Health care system based on artificial intelligence, health care control method and computer storage medium

Abstract

The invention relates to the technical field of intelligent health care, in particular to a health care system based on artificial intelligence, a health care control method and a computer storage medium. The health care system comprises a containing device used for providing a health care space for a user; the adjustable far infrared generation module is arranged in the accommodating device and comprises a first excitation unit and a second excitation unit; the data acquisition module is used for obtaining a first physiological time sequence signal and a second physiological time sequence signal; the intelligent control module comprises a signal fusion unit and an intelligent decision-making unit, and the signal fusion unit is used for performing fusion processing on the first physiological time sequence signal and the second physiological time sequence signal to obtain the heart rate and the respiration rate of the user; the intelligent decision-making unit is used for generating a control instruction through a pre-trained network model according to the heart rate, the breathing rate and the health record data of the user and the environmental parameters of the containing device, and the control instruction can regulate and control the environmental parameters in the containing device and / or the wavelength range and intensity of far infrared radiation.

Description

Technical Field

[0001] This application relates to the technical field of intelligent health care, and in particular to an artificial intelligence-based health care system, health care control method, and computer storage medium. Background Technology

[0002] As people's living standards improve, the demand for health management and wellness is increasing. Traditional wellness methods rely solely on human experience and often require professionals to perform complex operations or tactile tests, causing inconvenience and discomfort for users. For example, some wellness devices only offer fixed modes and cannot dynamically adjust based on the user's real-time physiological state and health records.

[0003] In the process of health and wellness management, real-time monitoring and feedback adjustments are crucial for achieving personalized care. However, existing systems cannot optimize care plans by incorporating users' real-time physiological parameters and health records. Therefore, a completely new health and wellness system is urgently needed. Summary of the Invention

[0004] To address the aforementioned technical issues, this application provides an artificial intelligence-based health and wellness system, a health and wellness control method, and a computer storage medium, thereby achieving precision health and wellness.

[0005] In a first aspect, this application provides an artificial intelligence-based health and wellness system, comprising: a housing device for providing a health and wellness space for a user; an adjustable far-infrared generation module disposed within the housing device, the adjustable far-infrared generation module comprising a first excitation unit and a second excitation unit, wherein the second excitation unit is used to excite far-infrared radiation with adjustable spectral characteristics through far-infrared rays generated by the first excitation unit; a data acquisition module comprising a first acquisition unit and a second acquisition unit, wherein the first acquisition unit is used to acquire a facial video stream of the user and obtain a first physiological time-series signal based on the facial video stream; the second acquisition unit is used to acquire a cardiac impact signal of the user and obtain a second physiological time-series signal based on the cardiac impact signal; the intelligent control module comprises a signal fusion unit and an intelligent decision-making unit, the signal fusion unit is used to fuse the first physiological time-series signal and the second physiological time-series signal to obtain the user's heart rate and respiratory rate; the intelligent decision-making unit is used to generate control commands through a pre-trained network model based on the user's heart rate, respiratory rate, health record data, and environmental parameters of the housing device, the control commands being able to regulate the environmental parameters within the housing device and / or the wavelength range and intensity of the far-infrared radiation.

[0006] Optionally, in some embodiments, the health and wellness system further includes: an environmental control module, disposed within the containment device, for monitoring and adjusting environmental parameters within the containment device; and a data storage module, for storing the user's health record data.

[0007] Optionally, in some embodiments, the first excitation unit is a graphene-based heating element, and the second excitation unit is a composite substrate containing negative ion powder, tourmaline powder, and / or terahertz material.

[0008] Optionally, in some embodiments, the health record data includes: basic information and health information of the user obtained through instrument detection, medical reports provided by the user, or consultation, wherein the health information includes: physiological and biochemical indicator data of the user.

[0009] Optionally, in some embodiments, the intelligent decision-making unit is further configured to: execute a personalized intervention strategy, the personalized intervention strategy including: identifying one or more target health improvement dimensions of the user based on the health record data; setting intervention weight coefficients based on each of the target health improvement dimensions, the intervention weight coefficients being determined based on the degree of deviation between the user's real-time physiological data and historical baseline data; and using the intervention weight coefficients as adjustment factors to regulate the range and intensity of the environmental parameters and / or the far-infrared radiation within the containment device.

[0010] Optionally, in some embodiments, regulating the environmental parameters within the containment device includes adjusting at least one of the following: temperature, humidity, airflow intensity, negative oxygen ion concentration, or light intensity within the containment device.

[0011] Optionally, in some embodiments, the housing is a cabin structure, which includes a bed for carrying a user and a cover movably connected to the bed.

[0012] Secondly, this application also provides an artificial intelligence-based health and wellness control method, applicable to the artificial intelligence-based health and wellness system described in the first aspect above. The health and wellness control method includes: acquiring a first physiological time-series signal and a second physiological time-series signal of the user through a data acquisition module; fusing the first physiological time-series signal and the second physiological time-series signal to obtain the user's heart rate and respiratory rate; and generating control commands through a pre-trained network model based on the user's heart rate, respiratory rate, health record data, and environmental parameters of the containment device. The control commands can regulate the environmental parameters within the containment device and / or the wavelength range and intensity of the far-infrared radiation.

[0013] Optionally, in some embodiments, generating control commands through a pre-trained network model includes: inputting time-series data consisting of the heart rate, the respiratory rate, the health record data, and the environmental parameters into a long short-term memory network model; outputting a predicted value for at least one physiological state of the user at the next moment through the long short-term memory network model; calculating an adjustment amount for the environmental parameters of the containment device and / or the wavelength range and intensity of the far-infrared radiation based on the difference between the predicted value and a preset target range corresponding to the physiological state; and generating the control commands based on the adjustment amount.

[0014] Thirdly, this application also provides a computer storage medium that stores programs or instructions that cause a computer to execute the artificial intelligence-based health and wellness control method described in the second aspect above.

[0015] The technical solution provided in this application has the following advantages compared with the prior art:

[0016] The AI-based health and wellness system provided in this application embodiment achieves non-contact monitoring primarily through a first and second data acquisition unit. Then, a signal fusion unit fuses the acquired first and second physiological time-series signals to obtain the user's heart rate and respiratory rate. An intelligent decision-making unit can generate control commands based on the user's heart rate, respiratory rate, health record data, and environmental parameters of the accommodation device, using a pre-trained network model. These control commands can adjust the environmental parameters and / or the wavelength range and intensity of far-infrared radiation within the accommodation device. This control command dynamically adjusts the health and wellness environment according to each user's health status, thereby achieving precise health and wellness care.

[0017] Furthermore, the data acquisition module primarily uses non-contact monitoring, thereby improving user comfort and experience. Attached Figure Description

[0018] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0019] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, those skilled in the art can obtain other drawings based on these drawings without any creative effort.

[0020] Figure 1 A schematic diagram of the structure of the AI-based elderly care system provided in the embodiments of this application;

[0021] Figure 2 A flowchart illustrating the AI-based elderly care control method provided in this application embodiment;

[0022] Figure 3 This is a schematic diagram of the structure of the Long Short-Term Memory network model provided in the embodiments of this application;

[0023] Figure 4 This is a schematic diagram of the structure of the computer storage medium provided in the embodiments of this application. Detailed Implementation

[0024] To better understand the above-mentioned objectives, features, and advantages of this application, the solution of this application will be further described below. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0025] Many specific details are set forth in the following description in order to provide a full understanding of this application, but this application may also be implemented in other ways different from those described herein; obviously, the embodiments in the specification are only some embodiments of this application, and not all embodiments.

[0026] The following description, in conjunction with the accompanying drawings, provides an exemplary account of the AI-based elderly care system, elderly care control method, and computer storage medium provided in the embodiments of this application. Figure 1 A schematic diagram of the structure of an artificial intelligence-based elderly care system provided in an embodiment of this application. (Refer to...) Figure 1 An AI-based health and wellness system 10 includes: a housing device 11 for providing a health and wellness space for users; an adjustable far-infrared generation module 12 disposed within the housing device 11, the adjustable far-infrared generation module 12 including a first excitation unit and a second excitation unit, wherein the second excitation unit is used to excite far-infrared radiation with adjustable spectral characteristics through the far-infrared rays generated by the first excitation unit; and a data acquisition module 13 including a first acquisition unit and a second acquisition unit, wherein the first acquisition unit is used to acquire the user's facial video stream and obtain a first physiological time-series signal based on the facial video stream.

[0027] The second acquisition unit is used to acquire the user's cardiac impact signal and obtain a second physiological time-series signal based on the cardiac impact signal; the intelligent control module 14 includes a signal fusion unit and an intelligent decision unit. The signal fusion unit is used to fuse the first physiological time-series signal and the second physiological time-series signal to obtain the user's heart rate and respiratory rate; the intelligent decision unit is used to generate control commands through a pre-trained network model based on the user's heart rate, respiratory rate, health record data, and environmental parameters of the containment device. The control commands can regulate the environmental parameters within the containment device and / or the wavelength range and intensity of the far-infrared radiation.

[0028] Specifically, the accommodation device 11 can be a closed or semi-closed space to facilitate health and wellness treatments. The accommodation device 11 can be, for example, an energy capsule, a health and wellness chair, a health and wellness bed, or a health and wellness cabin.

[0029] In one embodiment, the housing device 11 may be a cabin structure, which may include a bed for carrying a user and a cover movably connected to the bed.

[0030] Specifically, the housing device 11 can be a cabin structure that provides a comfortable and functional space for users to receive health and wellness treatments. This cabin structure may include a bed to support the user's body, ensuring comfort and stability during the treatment. The bed can be adjusted to accommodate different user body sizes and postural preferences. Furthermore, the cabin structure may include a cover that is movably connected to the bed. This cover can be transparent or semi-transparent, allowing light to pass through while protecting the user's privacy. The cover can be designed to be openable and closable for easy entry and exit, and can be completely closed when needed to create a relatively independent space for controlling environmental parameters such as temperature, humidity, and light.

[0031] The adjustable far-infrared generating module 12 is located inside the housing 11 and may include a first excitation unit and a second excitation unit. The first excitation unit may be a far-infrared light source such as a heating wire, a carbon fiber heater, an infrared LED array, or a halogen lamp, used to generate basic far-infrared radiation. The second excitation unit may be a graphene coating, an optical filter, a grating or diffraction grating, or an optical fiber system, etc. The second excitation unit can further excite and adjust the far-infrared radiation generated by the first excitation unit, thereby outputting far-infrared radiation with specific spectral characteristics. By combining the first and second excitation units, the adjustable far-infrared generating module 12 can control the wavelength, bandwidth, and intensity of far-infrared radiation to meet the personalized health and wellness needs of different users.

[0032] In one embodiment, the first excitation unit may be a graphene-based heating element, and the second excitation unit may be a composite substrate containing negative ion powder, tourmaline powder, and / or terahertz materials.

[0033] Specifically, the first excitation unit can be a graphene-based heating element. The graphene-based heating element utilizes the excellent electrothermal conversion efficiency and rapid thermal response characteristics of graphene to quickly generate far-infrared rays, which helps to promote blood circulation and accelerate metabolism.

[0034] The second excitation unit can be a composite substrate containing negative ion powder, tourmaline powder and / or terahertz materials. This composite substrate can not only enhance and regulate the far-infrared rays generated by the first excitation unit, but also improve the health and wellness effects by releasing negative ions and possible terahertz waves. At the same time, the addition of tourmaline powder can improve the therapeutic effect of far-infrared rays.

[0035] The data acquisition module 13 may include a first acquisition unit and a second acquisition unit. The core of the first acquisition unit is a visible light camera (such as an RGB camera).

[0036] The first acquisition unit may include a camera (such as an RGB camera). The first acquisition unit acquires the user's facial video stream in a non-contact manner. The video stream contains information on the slight color changes (changes in light absorption characteristics) of the facial skin area caused by the periodic changes in subcutaneous blood volume with heartbeat and respiration.

[0037] Then, the facial video stream can be processed using remote photoplethysmography (rPPG) to obtain the first physiological temporal signal. Specifically, key facial regions (such as the forehead and cheeks) can be located using computer vision algorithms, and the pixel color values ​​of these key regions can be tracked over time. By performing time-frequency analysis (such as blind source separation and filtering) on ​​the color change sequence, periodic optical intensity modulation signals corresponding to arterial pulsation (heartbeat) and respiratory movements can be extracted, which is the first physiological temporal signal. The first physiological temporal signal can reflect the optical pulse waves of cardiovascular and respiratory system activities.

[0038] The second acquisition unit can be, for example, a piezoelectric sensor, which can be installed in the load-bearing structure (such as a bed or seat) that the user comes into contact with, and collect cardiac impact signals (BCG) through the piezoelectric sensor.

[0039] Then, through signal conditioning circuitry (such as amplification and bandpass filtering), the BCG analog signal can be converted into a second physiological time-series signal. This second physiological time-series signal contains rhythmic characteristics synchronized with the cardiac mechanical activity cycle, as well as low-frequency amplitude variation information modulated by respiration. By analyzing the second physiological time-series signal (such as peak detection and time-frequency analysis), the heart rate interval sequence and respiratory rhythm information can be extracted, thereby calculating heart rate, heart rate variability, and respiratory rate.

[0040] The data acquisition module 13 performs detection primarily through non-contact methods and secondarily through contact methods, thereby improving user comfort.

[0041] The intelligent control module 14 is the core component of the health and wellness system 10. It achieves intelligent management and control of the entire health and wellness system 10 through the signal fusion unit and the intelligent decision-making unit.

[0042] The signal fusion unit is used to fuse the first and second physiological time-series signals to obtain the user's heart rate and respiratory rate. Specifically, the first and second physiological time-series signals can be time-aligned and their sampling rates unified, and their respective signal quality can be evaluated in real time to dynamically allocate fusion weights. Then, the first and second physiological time-series signals can be converted to the frequency domain and multiplied together. The spectral peaks shared by both signals and corresponding to the physiological rhythm are enhanced, while the noise peaks unique to each individual signal are relatively suppressed. The heart rate (typically peak-finding in the 0.67-3Hz range) and respiratory rate (typically peak-finding in the 0.1-0.5Hz range) extracted from the enhanced fused spectrum have better accuracy and anti-interference capabilities than those extracted from either individual spectrum.

[0043] In addition, cross-validation of the cardiac events of the first and second physiological time-series signals can be performed in the time domain to eliminate false positives and improve the accuracy of predicting the user's heart rate and respiratory rate.

[0044] The intelligent decision-making unit generates control commands through a pre-trained network model based on the user's heart rate and respiratory rate, combined with health record data and environmental parameters of the enclosure. These control commands can, for example, adjust environmental parameters (such as temperature and humidity) within the enclosure 11 via environmental control and far-infrared control interfaces, and / or regulate the radiation range and intensity of the far-infrared generating module 12, ensuring that the health and wellness environment meets the user's health requirements.

[0045] For example, regulating the environmental parameters within the containing device 11 may include adjusting at least one of the following: temperature, humidity, airflow intensity, negative oxygen ion concentration, or light intensity within the containing device 11.

[0046] Specifically, the control commands can adjust the environmental parameters within the containment device 11 to meet the personalized health and wellness needs of different users. These commands may include adjusting the temperature within the containment device 11 to ensure the ambient temperature is maintained within the most suitable range for the human body; adjusting humidity to help maintain air moisture and prevent excessively dry or humid air from affecting the user's respiratory and skin health; adjusting airflow intensity to control the speed of airflow, promote air circulation, keep the air fresh, and avoid direct airflow causing discomfort to the user; adjusting the concentration of negative oxygen ions to improve air quality, enhance the body's antioxidant capacity, help relieve stress, and improve sleep; and adjusting light intensity to simulate natural lighting conditions or adjust the light intensity and color according to the user's preferences and health and wellness needs, which is conducive to the user's physical and mental relaxation.

[0047] In one embodiment, the health and wellness system 10 may further include: an environmental control module, disposed within the housing 11, for monitoring and adjusting environmental parameters within the housing 11; and a data storage module, for storing the user's health record data.

[0048] Specifically, the environmental control module monitors and regulates the environmental conditions within the containment unit 11 to ensure a suitable health and wellness environment for users. This module may include a temperature and humidity convection control system, a lighting control system, and multiple environmental monitoring sensors, such as carbon dioxide and volatile organic compound (VOC) sensors. The temperature and humidity convection control system may consist of, for example, an air conditioner, a heating element, a humidifier, and a convection fan. The air conditioner controls the temperature and humidity reduction, the heating element increases the temperature, the humidifier increases the humidity, and the convection fan increases the airflow velocity within the chamber. The control system and sensors work together to monitor the temperature, humidity, air quality, and lighting conditions within the chamber in real time.

[0049] The data storage module may be one or more databases, used to store basic user information, medical history records, physiological parameter monitoring results, details of health care plans, and health care effect evaluation reports, etc.

[0050] For example, the health record data may include: the user's basic information and health information obtained through instrument detection, medical reports provided by the user, or consultation, and the health information may include: the user's physiological and biochemical indicator data.

[0051] Specifically, user health information can be obtained through instrument testing, user-provided medical reports, or consultations. Health information may include physiological and biochemical indicators such as heart rate, blood pressure, blood sugar, blood lipids, and uric acid. Basic information may include age, gender, height, weight, and medical history.

[0052] In one embodiment, the intelligent decision-making unit is further configured to: execute a personalized intervention strategy, the personalized intervention strategy including: identifying one or more target health improvement dimensions of the user based on the health record data; setting intervention weight coefficients based on each target health improvement dimension, the intervention weight coefficients being determined based on the degree of deviation between the user's current physiological data and historical baseline data; and using the intervention weight coefficients as adjustment factors to regulate the range and intensity of the environmental parameters and / or the far-infrared radiation within the containment device.

[0053] Specifically, personalized intervention strategies can include identifying one or more target health improvement dimensions based on user health record data (such as diagnosed disease "hypertension", abnormal physical examination item "high LDL cholesterol", self-reported symptoms "sleep disorder", etc.). For example, the aforementioned health record data can be used to identify dimensions such as cardiovascular load management, metabolic regulation, nerve relaxation, and sleep promotion.

[0054] For each identified dimension of health improvement, corresponding physiological baseline data can be obtained from the user's historical physiological data. For example, for the cardiovascular load management dimension, the baseline data might be the user's long-term monitored resting heart rate range and normal heart rate variability.

[0055] Based on the user's current physiological data (such as current heart rate, heart rate variability, etc.), the degree of deviation between the current physiological data and the baseline data of the corresponding dimension is calculated. The degree of deviation can be quantified as a relative percentage change, a multiple of standard deviation, or a gradation based on a preset threshold.

[0056] The intervention weighting coefficient is positively correlated with the degree of deviation. The greater the deviation, the more serious the user's current state deviates from their personal health baseline, and therefore a higher weighting coefficient is assigned. For example, when a user's heart rate is significantly higher than their resting baseline, the weighting coefficient for the cardiovascular load management dimension is increased.

[0057] Using the intervention weight coefficient as a regulating factor, a mapping relationship of regulation parameters is obtained. This parameter mapping relationship can define the environmental parameters and far-infrared radiation parameters for each health improvement dimension, thereby obtaining the target values ​​of environmental parameters (such as target temperature and target humidity) and far-infrared radiation parameters (such as target wavelength range and target intensity).

[0058] In summary, the AI-based health and wellness system provided in this application embodiment achieves non-contact monitoring primarily through the first and second acquisition units of the data acquisition module. Then, the acquired first and second physiological time-series signals are fused and processed by the signal fusion unit to obtain the user's heart rate and respiratory rate. The intelligent decision-making unit can generate control commands based on the user's heart rate, respiratory rate, health record data, and environmental parameters of the accommodation device through a pre-trained network model. These control commands can adjust the environmental parameters and / or the wavelength range and intensity of far-infrared radiation within the accommodation device. These control commands can dynamically adjust the health and wellness environment according to each user's health status, thereby achieving precise health and wellness care.

[0059] Furthermore, the data acquisition module primarily uses non-contact monitoring, thereby improving user comfort and experience.

[0060] Based on the same inventive concept, this application also provides an artificial intelligence-based health and wellness control method 20. Figure 2 This is a flowchart illustrating the artificial intelligence-based elderly care control method provided in an embodiment of this application. Figure 2 As shown, the health and wellness control method 20 includes the following steps:

[0061] S21. The user's first physiological time-series signal and second physiological time-series signal are collected through the data acquisition module 13.

[0062] Specifically, the user's first physiological time-series signal and second physiological time-series signal are collected by the data acquisition module 13, which has been described in detail in the above embodiments and will not be repeated here.

[0063] S22. The first physiological time sequence signal and the second physiological time sequence signal are fused to obtain the user's heart rate and respiratory rate.

[0064] For example, in practical applications, remote photoplethysmography (rPPG) and cardiac impulse (BCG) signals can be fused to obtain more accurate physiological data.

[0065] S23. Based on the user's heart rate, respiratory rate, health record data, and environmental parameters of the containment device, control commands are generated through a pre-trained network model. The control commands can regulate the environmental parameters within the containment device and / or the wavelength range and intensity of the far-infrared radiation.

[0066] Specifically, the user's current heart rate, respiratory rate, health record data obtained in step S22, and the environmental parameters of the containment device 11 are used as input data and fed into a pre-trained network model. The network model then generates specific control commands. The pre-trained network model can be a deep learning algorithm, such as a convolutional neural network (CNN) for image data, or a recurrent neural network (RNN) or long short-term memory network (LSTM) for time series data, to identify complex patterns and correlations in the data. After training, the network model can analyze the interactions between different data types and predict the most suitable intervention strategy for the user.

[0067] The control commands may include adjusting the radiation intensity and wavelength of the far-infrared generating module; they may also include adjusting environmental parameters within the containment device 11 to change temperature, humidity, or light intensity. These control commands are sent to corresponding execution modules, such as the far-infrared generator, air conditioning system, and lighting equipment, to achieve environmental control within the containment device 11.

[0068] For example, the step of generating control commands through a pre-trained network model includes: inputting time-series data consisting of the heart rate, the respiratory rate, the health record data, and the environmental parameters into a long short-term memory network model; outputting a predicted value for at least one physiological state of the user at the next moment through the long short-term memory network model; calculating an adjustment amount for the environmental parameters of the containment device and / or the wavelength range and intensity of the far-infrared radiation based on the difference between the predicted value and a preset target range corresponding to the physiological state; and generating the control commands based on the adjustment amount.

[0069] Specifically, the user's heart rate, respiratory rate, environmental parameters within the containment device (temperature, humidity, etc.), and static features obtained from health records (such as risk labels) are aligned using a unified timestamp to generate multi-dimensional time-series data. This time-series data is then input into a Long Short-Term Memory (LSTM) network model. The LSTM model can receive the time-series data using a sliding time window approach. For example, data points from the past 60 seconds, grouped per second (forming a 60×N dimensional matrix, where N is the total number of features), can be used as input.

[0070] The special gating structure of LSTM network models (input gate, forget gate, output gate) enables them to learn and remember long-term dependencies across time steps. For example, it can learn that when the temperature rises slowly within a certain range, a particular user's heart rate will typically begin to show a predictable upward dynamic pattern after about 30 seconds.

[0071] Based on learning from sequences within past time windows, LSTM models can output predicted values ​​(such as heart rate and respiratory rate) of a user's physiological state in the next moment (10 or 30 seconds). The predicted values ​​represent the most likely direction of the user's physiological state under all current input conditions, without intervention.

[0072] You can also set target ranges for each predictable physiological state. These target ranges can be determined based on the user's health record data (such as resting heart rate range) or wellness goals (such as respiratory rate range in a target relaxed state).

[0073] The predicted value output by the LSTM model is compared with the corresponding target range for physiological state, and the difference or degree of deviation is calculated. For example, if the predicted heart rate is 85 beats / min, while the user's target resting heart rate range is 60-75 beats / min, there is a deviation. Based on the magnitude and direction of the deviation, the required intervention level to bring the predicted value back to the target range is output, i.e., the adjustment amount of environmental parameters and far-infrared radiation parameters. Control commands can be generated based on the adjustment amount.

[0074] Figure 3 This is a schematic diagram of the structure of the Long Short-Term Memory network model provided in the embodiments of this application. Figure 3 In this framework, each parameter's subscript indicates the nth time segment. Ph represents physiological parameters, including heart rate, heart rate variability, and respiration. Env represents environmental parameters, including T, RH, WS, NOI, Lx, etc. Us represents basic user parameters, which are comprehensively scored by professionals after understanding user information, considering multiple dimensions (aging index, balance index, cardiovascular index, self-care level, brain function index, etc.). h represents the hidden layer of the LSTM neural network. Wrn represents whether the user's metabolism is excessive. Set represents the parameters for intelligently setting external devices, such as whether the air conditioner is set to a certain temperature, the humidifier is set to a certain level, the convection fan speed, and the heating element power level. Rs represents the user's recovery index and the comprehensive physical score expected at the next health and wellness stage. Pph represents the predicted physiological parameter values ​​(such as respiration and heart rate) at the next time segment.

[0075] In practical applications, the Long Short-Term Memory (LSTM) network model can be referenced. Figure 3 The LSTM model is used to process and analyze time series data to generate control commands and optimize health and wellness outcomes. In this LSTM model, each parameter corresponds to a specific number of time slices. For example, the total duration of a health and wellness session can be set to 1 hour, and each time slice can be set to 30 seconds, meaning there are 120 slices within the total duration of a health and wellness session. This LSTM network model can capture the changes in physiological and environmental parameters over time.

[0076] Furthermore, in the Long Short-Term Memory (LSTM) network model provided in this application embodiment, the network loss function is calculated using the following formula:

[0077] ;

[0078] In the formula, Let be the network loss function. The predicted recovery index at the end of the wellness retreat. This is a comprehensive rating of users by professionals. To adjust the parameter, the value is taken between 0.002 and 0.005, and this value can be obtained through a limited number of experiments by those skilled in the art. This is the physiological parameter value to be measured next time (there may be multiple sets, so take the average value). These are the predicted physiological parameter values ​​(there may be multiple sets, so the average value is taken). This represents the total number of time slices. This is the i-th time slice.

[0079] The AI-based health and wellness control method provided in the above embodiments can execute the AI-based health and wellness control system provided in the above embodiments and has the same or corresponding beneficial effects, which will not be described in detail here.

[0080] This application also provides a computer storage medium 400, Figure 4 A schematic diagram of a computer storage medium according to an embodiment of this application is shown. For example... Figure 4 As shown, the storage medium 400 stores a computer program 401, which, when executed by a processor, can implement the artificial intelligence-based elderly care control method described in any of the embodiments above. It should be understood that, in this embodiment, the aforementioned computer storage medium may be located at at least one of multiple network servers in a computer network. Optionally, in this embodiment, the aforementioned storage medium may include, but is not limited to, various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0081] It should be understood that, in this embodiment, the aforementioned computer storage medium may be located at at least one of the multiple network servers in a computer network. Optionally, in this embodiment, the aforementioned storage medium may include, but is not limited to, various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0082] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments.

[0083] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of those different embodiments or examples.

[0084] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0085] In this application, unless otherwise stated, directional terms such as "up" and "down" are generally used in relation to the direction shown in the accompanying drawings, or in relation to the vertical, perpendicular, or gravitational direction; similarly, for ease of understanding and description, "left" and "right" are generally used in relation to the left and right shown in the accompanying drawings; "inner" and "outer" refer to the inner and outer contours of each component itself, but the above directional terms are not intended to limit this application.

[0086] The above description is merely an exemplary embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope described in this application, and these should all be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. An artificial intelligence-based health and wellness system, characterized in that, The system comprises: a housing device for providing a health care space for a user; an adjustable far infrared generating module arranged in the housing device, the adjustable far infrared generating module comprising a first excitation unit and a second excitation unit, wherein the second excitation unit is configured to excite far infrared radiation with adjustable output spectral characteristics by far infrared generated by the first excitation unit; a data acquisition module comprising a first acquisition unit and a second acquisition unit, wherein the first acquisition unit is configured to acquire a facial video stream of the user and obtain a first physiological time series signal based on the facial video stream; the second acquisition unit is configured to acquire a ballistocardiogram of the user and obtain a second physiological time series signal based on the ballistocardiogram; an intelligent control module, the intelligent control module comprising a signal fusion unit and an intelligent decision unit, the signal fusion unit being configured to fuse and process the first physiological time series signal and the second physiological time series signal to obtain a heart rate and a respiration rate of the user; the intelligent decision unit being configured to generate a control instruction based on the heart rate, the respiration rate, health record data of the user, and environmental parameters of the housing device by using a pre-trained network model, the control instruction being capable of adjusting the environmental parameters in the housing device and / or the wavelength range and intensity of the far infrared radiation.

2. The AI-based health and wellness system of claim 1, wherein, The health care system further comprises: an environmental regulation module arranged in the housing device, configured to monitor and adjust the environmental parameters in the housing device; a data storage module configured to store the health record data of the user.

3. The health care system based on artificial intelligence according to claim 2, wherein the first excitation unit is a graphene-based heating element, and the second excitation unit is a composite substrate containing negative ion powder, tourmaline powder, and / or terahertz material.

4. The AI-based health and wellness system of claim 3, wherein, The health record data comprises: basic information and health information of the user obtained through instrument detection, medical reports or inquiries provided by the user, the health information comprising physiological and biochemical index data of the user.

5. The AI-based health and wellness system of claim 4, wherein, The intelligent decision unit is further configured to execute a personalized intervention strategy, the personalized intervention strategy comprising: identifying one or more target health improvement dimensions of the user based on the health record data; setting an intervention weight coefficient based on each target health improvement dimension, the intervention weight coefficient being determined based on the deviation degree between real-time physiological data and historical baseline data of the user; adjusting the environmental parameters in the housing device and / or the range and intensity of the far infrared radiation by using the intervention weight coefficient as an adjustment factor.

6. The AI-based health and wellness system of claim 5, wherein, The adjustment of the environmental parameters in the housing device comprises: adjusting at least one of the temperature, humidity, air flow intensity, negative oxygen ion concentration, or light intensity in the housing device.

7. The AI-based health and wellness system of claim 1, wherein, The housing device is a cabin structure, the cabin structure comprising a bed body for carrying the user and a cover body movably connected to the bed body.

8. A health care control method based on artificial intelligence, applicable to the health care system based on artificial intelligence according to any one of claims 1 to 7, the health care control method comprising: acquire a first physiological time sequence signal and a second physiological time sequence signal of a user through a data acquisition module; fuse the first physiological time sequence signal and the second physiological time sequence signal to obtain a heart rate and a breathing rate of the user; generate a control instruction through a pre-trained network model according to the heart rate, the breathing rate, health record data, and an environmental parameter of the accommodating device, the control instruction being capable of regulating the environmental parameter in the accommodating device and / or a wavelength range and intensity of the far infrared radiation.

9. The AI-based health and wellness control method of claim 8, wherein, The generating of the control instruction through the pre-trained network model comprises: inputting time sequence data composed of the heart rate, the breathing rate, the health record data, and the environmental parameter into a long short-term memory network model; outputting a prediction value of at least one physiological state of the user at a next moment through the long short-term memory network model; calculating an adjustment amount of the environmental parameter in the accommodating device and / or the wavelength range and intensity of the far infrared radiation based on a difference between the prediction value and a preset target range corresponding to the physiological state; generating the control instruction based on the adjustment amount.

10. A computer storage medium, characterized in that, The computer storage medium stores programs or instructions, and the programs or instructions enable the computer to execute the artificial intelligence-based health care control method according to any one of claims 8 to 9.

Images