Health monitoring method, device, headset and health monitoring system

Abstract

The application relates to the technical field of wearable devices, and discloses a health monitoring method and device, a head-mounted earphone and a health monitoring system. The method comprises the following steps: collecting the current temperature of a user wearing the device by a first collecting component, and determining a temperature change value at the current time according to the current temperature; collecting health influence data of the user wearing the device by a second collecting component when the temperature change value reaches a preset change threshold, wherein the health influence data comprises at least one of physiological data, electroencephalogram data and a facial image; taking the health influence data as an influence factor, adjusting the current temperature by the influence factor, and monitoring the health of the user wearing the device by the adjusted current temperature. Since at least one health influence data of the user is taken as an influence factor to adjust the temperature of the user, the measured temperature of the user can be prevented from being disturbed by other factors, so that the accuracy of health evaluation can be improved.

Description

Technical Field

[0001] This application relates to the field of wearable device technology, and more particularly to health monitoring methods, devices, headphones, and health monitoring systems. Background Technology

[0002] With the continuous advancement of technology, health monitoring technology plays an increasingly important role in people's daily lives. Body temperature, as a crucial indicator of human health, can reflect various physiological and pathological states. Therefore, existing health monitoring technologies can assess a user's health status by measuring their body temperature. However, while this method can assess a user's health status to some extent, body temperature is not only affected by disease conditions but is also easily interfered with by other factors, resulting in low accuracy and reliability of temperature measurements, and thus making it difficult to accurately assess a user's health status. Summary of the Invention

[0003] The main purpose of this application is to provide a health monitoring method, device, headset, and health monitoring system, which aims to solve the technical problem that existing health monitoring technologies are prone to inaccurate body temperature measurement due to interference, thereby affecting the accuracy of health assessment.

[0004] To achieve the above objectives, this application proposes a health monitoring method, which is applied to a headset in a health monitoring system. The headset is equipped with a first data acquisition component, and the health monitoring system further includes a second data acquisition component. The method includes: The first acquisition component acquires the wearer's current temperature and determines the temperature change value at the current moment based on the current temperature. When the temperature change value reaches a preset change threshold, the second acquisition component collects the health impact data of the wearer, which includes at least one of physiological data, electroencephalogram data, and facial images. The health impact data is used as an influencing factor to adjust the current temperature, and the adjusted current temperature is used to monitor the health of the wearer.

[0005] In one embodiment, the health impact data includes the physiological data, the electroencephalogram (EEG) data, and the facial image; The step of using the health impact data as an influencing factor to adjust the current temperature includes: The wearer's current heart rate is determined based on the physiological data, and the wearer's current emotion category is determined based on the electroencephalogram (EEG) data and the facial image. The current heart rate and the current emotion category are input as influencing factors into a preset neural network model for influence detection, and the influence detection results are obtained. Based on the impact detection results, it is determined whether the influencing factor affects the current temperature. If it does, the current temperature is adjusted using the influencing factor.

[0006] In one embodiment, the step of determining the current emotion category of the wearer based on the EEG data and the facial image includes: Feature extraction is performed on the EEG data to obtain multi-dimensional features of the EEG data, including frequency domain features and time domain features; Feature extraction is performed on the facial image to obtain the facial expression features of the facial image; The frequency domain features, the time domain features, and the facial expression features are concatenated to obtain the emotional features of the wearer. The emotional score of the wearer is determined based on the emotional characteristics, and the current emotional category of the wearer is determined based on the emotional score.

[0007] In one embodiment, the step of extracting features from the facial image to obtain facial expression features of the facial image includes: Perform face detection on the facial image to determine the face region in the facial image, and extract the global expression features of the face region; Identify the key expression regions of the face region and extract the local expression features of the key expression regions; The global expression features and the local expression features are fused to obtain the facial expression features of the facial image.

[0008] In one embodiment, prior to the step of determining the wearer's emotional score based on the emotional characteristics, the method further includes: A user profile is constructed based on the user's basic information, and the emotional category classification criteria for the user are determined based on the user profile. The step of determining the emotional score of the wearer based on the emotional characteristics includes: The current emotional category of the wearer is determined based on the emotional score and the emotional category classification criteria.

[0009] In one embodiment, the step of adjusting the current temperature using the influencing factor includes: The heart rate change rate and mood change value of the wearer at the current moment are determined by the influencing factors. Construct a real-time influencing factor vector for the wearer user based on the heart rate variability and the emotion variability values; The temperature adjustment amount for the wearer is determined based on the real-time influencing factor vector using a preset body temperature offset prediction model. The current temperature is adjusted according to the temperature adjustment amount.

[0010] In one embodiment, before the step of adjusting the current temperature according to the temperature adjustment amount, the method further includes: Extract key individual characteristics of the wearer from the user profile; Extract key features of influencing factors of the wearable user from the real-time influencing factor vector; Determine the first influence coefficient corresponding to the key characteristics of the individual, and the second influence coefficient corresponding to the key characteristics of the influencing factors; The dynamic adjustment coefficient is determined based on the individual key characteristics, the first influence coefficient, the key characteristics of the influencing factors, and the second influence coefficient. The step of adjusting the current temperature according to the temperature adjustment amount includes: The current temperature is adjusted according to the temperature adjustment amount and the dynamic adjustment coefficient.

[0011] Furthermore, to achieve the above objectives, this application also proposes a health monitoring device, the device comprising: The data acquisition module is used to acquire the current temperature of the wearer through the first data acquisition component, and determine the temperature change value at the current moment based on the current temperature; The acquisition module is also used to acquire health impact data of the wearer through the second acquisition component when the temperature change value reaches a preset change threshold. The health impact data includes at least one of physiological data, electroencephalogram data and facial images. The monitoring module is used to use the health impact data as an influencing factor, adjust the current temperature using the influencing factor, and monitor the health of the wearer using the adjusted current temperature.

[0012] Furthermore, to achieve the above objectives, this application also proposes a headset, which includes: a memory, a processor, and a health monitoring program stored in the memory and executable on the processor. When the health monitoring program is executed by the processor, it implements the steps of the health monitoring method described above.

[0013] In addition, to achieve the above objectives, this application also proposes a health monitoring system, which includes headphones as described above and a second data acquisition component.

[0014] This application provides a health monitoring method, device, headset, and health monitoring system. The method is applied to a headset in a health monitoring system. The headset is equipped with a first acquisition component, and the health monitoring system further includes a second acquisition component. The method includes: acquiring the current temperature of a wearer through the first acquisition component, and determining a temperature change value at the current moment based on the current temperature; when the temperature change value reaches a preset change threshold, acquiring health impact data of the wearer through the second acquisition component, the health impact data including at least one of physiological data, electroencephalogram (EEG) data, and facial images; using the health impact data as an influencing factor, adjusting the current temperature based on the influencing factor, and performing health monitoring on the wearer based on the adjusted current temperature.

[0015] This application can incorporate a first data acquisition component in a headset. In actual use, this first component can acquire the wearer's current temperature and determine the temperature change value at that moment. If the temperature change value reaches a preset threshold, a second data acquisition component can acquire at least one health impact data point from the wearer. This health impact data is then used as an influencing factor to adjust the current temperature, enabling health monitoring based on the adjusted current temperature. Because this application can acquire at least one health impact data point from the wearer when the temperature change value reaches the preset threshold, and adjust the current temperature based on this data, it avoids inaccurate temperature measurements due to interference from other factors, thereby improving the accuracy of health assessments. Attached Figure Description

[0016] 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.

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of the headphone structure of the hardware operating environment involved in the embodiments of this application; Figure 2 This is a flowchart illustrating the first embodiment of the health monitoring method of this application; Figure 3 This is a flowchart illustrating the second embodiment of the health monitoring method of this application; Figure 4 This is a flowchart illustrating the third embodiment of the health monitoring method of this application; Figure 5 This is a flowchart illustrating the overall process of health monitoring for users in the health monitoring method described in this application. Figure 6 This is a structural block diagram of the first embodiment of the health monitoring device of this application.

[0019] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0020] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0021] Reference Figure 1 , Figure 1 This is a schematic diagram of the headphone structure of the hardware operating environment involved in the embodiments of this application.

[0022] like Figure 1 As shown, the headset may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may be connected to a display screen; optionally, the user interface 1003 may include a standard wired interface or a wireless interface. In this application, the wired interface of the user interface 1003 may be a USB interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or non-volatile memory (NVM), such as a disk storage device. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0023] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on headphones and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0024] like Figure 1 As shown, the memory 1005, which is identified as a computer storage medium, may include an operating system, a network communication module, a user interface module, and a health monitoring program.

[0025] exist Figure 1 In the headset shown, the network interface 1004 is mainly used to connect to the backend server and communicate data with the backend server; the user interface 1003 is mainly used to connect to the user device; the headset calls the health monitoring program stored in the memory 1005 through the processor 1001 and executes the steps of the health monitoring method provided in the embodiments of this application.

[0026] It should be noted that with the continuous advancement of technology, health monitoring technology plays an increasingly important role in people's daily lives. Body temperature, as an important indicator of human health, can reflect various physiological and pathological states. Therefore, existing health monitoring technologies can assess a user's health status by measuring their body temperature. However, while this method can assess a user's health status to a certain extent, body temperature is not only affected by disease states but is also easily interfered with by other factors, resulting in low accuracy and reliability of temperature measurements, and thus making it difficult to accurately assess a user's health status.

[0027] Therefore, to address the aforementioned shortcomings, this embodiment incorporates a first data acquisition component within the headset. In actual use, this first component acquires the wearer's current temperature and determines the temperature change value at that moment. If this temperature change value reaches a preset threshold, a second data acquisition component can acquire at least one health impact data point from the wearer. This health impact data is then used as an influencing factor to adjust the current temperature, enabling health monitoring based on the adjusted current temperature. Because this embodiment can acquire at least one health impact data point from the wearer when the temperature change value reaches the preset threshold, and adjust the current temperature based on this data, it avoids inaccurate temperature measurements due to interference from other factors, thereby improving the accuracy of health assessments.

[0028] For ease of understanding, the following is combined with Figures 2 to 5 The health monitoring method provided in the embodiments of this application will be described in detail.

[0029] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the health monitoring method of this application. The first embodiment of the health monitoring method of this application is presented as follows: Figure 2 As shown, in this embodiment, the method is applied to a headset in a health monitoring system. The headset is equipped with a first data acquisition component, and the health monitoring system further includes a second data acquisition component. The specific method includes: Step S10: Collect the current temperature of the wearer through the first acquisition component, and determine the temperature change value at the current moment based on the current temperature.

[0030] It is understood that the method of this embodiment can be applied to the aforementioned headset, which may have functions such as data processing, program execution, and data acquisition. This embodiment does not limit this. The execution subject of this method embodiment can be the aforementioned headset to illustrate this embodiment and the following embodiments.

[0031] Understandably, the first data acquisition component can be a sensor or module in the headphones used to measure the user's body temperature, such as an infrared sensor, a contact temperature sensor, or a photoelectric sensor; this embodiment does not limit this. Specifically, an infrared sensor can measure body temperature by detecting infrared radiation emitted by the human body; a contact temperature sensor (such as a thermistor (NTC), thermocouple, etc.) can measure body temperature by directly contacting the user's skin; and a photoelectric sensor can indirectly reflect the user's body temperature by measuring changes in blood flow.

[0032] It should be understood that the aforementioned wearable user can be a user wearing the aforementioned headphones. In this embodiment, both the first and second acquisition components can be located within the headphones. If it is necessary to acquire the user's facial image through the second acquisition component, it can be located within the headphone case. Furthermore, the health monitoring system can also include another smart wearable device (such as a smart bracelet), with the first and second acquisition components respectively located within the headphones and the other smart wearable device. Subsequently, when the user wears both the headphones and the other smart wearable device simultaneously, health monitoring for that user begins.

[0033] It should also be understood that the aforementioned current temperature can be the body temperature of the wearer measured by the first acquisition component at the current moment. Here, "current moment" can be the specific point in time when the temperature measurement was performed, which is usually recorded using the system time or UTC time of the headset to mark the specific time of each measurement.

[0034] It should be noted that the aforementioned temperature change value can be the difference between the current temperature and the temperature at the previous moment, used to reflect the trend of the user's body temperature change. In this embodiment, the time difference between the current moment and the previous moment can be set according to the actual situation, such as setting it to 1 minute. For example, if the temperature measured at 14:30:00 on January 8, 2026 is 36.8℃, and the temperature measured at 14:29:00 is 36.6℃, then the user's temperature change value at this time is... .

[0035] Step S20: When the temperature change value reaches a preset change threshold, the health impact data of the wearer is collected by the second acquisition component. The health impact data includes at least one of physiological data, electroencephalogram data and facial images.

[0036] It should be noted that the aforementioned preset change threshold can be a pre-set temperature value used to determine whether a change in the user's body temperature is significant. In this embodiment, the preset change threshold can be set based on medical research and clinical data; for example, the preset change threshold can be set to 0.5℃. Furthermore, the preset change threshold can be dynamically adjusted based on the user's individual characteristics (such as basal body temperature and health status). In actual use, when the temperature change value actually measured by the first acquisition component reaches or exceeds the preset change threshold, further health data acquisition can be triggered.

[0037] It should be noted that the aforementioned second acquisition component can be a sensor or module used to collect health impact data of the user. In this embodiment, the second acquisition component may include, but is not limited to, physiological data acquisition components (such as heart rate sensors (PPG sensors), respiratory rate sensors, etc.), electroencephalogram (EEG) data acquisition components (such as ear EEG sensors), or cameras.

[0038] It should also be noted that the aforementioned health impact data can be various data that reflect the user's health status, including physiological data, electroencephalogram (EEG) data, and facial images. Physiological data can reflect the user's physiological state, such as heart rate, respiratory rate, and skin conductance; EEG data can reflect the user's psychological state, such as mood and stress level.

[0039] It should be understood that the aforementioned facial image may be a user's facial image captured by the headset through the second acquisition component (camera). In this embodiment, the headset can analyze the user's current facial expression through the facial image, thereby determining the user's emotion category.

[0040] In practical use, when a user wears the headset, the first data acquisition component (such as a contact temperature sensor) in the headset can periodically collect the user's current temperature and calculate the temperature change value based on the current temperature and the previous temperature. Simultaneously, the temperature change value is compared with a preset threshold. If the temperature change value reaches or exceeds the preset threshold, a second data acquisition component can be triggered to collect further data. Specifically, the headset can first initialize the corresponding sensors or modules to collect data based on the required type of health impact data, thereby obtaining at least one type of health impact data. For example, if it is necessary to collect the user's physiological data, the PPG sensor can be controlled to start data collection; if it is necessary to collect the user's electroencephalogram (EEG) data, the EEG sensor can be controlled to start collecting data related to the user's emotions; if it is necessary to collect the user's facial images, the camera can be controlled to start collecting facial images. Furthermore, the headset can also control the PPG sensor, EEG sensor, and camera to collect data simultaneously to obtain multiple types of health impact data.

[0041] Step S30: Use the health impact data as an impact factor, adjust the current temperature using the impact factor, and monitor the health of the wearer using the adjusted current temperature.

[0042] It is understood that the aforementioned influencing factors can be parameters or indicators used to reflect potential causes of changes in a user's body temperature, such as the user's heart rate, mood, etc., and this embodiment does not limit them.

[0043] It should be understood that the adjusted current temperature mentioned above can be a value obtained after correcting the current temperature. In practical applications, the headphones can input real-time collected health impact data as influencing factors into a neural network model. The neural network model can then predict the user's expected body temperature deviation based on these influencing factors. For example, the model might determine that "current emotional excitement usually leads to a temperature measurement that is 0.3℃ higher than the baseline body temperature" and output a body temperature deviation of 0.3℃. Subsequently, the headphones can adjust the current temperature based on the body temperature deviation output by the model and perform health monitoring on the user based on the adjusted temperature. For example, the adjusted temperature can be compared with the normal temperature range. If the adjusted temperature is still abnormal, it indicates that the user may have a pathological fever, and a health warning can be issued. If the adjusted temperature is within the normal body temperature range, it indicates that the user's current elevated body temperature may be caused by factors such as emotions, and no warning can be issued to reduce false alarms.

[0044] In this embodiment, a first data acquisition component can be installed in the headphones. During actual use, the first data acquisition component can collect the wearer's current temperature and determine the temperature change value at the current moment. If the temperature change value reaches a preset threshold, a second data acquisition component can collect at least one health impact data point from the wearer. This health impact data is then used as an influencing factor to adjust the current temperature, enabling health monitoring based on the adjusted current temperature. Because this embodiment can collect at least one health impact data point from the wearer when the temperature change value reaches the preset threshold, and adjust the current temperature based on this data, it avoids inaccurate temperature measurements due to interference from other factors, thereby improving the accuracy of health assessments.

[0045] Reference Figure 3 , Figure 3 This is a flowchart illustrating the second embodiment of the health monitoring method of this application. Based on the first embodiment described above, a second embodiment of the health monitoring method of this application is proposed.

[0046] To more comprehensively assess the impact of health data on user body temperature, this embodiment can comprehensively consider data from multiple dimensions (including physiological data, electroencephalogram data, and facial image data) to accurately determine the cause of body temperature changes, thereby providing more reliable health monitoring results. In this embodiment, such as Figure 5 As shown, the health impact data includes the physiological data, the electroencephalogram (EEG) data, and the facial image; prior to step S30, the method further includes: Step S31: Determine the current heart rate of the wearer based on the physiological data, and determine the current emotion category of the wearer based on the EEG data and the facial image.

[0047] It is understood that the aforementioned current heart rate can be the heart rate of the wearer at the current moment. In this embodiment, the headset can measure the wearer's heart rate through a photoelectric sensor or an electrocardiogram sensor installed therein.

[0048] It is also understood that the aforementioned current emotion category can be the emotional state of the wearer at the current moment. In this embodiment, the emotion category can be specifically divided into positive emotions (such as happiness, excitement), negative emotions (such as anger, anxiety), and neutral emotions (such as calm, relaxation). In practical applications, the headset can collect signals from the wearer using EEG sensors to obtain the user's EEG data, and extract features from the EEG data, such as frequency band energy (α, β, θ, etc.) and the complexity of the EEG waveform. At the same time, it can extract the user's facial expression features from facial images, and then use machine learning or deep learning models (such as CNN, LSTM) to classify the extracted features, thereby determining the user's current emotion category. For example, the headset can determine that the user is currently in a relaxed state by analyzing the increase in α frequency band energy.

[0049] Step S32: Input the current heart rate and the current emotion category as influencing factors into a preset neural network model to perform influence detection and obtain the influence detection results.

[0050] It should be noted that the aforementioned preset neural network model can be a deep learning model used to evaluate whether the input multi-dimensional data has an impact on the measured temperature. In this embodiment, the headset first collects historical data, including physiological data (such as heart rate), EEG data, facial images, and corresponding body temperature changes, wherein these data are labeled with whether they have an impact on body temperature. Then, the headset can extract key features from these raw data. For example, the heart rate features corresponding to the physiological data can include heart rate variability, average heart rate, etc.; the emotional features corresponding to the EEG data and facial images can include the encoding of emotion categories (such as happiness=1, anxiety=2, calm=0). Subsequently, the headset can select a suitable deep learning framework (such as TensorFlow, PyTorch) and train the neural network model based on the prepared data to finally obtain the aforementioned preset neural network model.

[0051] It should be understood that the above-mentioned impact detection results can be prediction results used to characterize whether health impact data has a significant impact on the change of current temperature. In this embodiment, the impact detection result can be a binary output, such as 0 or 1, where 0 can indicate that health impact data has no significant impact on the change of current temperature; and 1 can indicate that health impact data has a significant impact on the change of current temperature.

[0052] Step S33: Determine whether the influencing factor has an impact on the current temperature based on the impact detection results. If it does have an impact, adjust the current temperature using the influencing factor.

[0053] It should be noted that detecting whether health impact data affects the current temperature can refer to the process of analyzing the user's physiological data, electroencephalogram (EEG) data, and facial images to determine whether these factors have a significant impact on the currently measured body temperature. In practical applications, a user's body temperature is easily affected by various factors. For example, when a user is emotionally agitated or their heart rate increases due to exercise, their body temperature may rise. These changes are not necessarily related to the user's health condition; therefore, relying solely on a single body temperature measurement cannot accurately reflect the user's health status. Therefore, this embodiment can detect health impact data and comprehensively consider the influence of multiple factors on body temperature, thereby providing the user with a more accurate health assessment.

[0054] It should also be noted that this embodiment can analyze the relationship between health impact data and body temperature changes through a neural network model to detect whether the health impact data has a significant impact on the current temperature. Specifically, the headphones can input the real-time collected health impact data into the neural network model to predict the impact of the health impact data on the current temperature. For example, if the model predicts an impact exceeding a preset threshold, it indicates that the health impact data has a significant impact on the current temperature.

[0055] In this embodiment, if the recognition result output by the preset neural network model is 1 (or greater than the threshold 0.5), it indicates that the current heart rate and emotion category have a significant impact on the change in the current temperature. In this case, it may be necessary to adjust the current temperature or conduct further health monitoring. Conversely, if the recognition result is 0 (or less than or equal to the threshold 0.5), it indicates that the current heart rate and emotion category do not have a significant impact on the change in the current temperature. In this case, the current temperature can be directly used for health monitoring of the user.

[0056] It should be noted that this embodiment can analyze the relationship between health impact data and body temperature changes through a neural network model to detect whether the health impact data has a significant impact on the current temperature. Specifically, the headphones can input the user's current heart rate and current mood category as influencing factors into the neural network model, so that the model can predict the impact of health impact data on the current temperature based on the influencing factors. For example, if the model predicts an impact exceeding a preset threshold, it indicates that the health impact data has a significant impact on the change in the current temperature.

[0057] In practical applications, if detection reveals that health-related data influences the current temperature, it indicates that the measured temperature has been affected by other factors (such as emotional fluctuations or physiological activities), resulting in some error. In this case, the headset can predict the deviation of the current body temperature and adjust it accordingly to eliminate errors caused by other factors, obtaining an adjusted current temperature that more accurately reflects the user's true body temperature. Subsequently, the headset can perform a health assessment based on the adjusted current temperature, thereby improving the accuracy of the health assessment.

[0058] Understandably, if the detected health impact data of the user (such as physiological data and EEG data) does not significantly affect the current temperature, then the current temperature is considered a reliable health indicator. Therefore, the deviation of the current body temperature can be 0, meaning the headset can directly monitor the wearer's health based on the current temperature and provide health tips or warnings. Specifically, the headset can detect whether the current temperature is within the normal range (e.g., 36.5℃-37.5℃). If it is, it can indicate that the user's health is good; if not, it can generate corresponding health suggestions. For example, if the body temperature is high, it can suggest that the user rest, drink water, or seek medical attention; if the body temperature is low, it can suggest that the user take measures to keep warm or seek medical attention.

[0059] Further, the step of determining the current emotion category of the wearer based on the EEG data and the facial image includes: Step S311: Extract features from the EEG data to obtain multi-dimensional features of the EEG data, including frequency domain features and time domain features.

[0060] It should be understood that the aforementioned multidimensional features can be different types of features extracted from EEG data, and these features can describe the characteristics of the data from multiple perspectives. In this embodiment, the multidimensional features specifically include frequency domain features and time domain features.

[0061] It should also be understood that the aforementioned frequency domain features can be obtained through spectral analysis, reflecting the energy distribution of the signal at different frequencies. For example, the delta band (1-4Hz) is typically associated with deep sleep and relaxation; the theta band (4-8Hz) is typically associated with relaxation, meditation, and mild fatigue; the alpha band (8-12Hz) is typically associated with wakefulness and relaxation; the beta band (12-30Hz) is typically associated with alertness and focused attention; and the gamma band (30-100Hz) is typically associated with high concentration and cognitive processing. Time domain features can be extracted directly from time-series signals, reflecting the signal's characteristics over time, such as mean, variance, standard deviation, and zero crossover rate.

[0062] Step S312: Extract features from the facial image to obtain facial expression features of the facial image.

[0063] Understandably, the aforementioned facial expression features can be features used to characterize the user's current facial expression state, and can be used for emotion classification (such as happiness, anger, relaxation, fear, etc.).

[0064] Further, step S312 includes: performing face detection on the facial image, determining the face region in the facial image, and extracting global expression features of the face region; determining key expression regions of the face region, and extracting local expression features of the key expression regions; fusing the global expression features and the local expression features to obtain facial expression features of the facial image.

[0065] It should be understood that the aforementioned face region can be a rectangular region containing a face in a facial image; correspondingly, the aforementioned global expression features can be features extracted from the face region that reflect the user's overall facial expression, such as the overall shape and movement trend of the user's facial contours, eyes, mouth, and eyebrows. In this embodiment, the headphones can extract features from the face region using a deep learning model (such as a convolutional neural network CNN) to obtain global expression features, so that the user's overall emotional state can be determined based on these global expression features.

[0066] It should also be understood that the aforementioned key expression regions can be local areas on the face that contribute significantly to expression recognition, such as the eye area, eyebrow area, mouth area, and cheek area. This embodiment does not impose any limitations on this. Correspondingly, the aforementioned local expression features can be features extracted from the key expression regions that reflect local facial expressions of the user, such as the degree of eye opening and closing, the upward or downward movement of the corners of the mouth, and the raising or frowning of the eyebrows. In this embodiment, extracting local expression features helps to analyze subtle emotional changes in the user, thereby improving the accuracy of subsequent emotion classification of the user.

[0067] In practical applications, the headset first uses a face detection model to locate the face region in an image, and then inputs this face region into a pre-trained CNN model to extract global expression features. Simultaneously, the headset uses a facial landmark detection model to locate key points such as the eyes, eyebrows, and mouth within the face region, and crops key expression regions from the face region based on the coordinates of these key points. Then, features are extracted from each key expression region separately, and the extracted feature vectors are concatenated or weighted and fused to form local expression features. Subsequently, the headset can fuse the global and local expression features to generate a comprehensive facial expression feature vector.

[0068] Step S313: Concatenate the frequency domain features, the time domain features, and the facial expression features to obtain the emotional features of the wearer.

[0069] It is also understood that the aforementioned emotional features can be features related to emotional state extracted from EEG data, which may include the energy of key frequency bands (such as alpha band energy, beta band energy, theta band energy, etc.), statistical values ​​of key time-domain features (such as mean, variance, standard deviation, zero crossover rate, etc.), and facial expression features, etc.

[0070] In practical applications, after the headphones collect the user's electroencephalogram (EEG) data, they can use Fast Fourier Transform (FFT) to convert the time-series signal into a frequency-domain signal to extract the energy of key frequency bands and extract key time-domain features from the time-series signal. At the same time, facial expression features from the facial image are extracted. Then, the frequency-domain features, time-domain features, and facial expression features are spliced ​​and integrated to generate an emotion feature vector.

[0071] Step S314: Determine the emotional score of the wearer based on the emotional characteristics, and determine the current emotional category of the wearer based on the emotional score.

[0072] Understandably, the aforementioned emotion score can be a quantitative value reflecting the user's emotional state. In practical applications, after obtaining the emotional characteristics of the user, the headphones can input these characteristics into a pre-trained emotion scoring model to calculate an emotion score within a specific range (e.g., 0-1 or 0-100). In this embodiment, different emotion categories can correspond to an emotion score interval. After calculating the user's emotion score, the headphones can determine the specific emotion category represented by the target emotion score interval, thereby identifying that emotion category as the user's current emotion category.

[0073] Furthermore, in order to adapt to the personalized needs of different users and thus provide more accurate emotion monitoring and health assessment, before step S314, the method further includes: constructing a user profile corresponding to the wearer based on the wearer's basic user information, and determining the emotion category classification criteria of the wearer based on the user profile.

[0074] It should be understood that the aforementioned basic user information can be basic data related to the user, such as age, gender, health status, lifestyle, and emotional expression. Among these, health status can refer to the user's health background, such as whether they have chronic diseases or sleep quality; lifestyle can refer to the user's daily activity patterns, such as exercise habits and work stress; and emotional expression habits can refer to the way the user expresses emotions, such as whether they are easily agitated or introverted.

[0075] It should also be understood that the aforementioned user profile can be a model used to comprehensively describe the characteristics of the wearer. It can be built based on the user's basic information and historical data, and is used for personalized emotion monitoring and health assessment of the user. In this embodiment, the headset can collect the wearer's basic information and historical health data, extract the wearer's key features from the collected data, and then use machine learning or statistical methods to build a user profile model corresponding to the wearer.

[0076] It should be noted that the aforementioned emotion category classification criteria can be rules used to map emotion scores to specific emotion categories. These criteria can be threshold-based rules or learned through machine learning models, and this embodiment does not impose any limitations on them. In this embodiment, the specific form of the emotion category classification criteria can include, but is not limited to, fixed thresholds, dynamic thresholds, and machine learning models. Fixed thresholds can refer to setting a fixed threshold range for each emotion category; dynamic thresholds can refer to dynamically adjusting the threshold based on user profiles and historical data; and machine learning models can refer to using a classification model to directly learn the classification of emotion categories from the data.

[0077] In practical applications, individual user characteristics, such as age, gender, health status, and emotional expression habits, may affect the criteria for classifying emotions. Therefore, this embodiment can construct user profiles based on these user characteristics and adjust the criteria for classifying emotions according to the characteristics in the user profiles to adapt to the personalized needs of different users. For example, young people may be more likely to exhibit extreme emotions, while older people may experience smaller emotional fluctuations.

[0078] Step S314 includes: determining the current emotion category of the wearer based on the emotion score and the emotion category classification criteria.

[0079] In practical applications, after calculating the emotional score of the wearer, the headphones can map the emotional score to a specific emotional category according to the emotional category classification criteria, thereby determining the wearer's current emotional category. For example, if the emotional category classification criteria are: the threshold range for negative emotions (such as anger and anxiety) is 0-0.3; the threshold range for neutral emotions (such as calm and relaxation) is 0.3-0.7; and the threshold range for positive emotions (such as happiness and excitement) is 0.7-1, then the user's current emotional category can be directly obtained by mapping the calculated emotional score. Specifically, if the wearer's emotional score is 0.85, then according to the emotional category classification criteria, an emotional score of 0.85 falls within the threshold range for positive emotions, and therefore the wearer's current emotional category is positive.

[0080] Reference Figure 4 , Figure 4This is a flowchart illustrating the third embodiment of the health monitoring method of this application. Based on the above embodiments, the third embodiment of the health monitoring method of this application is proposed.

[0081] Considering that user body temperature is easily affected by various factors, leading to low accuracy in health assessments, this embodiment uses health impact data as an influencing factor to adjust the current temperature. This adjusted temperature accurately reflects the user's true body temperature status, resulting in more accurate health detection results. In this embodiment, as... Figure 4 As shown, the step of adjusting the current temperature using the influencing factor includes: Step S401: Determine the heart rate change rate and mood change value of the wearer at the current moment.

[0082] It is understood that the aforementioned heart rate variability rate can be the change in the wearer's heart rate at the current moment. In this embodiment, it can be calculated by measuring the wearer's heart rate value at the current moment. Heart rate value at the previous moment The difference, divided by the time interval. Obtain heart rate variability The corresponding calculation formula can be: / .

[0083] It is also understood that the aforementioned emotion change value can be the amount of change in the wearer's emotional state at the current moment, which can reflect changes in the user's psychological state, such as from calm to anxiety. In this embodiment, the emotion change value can be obtained by calculating the difference between the wearer's emotion score at the current moment and the previous moment.

[0084] Step S402: Construct a real-time influencing factor vector for the wearable user based on the heart rate change rate and the emotion change value.

[0085] It should be noted that the aforementioned real-time influencing factor vector can be a comprehensive vector describing the potential impact on a user's body temperature at the current moment. This vector may include heart rate variability and mood variability, and can be specifically represented as: Real-time Influencing Factor Vector .

[0086] Step S403: Determine the temperature adjustment amount for the wearer based on the real-time influencing factor vector using a preset body temperature offset prediction model.

[0087] It should be noted that the aforementioned preset body temperature offset prediction model can be a machine learning model used to predict the offset of the user's body temperature; correspondingly, the aforementioned temperature adjustment amount can be the offset of the user's body temperature predicted by the preset body temperature offset prediction model, and this offset can be used to adjust the currently measured body temperature value.

[0088] In practical applications, the preset body temperature shift prediction model can employ a hybrid neural network architecture, which can combine convolutional neural networks and long short-term memory networks to process multimodal data (physiological data and EEG data) and capture time-series features. During model training, the headset can collect a large amount of body temperature data from users under different physiological states (heart rate changes) and psychological states (emotional changes). This data needs to be labeled with the user's actual body temperature change as the training target value. Then, the headset can normalize the heart rate change rate and emotional change values ​​to a range between [0,1] or [-1,1], and divide the data into training, validation, and test sets for model training, with a ratio of 7:2:1. In this embodiment, the headset can use mean squared error as the loss function to measure the difference between the model-predicted temperature adjustment and the actual temperature adjustment. The formula for calculating this loss function is:

[0089] In the formula, For loss function, For the sample size, This is the temperature adjustment amount predicted by the model. This represents the actual temperature adjustment amount.

[0090] Step S404: Adjust the current temperature according to the temperature adjustment amount.

[0091] In this embodiment, the headset can input a pre-built real-time influencing factor vector into a preset body temperature offset prediction model. This model can then predict the user's body temperature offset based on the real-time influencing factor vector, which represents the user's temperature adjustment. The headset can then adjust the user's current temperature based on this adjustment to eliminate errors and obtain the adjusted current temperature. Based on this adjusted current temperature, health monitoring can be performed, providing a more accurate health assessment and recommendations.

[0092] Furthermore, before the step of adjusting the current temperature according to the temperature adjustment amount, the method further includes: extracting the individual key features of the wearer from the user profile; extracting the key features of the influencing factors of the wearer from the real-time influencing factor vector; determining the first influence coefficient corresponding to the individual key features and the second influence coefficient corresponding to the key features of the influencing factors; and determining the dynamic adjustment coefficient based on the individual key features, the first influence coefficient, the key features of the influencing factors, and the second influence coefficient. It should be understood that the aforementioned key individual characteristics can be user features in the user profile that can significantly affect thermoregulation. These characteristics can reflect the user's inherent physiological and health attributes. In this embodiment, key individual characteristics may include, but are not limited to, the user's age, gender, health status, lifestyle, and emotional expression habits. Specifically, regarding age, thermoregulation ability typically varies across different age groups, with older adults generally exhibiting weaker thermoregulation. Regarding gender, the thermoregulation mechanisms of men and women often differ; for example, women experience temperature fluctuations during their menstrual cycle. Regarding health status, conditions such as thyroid dysfunction can affect thermoregulation. Regarding lifestyle, exercise habits may influence thermoregulation; those who exercise regularly tend to have stronger thermoregulation abilities.

[0093] It should also be understood that the key features of the aforementioned influencing factors can be those features in the real-time influencing factor vector that can significantly affect thermoregulation. These features may include, but are not limited to, heart rate variability and mood variability. For example, significant changes in heart rate may predict changes in body temperature; emotional fluctuations (such as anger or fear) may lead to an increase in body temperature; changes in blood oxygen levels may affect thermoregulation, etc.

[0094] Understandably, the aforementioned first influence coefficient can be an indicator reflecting the degree of influence of an individual's key characteristics on body temperature regulation; correspondingly, the aforementioned second influence coefficient can be an indicator reflecting the degree of influence of the key characteristics of influencing factors on body temperature regulation. In this embodiment, both the first and second influence coefficients can be determined through experimental data or machine learning models.

[0095] It is also understood that the aforementioned dynamic adjustment coefficient can be a coefficient used to dynamically adjust the current temperature. In this embodiment, the calculation formula for the dynamic adjustment coefficient is specifically as follows:

[0096] In the formula, This represents the basic adjustment factor (usually 1). Indicates the first The first influence coefficient of key individual characteristics Indicates the first The values ​​of key features of an individual Indicates the first The second influence coefficient of key individual characteristics Indicates the first The values ​​of key features of each individual.

[0097] The step of adjusting the current temperature according to the temperature adjustment amount includes: adjusting the current temperature according to the temperature adjustment amount and the dynamic adjustment coefficient.

[0098] In this embodiment, the dynamic adjustment coefficient is calculated. Then, you can use the formula: Adjust the current temperature, where, This indicates the adjusted current temperature. The current temperature. This is the temperature adjustment amount.

[0099] In the specific implementation, refer to Figure 5 , Figure 5 This is a flowchart illustrating the overall process of health monitoring for users in the health monitoring method described in this application. Figure 5 As shown, when a user wears the headphones, the first acquisition component in the headphones continuously collects the user's temperature data and compares the current temperature data with the previous temperature data to determine if the user's temperature has changed too much. If the user's temperature changes too much, the second acquisition component collects the user's physiological data, EEG data, and facial images, extracts the user's heart rate from the physiological data, and identifies and classifies the user's current emotion based on features in the EEG data and facial images to determine the user's current emotion category. Then, the headphones can fuse the user's heart rate data and current emotion category and input them into a neural network model to determine whether the change in the user's body temperature is affected by other factors. If not, it indicates that the change in the user's body temperature is due to an abnormal physical condition, and an abnormality alert can be issued. If so, a preset body temperature offset prediction model can predict the user's temperature adjustment amount based on real-time influencing factor vectors. Based on the individual key features in the user profile and the real-time influencing factor vectors, combined with the corresponding influence coefficients, a dynamic adjustment coefficient is calculated, and the current temperature is adjusted according to the temperature adjustment amount and the dynamic adjustment coefficient.

[0100] In addition, refer to Figure 6 , Figure 6 This is a structural block diagram of the first embodiment of the health monitoring device of this application; as shown below. Figure 6 As shown in the embodiments of this application, a health monitoring device is also proposed, which includes: The acquisition module 601 is used to acquire the current temperature of the wearer through the first acquisition component, and determine the temperature change value at the current moment based on the current temperature; The acquisition module 601 is also used to acquire health impact data of the wearer through the second acquisition component when the temperature change value reaches a preset change threshold. The health impact data includes at least one of physiological data, electroencephalogram data and facial images. The monitoring module 602 is used to use the health impact data as an influencing factor, adjust the current temperature using the influencing factor, and monitor the health of the wearer using the adjusted current temperature.

[0101] Based on the first embodiment of the health monitoring device described in this application, a second embodiment of the health monitoring device of this application is proposed.

[0102] In this embodiment, the health impact data includes the physiological data, the electroencephalogram (EEG) data, and the facial image; the monitoring module 602 is further configured to determine the wearer's current heart rate based on the physiological data, and determine the wearer's current emotion category based on the EEG data and the facial image; input the current heart rate and the current emotion category as impact factors into a preset neural network model for impact detection, and obtain impact detection results; determine whether the impact factors affect the current temperature based on the impact detection results, and if they do, adjust the current temperature using the impact factors.

[0103] In one implementation, the monitoring module 602 is further configured to extract features from the EEG data to obtain multi-dimensional features of the EEG data, the multi-dimensional features including frequency domain features and time domain features; extract features from the facial image to obtain facial expression features of the facial image; concatenate the frequency domain features, the time domain features and the facial expression features to obtain the emotional features of the wearer; determine the emotional score of the wearer based on the emotional features, and determine the current emotional category of the wearer based on the emotional score.

[0104] In one implementation, the monitoring module 602 is further configured to perform face detection on the facial image, determine the face region in the facial image, and extract the global expression features of the face region; determine the key expression region of the face region, and extract the local expression features of the key expression region; and fuse the global expression features and the local expression features to obtain the facial expression features of the facial image.

[0105] In one implementation, the monitoring module 602 is further configured to construct a user profile corresponding to the wearer based on the wearer's basic user information, and determine the wearer's emotional category classification criteria based on the user profile; and determine the wearer's current emotional category based on the emotional score and the emotional category classification criteria.

[0106] Based on the above embodiments of the health monitoring device of this application, a third embodiment of the health monitoring device of this application is proposed.

[0107] In this embodiment, the monitoring module 602 is further configured to determine the heart rate change rate and mood change value of the wearer at the current moment through the influencing factors; construct a real-time influencing factor vector of the wearer based on the heart rate change rate and the mood change value; determine the temperature adjustment amount of the wearer based on the real-time influencing factor vector through a preset body temperature offset prediction model; and adjust the current temperature according to the temperature adjustment amount.

[0108] In one implementation, the monitoring module 602 is further configured to extract individual key features of the wearer from the user profile; extract key features of influencing factors of the wearer from the real-time influencing factor vector; determine a first influence coefficient corresponding to the individual key features and a second influence coefficient corresponding to the key features of influencing factors; determine a dynamic adjustment coefficient based on the individual key features, the first influence coefficient, the key features of influencing factors and the second influence coefficient; and adjust the current temperature according to the temperature adjustment amount and the dynamic adjustment coefficient.

[0109] Other embodiments or specific implementations of the health monitoring device described in this application can be found in the above-described method embodiments, and will not be repeated here.

[0110] Furthermore, embodiments of this application also propose a headset, which includes: a memory, a processor, and a health monitoring program stored in the memory and executable on the processor. When the health monitoring program is executed by the processor, it implements the steps of the health monitoring method described above.

[0111] In addition, this application also proposes a health monitoring system, which includes headphones as described above and a second data acquisition component.

[0112] The specific implementation of the health monitoring system in this embodiment can be referred to the description of the above method embodiments, and will not be repeated in this embodiment.

[0113] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0114] The sequence numbers of the embodiments in this application are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0115] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as a read-only memory image (ROM) / random access memory (RAM), magnetic disk, optical disk), and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of this application.

[0116] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A health monitoring method, characterized in that, The method is applied to a headset in a health monitoring system, the headset being equipped with a first data acquisition component, and the health monitoring system further including a second data acquisition component; The method includes: The first acquisition component acquires the wearer's current temperature and determines the temperature change value at the current moment based on the current temperature. When the temperature change value reaches a preset change threshold, the second acquisition component collects the health impact data of the wearer, which includes at least one of physiological data, electroencephalogram data, and facial images. The health impact data is used as an influencing factor to adjust the current temperature, and the adjusted current temperature is used to monitor the health of the wearer.

2. The method as described in claim 1, characterized in that, The health impact data includes the physiological data, the electroencephalogram (EEG) data, and the facial images; The step of using the health impact data as an influencing factor to adjust the current temperature includes: The wearer's current heart rate is determined based on the physiological data, and the wearer's current emotion category is determined based on the electroencephalogram (EEG) data and the facial image. The current heart rate and the current emotion category are input as influencing factors into a preset neural network model for influence detection, and the influence detection results are obtained. Based on the impact detection results, it is determined whether the influencing factor affects the current temperature. If it does, the current temperature is adjusted using the influencing factor.

3. The method as described in claim 2, characterized in that, The step of determining the current emotion category of the wearer based on the EEG data and the facial image includes: Feature extraction is performed on the EEG data to obtain multi-dimensional features of the EEG data, including frequency domain features and time domain features; Feature extraction is performed on the facial image to obtain the facial expression features of the facial image; The frequency domain features, the time domain features, and the facial expression features are concatenated to obtain the emotional features of the wearer. The emotional score of the wearer is determined based on the emotional characteristics, and the current emotional category of the wearer is determined based on the emotional score.

4. The method as described in claim 3, characterized in that, The step of extracting features from the facial image to obtain facial expression features includes: Perform face detection on the facial image to determine the face region in the facial image, and extract the global expression features of the face region; Identify the key expression regions of the face region and extract the local expression features of the key expression regions; The global expression features and the local expression features are fused to obtain the facial expression features of the facial image.

5. The method as described in claim 3, characterized in that, Before the step of determining the emotional score of the wearer based on the emotional characteristics, the method further includes: A user profile is constructed based on the user's basic information, and the emotional category classification criteria for the user are determined based on the user profile. The step of determining the emotional score of the wearer based on the emotional characteristics includes: The current emotional category of the wearer is determined based on the emotional score and the emotional category classification criteria.

6. The method as described in claim 5, characterized in that, The step of adjusting the current temperature using the influencing factor includes: The heart rate change rate and mood change value of the wearer at the current moment are determined by the influencing factors. Construct a real-time influencing factor vector for the wearer user based on the heart rate variability and the emotion variability values; The temperature adjustment amount for the wearer is determined based on the real-time influencing factor vector using a preset body temperature offset prediction model. The current temperature is adjusted according to the temperature adjustment amount.

7. The method as described in claim 6, characterized in that, Before the step of adjusting the current temperature according to the temperature adjustment amount, the method further includes: Extract key individual characteristics of the wearer from the user profile; Extract key features of influencing factors of the wearable user from the real-time influencing factor vector; Determine the first influence coefficient corresponding to the key characteristics of the individual, and the second influence coefficient corresponding to the key characteristics of the influencing factors; The dynamic adjustment coefficient is determined based on the individual key characteristics, the first influence coefficient, the key characteristics of the influencing factors, and the second influence coefficient. The step of adjusting the current temperature according to the temperature adjustment amount includes: The current temperature is adjusted according to the temperature adjustment amount and the dynamic adjustment coefficient.

8. A health monitoring device, characterized in that, The device includes: The data acquisition module is used to acquire the current temperature of the wearer through the first data acquisition component, and determine the temperature change value at the current moment based on the current temperature; The acquisition module is also used to acquire health impact data of the wearer through the second acquisition component when the temperature change value reaches a preset change threshold. The health impact data includes at least one of physiological data, electroencephalogram data and facial images. The monitoring module is used to use the health impact data as an influencing factor, adjust the current temperature using the influencing factor, and monitor the health of the wearer using the adjusted current temperature.

9. A type of headset, characterized in that, The headset includes: a memory, a processor, and a health monitoring program stored in the memory and executable on the processor, wherein the health monitoring program, when executed by the processor, implements the steps of the health monitoring method as described in any one of claims 1 to 7.

10. A health monitoring system, characterized in that, The system includes the headset as described in claim 9 and a second acquisition component.

Images