A mattress-based health monitoring method and device, mattress and medium

Abstract

The application discloses a health monitoring method and device based on a mattress, the mattress and a medium. The method comprises the following steps: determining a breathing signal of a target user on the mattress; inputting the breathing signal into a preset model to obtain a health confidence of the target user, or the health confidence and an abnormal level; if the preset model outputs multiple abnormal levels in a preset time period and the change trend of the abnormal levels meets a preset trend, sending health warning information to the target user. The technical scheme of the embodiment of the application uses the preset model to mine the correlation between the breathing signal of the user and health, realizes the health monitoring only at home of the user, and solves the problem that the economy and accuracy of the traditional health monitoring mode cannot be considered simultaneously.

Description

Technical Field

[0001] This invention relates to the field of smart mattress technology, and in particular to a health monitoring method, device, mattress, and medium based on a mattress. Background Technology

[0002] Regular and sufficient sleep plays a vital role in fighting viral invasion and boosting immunity. Sleep quality is one of the standards for measuring health, and with the continuous development of technology, sleep monitoring has become increasingly common, and people are paying more and more attention to the quality of their sleep.

[0003] Some diseases can affect a user's sleep. For example, some diseases can cause abnormal shaking, while others can lead to degeneration of the areas of the brain that control breathing, resulting in weakened respiratory muscles and sleep-disordered breathing. Health monitoring and early warning systems are essential for specific populations, such as the elderly. Currently, health monitoring mainly relies on specialized medical devices in hospitals or wearable electronic devices such as watches.

[0004] However, professional medical devices are expensive and have low availability, while home-use wearable electronic devices are not very accurate. Summary of the Invention

[0005] This invention provides a health monitoring method, device, mattress, and medium based on a mattress, to solve the problem that traditional health monitoring methods cannot simultaneously achieve both economy and accuracy.

[0006] In a first aspect, the present invention provides a health monitoring method based on a mattress, applied in a mattress, the method comprising:

[0007] Identify the target user's breathing signals on the mattress;

[0008] The respiratory signal is input into a preset model to obtain the health confidence score of the target user, or the health confidence score and the abnormality level, wherein the abnormality level is determined based on the magnitude of the respiratory signal and the health confidence score;

[0009] If, within a preset time period, the preset model outputs multiple abnormality levels and the trend of the abnormality level changes conforms to a preset trend, then a health warning message is sent to the target user.

[0010] In a second aspect, the present invention provides a mattress-based health monitoring device, configured within a mattress, the device comprising:

[0011] A breathing signal determination module is used to determine the breathing signal of a target user on the mattress;

[0012] The signal processing module is used to input the respiratory signal into a preset model to obtain the health confidence score of the target user, or the health confidence score and the abnormality level, wherein the abnormality level is determined based on the magnitude of the respiratory signal and the health confidence score;

[0013] The information sending module is used to send health warning information to the target user if the preset model outputs multiple abnormal levels within a preset time period and the changing trend of the abnormal levels conforms to a preset trend.

[0014] Thirdly, the present invention provides a mattress comprising:

[0015] The mattress body, multiple flexible piezoelectric sensors, and at least one processor are included, wherein the flexible piezoelectric sensors and the processor are communicatively connected.

[0016] and memory that is communicatively connected to at least one processor;

[0017] The memory stores a computer program that can be executed by at least one processor, which enables the at least one processor to perform the mattress-based health monitoring method of the first aspect described above.

[0018] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a processor to execute the mattress-based health monitoring method of the first aspect described above.

[0019] This invention provides a mattress-based health monitoring solution that determines the respiratory signals of a target user on the mattress, inputs these signals into a preset model, and obtains the target user's health confidence level, or the health confidence level and an abnormality level. The abnormality level is determined based on the magnitude of the respiratory signal and the health confidence level. If, within a preset time period, the preset model outputs multiple abnormality levels and the trend of these abnormality levels conforms to a preset trend, a health warning is sent to the target user. By employing this technical solution, by inputting the respiratory signal into a preset model, a health confidence level representing the (target) user's health status and an abnormality level representing the degree of abnormality are obtained. The preset model uncovers the correlation between the user's respiratory signals and health, enabling health monitoring to be conducted solely at the user's home, thus solving the problem of balancing economy and accuracy in traditional health monitoring methods.

[0020] It should be understood that the description in this section is not intended to identify key or essential features of the invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0021] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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.

[0022] Figure 1 This is a flowchart of a mattress-based health monitoring method according to Embodiment 1 of the present invention;

[0023] Figure 2 This is a flowchart of a mattress-based health monitoring method according to Embodiment 2 of the present invention;

[0024] Figure 3 This is a schematic diagram of a mattress-based health monitoring device according to Embodiment 3 of the present invention;

[0025] Figure 4 This is a structural schematic diagram of a mattress provided according to Embodiment 4 of the present invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. In the description of this invention, unless otherwise stated, "a plurality of" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist; for example, A and / or B can represent: A alone, A and B simultaneously, and B alone. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or mattress that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to these processes, methods, products, or mattresses.

[0028] Example 1

[0029] Figure 1 The flowchart of a mattress-based health monitoring method is provided in Embodiment 1 of the present invention. This embodiment is applicable to monitoring a user's health using a mattress. The method can be executed by a mattress-based health monitoring device, which can be implemented in hardware and / or software. The mattress-based health monitoring device can be configured in a mattress, which can be composed of two or more physical entities or a single physical entity. The mattress has a built-in flexible piezoelectric sensor.

[0030] like Figure 1 As shown, the health monitoring method based on a mattress provided in Embodiment 1 of the present invention specifically includes the following steps:

[0031] S101. Determine the breathing signals of the target user on the mattress.

[0032] In this embodiment, when a person lies on the mattress, their breathing causes changes in the mattress, such as slight undulations. Therefore, a preset sensor (such as a piezoelectric sensor) can be used to collect information about these changes. Through preset processing, such as signal separation, the target user's breathing signal can be extracted from this information. The target user can be understood as the person lying on the mattress. The breathing signal can include breathing frequency and amplitude, with the mattress undulation amplitude representing the breathing amplitude and the mattress undulation frequency representing the breathing frequency.

[0033] S102. Input the respiratory signal into a preset model to obtain the health confidence level of the target user, or the health confidence level and the abnormality level, wherein the abnormality level is determined based on the magnitude of the respiratory signal and the health confidence level.

[0034] In this embodiment, some diseases can affect a person's normal breathing, such as causing shortness of breath or irregular breathing. Therefore, by extracting features from the breathing signal, the health status and degree of unhealth of the target user can be monitored. The health confidence score can be used to represent the probability of the target user being healthy, and the anomaly level can be used to represent the degree of unhealth of the target user. When the health confidence score meets the preset requirements, such as when the health confidence score is 1, it can be said that the target user is healthy. In this case, the anomaly level does not need to be output, that is, only the health confidence score is output. When the health confidence score is not 1, both the health confidence score and the anomaly level can be output simultaneously.

[0035] S103. If, within a preset time period, the preset model outputs multiple abnormal levels and the changing trend of the abnormal levels conforms to a preset trend, then a health warning message is sent to the target user.

[0036] In this embodiment, if multiple abnormal levels are obtained within a preset time period, such as one week, and the trend of these abnormal levels conforms to a preset trend, such as an upward trend, it can be determined that the target user's health may be problematic and the problem may worsen. In this case, a health warning message can be sent to the target user and designated individuals, such as the target user and their relatives, to remind the target user to seek medical attention promptly. The health warning message may include information such as the target user's frequency of breathing, respiratory amplitude, health confidence level, and abnormal level within the preset time period.

[0037] The health monitoring method based on a mattress provided by this invention determines the breathing signal of a target user on the mattress, inputs the breathing signal into a preset model, and obtains the target user's health confidence level, or the health confidence level and an abnormality level. The abnormality level is determined based on the magnitude of the breathing signal and the health confidence level. If, within a preset time period, the preset model outputs multiple abnormality levels and the trend of the abnormality level changes conforms to a preset trend, a health warning message is sent to the target user. This invention's technical solution, by inputting the breathing signal into a preset model, obtains a health confidence level representing the user's health status and an abnormality level representing the degree of abnormality. It utilizes the preset model to uncover the correlation between the user's breathing signal and health, enabling health monitoring only at the user's home, and solving the problem of the inability to simultaneously achieve cost-effectiveness and accuracy in traditional health monitoring methods.

[0038] Example 2

[0039] Figure 2 This is a flowchart of a mattress-based health monitoring method provided in Embodiment 2 of the present invention. The technical solution of the present invention is further optimized based on the above optional technical solutions, and provides a specific way to monitor user health using a mattress.

[0040] Optionally, the step of inputting the respiratory signal into a preset model to obtain the health confidence score of the target user, or the health confidence score and the abnormality level, includes: inputting the respiratory signal into a preset model to obtain the health confidence score of the target user; if the health confidence score is less than a second preset value, the preset model also outputs the abnormality level of the target user. The advantage of this setting is that when the target user is relatively healthy, i.e., when the health confidence score is less than the second preset value, there is no need to calculate and output the user's abnormality level, thereby avoiding a waste of computing power and improving the efficiency of the model in processing data.

[0041] Optionally, the step of sending a health warning to the target user if the preset model outputs multiple abnormal levels within a preset time period and the trend of the abnormal levels conforms to a preset trend includes: if the preset model outputs multiple abnormal levels within a preset time period and the number of times the abnormal level is greater than a first preset level exceeds a preset number, then at least one health report containing first health warning information is sent to the target user; if the preset model outputs multiple abnormal levels within the preset time period, the abnormal level is greater than a second preset level, and the trend of the abnormal level conforms to an upward trend, then at least one health report containing second health warning information is sent to the target user, wherein the first preset level is greater than the second preset level, and the first health warning information is different from the second health warning information. The advantage of this setting is that it comprehensively covers the health abnormality information that the abnormality level can represent when the target user's health is abnormal, thus improving the accuracy of health monitoring.

[0042] Optionally, the method further includes: if the cumulative data volume of the respiratory signal is greater than a preset data volume, the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds a preset number, then at least one health report containing third health warning information is sent to the target user; if the cumulative data volume is greater than the preset data volume, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the trend of the abnormal level change conforms to an upward trend, then at least one health report containing fourth health warning information is sent to the target user, wherein the first preset level is greater than the second preset level, and the third health warning information is different from the fourth health warning information. The advantage of this setting is that it avoids the situation where the amount of identifiable respiratory signal data is too small due to the target user not using the mattress for a long time, resulting in missed health warning information.

[0043] like Figure 2 As shown, the second embodiment of the present invention provides a health monitoring method based on a mattress, which specifically includes the following steps:

[0044] S201. Determine the breathing signals of the target user on the mattress.

[0045] Optionally, determining the target user's breathing signal on the mattress includes: acquiring the piezoelectric signal corresponding to the target user on the mattress using a flexible piezoelectric sensor built into the mattress; and filtering and separating the piezoelectric signal to obtain the target user's breathing signal. The advantage of this setup is that by utilizing the flexible piezoelectric sensor built into the mattress, it avoids affecting the user's sleep during health monitoring, and by filtering and separating the piezoelectric signal, it accurately determines the target user's breathing signal.

[0046] Specifically, when the target user lies on the mattress, the flexible piezoelectric sensors built into the mattress can collect piezoelectric signals (electrical signals generated by pressure). These signals can then be filtered to remove interference, and the filtered signals can be further processed to obtain the target user's breathing signal. Typically, multiple flexible piezoelectric sensors are used, and they can be integrated into the mattress in the form of sleep monitoring strips.

[0047] S202. Input the respiratory signal into a preset model to obtain the health confidence level of the target user; if the health confidence level is less than a second preset value, the preset model also outputs the abnormality level of the target user.

[0048] For example, if the second preset value is 0.6, when the breathing signal is input to the preset model, if the health confidence level output by the preset model is 0.7, then the preset model will also output an abnormality level, such as level 1. If the health confidence level output by the preset model is 0.4, then the preset model will not output an abnormality level.

[0049] Optionally, the preset model includes a convolutional layer, an encoding layer constructed based on a multi-head attention mechanism, and a classification layer. The preset model is determined by: acquiring sample respiratory signals configured with sample labels, wherein the sample labels include a confidence label, or the confidence label and an anomaly level label; inputting the sample respiratory signals into the convolutional layer of the initial preset model to obtain a first feature vector; inputting the first feature vector into the encoding layer of the initial preset model constructed based on causal convolution to obtain a second feature vector, wherein the causal convolution is used to implement the multi-head attention mechanism; inputting the second feature vector into the classification layer of the initial preset model to obtain sample health information, and training the initial preset model based on the difference between the sample health information and the sample labels to determine the preset model, wherein the sample health information includes sample health confidence, or the sample health confidence and sample anomaly level. The advantage of this setup is that training the initial preset model using labeled samples can yield a preset model with relatively accurate prediction results after training.

[0050] Specifically, the convolutional layers in the pre-defined model can be used to extract respiratory features. The encoding layer, built based on a multi-head attention mechanism, can extract the relationship between respiratory features and time, enhancing the accuracy of the pre-defined model's output. The classification layer processes the feature vectors output by the encoding layer and outputs a health confidence score, or a health confidence score and an abnormality level. Sample respiratory signals can be obtained from publicly available clinical data. This clinical data contains respiratory signals from multiple patients whose conditions may differ but who all exhibit respiratory abnormalities. The clinical data also includes respiratory signals from healthy individuals with normal breathing. Sample labels for the respiratory signals can be pre-defined. For example, confidence labels can be 0 and 1, representing unhealthy and healthy conditions, respectively, and abnormality level labels can be 1, 2, 3, and 4, with higher levels indicating greater unhealthiness. The sample respiratory signals with sample labels can be input into the convolutional layer, which can extract a feature vector (the first feature vector) from the sample respiratory signals. This convolutional layer can contain multiple convolutional processing layers and at least one skip connection layer. The first feature vector is then input into an encoding layer built on causal convolution to extract the temporal dependencies between the first feature vectors, thus obtaining the second feature vector. This second feature vector is then input into the classification layer of the initial preset model to obtain sample health information. This health information may include sample health confidence and sample anomaly level, or only sample health confidence. Based on the difference between this sample health information and the sample label, the initial preset model is trained multiple times. When the difference is sufficiently small, the trained preset model is obtained.

[0051] Furthermore, the step of inputting the second feature vector into the classification layer of the initial preset model to obtain sample health information includes: inputting the second feature vector into the health classification layer of the initial preset model to obtain sample health confidence, wherein the health classification layer and the anomaly level prediction layer belong to the classification layer of the initial preset model, and the health classification layer includes a fully connected layer and an activation layer; if the sample health confidence is less than a first preset value, then the second feature vector is input into the anomaly level prediction layer to obtain the sample anomaly level, wherein the anomaly level prediction layer includes a fully connected layer. The advantage of this setup is that when the probability of a sample's respiratory signal being a healthy respiratory signal is high, there is no need to calculate the anomaly level of the respiratory signal, saving computational power and improving the model's data processing efficiency.

[0052] Specifically, the second feature vector is input into the health classification layer of the initial preset model to obtain the sample's health confidence score. This health classification layer may contain multiple fully connected layers and at least one activation layer. The confidence score can range from greater than or equal to 0 to less than or equal to 1. If the sample's health confidence score is less than the first preset value, it indicates that the probability of the sample's respiratory signal being a healthy respiratory signal is low. In this case, the second feature vector needs to be input into the anomaly level prediction layer to obtain the sample's anomaly level. This anomaly level prediction layer may contain multiple fully connected layers. The first preset value can be the same as or different from the second preset value.

[0053] S203. If, within a preset time period, the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds a preset number, then at least one health report containing the first health warning information is sent to the target user.

[0054] For example, if the preset time period is one week, the first preset level is level 2, and the preset number of occurrences is 3, then if the preset model outputs multiple abnormal levels within one week, and the number of times the abnormal level is greater than level 2 exceeds 3, then at least one health report containing the first health warning information can be sent to the target user. The purpose of the first health warning information is to alert the target user to potential health abnormalities. The first health warning information may include information such as the target user's frequency, breathing amplitude, health confidence level, and abnormality level within the preset time period. In addition to the first health warning information, the health report may also include information such as medical attention reminders and medical attention recommendations; the higher the abnormality level, the higher the medical attention recommendation.

[0055] S204. If, within the preset time period, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the trend of the abnormal level change conforms to an upward trend, then at least one health report containing the second health warning information is sent to the target user.

[0056] Wherein, the first preset level is greater than the second preset level, and the first health warning information is different from the second health warning information.

[0057] For example, as described above, if the second preset level is also level 2, and the preset model outputs multiple abnormal levels within a week, all of which are greater than level 2 and exhibit an upward trend, then at least one health report containing the second health warning information can be sent to the target user. The first and second preset levels can be the same or different. Since both the first and second health warning information can contain information such as the abnormal level, they can be the same or different. If the first and second preset levels are different, the abnormal levels in the first and second health warning information will differ.

[0058] S205. If the cumulative data volume of the respiratory signal is greater than the preset data volume, and the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds the preset number, then at least one health report containing third health warning information is sent to the target user.

[0059] For example, as described above, the preset data volume can be determined based on the cumulative data volume of respiratory signals per hour. If the cumulative data volume of respiratory signals per hour is 2 MB when the target user is on a mattress, the preset data volume can be set to 200 MB, indicating that the target user's time in bed is close to 200 hours. If the preset data volume is 200 MB, then when the cumulative data volume of respiratory signals exceeds 200 MB, and the preset model outputs multiple abnormal levels, and the number of times the abnormal level is greater than 3 exceeds 3, then at least one health report containing third health warning information can be sent to the target user. The third health warning information can be the same as, or different from, the first and second health warning information. For example, if the preset data volume is large (the duration corresponding to the preset data volume is greater than the duration of the preset time period), the third health warning information can be different from, the first and second health warning information.

[0060] S206. If the accumulated data volume is greater than the preset data volume, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the abnormal level change trend conforms to an upward trend, then at least one health report containing the fourth health warning information is sent to the target user.

[0061] Wherein, the first preset level is greater than the second preset level, and the third health warning information is different from the fourth health warning information.

[0062] For example, as described above, when the cumulative data volume of respiratory signals exceeds 200 MB, and the preset model outputs multiple abnormality levels, all of which are greater than level 2, and the trend of these abnormality levels is upward, then at least one health report containing fourth health warning information can be sent to the target user. The fourth health warning information can be the same as, or different from, the first, second, and third health warning information. For instance, if the preset data volume is large and the first and second preset levels are different, the first, second, third, and fourth health warning information can all be different.

[0063] Optionally, the cumulative data volume of the respiratory signal can be reset each time a health report containing the third or fourth health warning information is sent to the target user, but the data of the respiratory signal is retained, so that steps 205 and 206 can be executed whenever the cumulative data volume of the respiratory signal exceeds the preset data volume.

[0064] The health monitoring method based on a mattress provided in this invention uses a mattress to determine the (target) user's respiratory signal. It is suitable for daily continuous monitoring in a home environment, and is especially suitable for health monitoring of special groups such as the elderly. By inputting the respiratory signal into a preset model, a health confidence level representing whether the user is healthy is obtained. Only when the probability of the user being healthy is low (when the health confidence level is low) will the preset model output an abnormality level representing the degree of abnormality of the user, avoiding the waste of computing power and improving the efficiency of model data processing. By setting preset time periods, preset frequency, preset levels, preset trends, and preset data volumes, it comprehensively covers the health abnormality information that the abnormality level can represent when the target user's health is abnormal, improving the accuracy of health monitoring. At the same time, it also avoids the situation where the amount of identifiable respiratory signal data is too small due to the target user not using the mattress for a long time, resulting in missed health warning information. It realizes health monitoring that can be carried out only in the user's home, solving the problem of the inability to balance economy and accuracy in traditional health monitoring methods.

[0065] Example 3

[0066] Figure 3 This is a schematic diagram of a mattress-based health monitoring device provided in Embodiment 3 of the present invention. Figure 3 As shown, the device includes: a respiratory signal determination module 301, a signal processing module 302, and an information transmission module 303. The device can be configured in a mattress, wherein:

[0067] A breathing signal determination module is used to determine the breathing signal of a target user on the mattress;

[0068] The signal processing module is used to input the respiratory signal into a preset model to obtain the health confidence score of the target user, or the health confidence score and the abnormality level, wherein the abnormality level is determined based on the magnitude of the respiratory signal and the health confidence score;

[0069] The information sending module is used to send health warning information to the target user if the preset model outputs multiple abnormal levels within a preset time period and the changing trend of the abnormal levels conforms to a preset trend.

[0070] The mattress-based health monitoring device provided in this invention obtains the health confidence level and the abnormality level, which represent the degree of abnormality of the user, by inputting the respiratory signal into a preset model. The preset model is used to explore the correlation between the user's respiratory signal and health, realizing health monitoring only in the user's home, and solving the problem that the economy and accuracy of traditional health monitoring methods cannot be balanced.

[0071] Optionally, the preset model includes a convolutional layer, an encoding layer constructed based on a multi-head attention mechanism, and a classification layer. The preset model is determined by: acquiring a sample respiratory signal configured with sample labels, wherein the sample labels include a confidence label, or the confidence label and an abnormality level label; inputting the sample respiratory signal into the convolutional layer of the initial preset model to obtain a first feature vector; inputting the first feature vector into the encoding layer constructed based on causal convolution in the initial preset model to obtain a second feature vector, wherein the causal convolution is used to implement the multi-head attention mechanism; inputting the second feature vector into the classification layer of the initial preset model to obtain sample health information, and training the initial preset model based on the difference between the sample health information and the sample label to determine the preset model, wherein the sample health information includes a sample health confidence level, or the sample health confidence level and a sample abnormality level.

[0072] Furthermore, the step of inputting the second feature vector into the classification layer of the initial preset model to obtain sample health information includes: inputting the second feature vector into the health classification layer of the initial preset model to obtain sample health confidence, wherein the health classification layer and the anomaly level prediction layer belong to the classification layer of the initial preset model, and the health classification layer includes a fully connected layer and an activation layer; if the sample health confidence is less than a first preset value, then the second feature vector is input into the anomaly level prediction layer to obtain the sample anomaly level, wherein the anomaly level prediction layer includes a fully connected layer.

[0073] Optionally, the signal processing module includes:

[0074] A confidence determination unit is used to input the respiratory signal into a preset model to obtain the health confidence of the target user;

[0075] An anomaly level determination unit is used to output the anomaly level of the target user if the health confidence level is less than a second preset value.

[0076] Optionally, the information sending module includes:

[0077] The first sending unit is configured to send a health report containing first health warning information to the target user at least once if, within a preset time period, the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds a preset number.

[0078] The second sending unit is configured to send at least one health report containing second health warning information to the target user if, within the preset time period, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the abnormal level change trend conforms to an upward trend, wherein the first preset level is greater than the second preset level, and the first health warning information is different from the second health warning information.

[0079] Optionally, the device may also include:

[0080] The first sending module is used to send a health report containing third health warning information to the target user at least once if the cumulative data amount of the respiratory signal is greater than a preset data amount, the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds a preset number.

[0081] The second sending module is used to send a health report containing fourth health warning information to the target user at least once if the accumulated data volume is greater than the preset data volume, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the abnormal level change trend conforms to an upward trend. The first preset level is greater than the second preset level, and the third health warning information is different from the fourth health warning information.

[0082] Optionally, the respiratory signal determination module includes:

[0083] A piezoelectric signal determination unit is used to acquire the piezoelectric signal corresponding to the target user on the mattress using a flexible piezoelectric sensor built into the mattress.

[0084] The breathing signal determination unit is used to filter and separate the piezoelectric signal to obtain the breathing signal of the target user.

[0085] The mattress-based health monitoring device provided in this embodiment of the invention can execute the mattress-based health monitoring method provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects of the method.

[0086] Example 4

[0087] Figure 4 A schematic diagram of the structure of a mattress 40 that can be used to implement an embodiment of the present invention is shown. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the invention described and / or claimed herein.

[0088] like Figure 4 As shown, the mattress 40 includes a mattress body 41, a plurality of flexible piezoelectric sensors 42, and at least one processor 43. The flexible piezoelectric sensors 42 and the processor 43 are communicatively connected. A memory 44, such as a read-only memory (ROM) or a random access memory (RAM), is also communicatively connected to the at least one processor 43. The memory stores computer programs executable by the at least one processor. The processor 43 can perform various appropriate actions and processes based on the computer programs stored in the ROM or loaded from memory into the RAM. The RAM can also store various programs and data required for the operation of the mattress 40. The processor, ROM, and RAM are interconnected via a bus. Input / output (I / O) interfaces are also connected to the bus.

[0089] Multiple components in mattress 40 are connected to I / O interfaces, including: input units such as keyboards and mice; output units such as various types of monitors and speakers; storage units such as disks and optical discs; and communication units such as network cards, modems, and wireless transceivers. The communication units allow mattress 40 to exchange information / data with other mattresses through computer networks such as the Internet and / or various telecommunications networks.

[0090] Processor 43 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 43 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 43 performs the various methods and processes described above, such as mattress-based health monitoring methods.

[0091] In some embodiments, the mattress-based health monitoring method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as a storage unit. In some embodiments, part or all of the computer program may be loaded and / or installed on the mattress 40 via a ROM and / or communication unit. When the computer program is loaded into RAM and executed by the processor 43, one or more steps of the mattress-based health monitoring method described above may be performed. Alternatively, in other embodiments, the processor 43 may be configured to perform the mattress-based health monitoring method by any other suitable means (e.g., by means of firmware).

[0092] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), load-programmable logic mats (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0093] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0094] The computer mattress provided above can be used to perform the mattress-based health monitoring method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0095] Example 5

[0096] In the context of this invention, a computer-readable storage medium may be a tangible medium, and the computer-executable instructions, when executed by a computer processor, are used to perform a mattress-based health monitoring method, which can be applied to a mattress, the method comprising:

[0097] Identify the target user's breathing signals on the mattress;

[0098] The respiratory signal is input into a preset model to obtain the health confidence score of the target user, or the health confidence score and the abnormality level, wherein the abnormality level is determined based on the magnitude of the respiratory signal and the health confidence score;

[0099] If, within a preset time period, the preset model outputs multiple abnormality levels and the trend of the abnormality level changes conforms to a preset trend, then a health warning message is sent to the target user.

[0100] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use with or in conjunction with an instruction execution system, device, or mattress. The computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or mattresses, or any suitable combination thereof. Alternatively, the computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disc read-only memory (CD-ROM), optical storage mattresses, magnetic storage mattresses, or any suitable combination thereof.

[0101] The computer mattress provided above can be used to perform the mattress-based health monitoring method provided in any of the above embodiments, and has corresponding functions and beneficial effects.

[0102] It is worth noting that in the above embodiments of the health monitoring device based on a mattress, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0103] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A health monitoring method based on a mattress, characterized in that, When applied to a mattress, the method includes: Identify the target user's breathing signals on the mattress; The respiratory signal is input into a preset model to obtain the health confidence score of the target user, or the health confidence score and the abnormality level, wherein the abnormality level is determined based on the magnitude of the respiratory signal and the health confidence score; If the preset model outputs multiple abnormal levels within a preset time period and the trend of the abnormal level changes conforms to a preset trend, then a health warning message is sent to the target user; the health warning message includes the target user's frequency, breathing amplitude, health confidence level, and abnormal level information within the preset time period. The method further includes: If the cumulative data volume of the respiratory signal is greater than the preset data volume, and the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds the preset number, then at least one health report containing third health warning information will be sent to the target user. If the accumulated data volume is greater than the preset data volume, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the abnormal level change trend conforms to an upward trend, then at least one health report containing fourth health warning information is sent to the target user, wherein the first preset level is greater than the second preset level, and the third health warning information is different from the fourth health warning information.

2. The method according to claim 1, characterized in that, The preset model includes convolutional layers, an encoding layer constructed based on a multi-head attention mechanism, and a classification layer. The preset model is determined by the following methods: Acquire a sample respiratory signal configured with a sample label, wherein the sample label includes a confidence label, or the confidence label and an anomaly level label; The sample respiratory signal is input into the convolutional layer of the initial preset model to obtain the first feature vector; The first feature vector is input into the encoding layer constructed based on causal convolution in the initial preset model to obtain the second feature vector, wherein the causal convolution is used to implement the multi-head attention mechanism; The second feature vector is input into the classification layer of the initial preset model to obtain sample health information. Based on the difference between the sample health information and the sample label, the initial preset model is trained to determine the preset model. The sample health information includes sample health confidence, or the sample health confidence and sample anomaly level.

3. The method according to claim 2, characterized in that, The step of inputting the second feature vector into the classification layer of the initial preset model to obtain sample health information includes: The second feature vector is input into the health classification layer of the initial preset model to obtain the sample health confidence. The health classification layer and the anomaly level prediction layer belong to the classification layer of the initial preset model. The health classification layer includes a fully connected layer and an activation layer. If the health confidence of the sample is less than a first preset value, the second feature vector is input into the anomaly level prediction layer to obtain the anomaly level of the sample, wherein the anomaly level prediction layer includes a fully connected layer.

4. The method according to claim 1, characterized in that, The step of inputting the respiratory signal into a preset model to obtain the health confidence level of the target user, or the health confidence level and abnormality level, includes: The respiratory signal is input into a preset model to obtain the health confidence level of the target user; If the health confidence level is less than the second preset value, the preset model also outputs the abnormality level of the target user.

5. The method according to claim 4, characterized in that, If, within a preset time period, the preset model outputs multiple anomaly levels and the trend of these anomaly levels conforms to a preset trend, then a health warning message is sent to the target user, including: If, within a preset time period, the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds a preset number, then at least one health report containing the first health warning information is sent to the target user. If, within the preset time period, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the abnormal level changes in an upward trend, then at least one health report containing second health warning information is sent to the target user, wherein the first preset level is greater than the second preset level, and the first health warning information is different from the second health warning information.

6. The method according to any one of claims 1-5, characterized in that, The determination of the target user's breathing signals on the mattress includes: The flexible piezoelectric sensor built into the mattress is used to acquire the piezoelectric signal corresponding to the target user on the mattress; The piezoelectric signal is filtered and separated to obtain the respiratory signal of the target user.

7. A health monitoring device based on a mattress, characterized in that, The device, disposed in a mattress, includes: A breathing signal determination module is used to determine the breathing signal of a target user on the mattress; The signal processing module is used to input the respiratory signal into a preset model to obtain the health confidence score of the target user, or the health confidence score and the abnormality level, wherein the abnormality level is determined based on the magnitude of the respiratory signal and the health confidence score; The information sending module is used to send health warning information to the target user if the preset model outputs multiple abnormal levels within a preset time period and the changing trend of the abnormal levels conforms to a preset trend; the health warning information includes the target user's frequency, breathing amplitude, health confidence level and abnormal level information within the preset time period. The device further includes: The first sending module is used to send a health report containing third health warning information to the target user at least once if the cumulative data amount of the respiratory signal is greater than a preset data amount, the preset model outputs multiple abnormal levels and the number of times the abnormal level is greater than the first preset level exceeds a preset number. The second sending module is used to send a health report containing fourth health warning information to the target user at least once if the accumulated data volume is greater than the preset data volume, the preset model outputs multiple abnormal levels, the abnormal level is greater than the second preset level, and the abnormal level change trend conforms to an upward trend. The first preset level is greater than the second preset level, and the third health warning information is different from the fourth health warning information.

8. A mattress, characterized in that, The mattress includes: The mattress body, multiple flexible piezoelectric sensors, and at least one processor are included, wherein the flexible piezoelectric sensors and the processor are communicatively connected. and a memory communicatively connected to the at least one processor; The memory stores a computer program that can be executed by the at least one processor, which is then executed by the at least one processor to enable the at least one processor to perform the mattress-based health monitoring method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the mattress-based health monitoring method according to any one of claims 1-6.

Images