A sleep improvement method based on physiological data

Abstract

The application discloses a sleep improvement method based on physiological data and belongs to the technical field of massage mattresses, and comprises the following steps: in step S00, a sensor arranged on a mattress is used to collect physiological data of a human body, including heart rate, breathing rate, body movement and snoring sound; in step S10, a prediction and evaluation model is established based on the physiological data; in the prediction and evaluation model, sleep states are divided into a wake stage, a REM stage, a light sleep stage and a deep sleep stage based on the physiological data, and a sleep quality evaluation model is established; and in step S20, a massage improvement scheme is formulated based on sleep quality obtained through evaluation, a massage air bag is arranged in the mattress body, a proportional adjusting valve is arranged at an air inlet end of the massage air bag, and when snoring sound or apnea is detected, the massage air bag is started to adjust a human body posture so as to ensure smooth breathing, sleep quality is improved after massage, physiological data before and after massage is collected, and massage effect is evaluated through comparison.

Description

Technical Field

[0001] This invention relates to the field of massage mattress technology, and more particularly to a method for improving sleep based on physiological data. Background Technology

[0002] Sleep occupies roughly one-third of a person's life, and its quality is crucial to health. Sleep quality assessment is the basis for diagnosing and treating sleep-related disorders. Current sleep quality assessments are mainly divided into subjective and objective aspects. Subjective assessments rely heavily on various scales; objective assessments analyze various physiological, biochemical, and behavioral indicators related to sleep using instruments and equipment. Subjective assessments require manual interpretation and visual analysis by sleep experts, which is inefficient and prone to misjudgment. Collecting physiological data during sleep and using algorithmic models for sleep staging is an effective and objective method for automatically assessing sleep quality using modern signal processing technology. Regarding sleep monitoring and data collection: PSG is an authoritative monitoring method that can simultaneously record multiple physiological data points throughout the night's sleep; however, the monitoring process is complex, requiring multiple monitoring patches to be attached to the body. Smart bracelets use built-in accelerometers, optical sensors, speakers / microphones, and other devices to record signals such as heart rate vibrations, displacement changes, light intensity changes, and sound changes throughout the night, generating reports.

[0003] Existing mattresses do not significantly improve sleep. Summary of the Invention

[0004] To address the aforementioned problems, this invention aims to solve the problems described above. One object of this invention is to provide a sleep improvement method based on physiological data to solve the above problems.

[0005] The solution adopted in this embodiment is a sleep improvement method based on physiological data, which includes the following steps: Step S00: Collect several physiological data of human body, such as heart rate, respiratory rate, body movement and snoring, through sensors set on the mattress; Step S10: Based on the above physiological data, establish a predictive assessment model. The predictive assessment model divides sleep state into wakefulness stage, REM stage, light sleep stage, and deep sleep stage based on physiological data, and establishes a sleep quality assessment model. Step S20: Based on the assessed sleep quality, a massage improvement plan is developed. A massage airbag is built into the mattress body. The air inlet of the massage airbag is equipped with a proportional adjustment valve. When snoring is detected, the massage airbag is activated to adjust the body posture and improve sleep quality through massage.

[0006] A preferred technical solution is that, in step S00, the sensors installed on the mattress include a piezoelectric film sensor and a piezoelectric electrocardiogram and fabric sensor. The piezoelectric film sensor is integrated into the mattress body to use the tiny electrical signals generated by the sound wave vibration acting on the piezoelectric film to sense snoring. By integrating the piezoelectric electrocardiogram and fabric sensor array, multimodal physiological information of human sleeping posture and sleep duration is captured.

[0007] The preferred technical solution is that, in step S10, the body movement signals detected by the sensor are classified as follows: when there are many body movements and they are concentrated in large movements, it is determined to be the awake stage; when there are few body movements but they are mainly large movements, it is determined to be the REM stage; when large movements decrease and they are concentrated in small body twitches, it is determined to be the light sleep stage; and when there are only slight small movements, it is determined to be the deep sleep stage.

[0008] Furthermore, in step S10, the sleep state is further corrected by combining other physiological data. The length of the heart rate and respiratory rate cycle is used to determine whether the state is sleep or wakefulness, and the transition process of sleep stages is determined based on the heart rate change cycle, so as to determine the current sleep state.

[0009] An alternative approach is to use a hybrid method of Hidden Markov Model (HMM) and Backpropagation Neural Network (BPNN) to stage sleep based on the acquired physiological information in step S10. The HMM is used to model the heart rate and respiratory rate signals, train the HMM for each sleep state, perform the initial sleep stage calculation, and then use the discrimination ability of the BPNN to train the memory of the HMM stage calculation results.

[0010] Furthermore, during model training, a neural network model is used to memorize the incorrect matching results and correct labeling results of the HMM model. When new sleep data is input, the BP neural network corrects the calculation results of the HMM stage and maps them to the correct sleep phases to improve the accuracy of sleep stage calculation.

[0011] The preferred technical solution is that, in step S10, a sleep quality assessment model is established, and for each sleep stage, the total duration H of that stage is calculated. x Add up the duration of each stage to get the total sleep time h; The adjustment coefficients for the waking stage, REM stage, light sleep stage, and deep sleep stage are set as a, b, c, and d, respectively. The sleep score is calculated using the formula (aH1+bH2+cH3+dH4) / h, and the massage plan is adjusted based on H1, H2, H3, H4 and the sleep score.

[0012] The preferred technical solution is that, in step S20, when adjusting the intake air volume, the proportional control valve multiplies the adjustment by the corresponding adjustment coefficients a, b, c, and d based on the detected human body's awake stage, REM stage, light sleep stage, and deep sleep stage.

[0013] The preferred technical solution is that, in step S21, physiological data before and after massage are collected, the massage plan is compared and analyzed to improve the quality of massage, and the space for further improvement is analyzed based on theoretical models. Adjustment plans are made for different massage areas, massage time, massage techniques and massage intensity.

[0014] The preferred technical solution is that the massage airbag in step S20 is provided with multiple layers of airbags, which are stacked sequentially from top to bottom, and the multiple airbags of each massage airbag are interconnected.

[0015] The sleep improvement method based on physiological data of this invention has the following technical effects: 1. This application discloses a sleep improvement method based on physiological data, comprising the following steps: Step S00: Collecting physiological data such as heart rate, respiratory rate, body movement, and snoring sound from sensors installed on the mattress; Step S10: Establishing a predictive evaluation model based on the aforementioned physiological data. The predictive evaluation model divides sleep states into wakefulness, REM sleep, light sleep, and deep sleep stages based on the physiological data, thus establishing a sleep quality evaluation model; Step S20: Developing a massage improvement plan based on the evaluated sleep quality. A massage airbag is built into the mattress body, and the air inlet of the massage airbag is equipped with a proportional adjustment valve. When snoring is detected, the massage airbag is activated to adjust the body's posture, thereby improving sleep quality through massage. By detecting the body's sleep state, different massages are performed in different states, thereby improving sleep.

[0016] 2. The air inlet of the massage airbag is equipped with a proportional adjustment valve. When adjusting the air intake, the valve is multiplied by the corresponding adjustment coefficients a, b, c, and d, based on the detected stage of the human body: wakefulness, REM sleep, light sleep, and deep sleep. This allows for different air intake adjustments at different stages, preventing the massage from disrupting the user's sleep. 3. The massage airbag in step S20 has multiple layers of airbags stacked sequentially from top to bottom, with the multiple airbags of each massage airbag interconnected. By synchronously releasing gas through multiple airbags, the incoming gas is dispersed and released, resulting in a gentle massage motion. Furthermore, the multiple layers of airbags allow for various massage techniques, enhancing the massage effect.

[0017] Other features and advantages of the invention will become clear when reading the following description of exemplary embodiments with reference to the accompanying drawings. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and, together with the description, serve to explain the principles of the invention. In these drawings, similar reference numerals are used to denote similar elements. The drawings described below are some embodiments of the invention, but not all embodiments. Other drawings will be readily available to those skilled in the art based on these drawings without any inventive effort.

[0019] Figure 1 This is a schematic diagram of the overall structure of a bionic finger smart massage mattress provided in a specific embodiment of the present invention; Figure 2 This is a schematic diagram of model training provided in a specific embodiment of the present invention; Figure 3 This is a schematic diagram of the distribution of massage airbags in a mattress according to a specific embodiment of the present invention; In the picture: 1. Mattress body; 2. Sensor; 3. Massage airbag; 31. Bag body. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0021] Definition: HMM stands for Hidden Markov Model.

[0022] The following description, in conjunction with the accompanying drawings and embodiments, details the bionic finger smart massage mattress.

[0023] like Figure 1-3As shown, the present invention provides a sleep improvement method based on physiological data, comprising the following steps: Step S00: Collecting several physiological data of the human body, including heart rate, respiratory rate, body movement, and snoring, through sensors 2 installed on the mattress; the sensors 2 installed on the mattress include a piezoelectric film sensor and a piezoelectric ECG and fabric sensor. The piezoelectric film sensor is integrated into the mattress body 1 to detect snoring by using the tiny electrical signals generated by the vibration of sound waves acting on the piezoelectric film; by integrating the piezoelectric ECG and fabric sensor array, capturing multimodal physiological information of the human body's sleeping posture and sleep duration. Alternatively, human physiological data can be directly collected through a fitness tracker and transmitted to the control terminal of the massage mattress. The collection of human physiological data is mainly for monitoring human sleep quality and detecting the human body's sleep state, which facilitates targeted adjustments when regulating human sleep quality and ensures effective improvement of sleep quality.

[0024] Step S10: Based on the above physiological data, establish a predictive evaluation model. The predictive evaluation model divides the sleep state into the wakefulness stage, REM stage, light sleep stage, and deep sleep stage based on the physiological data, and establishes a sleep quality evaluation model. The human body responds differently to the external environment in different sleep stages. In order to avoid the massage airbag 3 affecting sleep when it is working, it is necessary to monitor and distinguish the human sleep state and use different feedback regulation mechanisms for different stages in order to better improve sleep quality.

[0025] Step S20: Based on the assessed sleep quality, a massage improvement plan is developed. A massage airbag 3 is built into the mattress body 1. The airbag 3 has a proportional adjustment valve at its air inlet. When snoring is detected, the massage airbag 3 is activated to adjust the body's posture, improving sleep quality through massage. The proportional adjustment valve allows for more precise control of the air intake, enabling extremely small air intake adjustments. Even after the user falls asleep, the massage airbag 3 can be adjusted very gently through inflation without waking the user. When snoring is detected, the massage airbag 3 lifts the neck, reducing pressure on the trachea and preventing snoring, rather than directly pushing the user to change their sleep state, resulting in better sleep quality. Besides neck elevation to clear the airway, the inflation and deflation of the massage airbag 3 on one side of the body also helps change the sleeping position, such as turning from supine to side-lying, thus improving snoring. In case of sleep apnea: the massage airbag 3 is raised to clear the airway, and then a buzzer alarm is activated. The buzzer alarm is located at the neck position.

[0026] In order to accurately determine the sleep stage of the human body, in step S10, the body movement signals detected by sensor 2 are used to determine the awake stage when there are many body movements and they are concentrated in large movements; the REM stage when there are few body movements but they are mainly large movements; the light sleep stage when large movements decrease and they are concentrated in small body twitches; and the deep sleep stage when there are only slight small movements.

[0027] Furthermore, in step S10, the sleep state is further corrected by combining other physiological data. The length of the heart rate and respiratory rate cycle is used to determine whether the person is asleep or awake, and the transition process of the sleep stage is determined based on the heart rate change cycle to determine the current sleep state. By combining body movement and the cycles of heart rate and respiratory rate to comprehensively determine the sleep stage, the sleep stage can be determined more accurately. After accurate determination, a more precise adjustment plan for the massage airbag 3 can be provided to ensure effective improvement of the user's sleep quality.

[0028] In terms of data processing, in step S10, the acquired physiological information is used to perform sleep staging using a hybrid algorithm of Hidden Markov Model and Backpropagation Neural Network. The Hidden Markov Model is used to model the heart rate signal and respiratory rate signal, and the HMM for each sleep state is trained to perform the initial sleep staging calculation. Then, the discrimination ability of the Backpropagation Neural Network is used to train the memory of the HMM staging calculation results.

[0029] Furthermore, during model training, a neural network model is used to memorize and train the incorrect matching and correct labeling results of the HMM model. When new sleep data is input, the BP neural network corrects the calculation results of the HMM stage, mapping them to the correct sleep phases to improve the accuracy of sleep stage calculation. By using a hybrid algorithm combining Hidden Markov Model and BP neural network, the accuracy of human sleep stage determination can be greatly guaranteed.

[0030] To better adjust the massage based on sleep patterns, in step S10, a sleep quality assessment model is established, and the total duration H of each sleep stage is calculated. x The duration of each stage is summed to obtain the total sleep time h. Adjustment coefficients a, b, c, and d are set for the waking stage, REM sleep stage, light sleep stage, and deep sleep stage, respectively. The sleep score is calculated using the formula (aH1 + bH2 + cH3 + dH4) / h. The massage plan is then adjusted based on H1, H2, H3, H4, and the overall sleep score. Where a = 8, b = 17, c = 36, and d = 39; when the sleep durations of the four stages are 2 hours, 1 hour, 5 hours, and 1 hour, respectively, the sleep score is 28. Then, a suitable massage plan is specified based on the duration of each stage and the overall sleep score.

[0031] Furthermore, in step S20, when adjusting the air intake, the proportional control valve multiplies the adjustment by corresponding adjustment coefficients a, b, c, and d based on the detected stage of wakefulness, REM sleep, light sleep, and deep sleep. The setting of these adjustment coefficients allows for more targeted adjustments by the proportional control valve. Applying the same massage technique at different stages will produce different effects. During the REM sleep stage, a gentler massage is used because the body is easily startled, and sudden changes in the massager's intensity can easily wake the user. This solution uses a proportional control valve to maintain a stable flow rate, and the coefficient adjustment makes the massage gentler, allowing the user to more easily enter deep sleep.

[0032] To facilitate user adjustment and control of the massager, a mobile app is connected to the massage mattress. In step S21, physiological data before and after the massage is collected, and the improvement effect of the massage program on massage quality is compared and analyzed. Based on a theoretical model, potential for further improvement is analyzed, and adjustments are made to the massage plan for different massage areas, massage time, massage techniques, and massage intensity. The data before and after the massage can be clearly viewed on the mobile phone, and the massage effect can be evaluated by comparing changes in physiological data. Users can also select massage methods and durations through the app, or directly use the system's recommended massage methods.

[0033] Preferably, the massage airbag 3 in step S20 is provided with multiple layers of airbags 31, which are stacked sequentially from top to bottom, and the multiple airbags 31 of each massage airbag 3 are interconnected. By releasing gas through multiple airbags 31 simultaneously, the gas is dispersed and released, the massage action is soothing, and the multiple layers of airbags 31 can realize a variety of massage techniques, resulting in a better massage effect. The multiple layers of airbags 31 can realize multi-level massage.

[0034] The massage mattress also has a wake-up function. When the preset wake-up time arrives, the mattress will activate the massage airbags 3 to gently massage the body, thus gently waking the body. This wake-up mode will not produce noise that disturbs others.

[0035] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that an article or apparatus comprising a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such an article or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of other identical elements in the article or apparatus that includes that element.

[0036] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. The present invention has been described in detail with reference to preferred embodiments. Those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the present invention, and all such modifications and substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A sleep improvement method based on physiological data, characterized in that: Includes the following steps: Step S00: Collect several physiological data of the human body, such as heart rate, respiratory rate, body movement, and snoring, through sensors installed on the mattress; Step S10: Based on the above physiological data, establish a predictive evaluation model. The predictive evaluation model divides sleep state into wakefulness stage, REM stage, light sleep stage and deep sleep stage based on physiological data, and establishes a sleep quality evaluation model. Step S20: Based on the sleep quality assessment, a massage improvement plan is developed. A massage airbag is built into the mattress body. The air inlet of the massage airbag is equipped with a proportional adjustment valve. When snoring is detected, the massage airbag is activated to adjust the body posture and improve sleep quality through massage. In step S10, a sleep quality assessment model is established, and for each sleep stage, the total duration H of that stage is calculated. x Add up the duration of each stage to get the total sleep time h; The adjustment coefficients for the waking stage, REM stage, light sleep stage, and deep sleep stage are set as a, b, c, and d, respectively. The sleep score is calculated using the formula (aH1+bH2+cH3+dH4) / h, and the massage plan is adjusted based on H1, H2, H3, H4 and the sleep score. In step S20, when adjusting the intake air volume, the proportional control valve multiplies the adjustment by the corresponding adjustment coefficients a, b, c, and d based on the detected human body's wakefulness stage, REM stage, light sleep stage, and deep sleep stage. Step S21: Collect physiological data before and after massage, compare and analyze the effect of massage plan on improving massage quality, analyze the room for further improvement based on theoretical model, and adjust the massage plan for different massage areas, massage time, massage techniques and massage intensity.

2. The sleep improvement method based on physiological data as described in claim 1, characterized in that: In step S00, the sensors installed on the mattress include a piezoelectric film sensor and a piezoelectric ECG and fabric sensor. The piezoelectric film sensor is integrated into the mattress body to detect snoring by using the tiny electrical signals generated by the vibration of sound waves acting on the piezoelectric film. By integrating the piezoelectric ECG and fabric sensor array, multimodal physiological information of human sleeping posture and sleep duration is captured.

3. The sleep improvement method based on physiological data as described in claim 1, characterized in that: In step S10, the body movement signals detected by the sensor are classified as follows: when there are many body movements and they are concentrated in large movements, it is determined to be the awake stage; when there are few body movements but they are mainly large movements, it is determined to be the REM stage; when large movements decrease and they are concentrated in small body twitches, it is determined to be the light sleep stage; and when there are only slight small movements, it is determined to be the deep sleep stage.

4. The sleep improvement method based on physiological data as described in claim 3, characterized in that: In step S10, the sleep state is further corrected by combining other physiological data. The length of the heart rate and respiratory rate cycle is used to determine whether the state is sleep or wakefulness. The transition process of sleep stages is determined based on the heart rate change cycle to determine the current sleep state.

5. The sleep improvement method based on physiological data as described in claim 1, characterized in that: In step S10, the acquired physiological information is used to perform sleep staging using a hybrid algorithm of Hidden Markov Model and Backpropagation Neural Network. The Hidden Markov Model is used to model the heart rate and respiratory rate signals, and the HMM for each sleep state is trained to perform the initial sleep staging calculation. Then, the discrimination ability of the Backpropagation Neural Network is used to train the memory of the HMM staging calculation results.

6. The sleep improvement method based on physiological data as described in claim 5, characterized in that: During model training, a neural network model is used to memorize the incorrect matching results and correct labeling results of the HMM model. When new sleep data is input, the BP neural network corrects the calculation results of the HMM stage and maps them to the correct sleep phase to improve the accuracy of sleep stage calculation.

7. The sleep improvement method based on physiological data as described in claim 1, characterized in that: The massage airbag in step S20 is provided with multiple layers of airbags, which are stacked sequentially from top to bottom, and the multiple airbags of each massage airbag are interconnected.

Images