Sleep assistance mattress

[0044] A detailed description will be given below in conjunction with the accompanying drawings.

[0045] figure 1 A schematic diagram of the functional modules of the mattress of the present invention is shown. like figure 1 As shown, the functional modules of the mattress of the present invention include a power supply module 101 on the mattress, a data acquisition unit 102, a single-chip processing unit 103, a data transmission unit 104, an alarm unit 105, a display module 106 and a data storage unit 109. The mattress function module also includes a remote mobile terminal 107 and a server/cloud platform 108 .

[0046] The data collection unit 102 includes a pressure sensor 102a for collecting pressure data, a humidity sensor 102b for collecting humidity data, a temperature sensor 102c for collecting temperature data, a heart sound sensor 102d for collecting heart sound data, and a heart sound sensor 102d for collecting heart sound data. Amplifying circuit 102e and A/D converting circuit 102f for signal amplification processing. The server/cloud platform 108 includes a data preprocessing module 108a, a scheme database 108b and a comprehensive data processing module 108c. The scheme database 108b has data processing schemes for different groups of people, such as data processing schemes for children, adults and the elderly, and treatment schemes for patients.

[0047] The surface of the mattress is divided into a plurality of regions according to a rectangular grid, and each region has the same rectangular shape. For example, the mattress can be divided into a region with nine rectangular structures in the horizontal direction and a region with twelve rectangular structures in the vertical direction. Each area of ​​the mattress is provided with a pressure sensor 102a, a humidity sensor 102b, a temperature sensor 102c and a heart sound sensor 102d. At the same time, each area has an adjustment mechanism for adjusting the hardness of the area through inflation or other means.

[0048] The power supply module 101 is respectively connected with the data acquisition unit 102 and the single-chip processing unit 103, and is used to realize power supply for each functional module located on the mattress. The data acquisition unit 102 is connected with a single chip processing unit 103 . The data collection unit 102 is used to realize the data collection of one layer of data including pressure data, humidity data, temperature data and heart sound data.

[0049] The single-chip processing unit 103 is used to perform preliminary processing including classification and storage on the collected first-tier data, and form second-tier data. The single-chip processing unit 103 classifies and stores the one-layer data received by different channels, and links the category information as tag information to the sensory data records corresponding to each sensor in the one-layer data, and will contain the data corresponding to each sensor. The sensing data record and the tag information linked to the sensing data record corresponding to each sensor in the first-tier data are regarded as the second-tier data.

[0050] The single-chip processing unit 103 is connected to a data storage unit 109 for storing data related to the single-chip processing unit 103 . The single-chip processing unit 103 is also connected with the alarm unit 105 located on the mattress. When the server/cloud platform 108 detects the abnormal situation of the human body signal, it will trigger the alarm unit 105, and the alarm unit 105 will send an alarm prompt to the display module 106 in addition to sending out warning sounds and buzzers, and the server/cloud platform 108 will report to the police simultaneously. The message is sent to at least one associated object of the user.

[0051] The process of the server/cloud platform 108 sending the alarm information to at least one associated object includes sending the alarm information to at least one associated object that is geographically associated and/or logically associated with the user, and according to the The association strength values ​​of are sent to at least one association object in descending order.

[0052] Wherein, the at least one associated object geographically associated with the user includes at least one associated object whose physical distance from the user is less than or equal to a preset threshold. Wherein, at least one associated object logically associated with the user includes at least one associated object having a kinship relationship, an affiliation relationship, and a rescue relationship with the user. The affiliation relationship includes that the user belongs to a certain nursing institution, a certain community and/or is managed by a certain property company, and belongs to a certain street office, etc. The rescue relationship includes the rescue relationship between the user and a certain hospital and/or a certain medical institution.

[0053] At the same time, the impact strength and/or correlation strength of the alarm information on the associated objects is also determined for different correlation strength values, and the associated objects with stronger correlations respond first.

[0054] Wherein, the at least one associated object geographically associated with the user includes at least one associated object whose physical distance from the user is less than or equal to a preset threshold. For example, the threshold may be set as 100m, and within the threshold distance range, associated objects closer to the user have larger associated values. For example, when the distance to the user is less than or equal to 1m, the correlation value is 100; when the distance to the user is greater than 1m and less than or equal to two meters, the correlation value is 95; Small.

[0055] Wherein, at least one associated object logically associated with the user includes at least one associated object having a kinship relationship, an affiliation relationship, and a rescue relationship with the user. For the logical association value, when the alarm data is temperature, humidity and/or pressure information, the association object belonging to the affiliation relationship has a larger association value, the association object related to the kinship relationship has a smaller association value, and the association object related to the rescue relationship has a smaller association value. Associated objects have a minimum associated value. When the alarm data is heart sound information, the associated object related to rescue has a larger associated value, and the associated object related to membership has a smaller associated value. According to a preferred embodiment, the feedback information sent to the associated objects is preferentially sent to at least one associated object related to the logical association.

[0056] According to a preferred embodiment, the alarm unit 105 also includes a camera. The camera device is used for monitoring the abnormal state and leaving the bed. When the user is in an abnormal state, the server/cloud platform 108 will receive an alarm signal from the mattress, and will simultaneously determine the level of the abnormal state. When the user is in an emergency state, the The associated object set by the user will be contacted, and the associated object can remotely open the camera device. When the data of the temperature sensor 102c and the heart sound sensor 102d are all reset to zero for a certain period of time, for example, 10 seconds. The camera will automatically turn on for identification. If no human body is found, it will be determined that the user has left the bed. Otherwise, it will be determined that the user is in an emergency, such as respiratory arrest, cardiac arrest, etc.

[0057] Abnormal state level judgment: When the user sends out an abnormal state, an alarm will be issued for different levels of abnormal state and combined with the physiological information provided by the user. For example, the levels of the abnormal state include: poor sleep, disease prevention, need for rescue, and the like. When the user just flips over, it can be determined that the user is in good health but not sleeping well. When the user turns around, it can be determined that the user needs to observe, and the server/cloud platform 108 will send information to the associated object of the kinship or affiliation set by the user, and the associated object can remotely open the camera on the mobile terminal 107 or the server/cloud platform 108 The device realizes video observation. When the user needs rescue, for example, the temperature sensor 102c and the heart sound sensor 102d data suddenly return to zero, and the camera is turned on to determine that the user is still in bed, the server/cloud platform 108 will send a command message to contact the associated object, and the associated object can be on the mobile terminal. Or the cloud platform can remotely turn on the camera for status confirmation and realize user rescue.

[0058] The single-chip processing unit 103 is connected to the mobile terminal 107 via the data transmission unit 104, so as to transmit the layer-2 data to the mobile terminal 107 in a wired and/or wireless manner, and is used to receive data transmitted by the mobile terminal 107 to the single-chip microcomputer for processing. Feedback information or data from unit 103. The mobile terminal 107 can be used to display the data of the second layer, and the user can view various sensory data collected by the data acquisition unit 102 and the data information processed by the single-chip processing unit 103 on the data of the first layer through the mobile terminal 107 . At the same time, the user can check the feedback data of the server/cloud platform 108 through the mobile terminal 107 . The mobile terminal 107 is also used to implement the input of plan data and user personal information in the server/cloud platform 108 . The mobile terminal 107 is connected to a server/cloud platform 108 . The server/cloud platform 108 is used to implement the processing and feedback process of the Layer 2 data transmitted to the mobile terminal 107 . The mobile terminal includes but is not only a mobile phone, a tablet computer, and all devices capable of communicating with 2G/3G/4G and other 3GPP protocols can be considered as the mobile terminal 107 .

[0059]The data preprocessing module 108a in the server/cloud platform 108 is connected to the mobile terminal 107 and the scheme database 108b, and performs data preprocessing on the received Layer 2 data transmitted by the mobile terminal 107. The data preprocessing module 108a is also connected to the comprehensive data processing module 108c. The data preprocessing module 108a sends the three-layer data formed after processing the two-layer data to the integrated data processing module 108c.

[0060] The data preprocessing includes: confirming the type of sensing data in the received Layer 2 data. And based on the type information of one or more sensor data contained in the second-tier data, the data classification scheme stored in the scheme database 108b is retrieved, and the numerical range classification of various sensor data in the second-tier data is completed. And the data preprocessing module 108a sends the three-layer data formed after the numerical range classification of the two-layer data is completed to the comprehensive data processing module 108c.

[0061] The comprehensive data processing module 108c sends the feedback data formed after three-layer data processing to the mobile terminal 107 and/or the single-chip processing unit 103 .

[0062] The comprehensive data processing module 108c completes the analysis of the user's current physiological state, sleep posture and bed-leaving situation based on the received user pressure information data, humidity information data, temperature information data and heart sound data, and based on the user's current physiological state, sleep posture , the analysis result of getting out of bed and the comprehensive processing plan retrieved from the program database 108b to realize the confirmation of the user's physiological safety level, sleeping posture and getting out of bed.

[0063] And based on the user's physiological safety level, sleeping posture and bed-leaving situation, feedback data including suggestion information is formed. The feedback data includes result display data and/or alarm data and/or command data, wherein the display data is sent to the display module 106 and the mobile terminal 107 to be displayed to the user and includes text, number lists and images At least one of the data in . The alarm data includes data sent to the mobile terminal 107 for alarming through the control device and/or vibration device and data sent to the single-chip processing unit 103 located in the mattress for alarming by controlling the alarm unit 105 and the display module 106 . The command data is command information sent to the single-chip processing unit 103 for controlling at least one sensor in the data acquisition unit 102 to perform data re-acquisition.

[0064] According to a preferred embodiment, the feedback data includes at least command data, and the command data is a command sent to the single-chip processing unit 103 for controlling at least one sensor in the data acquisition unit 102 to perform secondary acquisition of sensory data information. The secondary collection of the sensing data is the pressure information data and/or humidity information data and/or humidity information data sent by the comprehensive data processing module 109c in the server/cloud platform 108 when the pressure sensor 102a monitors that the user's sleeping posture changes. /or secondary collection of temperature information data and/or heart sound information data. During the process of data collection by the data collection unit 102 of the device, all the sensors on the mattress do not collect data at the same time, thereby reducing the data processing pressure of the server/cloud platform 108 . Simultaneously, each sensor in the described data acquisition unit 102 only realizes data acquisition according to the control command of the single-chip microcomputer processing unit, and does not have to maintain working state all the time thereby increases the service life of each sensor, and has reduced mattress energy consumption, has reduced unnecessary energy used.

[0065] The sensing data collected by the sensing unit 102 for the second time is processed by the server/cloud platform 108 . The secondary collection of the humidity information data and/or temperature information data and/or heart sound information data is to perform secondary collection of pressure information based on the pressure sensor 102a and analyze and confirm the user's sleeping posture and the user's relationship with the user through the server/cloud platform 108. The contact area of ​​the mattress realizes secondary collection of humidity information data and/or temperature information data and/or heart sound information data of the corresponding area. The secondary collection of humidity information data and/or temperature information data and/or heart sound information data of the mattress is performed after the pressure sensor 102a re-confirms the user's sleeping posture and the contact position with the mattress, so that Humidity, temperature and heart sound sensors are all based on data re-acquisition based on the new contact area or position between the user and the mattress, which avoids opening the sensors on the entire mattress for data collection and data analysis, and avoids the need for the mattress to interact with the user. Unnecessary data collection by sensors in the non-contact area reduces the data processing pressure on the server/cloud platform 108 .

[0066] The command data also includes temperature adjustment data based on the collected user temperature information and mattress area firmness adjustment data based on the user's sleeping posture. The temperature adjustment data is used to control the temperature adjustment unit located on the mattress to achieve temperature adjustment. The mattress area hardness adjustment data is based on the user's sleeping posture and the contact area between the user and the mattress to complete the hardness adjustment of the corresponding area. The adjustment of the firmness is determined based on the height of the pillow and the weight of the user. The hardness of the mattress is adjusted to realize that when the user is lying on his back or sleeping on his side, the height difference between the user's head force area and the user's shoulder force area in the vertical direction is 10cm to 15cm. For example, if the height of the user's sleeping pillow is 8cm, the deformation range of the mattress is 2cm to 7cm. The deformation range is 2cm to 7cm. The temperature information detected by the mattress device through the data sensing unit 102 can be used to adjust the temperature of the mattress through the comprehensive data processing unit of the server/cloud platform 108 to improve the user's sleep examination. At the same time, through the sleeping posture analyzed by the user on the pressure information of the mattress, the mattress can also adjust the hardness of the mattress based on the different sleeping postures of the user, so as to achieve the purpose of protecting the cervical spine of the human body.

[0067] figure 2 The connection relationship between the data acquisition unit 102 and the single-chip processing unit 103 is shown. Wherein, the pressure sensor 102a is connected with the amplification circuit 102g, and is used for collecting pressure information of the mattress and completing signal amplification. The pressure sensor 102a may be a semiconductor piezoresistive sensor, an electrostatic capacitance type pressure sensor and a diffused silicon pressure transmitter. The humidity sensor 102b is connected with the amplifier circuit 102g, and is used for collecting the humidity information of the mattress and completing the signal amplification. The humidity sensor may be one or more of a resistive lithium chloride hygrometer, a dew point lithium chloride hygrometer, a carbon humidity sensitive hygrometer, an alumina hygrometer and a ceramic humidity sensor. The temperature sensor 102c is connected with the amplification circuit 102g, and is used for collecting temperature information of the mattress and completing signal amplification. The temperature sensor 102c may be a contact or non-contact thermometer, such as a contact temperature sensor including a pressure thermometer, a resistance thermometer, a thermistor and a thermocouple thermometer. The heart sound sensor 102d is connected with the amplifying circuit 102g, and is used for collecting the user's heart sound information by the mattress and amplifying the signal. The heart sound sensor 102d may be one or more of an infrared pulse sensor, a heart sound pulse sensor, a photoelectric pulse sensor, a digital pulse sensor, a heart sound pulse sensor and an integrated pulse sensor. The amplifying circuit 102g sends the amplified data including pressure information data, humidity information data, temperature information data, and heart sound information to the single-chip processing unit 103 through the A/D conversion circuit for preliminary processing of one layer of data.

[0068] According to a preferred embodiment, the mattress monitors the posture of the user on the bed through the pressure sensor 102a. Because it is difficult to predict the posture of the user on the bed, it is necessary to collect user data through multiple channels. At the same time, the pressure sensing system needs to be selected to obtain the most suitable channel for data analysis. When the user's sleeping posture changes in bed, such as turning over and tossing, it is necessary to change the channel used for data processing, and record the user's abnormal state, such as the number of times of turning over, etc., to achieve preliminary sleep quality analysis of the user. When the user is a group that cannot take care of themselves, such as infants and the elderly, an alarm message can be sent in time.

[0069] Furthermore, the user lies flat on the bed, and the pressure sensors 102a on all channels start to collect pressure change signals at the same time, and after a period of time, for example, 6 seconds, the channels start to be screened. When the data collected in a channel meets certain conditions, for example, when the number of points below/above a certain threshold is the largest, the channel is selected for data processing, and the user’s physiological signals can be obtained by performing data processing on the selected channel, such as breathing, Heartbeat, temperature and humidity, etc., the physiological signals are respectively collected by the humidity sensor 102b, the temperature sensor 102c and the heart sound sensor 102d. At the same time, turn off the sensors corresponding to the channels not related to data acquisition.

[0070] When the user changes the sleeping position, some channels distributed on the bed will have a large signal change, and the system will select the channel again. The comprehensive data processing module 108c of the server/cloud platform 108 completes the analysis of the user's posture change based on the pressure changes collected by the multiple pressure sensors 102a in the data acquisition unit 102, and controls the data acquisition unit 102 to replace the data acquisition channel or based on the analysis result. Maintain the original data collection channel to continue data collection. When more than 1/10 of the data collection by the pressure sensor 102a disappears or decreases significantly, for example, the pressure value decreases by 1/5, it is determined that the user's bed posture has changed significantly, and the data collection unit 102 needs to be controlled to change the data collection channel for data collection. Collect again.

[0071] The process of changing the data collection channel is that when the server/cloud platform 108 monitors that the bed posture of the human body changes through the data collection unit 102, the pressure distributed on the entire mattress surface in the data collection unit 102 is controlled by the single-chip processing unit 103. The sensor 102a realizes pressure data re-acquisition, so as to determine the new posture of the human body in bed. After the user’s bed posture is determined, the original sensor acquisition channel is switched to the acquisition channel of the user in contact with the mattress or the force-bearing area sensor, and is controlled by the corresponding channel. It includes a pressure sensor 102a, a humidity sensor 102b, a temperature sensor 102c and a heart sound sensor 102d to collect sensory data from the user.

[0072] If the pressure signal collected by the pressure sensor 102a does not change significantly, channel switching is not performed. For example, when a person moves his body slightly, channel switching is not performed at this time, so as to reduce data loss caused by channel switching.

[0073] When the user is in an abnormal state, the user's abnormal state type and times will be recorded. Take turning over as an example here. When the user turns over, the data collected by part of the pressure sensor 102a will change. For example, the sensor that was originally pressed is no longer under force. When the change meets the set conditions, such as some sensors The data is reset to zero. At the same time, when the sensor with no data receives data, it will record the user's turning over state and record the number of times.

[0074] When the user is no longer on the mattress, the original force sensor data changes. When the waveform meets the set conditions, for example, the sensor signal that originally had a value quickly returns to zero. The difference from turning over is the speed of zero return, and When all sensors are reset to zero, the system determines that the user has left the bed. If the user is unable to take care of himself, the system will send an alarm signal after the state of leaving the bed reaches a certain length of time, such as 10 minutes.

[0075] Further, the method for calculating the respiration rate is generally the waveform method, and the respiration rate is obtained by addressing the effective peak and trough calculation period adjacent to the respiration wave. At present, many studies in this field focus on the mattress-based physiological signal monitoring system. For example, French scientist J.MOLET first used wavelet transform when analyzing seismic waves. Heartbeat and respiration signals were extracted by wavelet transform. Wavelet refers to a vibration waveform with a certain amplitude and frequency. The average value of the waveform is zero, and the amplitude alternates between positive and negative. The wavelet transform is such a transformation: use the wavelet basis formed by the shifted wavelets of different frequencies to compose or decompose the time domain signal. The ratio of center frequency and bandwidth determines the difference of wavelet. The process of wavelet transform is very similar to Fourier transform.

[0076]For example, American scholar Norden E.Huang et al. proposed a non-stationary signal processing method in the time domain - Empirical Mode Decomposition Algorithm, referred to as EMD. Any complex data sequence can be decomposed into a finite number, usually several Intrinsic Mode Functions (IMF for short). This method has strong self-adaptability and high efficiency, and because this decomposition is based on the local characteristics of the data time scale, it is suitable for nonlinear and non-stationary processing. The derivation of the intrinsic mode function makes the significance of the instantaneous frequency more prominent. At the same time, the introduction of the concept of instantaneous frequency of complex data sequences effectively avoids the disadvantages of using spurious harmonics to describe nonlinear and unstable signals. From the perspective of signal processing, EMD is a process of gradually decomposing signals from high frequency to low frequency. It embodies the characteristics of multi-resolution. Both in terms of concept and signal analysis method, it is an innovative breakthrough in non-stationary signal processing and opens up new ideas. Compared with the wavelet transform, the EMD method does not need to select the basis function, but according to the characteristics of the signal itself, it adaptively decomposes the signal into a limited number of intrinsic mode functions with frequencies from high to low, and different IMF components also reflect The characteristics of the signal on different time scales. By analyzing and judging the spectrum of each IMF component, and then returning to the time domain to separate and reconstruct the breathing and heartbeat signals of the newborn, this can avoid interference such as noise and harmonics of the breathing signal.

[0077] The device of the present invention can also realize the statistics of the user's respiration rate through the wavelet algorithm or the EMD algorithm. The sampled signal is a mixed signal of noise such as respiration, heartbeat and body movement. The signal contains harmonic components of different frequencies, and further signal processing is required to obtain accurate heartbeat and respiration signals. The EMD algorithm is used to decompose the signal into a finite sum of intrinsic mode functions, and the respiration and heartbeat waveforms are reconstructed according to the frequency band range of the respiration rate and heart rate. First, Fast Fourier Transform (FFT) is performed on the mixed signal of breathing and heartbeat to find out the frequency corresponding to the maximum spectral peak, thereby estimating the frequency band range of breathing and heartbeat. Then EMD is performed on the same signal, which can be decomposed into several harmonic components with different frequencies. Some harmonic components are components of the breathing or heartbeat signal. By performing FFT on each harmonic component, the proportion of energy in the breathing and heartbeat frequency range to a certain harmonic component is calculated. When the ratio is greater than 60% (an empirical parameter), it can be considered as a component of breathing and heartbeat. Respiration and heartbeat signals can be reconstructed after calculation of all harmonic components. Perform FFT on the breathing and heartbeat signals respectively, and calculate the frequency corresponding to the highest spectral peak, which is the breathing rate and heart rate.

[0078] Furthermore, the respiration and heart rate of different people will be very different, and the respiration and heart rate of the same person in different time periods are also different. In special cases, the respiration rate can be as high as 150 times per minute (2.5Hz). will coincide with the heart rate range under normal conditions. Therefore, it is often impossible to accurately measure physiological information under various abnormal conditions by setting a fixed respiration rate and heart rate frequency range to separate the respiration signal and the heartbeat signal. Since the beating signal of the mixed signal collected by the signal collector is relatively weak, the energy of the respiration signal is the largest. Therefore, first perform fast Fourier transform on the initial mixed signal, and calculate the frequency value corresponding to the spectral peak with the largest energy, which is the estimated value fc of the respiration rate. Taking advantage of the fact that there are differences in the frequency domain between respiration and heartbeat of the human body at the same time and the heart rate is generally higher than the respiration rate, choose fc as the median and the frequency within 0.2Hz as the frequency band range of the respiration signal, fc+0.2 Hz to 3Hz is the frequency band range of the heartbeat signal. Through dynamically selected filter frequency bands of breathing and heartbeat signals, the real-time breathing rate and heart rate can be calculated more accurately without being affected by the abnormal physiological conditions of the subject.

[0079] Taking the respiration monitoring realized by the mattress as an example, when the human body is lying on the mattress, the central sound sensor 102d of the data collection unit 102 starts to collect parameters, and converts the collected information into electrical signals. After the amplification circuit 102e, the charge signal can be converted into a voltage signal, and after the signal is denoised, it is sent to the A/D conversion circuit 102f, and the mixed voltage signal with the respiration and heartbeat signal is denoised, and the respiration signal and heartbeat are separated Signal. The voltage signal is converted into a digital signal by the A/D conversion circuit 102f, and then the data is transmitted to the single-chip processing unit 103 through serial communication for digital processing, and the calculated respiration rate and other information are displayed in real time through the display unit. When there is a suffocation situation, the control speaker sends an alarm signal.

[0080] Wherein, the A/D conversion circuit 102f may use a 30s data sliding window to process data, and update the window every 1s. like image 3 As shown, the abscissa in the figure is time, the unit is second, and the ordinate is the voltage value count of the A/D conversion circuit 102f. The mattress has multiple data collection channels according to the contact area between the user and the mattress, image 3 Plot the values ​​for the 6 channels selected. The signal collected by the heart sound sensor 102d is smoothed and filtered by the amplifier circuit 102e to remove the noise in the signal, Figure 4 Plot the values ​​of the 6 channels after sliding filtering. from Figure 4 In each data acquisition channel, the channel corresponding to the maximum value of the peak or trough distance of 2048 is selected as the channel for judging breathing. The abscissa in the figure is time in seconds, and the ordinate is the voltage value count of the A/D conversion circuit 102f. The center sound sensor 102d itself will produce voltage changes due to pressure changes. After the amplification of the data acquisition unit 102 and the voltage is too high, the voltage change threshold is raised to (0V, +3V), and 2048 corresponds to 1.5V , 4096 corresponds to 3V. Numerical meanings such as 2048 or 4096 are the result of sampling the total A/D conversion circuit 102f of the sensing unit 102. When the sensor is in a static state, the voltage collected by the system is at 1.5V, which is around 2048 points. Figure 5 Channel 4 is selected for respiration judgment, the abscissa in the figure is time, the unit is second, and the ordinate is the count of the voltage value of the A/D conversion circuit 102f. Image 6 A graph of respiratory monitoring. If it crosses the 2048 line, it is recorded as 1, and if it crosses the 2048 line, it is recorded as 0. Count the number of breaths within 30 seconds and multiply by 2 to get the number of breaths per minute.

[0081] According to a preferred embodiment, taking the data collected by the pressure sensor 102a as an example, during the channel selection process, calculate the square of the difference between the sampling values ​​of all channels within 5 seconds and 2048, sort the square of the difference, and select the two largest channels as The currently selected channel, if the channel is switched, it means that the object has turned over once. like Figure 7 As shown, the abscissa in the figure is time, the unit is second, and the ordinate is the voltage value count of the A/D conversion circuit 102f. The selected channel is inside the dotted line box, and the original 2 channels have been switched to 2 new channels, indicating that the subject has turned over once. Compared with the actual situation, turning over has indeed occurred.

[0082] According to a preferred embodiment, taking the data collected by any sensor of the sensing unit 102 as an example, when it is detected that the signals of all channels tend to 2048, that is, no physiological signal is sensed, it is considered that the test subject has left the bed. like Figure 8 As shown, the abscissa in the figure is time, the unit is second, and the ordinate is the voltage value count of the A/D conversion circuit 102f. After the voltage curve (at 7 seconds), all signal values ​​tend to 2048, and the subject is considered to have left the bed.

[0083] Further, through the mattress of the present invention, users can intelligently realize the perceptual recording of life activities: use high-sensitivity sensors to detect basic life activity information elements such as heartbeat, respiration, and body movement; use data model analysis and calculation and filter analysis to separate Heartbeat, breathing and abnormal body movement events.

[0084] Through the mattress of the present invention, the user can intelligently realize the monitoring and alarming of the critical value of life activities: the system can set the normal value of heartbeat, and provide data monitoring function, and alarm when it exceeds the set range; the system can set Respiratory rate is normal, and provide data monitoring function, alarm when it exceeds the set range; the system can record various sensory states and life activity states that can be analyzed and measured, and provide data monitoring function, alarm when it exceeds the set range.

[0085] Through the mattress of the present invention, the user can intelligently realize the analysis of basic life activity events: determine the normal active state: such as lying in bed; determine the inactive state: such as leaving the bed (conforming to the preset plan for leaving the bed); Activity state: such as the disappearance of life activities (conforming to the specific inactive state data preset scheme); determination of abnormal activity state: such as continuous coughing, rolling, etc. Determination of abnormal activity state: such as falling bed (conforming to the falling bed data preset scheme); Determination of abnormal activity status: such as frequent getting out of bed at night, etc.

[0086] Through the mattress described in the present invention, the user can intelligently realize the special state alarm of the basic life activity event: the server/cloud platform 108 can set the threshold value of various measurement states, and will provide notification, warning and alarm of various measurement states Functions, such as: bed-leaving alarm: define alarm requirements for some patients who cannot leave the bed, when the system detects that the state is leaving the bed, the alarm is used to prevent falling from the bed; life activity disappears alarm. Special activity records: the system analyzes and processes data, analyzes and records possible activity states, such as abnormal heart rate, getting out of bed, rolling, etc.; night activity records: it will focus on capturing nocturnal life activity events, which can provide assistance for clinical medical analysis.

[0087] Through the mattress of the present invention, users can intelligently realize the setting and reminder of trigger time: the system will provide trigger event management functions, such as: trigger by time, trigger by admission time, trigger by time in bed, and time to leave bed Trigger, trigger by respiration or heartbeat, etc.; all trigger events will provide display, warning and alarm, and the portable monitoring terminal will provide services synchronously.

[0088] Through the mattress of the present invention, users can intelligently realize emergency calls: provide bedside close call buttons, and provide emergency call services.

[0089] Through the mattress described in the present invention, the associated object can realize intelligent monitoring of users: provide a mobile life activity sensing data browsing terminal; provide different network login and data access modes such as WIFI, 3G, and 4G.

[0090] Therefore, through the mattress of the present invention, users can intelligently monitor their physiological conditions, and implement feedback alarms or sleep suggestion feedback based on the monitoring results, thereby helping users understand their own sleep conditions and improve their sleep quality. During the process of data collection by the data collection unit 102, all sensors do not collect data at the same time, thereby reducing the data processing pressure of the server/cloud platform 108 of the mattress. At the same time, each sensor in the data acquisition unit 102 only realizes data acquisition according to the control command of the single-chip processing unit 103, without having to maintain a working state all the time, thereby increasing the service life of each sensor, reducing energy consumption of the mattress, and reducing unnecessary power consumption. Therefore, the mattress of the present invention is not only applicable to places such as hospitals and nursing centers, but also applicable to families, and is widely used.

[0091] It should be noted that the above-mentioned specific implementation methods are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the scope of the disclosure of the present invention and fall within the scope of this disclosure. within the scope of protection of the invention. Those skilled in the art should understand that the description and drawings of the present invention are illustrative rather than limiting to the claims. The protection scope of the present invention is defined by the claims and their equivalents.