An adaptive sleep evaluation device and method

Abstract

The application provides a kind of adaptive sleep evaluation device and method.The device includes: sleep monitoring information acquisition module, model calculation module and visualization module.Sleep monitoring information acquisition module includes the wearable device being set on user, and the sleep information and vital activity information of user are collected.Model calculation module is based on the BSR and ASR according to the sleep information and vital activity information of user according to the iterative operation of setting, obtains the SAV value of user day, and sleep rating is carried out to user according to SAV value.Visualization module is through the terminal of user to the SAV value and sleep rating result of user day is shown.The method of the application is aimed at improving sleep monitoring accuracy, combines evolutionary algorithm optimization mathematical model sleep rating standard, accurately and adaptively monitors personnel sleep state, evaluates individual sleep grade, so that personnel can more clearly understand their own sleep condition and health status.

Description

Technical Field

[0001] This invention relates to the field of sleep monitoring technology, and more particularly to an adaptive sleep evaluation device and method. Background Technology

[0002] With the emergence of declining sleep quality and increased attention to sleep, sleep monitoring technology products are gradually becoming necessities for insomniacs. However, existing sleep monitoring solutions mainly fall into two categories:

[0003] ① Sleep health monitoring products: Representative products in this category include sleep monitoring bracelets, smartwatches and other smart wearable devices. Their functions generally include a comprehensive evaluation of sleep quality based on sleep stages. Smartwatches also include sports and health monitoring, such as calories burned, daily steps, running or cycling distance, etc.

[0004] ② Sleep Health Protection Products: Representative products in this category include sleep aid lamps, aromatherapy diffusers, noise control devices, smart eye masks, and earplugs (excluding pharmaceutical products). Their function is generally to improve the sleep environment, but they cannot provide comprehensive monitoring of a person's physical health. While both of these product categories fulfill the basic function of sleep monitoring and protection, the increasing prevalence of sleep quality problems, the diverse causes of these problems, and the significant differences in individual physical conditions all contribute to the challenges.

[0005] One existing sleep monitoring method based on smartwatches involves dividing the smartwatch usage phase into an initial phase and a long-term phase, addressing the physiological or psychological impact caused by emotions and wearing habits during the initial phase, and improving the monitoring accuracy during the initial phase.

[0006] The disadvantages of the existing smartwatch-based sleep monitoring methods mentioned above include:

[0007] ① The technical solution of using a uniform standard to evaluate the sleep of all people is gradually losing its accuracy. Different people with the same sleep time and sleep evaluation show differences in their mental state during the day, which indicates that a uniform standard cannot accurately reflect the true sleep situation of an individual.

[0008] ② Considering the issue of sleep adaptation in abnormal environments, people's adaptability to sleep varies in different environments. For example, sleep quality is generally worse at high altitudes than at low altitudes, and blood oxygen concentration is unstable. Therefore, using the same sleep evaluation standards will affect the scientific validity of the evaluation indicators. The evaluation method should be tailored to the sleep adaptation process of individuals in abnormal environments. Summary of the Invention

[0009] Embodiments of the present invention provide an adaptive sleep evaluation device and method to effectively improve the quality of sleep monitoring for users.

[0010] To achieve the above objectives, the present invention adopts the following technical solution.

[0011] According to another aspect of the present invention, an adaptive sleep evaluation device is provided, comprising: a sleep monitoring information acquisition module, a model calculation module, and a visualization module;

[0012] The sleep monitoring information acquisition module includes a wearable device installed on the user, which collects the user's sleep information, pushes a questionnaire to the user, obtains the user's vital sign activity information based on the user's questionnaire answers, and transmits the user's sleep information and vital sign activity information to the model calculation module.

[0013] The model calculation module is used to set the Basic Sleep Rating Standard (BSR) and the Optimized Sleep Rating Standard (ASR). Based on the BSR and ASR, iterative calculations are performed according to the received sleep information and vital sign activity information of the user to obtain the user's sleep score (SAV) value for the day. The user's sleep is rated according to the SAV value, and the user's SAV value and sleep rating result for the day are transmitted to the visualization module.

[0014] The visualization module is used to display the user's SAV value and sleep rating results for the day on the user's terminal.

[0015] Preferably, the sleep monitoring information acquisition module includes: a wearable device equipped with a sleep monitoring device and a vital sign monitoring device;

[0016] The sleep monitoring device is used to collect sleep data from users through internal sensors. The sleep data includes the percentage of deep sleep, total sleep time, and continuity of deep sleep.

[0017] The aforementioned vital sign monitoring device is used to push questionnaires to users and obtain users' vital sign data based on the users' questionnaire answers. The vital sign data includes heart rate, daily steps, and self-report questionnaires.

[0018] Preferably, the wearable device includes a smart bracelet, and the internal sensors of the wearable monitoring device include a heart rate sensor, a gyroscope, an accelerometer, a barometric pressure sensor, a temperature sensor, an ambient light sensor, a capacitive touch sensor, a battery status sensor, a skin conduction sensor, and a blood oxygen sensor.

[0019] Preferably, the BSR is a percentage-based evaluation standard, corresponding to a pair of (x, y) arrays, where x and y represent the score thresholds in the evaluation standard, corresponding to three score levels respectively. The evaluation standard is divided into: For excellent sleep, Normal sleep and y represents the three-level standard for poor sleep, and ASR is a percentage-based evaluation standard, representing the result of iteratively updating the array (x, y) of BSR.

[0020] Preferably, the model calculation module includes:

[0021] The Basic Sleep Value (SBV) calculation module is used to determine a user's SBV score based on the percentage of deep sleep, total sleep time, and deep sleep continuity. The SBV score includes 0.3, 0.5, and 0.7 points, corresponding to the three states of "full of energy," "normal state," and "poor state," respectively.

[0022] The Daily Activity Rating (DAV) calculation module is used to perform heart rate variability analysis based on vital sign data. It performs linear combination weighting and summation on three indicators: heart rate variability analysis, daily steps, and self-reported questionnaire results, to calculate the user's DAV value.

[0023] Formula 1

[0024] Among them, HRV is heart rate variability analysis, measured as a percentage; Step is daily steps, measured as steps; SR is self-reported questionnaire results, measured as three-level scores of 0.3, 0.5, and 0.7; a, b, and c are the corresponding indicator weights; and DAV is measured as a percentage score.

[0025] The SBV and DAV iterative optimization calculation modules are used to... The initial value of SBV is input into the sleep evaluation mathematical model. SBV is defined as the optimization parameter in the b-direction, and DAV is the optimization parameter in the w-direction. Mini-batch is defined as a small-batch iteration, where each iteration's dataset consists of all data within 30 days from the user's current date. The input data for parts 2-7 are SBV, DAV, and the initial value of SBV. The model iteratively updates and outputs daily ASR and SAV. In the t-th iteration, d is calculated on the current mini-batch. d :

[0026] Formula 2

[0027] Formula 3

[0028] Formula 4

[0029] Formula 5

[0030] Formula 6

[0031] Formula 7

[0032] Formula 2-3 is used to calculate the squared gradient. and These represent the squared gradient momentum accumulated by the loss function during the first t-1 iterations; This is the gradient accumulation exponent, typically taken as 0.9; =10 -8 Formulas 4-5 are for numerical update calculations. The learning rate is determined based on the specific circumstances; Formulas 6-7 are used to calculate the gradient. and These represent the gradients of the latest SBV and DAV data, respectively; Loss is the loss function.

[0033] After iterative calculations using formulas 2-7, the optimized and updated SBV and DAV values ​​are output. These optimized and updated SBV and DAV values ​​are then used as input values ​​for formulas 8-10. The optimized ASR and loss function are calculated as follows:

[0034] Formula 8

[0035] Formula 9

[0036] Where d and e are the corresponding indicator weights; when calculating a new ASR each day, the evaluation standard for the previous day is BSR.

[0037] The Sleep Adaptive Rating (SAV) calculation module is used to calculate the Sleep Adaptive Rating (SAV) value based on the daily optimized and updated SBV and DAV values.

[0038] Formula 10

[0039] SAV value is the user's sleep score on a 100-point scale for that day;

[0040] Users are rated on their sleep based on their SAV (Sleep Activity Value), which includes excellent sleep, normal sleep, and poor sleep.

[0041] According to another aspect of the present invention, an adaptive sleep assessment method is provided, comprising:

[0042] The sleep information of the user is collected by a wearable device set on the user. The sleep monitoring device pushes a questionnaire to the user and obtains the user's vital sign activity information based on the user's questionnaire answers.

[0043] Based on the set BSR and ASR, iterative calculations are performed on the user's sleep information and vital sign activity information to obtain the user's SAV value for the day, and the user's sleep rating is performed based on the SAV value.

[0044] The user's SAV value and sleep rating results for the day are displayed on the user's terminal.

[0045] Preferably, the method involves collecting the user's sleep information via a wearable device mounted on the user, wherein the sleep monitoring device sends a questionnaire to the user, and obtains the user's vital sign activity information based on the user's returned questionnaire answers, including:

[0046] The wearable device worn by the user is equipped with sleep monitoring and vital sign monitoring devices.

[0047] The sleep monitoring device collects the user's sleep data through internal sensors, including the percentage of deep sleep, total sleep time, and continuity of deep sleep.

[0048] The vital signs monitoring device sends a questionnaire to the user and obtains the user's vital signs data based on the user's questionnaire answers. The vital signs data includes heart rate, daily steps, and self-report questionnaires.

[0049] Preferably, the step of iteratively calculating the user's SAV value based on the set BSR and ASR according to the user's sleep information and vital sign activity information, and then rating the user's sleep based on the SAV value, includes:

[0050] The user's SBV score is determined based on the percentage of deep sleep, total sleep time, and deep sleep continuity.

[0051] Heart rate variability analysis was performed based on vital signs data. The three indicators of heart rate variability analysis, daily steps and self-report questionnaire results were linearly combined, weighted and summed to calculate the user's DAV value.

[0052] Formula 1

[0053] Among them, HRV is heart rate variability analysis, measured as a percentage; Step is daily steps, measured as steps; SR is self-reported questionnaire results, measured as three-level scores of 0.3, 0.5, and 0.7; a, b, and c are the corresponding indicator weights; and DAV is measured as a percentage score.

[0054] Will The initial value of SBV is input into the sleep evaluation mathematical model. SBV is defined as the optimization parameter in the b-direction, and DAV is the optimization parameter in the w-direction. Mini-batch is defined as a small-batch iteration, where each iteration's dataset consists of all data within 30 days from the user's current date. The input data for parts 2-7 are SBV, DAV, and the initial value of SBV. The model iteratively updates and outputs daily ASR and SAV. In the t-th iteration, d is calculated on the current mini-batch. d :

[0055] Formula 2

[0056] Formula 3

[0057] Formula 4

[0058] Formula 5

[0059] Formula 6

[0060] Formula 7

[0061] Formula 2-3 is used to calculate the squared gradient. and These represent the squared gradient momentum accumulated by the loss function during the first t-1 iterations; This is the gradient accumulation exponent, typically taken as 0.9; =10 -8 Formulas 4-5 are for numerical update calculations. The learning rate is determined based on the specific circumstances; Formulas 6-7 are used to calculate the gradient. and These represent the gradients of the latest SBV and DAV data, respectively; Loss is the loss function.

[0062] After iterative calculations using formulas 2-7, the optimized and updated SBV and DAV values ​​are output. These optimized and updated SBV and DAV values ​​are then used as input values ​​for formulas 8-10. The optimized ASR and loss function are calculated as follows:

[0063] Formula 8

[0064] Formula 9

[0065] Where d and e are the corresponding indicator weights; when calculating a new ASR each day, the evaluation standard for the previous day is BSR.

[0066] The SAV value is calculated based on the daily optimized and updated SBV and DAV values;

[0067] Formula 10

[0068] SAV value is the user's sleep score on a 100-point scale for that day;

[0069] Users are rated on their sleep based on their SAV (Sleep Activity Value), which includes excellent sleep, normal sleep, and poor sleep.

[0070] As can be seen from the technical solutions provided by the embodiments of the present invention above, the method of the present invention aims to improve the accuracy of sleep monitoring. It combines an evolutionary algorithm to optimize the mathematical model of sleep rating standards, accurately and adaptively monitors the sleep state of individuals, evaluates individual sleep levels, and allows individuals to more clearly understand their sleep conditions and health status. It achieves self-learning, self-adaptation, and personalization of sleep rating, and can be applied to any sleep monitoring device, improving the accuracy of sleep monitoring and evaluation.

[0071] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and will become apparent from the description or may be learned by practice of the invention. Attached Figure Description

[0072] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0073] Figure 1 A schematic diagram illustrating the implementation principle of an adaptive sleep evaluation device provided in an embodiment of the present invention;

[0074] Figure 2 A schematic diagram of the structure of an adaptive sleep evaluation device provided in an embodiment of the present invention;

[0075] Figure 3 This is a flowchart illustrating an adaptive sleep evaluation method provided in an embodiment of the present invention. Detailed Implementation

[0076] Embodiments of the present invention are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0077] Those skilled in the art will understand that, unless specifically stated otherwise, the singular forms “a,” “an,” “the,” and “the” used herein may also include the plural forms. It should be further understood that the term “comprising” as used in this specification means the presence of the stated features, integers, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof. It should be understood that when we say an element is “connected” or “coupled” to another element, it can be directly connected or coupled to the other element, or there may be intermediate elements. Furthermore, “connected” or “coupled” as used herein can include wireless connections or couplings. The term “and / or” as used herein includes any and all combinations of one or more of the associated listed items.

[0078] It will be understood by those skilled in the art that, unless otherwise defined, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. It should also be understood that terms such as those defined in general dictionaries should be understood to have the same meaning as in the context of the prior art, and should not be interpreted in an idealized or overly formal sense unless defined as herein.

[0079] To facilitate understanding of the embodiments of the present invention, the following will provide further explanation and description with reference to the accompanying drawings and several specific embodiments. These embodiments do not constitute a limitation on the embodiments of the present invention.

[0080] The present invention provides a schematic diagram of the implementation principle of an adaptive sleep evaluation device, as shown in the figure below. Figure 1 As shown, the structure is as follows Figure 2 As shown, it includes: a sleep monitoring information acquisition module, a model calculation module, and a visualization module.

[0081] The sleep monitoring information acquisition module includes a wearable device installed on the user, which collects the user's sleep information, pushes a questionnaire to the user, obtains the user's vital sign activity information based on the user's questionnaire answers, and transmits the user's sleep information and vital sign activity information to the model calculation module.

[0082] The model calculation module is used to set the Basic Sleep Rating Standard (BSR) and the Optimized Sleep Rating Standard (ASR). Based on the BSR and ASR, iterative calculations are performed according to the received sleep information and vital sign activity information of the user to obtain the user's sleep score (SAV) value for the day. The user's sleep is rated according to the SAV value, and the user's SAV value and sleep rating result for the day are transmitted to the visualization module.

[0083] The visualization module is used to receive the user's SAV value and sleep rating results for the day from the model calculation module through the user's terminal device, and then present them to the user in a visual format.

[0084] The sleep monitoring information acquisition module includes: a wearable device equipped with sleep monitoring equipment and vital sign monitoring equipment;

[0085] The sleep monitoring device is used to collect sleep data from users through internal sensors. The sleep data includes the percentage of deep sleep, total sleep time, and continuity of deep sleep.

[0086] The aforementioned vital sign monitoring device is used to push questionnaires to users and obtain users' vital sign data based on the users' questionnaire answers. The vital sign data includes heart rate, daily steps, and self-report questionnaires.

[0087] The wearable device includes a smart bracelet. The internal sensors of the wearable monitoring device include, but are not limited to, heart rate sensors, gyroscopes, accelerometers, barometric pressure sensors, temperature sensors, ambient light sensors, capacitive touch sensors, battery status sensors, skin conduction sensors, and blood oxygen sensors. These sensors are used to collect personal data and to classify sleep stages. The percentage of deep sleep, total sleep time, and continuity of deep sleep are selected as the main indicators for sleep rating.

[0088] The BSR is a percentage-based evaluation standard, corresponding to a pair of arrays (x, y), including three levels of standards: excellent sleep, normal sleep, and poor sleep, corresponding to three score ranges: 85-100 points, 70-85 points, and below 70 points, respectively. The ASR is a percentage-based evaluation standard, representing the result after iterative updates of the array (x, y).

[0089] The model calculation module is used to receive raw data from the sleep monitoring module, classify, store, and process the raw data, apply mathematical models and optimization algorithms to calculate accurate adaptive sleep ratings, and implement self-learning iterative calculations. Specifically, it includes:

[0090] The Basic Sleep Value (SBV) calculation module is used to determine a user's SBV score based on the percentage of deep sleep, total sleep time, and deep sleep continuity. The SBV score includes 0.3, 0.5, and 0.7 points, corresponding to the three states of "full of energy," "normal state," and "poor state," respectively.

[0091] The Daily Activity Rating (DAV) calculation module is used to perform HRV (Heart Rate Variability) analysis based on vital sign data. It selects HRV analysis of daily activities, step count, and self-report questionnaires as the main indicators for daily activity rating. The self-report questionnaire is divided into three levels: "Full of Energy," "Normal," and "Poor," and users are required to submit a single score daily. The DAV value is calculated by linearly combining and weighting the three indicators—heart rate variability analysis, daily steps, and self-report questionnaire results—and summing them.

[0092] Formula 1

[0093] Among them, HRV is heart rate variability analysis, measured as a percentage; Step is daily steps, measured as steps; SR is self-reported questionnaire results, measured as three-level scores of 0.3, 0.5, and 0.7; a, b, and c are the corresponding indicator weights; and DAV is measured as a percentage score.

[0094] The SBV and DAV iterative optimization calculation modules are used to... The initial value of SBV is input into the sleep evaluation mathematical model. SBV is defined as the optimization parameter in the b-direction, and DAV is the optimization parameter in the w-direction. Mini-batch is defined as a small-batch iteration, where each iteration's dataset consists of all data within 30 days from the user's current date. The input data for parts 2-7 are SBV, DAV, and the initial value of SBV. The model iteratively updates and outputs daily ASR and SAV. In the t-th iteration, d is calculated on the current mini-batch. d :

[0095] Formula 2

[0096] Formula 3

[0097] Formula 4

[0098] Formula 5

[0099] Formula 6

[0100] Formula 7

[0101] Formula 2-3 is used to calculate the squared gradient. and These represent the squared gradient momentum accumulated by the loss function during the first t-1 iterations; This is the gradient accumulation exponent, typically taken as 0.9; =10 -8 Formulas 4-5 are for numerical update calculations. The learning rate is determined based on the specific circumstances; Formulas 6-7 are used to calculate the gradient. and represents the gradients of the latest SBV and DAV data, respectively; Loss is the loss function.

[0102] ASR and BSR are evaluation criteria, each corresponding to a pair of (x, y) arrays, where x and y represent the score thresholds in the evaluation criteria, corresponding to three score levels. The evaluation criteria are divided into: For excellent sleep, Normal sleep and y represents the three-level standard for poor sleep. For example, the initial BSR value is (x=85, y=70), where 85-100 points correspond to excellent sleep, 70-85 points correspond to normal sleep, and below 70 points correspond to poor sleep. ASR represents the result after iterative updates to the array (x, y). SBV is only used as an intermediate iteration term; it is the unstandardized result obtained from the previous day's SAV after new iteration calculation, and is a percentage score used for standardization conversion in Formula 10.

[0103] After iterative calculations using formulas 2-7, the optimized and updated SBV and DAV values ​​are output. These optimized and updated SBV and DAV values ​​are then used as input values ​​for formulas 8-10. The optimized ASR and loss function are calculated as follows:

[0104] Formula 8

[0105] Formula 9

[0106] Where d and e are the corresponding indicator weights; when calculating a new ASR each day, the evaluation standard for the previous day is BSR.

[0107] The Sleep Adaptive Rating (SAV) calculation module is used to calculate the Sleep Adaptive Rating (SAV) value based on the daily optimized and updated SBV and DAV values.

[0108] Formula 10

[0109] SAV value is the user's sleep score on a 100-point scale for that day;

[0110] Users are rated on their sleep based on their SAV (Sleep Activity Value), which includes excellent sleep, normal sleep, and poor sleep.

[0111] This invention provides a specific processing flow for an adaptive sleep evaluation method, as follows: Figure 3 As shown, the processing steps include the following:

[0112] Step S1: Place wearable devices, such as smart bracelets that include sleep monitoring devices and vital sign monitoring devices, on the user. The sleep monitoring device collects the user's sleep data through internal sensors. This sleep data includes the percentage of deep sleep, total sleep time, and deep sleep continuity. Based on the sleep data, a basic sleep rating is calculated according to the basic sleep rating standard (out of 100).

[0113] The SBV system sends three questions directly to users daily via a questionnaire, and users receive an SBV score based on their answers.

[0114] Step S2: Collect the user's vital signs data through the vital signs monitoring device. The vital signs data includes heart rate, daily steps and self-report questionnaire. Perform heart rate variability analysis based on the vital signs data, and perform linear combination weighting and summation on the three indicators of heart rate variability analysis, daily steps and self-report questionnaire results to calculate the user's DAV rating (three-level standard).

[0115] The calculation formula is:

[0116] Formula 1

[0117] Among them, HRV is the heart rate variability analysis, Step is the number of steps taken per day, SR is the self-reported questionnaire results, and a, b, and c are the corresponding indicator weights.

[0118] Step S3: Rate daily activity The initial values ​​of the Basic Sleep Rating (SBV) are input into the sleep evaluation mathematical model. Combined with the ASR and optimization algorithms, such as RMSprop, the sleep evaluation parameters are optimized and iterated. The ASR parameters are updated / iterated to obtain the optimized ASR.

[0119] The aforementioned sleep evaluation mathematical model is a gradient descent model in deep learning, selecting input terms and iterative update terms as basic model parameters. Input terms include: ASR, SBV, ... The cyclically updated items include: ASR and SAV (Sleep Adapted Value).

[0120] Define SBV as the optimization parameter in the b-direction and DAV as the optimization parameter in the w-direction. Define mini-batch as the small-batch iteration, i.e., the size of the dataset for each iteration, and specify that the dataset for each iteration consists of all data within 30 days from the user's current date. The input data for Equations 2-7 are SBV, DAV, and the initial value SBV. After iteration, the updated SBV and DAV values ​​are output, which are used in Equations 8-10 to iteratively update and output the daily ASR and SAV. In the t-th iteration, d is calculated on the current mini-batch. d :

[0121] Formula 2

[0122] Formula 3

[0123] Formula 4

[0124] Formula 5

[0125] Formula 6

[0126] Formula 7

[0127] Formula 2-3 is used to calculate the squared gradient. and These represent the squared gradient momentum accumulated by the loss function during the first t-1 iterations; This is the gradient accumulation exponent, typically taken as 0.9; =10 -8 Formulas 4-5 are for numerical update calculations. The learning rate is determined based on the specific circumstances; Formulas 6-7 are used to calculate the gradient. and represents the gradients of the latest SBV and DAV data, respectively; Loss is the loss function.

[0128] The optimized ASR and loss function are calculated as follows:

[0129] Formula 8

[0130] Formula 9

[0131] Where d and e are the corresponding indicator weights; when calculating a new ASR each day, the evaluation standard for the previous day is BSR.

[0132] Step S4: Calculate SAV based on the optimized SBV, DAV, and Basic Sleep Rating Scale (BSR).

[0133] Formula 10

[0134] SAV is the final output value, which is the user's sleep score (out of 100) for that day.

[0135] Step 5: Classify sleep based on SAV values, corresponding to excellent sleep, normal sleep, and poor sleep, and push the results to the user's terminal.

[0136] A sleep rating self-learning algorithm is developed using a mathematical model for sleep evaluation to calculate and converge evaluation criteria. This algorithm is then repeated iteratively multiple times to evaluate SBV, DAV, BSR, and other metrics. By optimizing parameters and continuously correcting the sleep quality evaluation method, the rating results of the sleep evaluation mathematical model gradually approach the actual sleep situation of individuals.

[0137] In summary, this invention combines daily activity level and basic sleep monitoring data to provide an adaptive sleep evaluation method, including a mathematical model and algorithmic logic for sleep evaluation. It continuously adjusts the sleep evaluation standards based on the actual sleep situation of individuals, thereby calculating evaluation results with individual differences. This makes the evaluation results more scientific, accurate, and realistic, and can accurately reflect the sleep situation of people in abnormal environments or with unstable sleep environments.

[0138] This invention aims to improve the accuracy of sleep monitoring by combining an evolutionary algorithm to optimize the mathematical model of sleep rating standards. It accurately and adaptively monitors an individual's sleep state, evaluates their individual sleep level, and allows them to better understand their sleep patterns and health status. The method achieves self-learning, adaptive, and personalized sleep rating, and can be applied to any sleep monitoring device, improving the accuracy of sleep monitoring and evaluation.

[0139] Those skilled in the art will understand that the accompanying drawings are merely schematic diagrams of one embodiment, and the modules or processes shown in the drawings are not necessarily essential for implementing the present invention.

[0140] As can be seen from the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments of the present invention.

[0141] The various embodiments in this specification are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, for apparatus or system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and relevant parts can be referred to the description of the method embodiments. The apparatus and system embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0142] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. An adaptive sleep assessment device, characterized in that, include: The module includes a sleep monitoring information acquisition module, a model calculation module, and a visualization module. The sleep monitoring information acquisition module includes a wearable device installed on the user, which collects the user's sleep information, pushes a questionnaire to the user, obtains the user's vital sign activity information based on the user's questionnaire answers, and transmits the user's sleep information and vital sign activity information to the model calculation module. The model calculation module is used to set the Basic Sleep Rating Standard (BSR) and the Optimized Sleep Rating Standard (ASR). The BSR is a percentage-based evaluation standard, corresponding to a pair of arrays (x, y), where x and y represent the score thresholds in the evaluation standard. The ASR is also a percentage-based evaluation standard, representing the result of iteratively updating the BSR array (x, y). Based on the BSR and ASR, iterative calculations are performed according to the received sleep information and vital sign activity information of the user to obtain the user's SAV value for the day. The user's sleep is rated based on the SAV value, and the user's SAV value and sleep rating result for the day are transmitted to the visualization module. The visualization module is used to display the user's SAV value and sleep rating results for the day on the user's terminal.

2. The apparatus according to claim 1, characterized in that, The sleep monitoring information acquisition module includes: a wearable device equipped with sleep monitoring equipment and vital sign monitoring equipment; The sleep monitoring device is used to collect sleep data from users through internal sensors. The sleep data includes the percentage of deep sleep, total sleep time, and continuity of deep sleep. The aforementioned vital sign monitoring device is used to push questionnaires to users and obtain users' vital sign data based on the users' questionnaire answers. The vital sign data includes heart rate, daily steps, and self-report questionnaires.

3. The apparatus according to claim 1, characterized in that, The wearable device includes a smart bracelet, and the internal sensors of the wearable monitoring device include a heart rate sensor, a gyroscope, an accelerometer, a barometric pressure sensor, a temperature sensor, an ambient light sensor, a capacitive touch sensor, a battery status sensor, a skin conduction sensor, and a blood oxygen sensor.

4. The apparatus according to claim 1, 2, or 3, characterized in that, The thresholds correspond to three score levels, with the evaluation criteria divided into: For excellent sleep, Normal sleep and y represents the third level of poor sleep.

5. The apparatus according to claim 4, characterized in that, The model calculation module includes: The Basic Sleep Value (SBV) calculation module is used to determine a user's SBV score based on the percentage of deep sleep, total sleep time, and deep sleep continuity. The SBV score includes 0.3, 0.5, and 0.7 points, corresponding to three states: "full of energy," "normal state," and "poor state," respectively. The Daily Activity Rating (DAV) calculation module is used to perform heart rate variability analysis based on vital sign data. It performs linear combination weighting and summation on three indicators: heart rate variability analysis, daily steps, and self-reported questionnaire results, to calculate the user's DAV value. Official 1 Among them, HRV is heart rate variability analysis, measured as a percentage; Step is daily steps, measured as steps; SR is self-reported questionnaire results, measured as three-level scores of 0.3, 0.5, and 0.7; a, b, and c are the corresponding indicator weights; and DAV is measured as a percentage score. The SBV and DAV iterative optimization calculation modules are used to... The initial value of SBV is input into the sleep evaluation mathematical model. SBV is defined as the optimization parameter in the b-direction, and DAV is the optimization parameter in the w-direction. Mini-batch is defined as a small-batch iteration, where each iteration's dataset consists of all data within 30 days from the user's current date. The input data for parts 2-7 are SBV, DAV, and the initial value of SBV. The model iteratively updates and outputs daily ASR and SAV. In the t-th iteration, d is calculated on the current mini-batch. d : Official 2 Official 3 Official 4 Official 5 Official 6 Official 7 Formula 2-3 is used to calculate the squared gradient. and These represent the squared gradient momentum accumulated by the loss function during the first t-1 iterations; This is the gradient accumulation exponent, typically taken as 0.9; =10 -8 Formulas 4-5 are for numerical update calculations. The learning rate is determined based on the specific circumstances; Formulas 6-7 are used to calculate the gradient. and These represent the gradients of the latest SBV and DAV data, respectively; Loss is the loss function. After iterative calculations using formulas 2-7, the optimized and updated SBV and DAV values ​​are output. These optimized and updated SBV and DAV values ​​are then used as input values ​​for formulas 8-10. The optimized ASR and loss function are calculated as follows: Official 8 Official 9 Where d and e are the corresponding indicator weights; when calculating a new ASR each day, the evaluation standard for the previous day is BSR. The Sleep Adaptive Rating (SAV) calculation module is used to calculate the Sleep Adaptive Rating (SAV) value based on the daily optimized and updated SBV and DAV values. Official 10 SAV value is the user's sleep score on a 100-point scale for that day; Users are rated on their sleep based on their SAV (Sleep Activity Value), which includes excellent sleep, normal sleep, and poor sleep.

6. An adaptive sleep assessment method, characterized in that, include: The sleep monitoring device collects the user's sleep information by placing a wearable device on the user, pushes a questionnaire to the user, and obtains the user's vital signs activity information based on the user's questionnaire answers. A Basic Sleep Rating Standard (BSR) and an Optimized Sleep Rating Standard (ASR) are defined. The BSR is a percentage-based evaluation standard, corresponding to a pair of arrays (x, y), where x and y represent the threshold scores in the evaluation standard. The ASR is also a percentage-based evaluation standard, representing the result of iteratively updating the BSR array (x, y). Based on the defined BSR and ASR, iterative calculations are performed according to the user's sleep information and vital sign activity information to obtain the user's daily SAV value. The user's sleep is then rated based on the SAV value. The user's SAV value and sleep rating results for the day are displayed on the user's terminal.

7. The method according to claim 6, characterized in that, The aforementioned method involves collecting sleep information from a wearable device worn by the user. The sleep monitoring device sends a questionnaire to the user and obtains the user's vital sign activity information based on the user's responses, including: The wearable device worn by the user is equipped with sleep monitoring and vital sign monitoring devices. The sleep monitoring device collects the user's sleep data through internal sensors, including the percentage of deep sleep, total sleep time, and continuity of deep sleep. The vital signs monitoring device sends a questionnaire to the user and obtains the user's vital signs data based on the user's questionnaire answers. The vital signs data includes heart rate, daily steps, and self-report questionnaires.

8. The method according to claim 7, characterized in that, The aforementioned sleep rating based on the set BSR and ASR is calculated iteratively according to the user's sleep information and vital sign activity information to obtain the user's SAV value for the day, and the user's sleep is rated based on the SAV value, including: The user's SBV score is determined based on the percentage of deep sleep, total sleep time, and deep sleep continuity. Heart rate variability analysis was performed based on vital signs data. The three indicators of heart rate variability analysis, daily steps and self-report questionnaire results were linearly combined, weighted and summed to calculate the user's DAV value. Official 1 Among them, HRV is heart rate variability analysis, measured as a percentage; Step is daily steps, measured as steps; SR is self-reported questionnaire results, measured as three-level scores of 0.3, 0.5, and 0.7; a, b, and c are the corresponding indicator weights; and DAV is measured as a percentage score. Will The initial value of SBV is input into the sleep evaluation mathematical model. SBV is defined as the optimization parameter in the b-direction, and DAV is the optimization parameter in the w-direction. Mini-batch is defined as a small-batch iteration, where each iteration's dataset consists of all data within 30 days from the user's current date. The input data for parts 2-7 are SBV, DAV, and the initial value of SBV. The model iteratively updates and outputs daily ASR and SAV. In the t-th iteration, d is calculated on the current mini-batch. d : Official 2 Official 3 Official 4 Official 5 Official 6 Official 7 Formula 2-3 is used to calculate the squared gradient. and These represent the squared gradient momentum accumulated by the loss function during the first t-1 iterations; This is the gradient accumulation exponent, typically taken as 0.9; =10 -8 Formulas 4-5 are for numerical update calculations. The learning rate is determined based on the specific circumstances; Formulas 6-7 are used to calculate the gradient. and These represent the gradients of the latest SBV and DAV data, respectively; Loss is the loss function. After iterative calculations using formulas 2-7, the optimized and updated SBV and DAV values ​​are output. These optimized and updated SBV and DAV values ​​are then used as input values ​​for formulas 8-10. The optimized ASR and loss function are calculated as follows: Official 8 Official 9 Where d and e are the corresponding indicator weights; when calculating a new ASR each day, the evaluation standard for the previous day is BSR. The SAV value is calculated based on the daily optimized and updated SBV and DAV values; Official 10 SAV value is the user's sleep score on a 100-point scale for that day; Users are rated on their sleep based on their SAV (Sleep Activity Value), which includes excellent sleep, normal sleep, and poor sleep.

Images