative rather than limiting, and those skilled in the art will appreciate that references to "one or more" are intended to be exemplary unless the context clearly indicates otherwise.

The names of messages or information exchanged between devices in the embodiments of the present application are for illustrative purposes only, and are not intended to limit the scope of the messages or information.

According to the rule of human body activity in a sleep state, the whole sleep process of a human body is divided into a Wake period (waking state period), a REM period (rapid eye movement period) and a Deep sleep period (Deep sleep period), wherein the Wake period is characterized by large activity frequency and amplitude of the whole human body and often accompanied by large movement of wrists, the REM period is a shallow sleep period and is characterized by weak activity signals, the activity frequency of wrists is low, the amplitude of wrists is small, and Deep sleep period is provided, so that the human body keeps calm and basically has no activity information. A standard sleep cycle may have a standard duration of 1.5 hours, and the ratio of Wake, REM, and Deep periods in each sleep cycle may have a standard ratio. The method and the device can predict and store the real-time sleep state (namely, the sleep state belongs to a Wake period, a REM period or a Deep period in a sleep cycle) in the sleep process, call the stored data after the sleep is finished or the sleep cycle is finished, and compare a plurality of data corresponding to the sleep cycle or the complete sleep process with a standard sleep cycle, thereby obtaining the evaluation result of the sleep quality.

It should be noted that the sleep state partition also includes other kinds of partitions, such as a Wake period, a REM period, and an NREM period (non-rapid eye movement period), which are similar to the disclosure and belong to the protection scope of the present application.

The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.

Fig. 1 is a schematic diagram of an application scenario of a sleep quality assessment method according to some embodiments of the present application.

In the application scenario of fig. 1, first, the computing device 101 may acquire wrist movement data 102, pulse blood oxygen saturation data 103, melatonin data 104, sympathetic excitement data 105 at the current time;

second, computing device 101 may generate cross ratio data 106 based on the wrist motion data 102 and pulse oximetry data 103;

again, the computing device 101 may input the wrist movement data 102, the pulse blood oxygen saturation data 103, the melatonin data 104, the sympathetic excitement data 105, and the cross ratio data 106 into a pre-trained sleep quality assessment model 107, generate and store sleep state information 108 for the current time;

finally, the computing device 101 may obtain a plurality of pieces of sleep state information 108 (i.e., n in fig. 1, n is a positive integer, n is greater than or equal to 2) corresponding to each time in a target sleep cycle, and generate a sleep quality evaluation result 109 of the target sleep cycle based on the plurality of pieces of sleep state information 108, where the sleep cycle is a complete sleep cycle or an incomplete sleep cycle.

The computing device 101 may be hardware or software. When the computing device is hardware, it may be implemented as a distributed cluster composed of multiple servers or terminal devices, or may be implemented as a single server or a single terminal device. When the computing device is embodied as software, it may be installed in the hardware devices enumerated above. It may be implemented, for example, as multiple software or software modules to provide distributed services, or as a single software or software module. And is not particularly limited herein.

It should be understood that the number of computing devices in FIG. 1 is merely illustrative. There may be any number of computing devices, as the implementation requires.

With continued reference to fig. 2, a flow 200 of some embodiments of a sleep quality assessment method according to the present application is shown. The method may be performed by computing device 101 in fig. 1. The sleep quality evaluation method comprises the following steps:

step 201, obtaining wrist movement data, pulse blood oxygen saturation data, melatonin data and sympathetic nerve excitation data of the current time, wherein the wrist movement data comprises at least two types of movement sub-data.

In some embodiments, the subject performing the sleep quality assessment method (e.g., computing device 101 shown in fig. 1) may connect to the target device via a wired connection or a wireless connection, and then obtain wrist movement data, pulse blood oxygen saturation data, melatonin data, and sympathetic nerve excitation data at the current time.

It should be noted that the wireless connection means may include, but is not limited to, a 3G/4G/5G connection, a WiFi connection, a bluetooth connection, a WiMAX connection, a Zigbee connection, an UWB (ultra wideband) connection, and other wireless connection means now known or developed in the future.

The wrist movement data may refer to information about detected wrist movements during sleep of the human body. By way of example, wrist motion information may include yaw angle data, roll angle data, pitch angle data, horizontal axis acceleration data, vertical axis acceleration data, and the like for wrist motion. It is noted that the wrist motion data may include at least two kinds of motion sub data. The motion subdata may refer to one of data centers such as yaw angle data, roll angle data, pitch angle data, horizontal axis acceleration data, vertical axis acceleration data and the like in the wrist motion information. The pulse oximetry data may refer to data related to detected changes in the concentration of blood oxygen in the blood during sleep of the person. The pulse oximetry data may estimate oxygenation of the lungs and the hemoglobin oxygen carrying capacity. The ambient light intensity data may refer to data relating to the illumination intensity of the ambient light. Melatonin data may refer to data related to melatonin that may be detected as being present in the human body. Sympathetic excitability data may refer to data related to skin conductance abnormalities caused by abnormal excitation in electrical brain stimulation.

Step 202, generating cross ratio data based on the at least two kinds of motion sub-data and the pulse oximetry data.

In some embodiments, the execution subject may generate cross ratio data based on at least two motion sub-data of the wrist motion data and the pulse oximetry protection data. It should be noted that the cross ratio data may refer to data obtained by combining any two of the above three data and calculating the ratio. By calculating the cross ratio data, the feature dimension of the data can be increased, so that a more accurate result can be predicted when machine learning is subsequently performed.

And 203, inputting the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitability data and the cross ratio data into a pre-trained sleep quality evaluation model, generating and storing the sleep state information at the current time.

In some embodiments, the executing subject may input the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic excitement data, and the cross ratio data into a pre-trained sleep quality assessment model, generate and store the sleep state information of the current time. The evaluation model can be various machine learning models and is used for predicting the evaluation result of the sleep quality based on wrist movement data, pulse blood oxygen saturation data, melatonin data, sympathetic nerve excitability data and the cross ratio data. As an example, the evaluation model may be a model based on different algorithms, such as a decision tree model, a neural network model, a support vector machine model, or a clustering algorithm model. The result predicted by the evaluation model may be a sleep state at the current time. The sleep state may be data in an integer, percentage, real number, or other representation form that conforms to a preset range, and the sleep state may be classified into different sleep states (Wake period, REM period, or Deep period) based on preset classification. The result may also be an identifier based on a word expression, such as english, chinese, or a special symbol, which may indicate a sleep state (Wake phase, REM phase, or Deep phase), and is not specifically limited herein.

In addition, when the sleep state information is saved, the sleep state information can be saved to a local storage structure, other storage structures connected in a wired or wireless connection mode, a cloud storage structure and other common storage structures.

Step 204, acquiring a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle, and generating a sleep quality evaluation result of the target sleep cycle based on the plurality of pieces of sleep state information, wherein the sleep cycle is a complete sleep cycle or an incomplete sleep cycle.

In some embodiments, the execution subject may acquire a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle, and generate a sleep quality evaluation result of the target sleep cycle based on the plurality of pieces of sleep state information, where the sleep cycle is a complete sleep cycle or an incomplete sleep cycle. The target sleep cycle may refer to a sleep cycle selected for evaluating sleep quality. It should be noted that, because the sleep cycle is interrupted due to a half-wake or sleep termination of the human sleep, and an incomplete cycle is generated, the target sleep cycle may be a complete sleep cycle or an incomplete sleep cycle.

The beneficial effects of one of the above embodiments of the present application at least include: the sleep state information is judged by comprehensively using wrist movement data, the pulse blood oxygen saturation, the melatonin, the pulse blood oxygen saturation and the sympathetic nerve excitation degree, and the sleep quality evaluation is carried out by combining the sleep state information corresponding to the sleep process or the sleep cycle, so that the accuracy of the sleep quality evaluation is greatly increased.

In some embodiments, the wrist motion data comprises at least two motion sub-data of yaw angle data, roll angle data, pitch angle data, lateral axis acceleration data, longitudinal axis acceleration data, and vertical axis acceleration data; the step of executing the subject to acquire the wrist movement data and the pulse oximetry data includes:

in the first step, the executing body can obtain pulse blood oxygen saturation signals and wrist movement signals, wherein the wrist movement signals comprise at least two motion sub-signals of yaw angle signals, roll angle signals, pitch angle signals, transverse axis acceleration signals, longitudinal axis acceleration signals and vertical axis acceleration signals. The yaw angle signal, the roll angle signal, the pitch angle signal, the horizontal axis acceleration signal, the vertical axis acceleration signal, and the vertical axis acceleration signal may be related signals of the acquired information about the wrist movement. The yaw angle signal, the roll angle signal, the pitch angle signal, the horizontal axis acceleration signal, the vertical axis acceleration signal, and the vertical axis acceleration signal are all common signals (data) for calculating relative motion information, and are not described in detail again. Since some motion sub-signals have lower weights and can even be omitted when performing model calculations, in some cases only part of the motion sub-signals can be obtained.

And secondly, the executing body can perform Gaussian filtering processing on the pulse blood oxygen saturation degree signal and the at least two motion sub-signals to obtain at least two motion sub-signals after pulse blood oxygen saturation filtering.

And a third step, the execution main body may use the current time as a base point, and expand a first preset time length before the base point and a second preset time length after the base point to generate the first sampling window.

Fourthly, the executing body can calculate the mean square difference data, the maximum positive value data and the maximum negative value data of the waveform amplitude of the filtered pulse blood oxygen saturation signal in the first sampling window to obtain the pulse blood oxygen saturation data.

And fifthly, the executing body may respectively calculate mean square difference data, maximum positive value data, and maximum negative value data of the waveform amplitudes of the at least two filtered motion sub-signals in the first sampling window, so as to obtain the wrist motion data.

At least two signals of the pulse blood oxygen saturation signal, the yaw angle signal, the roll angle signal, the pitch angle signal, the horizontal axis acceleration signal, the vertical axis acceleration signal and the vertical axis acceleration signal are processed, and data in a certain time window needs to be combined for comprehensive processing, so that data with better data characteristics are obtained. In the embodiment, a time interval with a certain proportion is extended forwards or backwards by taking a current time point as a reference, and change data in the interval is calculated to obtain the data characteristic of the current time. The mean square difference data may be a variance value of the waveform amplitudes in a sampling window, and the maximum positive value data and the maximum negative value data may refer to the maximum positive value data and the maximum negative value data in the sampling window. By setting a fixed time window for calculation, better data characteristics can be obtained, and further more accurate prediction results can be obtained when model prediction is carried out.

In some embodiments, the executing body may generate the cross-ratio data based on the at least two kinds of motion sub-data and the pulse oximetry data by: screening at least one data pair from the pulse blood oxygen saturation data and the at least two types of motion subdata based on a preset data pair screening index, wherein each data pair is composed of two different types of data, and the different data pairs comprise the same or different data; and calculating the ratio of the two data in each data pair to obtain the cross ratio data.

Since the wrist movement data plus the pulse blood sample saturation data has at least three (or all seven) data, pairwise pairing can increase data dimensionality, but does not require any pairwise cross ratio data, as excessive data dimensionality can increase meaningless resource consumption. Therefore, one or more cross ratio data with larger weight (larger influence degree) can be obtained through experiments, the data pair screening index is set according to the one or more cross ratio data, and then the cross ratio data based on the screening index is obtained through calculation. Through the screening, can be increasing under the prerequisite of data characteristic, guarantee the prediction precision, the running efficiency of this application is promoted to the minimum resource consumption that tries to get one's best.

In some embodiments, the step of executing the subject to acquire the melatonin data includes: acquiring a current signal generated by the reaction of the detection part and melatonin in sweat; performing signal amplification processing on the current signal to obtain an amplified current signal; performing analog-to-digital conversion processing on the amplified current signal to obtain a digital melatonin signal; and determining the melatonin data of the digital melatonin signal at the current time as the melatonin data.

In some embodiments, the performing agent may obtain the sympathetic excitement data based on: acquiring abnormal electrical signals detected from the skin surface; filtering the abnormal electric signal aiming at the high-frequency interference signal to obtain the filtered abnormal electric signal; performing empirical mode decomposition on the filtered abnormal electric signal to obtain a skin conductance level signal and a skin conductance response signal, wherein the frequency of the skin conductance level signal is lower than that of the skin conductance response signal; taking the current time as a base point, expanding a first preset time length before the base point, expanding a second preset time length after the base point, and generating a sampling window; calculating skin conductance baseline data of the skin conductance level signal in the sampling window; calculating skin conductance maximum data of the skin conductance response signal in the sampling window; generating the sympathetic excitement data based on the skin conductance baseline data and the skin conductance maximum data.

Similar to the foregoing processing of wrist movement data, for the processing of sympathetic nerve excitation data, it is necessary to combine data within a certain time window for comprehensive processing to obtain data with better data characteristics. In the embodiment, a time interval with a certain proportion is extended forwards or backwards by taking a current time point as a reference, and change data in the interval is calculated to obtain the data characteristic of the current time. The abnormal electrical signal may refer to a mixed signal of detected abnormalities of the skin. The abnormal electrical signal may be a mixture of a noise signal, a direct current signal, and other signals generated by the body for the direct current signal and reflecting frequencies. After the abnormal electric signals are decomposed, skin conductance level signals which are similar to straight lines and represent the direct current signals and skin conductance response signals which represent response frequencies can be obtained; obviously, the skin conductance level signal changes very slowly, while the skin conductance response signal changes more in magnitude. The skin conductance baseline data may refer to an average of the skin conductance response signal over the sampling window, approximately representing the value of the dc signal. The skin conductance maximum value data may refer to a maximum amplitude of the skin conductance-responsive signal in the sampling window. By decomposing abnormal electric signals and calculating through a set time window, more accurate sympathetic nerve excitation data can be obtained.

In some embodiments, after generating the cross ratio data based on the at least two kinds of motion sub-data and the pulse oximetry data, further comprising: calibrating melatonin data based on a preset coefficient; and performing coupling denoising processing on the periodic interference data and the Gaussian white noise data according to the wrist motion data, the pulse blood oxygen saturation data, the sympathetic nerve excitation data and the cross ratio data and the calibrated melatonin data, and determining the wrist motion data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the cross ratio data which are subjected to the coupling denoising processing as input data of the pre-trained sleep quality assessment model.

Since the data processing involves amplification and other processing of the signal, resulting in proportional distortion of the generated melatonin data, correction of the generated melatonin data is required. The preset coefficient may refer to a correction coefficient for melatonin. The range of the coefficient is set according to actual conditions. The correction coefficient may be determined by experiment or by standard data in the field, and is not limited herein. By way of example, the coefficient may range from 0.1 to 0.5. The coefficient is multiplied by the melatonin data to be generated.

In some embodiments, the pre-trained sleep quality assessment model includes an auto-encoder model and a regression model.

The sleep quality assessment model is a combination of an auto-encoder model and a regression model. Data can be firstly imported into the self-encoder model, and then the output result is imported into the regression model, so that the final prediction result, namely the sleep state information of the current time is obtained. When the model is selected, the calculation consumption and the accuracy of the model need to be considered, and through the test of various models (or model combinations), the self-encoder model and the regression model are the better model combination aiming at the application, so that the required accuracy can be achieved, and the calculation consumption can be reduced.

It should be noted that the regression model may be any one of the following regression models: a linear regression model, a stepwise regression model, a neural network model, and the like. The preferred regression model for this application is a linear regression model.

In some embodiments, the training of the execution subject for the sleep quality assessment model includes:

in the first step, the executing body can import the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the crossing ratio data of the known sleep state information into a preset original self-encoder model, and generate predicted comparison wrist movement data, comparison pulse blood oxygen saturation data, comparison melatonin data, comparison sympathetic nerve excitation data and comparison crossing ratio data. The original self-encoder model may refer to a pre-established model corresponding to the present application without parameter optimization.

And secondly, the executing subject needle can perform iterative training on the self-encoder model based on wrist movement data, pulse blood oxygen saturation data, melatonin data, sympathetic nerve excitation data and the crossing ratio data of the known sleep state information, comparison wrist movement data, comparison pulse blood oxygen saturation data, comparison melatonin data, comparison sympathetic nerve excitation data and the comparison crossing ratio data, and a preset first loss function to obtain a trained self-encoder model. The first penalty function may refer to a penalty function for the self-encoder model for calculating penalty values for the predicted and true results to optimize parameters of the model.

And thirdly, the executing main body needle can lead the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitability data and the cross ratio data of the known sleep state information into a trained self-encoder model to generate a hidden variable.

And fourthly, the executing subject can lead the hidden variable into a preset original regression model, and iterative training is carried out based on a preset second loss function, so that a trained regression model is obtained. The original regression model may refer to a model in which parameters for predicting a sleep state based on implicit variables are not optimized. The second loss function may refer to a loss function of the original regression model for calculating loss values of the predicted result and the true result to optimize parameters of the model.

And fifthly, the executing subject needle may obtain the trained sleep quality assessment model based on the trained self-encoder model and the trained regression model.

In some embodiments, the obtaining a plurality of pieces of sleep state information corresponding to a target sleep cycle and generating a sleep quality evaluation result of the target sleep cycle based on the plurality of pieces of sleep state information includes: acquiring a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle and standard proportional coefficients of a Wake period, a REM period and a Deep period in a standard sleep cycle; generating target proportional coefficients of a Wake period, a REM period and a Deep period in the target sleep cycle based on a plurality of pieces of sleep state information corresponding to each time in the target sleep cycle; and calculating to obtain a sleep quality evaluation result of the target sleep cycle based on a preset proportion calculation formula, the target proportion coefficient and the standard proportion coefficient. It should be understood that the plurality of sleep state information corresponding to each time of the target sleep cycle may be sleep state information of the current time obtained at different times.

The preset proportion calculation formula may be a calculation formula which is pre-established and used for calculating a ratio between the ratio of each sleep state in the target sleep cycle and the ratio of each sleep state in the standard sleep cycle. For example, assume that the standard ratios are 33.33%, 44.44%, and 22.22% in the Wake phase, REM phase, and Deep phase, respectively. The proportions of each period in the target sleep cycle are 20%, 50% and 30%, respectively, then (10/33.33 +60/44.44+ 30/22.22)/3 = (0.3 +1.35+ 1.4)/3 = (1.02) can be calculated based on the calculation formula of the evaluation score = (target data of each stage/standard data of each stage)/3. It can be seen that the ratio of target sleep to standard sleep is 1.02. In addition, different weights can be added to different stages to calculate the scores, and the calculated scores can be multiplied by other preset coefficients. Or, the occupation ratio of each sleep state in the target sleep cycle can be imported into a preset machine learning model for calculating a sleep evaluation result, and the like. If necessary, the method is not limited herein.

In some embodiments, the present application further comprises: the execution main body can acquire a sleep quality evaluation result of at least one sleep cycle in the complete sleep process and generate a sleep quality evaluation result corresponding to the complete sleep process. After the target sleep cycle is calculated, the score of each sleep cycle in the complete sleep process may be processed (e.g., added, multiplied, given different calculation coefficients according to different time periods, or processed in other calculation manners) to obtain a composite score of the complete sleep process.

With continued reference to fig. 3, a flow 300 of further embodiments of a sleep quality assessment method according to the present application is shown, which may be performed by the computing device 101 of fig. 1. The sleep quality evaluation method comprises the following steps:

step 301, obtaining wrist movement data, pulse blood oxygen saturation data, melatonin data and sympathetic nerve excitation data of the current time, wherein the wrist movement data comprises at least two types of movement sub-data.

Step 302, generating cross ratio data based on the at least two kinds of motion sub-data and the pulse oximetry data.

And step 303, inputting the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the cross ratio data into a pre-trained sleep quality evaluation model, generating and storing the sleep state information of the current time.

Step 304, acquiring a plurality of sleep state information corresponding to each time in the target sleep cycle, and standard proportionality coefficients of a Wake period, a REM period and a Deep period in the standard sleep cycle.

Step 305, generating target proportional coefficients of a Wake period, a REM period and a Deep period in the target sleep cycle based on a plurality of pieces of sleep state information corresponding to each time in the target sleep cycle.

And step 306, calculating to obtain a sleep quality evaluation result of the target sleep cycle based on a preset proportional calculation formula, the target proportional coefficient and the standard proportional coefficient.

And 307, acquiring a sleep quality evaluation result of at least one sleep cycle in the complete sleep process, and generating a sleep quality evaluation result corresponding to the complete sleep process.

In some embodiments, the specific implementation of steps 301 to 307 and the technical effect brought by the implementation may refer to the steps in the embodiments corresponding to fig. 2, which are not described herein again.

All the above optional technical solutions may be combined arbitrarily to form optional embodiments of the present application, and are not described herein again.

The following are embodiments of the apparatus of the present application that may be used to perform embodiments of the method of the present application. For details which are not disclosed in the embodiments of the apparatus of the present application, reference is made to the embodiments of the method of the present application.

With further reference to fig. 4, as an implementation of the above-described methods for the above-described figures, the present application provides some embodiments of a sleep quality assessment apparatus, which correspond to those method embodiments described above for fig. 2.

As shown in fig. 4, the sleep quality evaluation apparatus 400 of some embodiments includes:

an obtaining module 401, configured to obtain wrist motion data, pulse blood oxygen saturation data, melatonin data, and sympathetic nerve excitation data at a current time, where the wrist motion data includes at least two kinds of motion sub-data;

a first generating module 402 for generating cross ratio data based on the at least two motion sub-data and the pulse oximetry data;

a second generating module 403, configured to input the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data, and the cross ratio data into a pre-trained sleep quality assessment model, generate and store sleep state information at the current time;

a third generating module 404, configured to obtain multiple pieces of sleep state information corresponding to a target sleep cycle, and generate a sleep quality evaluation result of the target sleep cycle based on the multiple pieces of sleep state information, where the sleep cycle is a complete sleep cycle or an incomplete sleep cycle.

In some optional implementations of some embodiments, the wrist motion data includes at least two motion subdata of yaw angle data, roll angle data, pitch angle data, lateral axis acceleration data, longitudinal axis acceleration data, and vertical axis acceleration data; a step of acquiring said wrist movement data and said pulse oximetry data, comprising: acquiring pulse blood oxygen saturation signals and wrist movement signals, wherein the wrist movement signals comprise at least two movement sub-signals of yaw angle signals, roll angle signals, pitch angle signals, transverse axis acceleration signals, longitudinal axis acceleration signals and vertical axis acceleration signals; performing Gaussian filtering processing on the pulse blood oxygen saturation degree signal and the at least two motion sub-signals to obtain at least two motion sub-signals after filtering pulse blood oxygen saturation; the method comprises the steps that a current time is used as a base point, a first preset time length is expanded to the front of the base point, a second preset time length is expanded to the rear of the base point, and a first sampling window is generated; calculating mean square difference data, maximum positive value data and maximum negative value data of the waveform amplitude of the filtered pulse blood oxygen saturation signal in the first sampling window to obtain the pulse blood oxygen saturation data; and respectively calculating mean square difference data, maximum positive value data and maximum negative value data of the waveform amplitudes of the at least two filtered motion sub-signals in the first sampling window to obtain the wrist motion data.

In some optional implementations of some embodiments, the generating cross ratio data based on the at least two types of motion sub-data and pulse oximetry data comprises: screening at least one data pair from the pulse blood oxygen saturation data and the at least two types of motion subdata based on a preset data pair screening index, wherein each data pair is composed of two different types of data, and the different data pairs comprise the same or different data; and calculating the ratio of the two data in each data pair to obtain the cross ratio data.

In some optional implementations of some embodiments, the step of obtaining the melatonin data comprises: acquiring a current signal generated by the reaction of the detection part and melatonin in sweat; performing signal amplification processing on the current signal to obtain an amplified current signal; performing analog-to-digital conversion processing on the amplified current signal to obtain a digital melatonin signal; and determining the melatonin data of the digital melatonin signal at the current time as the melatonin data.

In some optional implementations of some embodiments, the step of obtaining the sympathetic excitability data comprises: acquiring abnormal electrical signals detected from the skin surface; filtering the abnormal electric signal aiming at the high-frequency interference signal to obtain the filtered abnormal electric signal; performing empirical mode decomposition on the filtered abnormal electric signal to obtain a skin conductance level signal and a skin conductance response signal, wherein the frequency of the skin conductance level signal is lower than that of the skin conductance response signal; taking the current time as a base point, expanding a third preset time length before the base point, expanding a fourth preset time length after the base point, and generating a second sampling window; calculating skin conductance baseline data of the skin conductance level signal in the second sampling window; calculating skin conductance maximum data of the skin conductance response signal in the second sampling window; generating the sympathetic arousal data based on the skin conductance baseline data and the skin conductance maximum data.

In some optional implementations of some embodiments, after generating the cross ratio data based on the at least two types of motion sub-data and the pulse oximetry data, further comprising: calibrating melatonin data based on a preset coefficient; and performing coupling denoising processing on the periodic interference data and the Gaussian white noise data according to the wrist motion data, the pulse blood oxygen saturation data, the sympathetic nerve excitation data and the cross ratio data and the calibrated melatonin data, and determining the wrist motion data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the cross ratio data which are subjected to the coupling denoising processing as input data of the pre-trained sleep quality assessment model.

In some optional implementations of some embodiments, the pre-trained sleep quality assessment model includes an auto-encoder model and a regression model.

In some optional implementations of some embodiments, the training of the sleep quality assessment model comprises: importing wrist movement data, pulse blood oxygen saturation data, melatonin data, sympathetic nerve excitation data and the cross ratio data of known sleep state information into a preset original self-encoder model to generate predicted comparison wrist movement data, comparison pulse blood oxygen saturation data, comparison melatonin data, comparison sympathetic nerve excitation data and comparison cross ratio data; performing iterative training on the self-encoder model based on wrist movement data, pulse blood oxygen saturation data, melatonin data, sympathetic nerve excitation data and the cross ratio data of the known sleep state information, comparing the wrist movement data with the pulse blood oxygen saturation data, comparing the melatonin data with the sympathetic nerve excitation data with the cross ratio data, and a preset first loss function to obtain a trained self-encoder model; importing the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitability data and the cross ratio data of the known sleep state information into a trained self-encoder model to generate a hidden variable; importing the hidden variable into a preset original regression model, and performing iterative training based on a preset second loss function to obtain a trained regression model; and obtaining the trained sleep quality evaluation model based on the trained self-encoder model and the trained regression model.

In some optional implementation manners of some embodiments, the obtaining multiple pieces of sleep state information corresponding to a target sleep cycle and generating a sleep quality evaluation result of the target sleep cycle based on the multiple pieces of sleep state information includes: acquiring a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle and standard proportional coefficients of a Wake period, a REM period and a Deep period in a standard sleep cycle; generating target proportional coefficients of a Wake period, a REM period and a Deep period in the target sleep cycle based on a plurality of sleep state information corresponding to each time in the target sleep cycle; and calculating to obtain a sleep quality evaluation result of the target sleep cycle based on a preset proportion calculation formula, the target proportion coefficient and the standard proportion coefficient.

In some optional implementations of some embodiments, the sleep quality assessment apparatus 400 further comprises: and the fourth generation module is used for acquiring a sleep quality evaluation result of at least one sleep cycle in the complete sleep process and generating a sleep quality evaluation result corresponding to the complete sleep process.

It is to be understood that the modules recited in the sleep quality assessment apparatus 400 correspond to the respective steps in the method described with reference to fig. 2. Thus, the operations, features and beneficial effects described above for the method are also applicable to the sleep quality assessment apparatus 400 and the modules included therein, and are not described herein again.

As shown in fig. 5, the central processing apparatus 500 may include a processing device (e.g., central processing apparatus, graphics processor, etc.) 501 that may perform various appropriate actions and processes according to a program stored in a Read Only Memory (ROM) 502 or a program loaded from a storage device 508 into a Random Access Memory (RAM) 503. In the RAM 503, various programs and data necessary for the operation of the central processing apparatus 500 are also stored. The processing device 501, the ROM 502, and the RAM 503 are connected to each other through a bus 504. An input/output (I/O) interface 505 is also connected to bus 504.

Generally, the following devices may be connected to the I/O interface 505: input devices 506 including, for example, a touch screen, touch pad, keyboard, mouse, camera, microphone, accelerometer, gyroscope, etc.; output devices 507 including, for example, a Liquid Crystal Display (LCD), a speaker, a vibrator, and the like; storage devices 508 including, for example, magnetic tape, hard disk, etc.; and a communication device 509. The communication means 509 may allow the central processing apparatus 500 to perform wireless or wired communication with other apparatuses to exchange data. While fig. 5 illustrates a central processing facility 500 having various means, it is to be understood that not all illustrated means are required to be implemented or provided. More or fewer devices may alternatively be implemented or provided. Each block shown in fig. 5 may represent one device or may represent multiple devices as desired.

In particular, according to some embodiments of the application, the processes described above with reference to the flow diagrams may be implemented as computer software programs. For example, some embodiments of the present application include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In some such embodiments, the computer program may be downloaded and installed from a network via the communication means 509, or installed from the storage means 508, or installed from the ROM 502. Which when executed by the processing means 501 performs the above-described functions as defined in the methods of some embodiments of the present application.

It should be noted that the computer readable medium described above in some embodiments of the present application may be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In some embodiments of the application, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In some embodiments of the present application, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.

In some embodiments, the clients, servers may communicate using any currently known or future developed network Protocol, such as HTTP (HyperText Transfer Protocol), and may be interconnected with any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), the Internet (e.g., the Internet), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future developed network.

The computer readable medium may be embodied in the apparatus described above; or may be separate and not incorporated into the central processing facility. The computer readable medium carries one or more programs which, when executed by the central processing apparatus, cause the central processing apparatus to:

acquiring wrist movement data, pulse blood oxygen saturation data, melatonin data and sympathetic nerve excitation data at the current time, wherein the wrist movement data comprises at least two kinds of movement subdata;

the method comprises the steps of obtaining a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle, and generating a sleep quality evaluation result of the target sleep cycle based on the sleep state information, wherein the sleep cycle is a complete sleep cycle or an incomplete sleep cycle.

Computer program code for carrying out operations for embodiments of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).

The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.

The modules described in some embodiments of the present application may be implemented by software or hardware. The described modules may also be provided in a processor, which may be described as:

the device comprises an acquisition module, a first generation module, a second generation module and a third generation module. For example, the acquisition module may also be described as a "module that acquires wrist movement data, pulse oximetry data, melatonin data, sympathetic excitement data for the current time".

The functions described herein above may be performed, at least in part, by one or more hardware logic components. For example, without limitation, exemplary types of hardware logic components that may be used include: field Programmable Gate Arrays (FPGAs), application Specific Integrated Circuits (ASICs), application Specific Standard Products (ASSPs), system on a chip (SOCs), complex Programmable Logic Devices (CPLDs), and the like.

In some embodiments, the present application further provides a wearable device comprising a wrist movement detection means, a pulse oximetry detection means, a melatonin detection means in contact with the skin and a sympathetic excitement detection means, and a central processing device, wherein,

the central processing device is configured to perform the steps in those embodiments corresponding to fig. 2.

It should be noted that the central processing device may be disposed in the wearable device, or may be disposed in another server or processing device that can be in wired or wireless communication with the wearable device. When the central processing device is disposed in the wearable device, the central processing device may be disposed in the same physical structure in the wearable device as any one or more of the wrist movement detection means, the pulse oximetry detection means, the melatonin detection means and the sympathetic excitement detection means, or may be disposed in one of the physical structures in the wearable device separately. The central processing device may be an integrated arrangement or a distributed arrangement. The number of central processing devices may be 1 or more. The above-described setting is performed as needed, and is not particularly limited herein.

In addition, the wearable device may be a wearable device that can be attached to skin, such as a bracelet, an arm ring, a neck ring, or a head ring, or may be a garment provided with a detection device that is attached to skin, and the like, and is not limited specifically here.

In some embodiments, the melatonin detecting device comprises a first sensing patch, a first data amplifying unit and a first analog-to-digital conversion unit, wherein the first sensing patch is made of FeCo bimetallic alloy embedded carbon nano fibers; the first induction patch is in contact with the skin and used as a detection part to generate an electrooxidation reaction with melatonin in sweat, and a first oxidation-reduction current is generated between a positive electrode and a negative electrode of the first induction patch; the first data amplification unit is connected with the first induction patch and is used for amplifying the first oxidation-reduction current to obtain an amplified first oxidation-reduction current; the first analog-to-digital conversion unit is connected with the first data amplification unit and used for converting the amplified first oxidation reduction current to obtain a filtered melatonin signal, and melatonin data of the filtered melatonin signal at the current time is the melatonin data. It should be noted that the FeCo bimetallic alloy may be replaced by other two metals or other materials having the same properties and reacting with melatonin in sweat, which may be present or discovered in the future, and all fall within the scope of the present application.

In some embodiments, the sympathetic excitement detection device includes a second inductive patch filtering unit contacting the skin; the second sensing patch is used for detecting abnormal electrical data from the surface of the skin; the filtering unit is connected with the second induction patch and used for filtering the abnormal electrical data aiming at high-frequency interference data to obtain the sympathetic nerve excitation signal.

Referring to fig. 6, in some embodiments, the wearable device is a bracelet. Skin one side is pressed close to the bracelet can be provided with melatonin detection device array and the sympathetic nerve excitability detection device array that interval set up in turn, wherein, melatonin detection device array includes two at least melatonin detection device, sympathetic nerve excitability detection device array includes two at least sympathetic nerve excitability detection device. The same detection device can be for radially setting up along the bracelet inboard, also can be based on the inboard axial setting of bracelet, can also set up or the interval sets up etc. through the slant that accords with preset law, selects as required, does not do specific restriction here. Through setting up the detection device array, can increase the area that detects to improve the precision and the detection efficiency that detect.

The foregoing description is only exemplary of the preferred embodiments of the present application and is provided for the purpose of illustrating the general principles of the technology. It will be appreciated by those skilled in the art that the scope of the invention in the embodiments of the present application is not limited to the specific combination of the above-mentioned features, but also encompasses other embodiments in which any combination of the above-mentioned features or their equivalents is made without departing from the inventive concept as defined above. For example, the above features and (but not limited to) technical features with similar functions disclosed in the embodiments of the present application are mutually replaced to form the technical solution.

Claims

1. A sleep quality assessment method, comprising:

screening at least one data pair with larger weight from the at least two types of motion subdata and pulse oximetry data as a screening index;

screening at least one data pair from the pulse blood oxygen saturation data and the at least two types of motion subdata based on the screening index, wherein each data pair is composed of two different types of data, and the different data pairs comprise the same or different data;

calculating the ratio of the two data in each data pair to obtain cross ratio data;

2. The method of claim 1, wherein the wrist motion data comprises at least two motion data of yaw angle data, roll angle data, pitch angle data, lateral axis acceleration data, longitudinal axis acceleration data, and vertical axis acceleration data; a step of acquiring said wrist movement data and said pulse oximetry data, comprising:

acquiring pulse blood oxygen saturation signals and wrist movement signals, wherein the wrist movement signals comprise at least two motion sub-signals of yaw angle signals, roll angle signals, pitch angle signals, transverse axis acceleration signals, longitudinal axis acceleration signals and vertical axis acceleration signals;

performing Gaussian filtering processing on the pulse blood oxygen saturation degree signal and the at least two motion sub-signals to obtain at least two motion sub-signals after filtering pulse blood oxygen saturation;

taking the current time as a base point, expanding a first preset time length before the base point, expanding a second preset time length after the base point, and generating a first sampling window;

calculating mean square difference data, maximum positive value data and maximum negative value data of the waveform amplitude of the filtered pulse blood oxygen saturation signal in the first sampling window to obtain the pulse blood oxygen saturation data;

and calculating the mean square difference data, the maximum positive value data and the maximum negative value data of the waveform amplitudes of the at least two filtered motion sub-signals in the first sampling window respectively to obtain the wrist motion data.

3. The method of claim 1, wherein the step of obtaining melatonin data comprises:

acquiring a current signal generated by the reaction of the detection part and melatonin in sweat;

performing signal amplification processing on the current signal to obtain an amplified current signal;

performing analog-to-digital conversion processing on the amplified current signal to obtain a digital melatonin signal;

and determining the melatonin data of the digital melatonin signal at the current time as the melatonin data.

4. The method of claim 1, wherein the step of obtaining the sympathetic excitability data comprises:

acquiring abnormal electrical signals detected from the skin surface;

filtering the abnormal electric signal aiming at the high-frequency interference signal to obtain the filtered abnormal electric signal;

performing empirical mode decomposition on the filtered abnormal electric signal to obtain a skin conductance level signal and a skin conductance response signal, wherein the frequency of the skin conductance level signal is lower than that of the skin conductance response signal;

with the current time as a base point, expanding a third preset time length before the base point, and expanding a fourth preset time length after the base point to generate a second sampling window;

calculating skin conductance baseline data of the skin conductance level signal in the second sampling window;

calculating skin conductance maximum data of the skin conductance response signal in the second sampling window;

generating the sympathetic arousal data based on the skin conductance baseline data and the skin conductance maximum data.

5. The method of claim 1, further comprising, after generating cross-ratio data based on the at least two types of motion sub-data and pulse oximetry data:

calibrating melatonin data based on a preset coefficient;

performing coupling denoising processing on the periodic interference data and the Gaussian white noise data according to the wrist motion data, the pulse blood oxygen saturation data, the sympathetic nerve excitation data, the cross ratio data and the calibrated melatonin data, and determining the wrist motion data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the cross ratio data which are subjected to the coupling denoising processing as input data of the pre-trained sleep quality evaluation model.

6. The method of claim 1, wherein the pre-trained sleep quality assessment model comprises an auto-encoder model and a regression model.

7. The method of claim 6, wherein the training step of the sleep quality assessment model comprises:

importing wrist movement data, pulse blood oxygen saturation data, melatonin data, sympathetic nerve excitation data and the cross ratio data of known sleep state information into a preset original self-encoder model to generate predicted comparison wrist movement data, comparison pulse blood oxygen saturation data, comparison melatonin data, comparison sympathetic nerve excitation data and comparison cross ratio data;

performing iterative training on the self-encoder model based on wrist movement data, pulse blood oxygen saturation data, melatonin data, sympathetic nerve excitation data and the cross ratio data of the known sleep state information, comparing the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitation data and the cross ratio data, and a preset first loss function to obtain a trained self-encoder model;

importing the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitability data and the cross ratio data of the known sleep state information into a trained self-encoder model to generate a hidden variable;

importing the hidden variable into a preset original regression model, and performing iterative training based on a preset second loss function to obtain a trained regression model;

and obtaining the trained sleep quality evaluation model based on the trained self-encoder model and the trained regression model.

8. The method according to any one of claims 1 to 7, wherein the obtaining a plurality of sleep state information corresponding to a target sleep cycle and generating a sleep quality assessment result of the target sleep cycle based on the plurality of sleep state information comprises:

acquiring a plurality of pieces of sleep state information corresponding to each time in a target sleep cycle and standard proportional coefficients of a Wake period, a REM period and a Deep period in a standard sleep cycle;

generating target proportional coefficients of a Wake period, a REM period and a Deep period in the target sleep cycle based on a plurality of sleep state information corresponding to each time in the target sleep cycle;

and calculating to obtain a sleep quality evaluation result of the target sleep cycle based on a preset proportion calculation formula, the target proportion coefficient and the standard proportion coefficient.

9. The method of claim 8, further comprising:

and acquiring a sleep quality evaluation result of at least one sleep cycle in the complete sleep process, and generating a sleep quality evaluation result corresponding to the complete sleep process.

10. A sleep quality evaluation apparatus, characterized by comprising:

the device comprises an acquisition module, a display module and a display module, wherein the acquisition module is used for acquiring wrist motion data, pulse blood oxygen saturation data, melatonin data and sympathetic nerve excitation data at the current time, and the wrist motion data comprises at least two types of motion subdata;

the first generation module is used for screening out at least one data pair with larger weight from the at least two types of motion sub-data and pulse blood oxygen saturation data as a screening index; screening at least one data pair from the pulse blood oxygen saturation data and the at least two types of motion subdata based on the screening index, wherein each data pair is composed of two different types of data, and the different data pairs comprise the same or different data; calculating the ratio of two data in each data pair to obtain cross ratio data;

the second generation module is used for inputting the wrist movement data, the pulse blood oxygen saturation data, the melatonin data, the sympathetic nerve excitability data and the cross ratio data into a pre-trained sleep quality evaluation model, and generating and storing the sleep state information of the current time;

11. A wearable device is characterized by comprising a wrist movement detection device, a pulse blood oxygen saturation detection device, a melatonin detection device and a sympathetic nerve excitation degree detection device which are in contact with the skin, and a central processing device, wherein,

the wrist movement detection device is used for detecting wrist movement related signals of the wrist and sending the wrist movement related signals to the central processing equipment;

the sympathetic nerve excitability detection device is used for detecting an abnormal electric signal of the skin, generating a sympathetic nerve excitability signal and sending the sympathetic nerve excitability signal to the central processing equipment;

the central processing device is configured to perform the method of any of claims 1 to 9.

12. The wearable device of claim 11, wherein the melatonin detecting device comprises a first sensing patch, a first data amplifying unit and a first analog-to-digital converting unit, wherein the first sensing patch is made of FeCo bimetallic alloy embedded carbon nano fibers;

the first induction patch is in contact with the skin and is used as a detection part to generate an electrooxidation reaction with melatonin in sweat, and a first redox current is generated between the positive electrode and the negative electrode of the first induction patch;

the first data amplification unit is connected with the first induction patch and is used for amplifying the first oxidation-reduction current to obtain an amplified first oxidation-reduction current;

the first analog-to-digital conversion unit is connected with the first data amplification unit and used for converting the amplified first oxidation reduction current to obtain a filtered melatonin signal, and melatonin data of the filtered melatonin signal at the current time is the melatonin data.

13. The wearable device according to claim 11, wherein the sympathetic excitement detecting means comprises a second inductive patch filtering unit that contacts the skin;

the second sensing patch is used for detecting abnormal electrical data from the surface of the skin;

the filtering unit is connected with the second induction patch and used for filtering the abnormal electrical data aiming at high-frequency interference data to obtain the sympathetic nerve excitation signal.

14. Wearing device according to any one of claims 11 to 13, characterized in that the wearing device is a bracelet.

15. The wearable device according to claim 14, wherein the bracelet is provided with an array of melatonin detection devices and an array of sympathetic excitement detection devices alternately arranged at intervals on a side close to the skin, wherein the array of melatonin detection devices comprises at least two melatonin detection devices, and the array of sympathetic excitement detection devices comprises at least two sympathetic excitement detection devices.

16. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 9.