WO2019191864A1 - 一种确定压力值的方法及装置 - Google Patents

一种确定压力值的方法及装置 Download PDF

Info

Publication number
WO2019191864A1
WO2019191864A1 PCT/CN2018/081571 CN2018081571W WO2019191864A1 WO 2019191864 A1 WO2019191864 A1 WO 2019191864A1 CN 2018081571 W CN2018081571 W CN 2018081571W WO 2019191864 A1 WO2019191864 A1 WO 2019191864A1
Authority
WO
WIPO (PCT)
Prior art keywords
pressure value
calibration
feature vector
category
calibration information
Prior art date
Application number
PCT/CN2018/081571
Other languages
English (en)
French (fr)
Inventor
许培达
李彦
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to CN201880072336.5A priority Critical patent/CN111315296B/zh
Priority to PCT/CN2018/081571 priority patent/WO2019191864A1/zh
Publication of WO2019191864A1 publication Critical patent/WO2019191864A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/16—Devices for psychotechnics; Testing reaction times ; Devices for evaluating the psychological state

Definitions

  • the present application relates to the field of intelligent terminals, and in particular, to a method and apparatus for determining a pressure value.
  • the current products for testing mental stress involve two broad categories: calibration and non-calibration.
  • Embodiments of the present application provide a method and apparatus for determining a pressure value, which may be applied to a device for determining a pressure value, which may include, but is not limited to, a mobile phone and a wearable device (eg, a watch, a wristband,
  • a device for determining a pressure value which may include, but is not limited to, a mobile phone and a wearable device (eg, a watch, a wristband,
  • the pressure value in the embodiment of the present application is only the psychological pressure value of the mental level.
  • the pressure value of the user's psychology can be measured by the device to improve the accuracy of the measured pressure value.
  • the embodiment of the present application provides a method for determining a pressure value, comprising: collecting, by a device, a physiological signal of a user, the physiological signal being a bioelectric signal generated by a joint control between a human body system and an autonomic nervous system, the physiological signal
  • the signal may be an EEG signal, an EMG signal, an ECG signal, a pulse signal, etc.
  • the device may obtain a target feature vector according to the physiological signal to determine whether calibration information exists, and the calibration information is used for calibrating the output pressure value.
  • the reference information if there is calibration information, the calibration model can be used to determine the calibrated target pressure value according to the calibration information and the target feature vector.
  • the calibration model includes the correspondence between the feature vector and the pressure value, and the calibration information is used to calibrate the model output target pressure value.
  • the reference information is used; then, the target pressure value is output; in the embodiment of the present application, the calibration information is reference information used for calibrating the output pressure value, for example, the calibration information may be after actually collecting the physiological signal of the user for a period of time.
  • the calibration information is related to each individual user, the human body is different, the calibration information may be different, the target feature vector is used as the input of the calibration model, and the calibration target pressure value is output through the calibration model, without the user's initiative Participation, that is, the calibration of the output pressure value, improve the accuracy of the output pressure value, and in this process, the user has no feeling, the experience is good.
  • the target feature vector is input of the classification model, and the classification model is used to determine the category pressure value corresponding to the pressure category according to the target feature vector, and the classification model includes the mapping relationship between the feature vector and the pressure category;
  • the difference parameter between the value and the target pressure value updates the calibration information.
  • the calibration information is continuously updated, and the update process does not require the user to actively participate, but continuously updates the calibration information according to the difference parameter between the output value of the classification model and the output value of the calibration model, and improves the calibration information. The accuracy of the target pressure value of the output is thus improved.
  • the classification model is used to determine the category pressure value corresponding to the pressure category according to the target feature vector, and the classification model includes the correspondence between the feature vector and the pressure category; the category pressure value and the target feature are The vector is saved as calibration information; then, the category pressure value is output.
  • the device displays the category pressure value output by the classification model.
  • the calibration information includes a calibration pressure value and a calibration feature vector
  • the calibration model determines the corrected target pressure value according to the calibration information and the target feature vector, which may be: determining the target feature vector and the calibration feature vector.
  • the difference feature vector between the two uses the calibration model to determine the difference pressure value corresponding to the difference feature vector according to the difference vector, and then determines the target pressure value according to the calibration pressure value and the differential pressure value.
  • the calibration feature can be based on the target feature vector. The basis of the vector is whether to increase (increment) or decrease (decrement), and to determine the differential pressure value according to the increment or decrement.
  • the calibration is performed.
  • the differential pressure value is increased based on the pressure value. If the target feature vector is reduced based on the calibration feature vector, the differential pressure value is reduced based on the calibration pressure value, thereby obtaining the calibrated target pressure value.
  • the specific manner of updating the calibration information according to the difference parameter between the category pressure value and the target pressure value may be: determining a difference parameter between the category pressure value and the target pressure value, and determining whether the difference parameter is greater than Threshold value, if the difference parameter is greater than the threshold value, the value recorded by the counter is increased by 1. When the value recorded by the counter reaches the first preset value, the calibration information is deleted, and the value recorded by the counter is set to zero. In the embodiment of the present application, If the difference is greater than the threshold, it indicates that the pressure difference of the model output is too large, the calibration information may be inaccurate, and the calibration information needs to be updated, but in order to avoid the difference between the category pressure value and the target pressure value is greater than the threshold value.
  • the value recorded by the counter is incremented by 1.
  • the calibration information is triggered to be deleted, the calibration information is updated, and the value recorded by the counter is set to zero to improve the accuracy of the calibration information. Sex.
  • determining whether the duration between the current time and the time when the calibration information was last updated is greater than a second preset value; if the current time and the last updated calibration information The time between the moments is greater than the second preset value, the calibration information is deleted, and the value recorded by the counter is set to zero. Since the human body condition is gradual and not transient, a calibration can only reflect a section of the user. The physiological signal to pressure mapping in time, this relationship will change as the user's physical condition changes, so setting a calibration interval mechanism to achieve automatic recalibration, can be done without the user's active intervention. The device itself captures the user's current status and updates the calibration information to improve the accuracy of the calibration information.
  • the device includes an acceleration sensor, and the acceleration sensor is used to detect a state of the user, the state includes a resting state, and when the state of the user is a resting state, the classification model is used to determine the pressure category according to the resting state.
  • the acceleration sensor determines that the user is in a resting state
  • the device displays the category pressure value corresponding to the normal pressure.
  • the user is added.
  • the pressure category output by the classification model is normal pressure.
  • the calibration information is not accurate, the calibration count will be accumulated, and the new calibration will be restarted after several times (update calibration information) ), no user involvement is required to automatically initiate the calibration process.
  • an auxiliary means for identifying the pressure of the classification model is added. When the user is detected to be in a resting state, the classification model is used to directly output the pressure value corresponding to the normal pressure, thereby improving the identification of the classification model. effectiveness.
  • an embodiment of the present invention provides a computer storage medium for storing computer software instructions for use in the foregoing apparatus, including a program designed to perform the above aspects.
  • an embodiment of the present invention provides a device for determining a pressure value, which has the functions performed by the device in the actual method.
  • This function can be implemented in hardware or in hardware by executing the corresponding software.
  • the hardware or software includes one or more modules corresponding to the functions described above.
  • the structure of the apparatus for determining the pressure value includes a memory and a processor.
  • the memory is used to store computer executable program code.
  • the program code includes instructions that, when executed by the processor, cause the apparatus to perform the information or instructions involved in the above method.
  • FIG. 1 is a schematic flow chart of steps of an embodiment of a method for determining a pressure value according to an embodiment of the present application
  • FIG. 2 is a schematic diagram of a periodic waveform of an electrocardiogram signal in an embodiment of the present application
  • FIG. 3 is a schematic diagram of a periodic waveform of a pulse signal in an embodiment of the present application.
  • FIG. 4 is a schematic flow chart of steps of another embodiment of a method for determining a pressure value according to an embodiment of the present application
  • FIG. 5 is a schematic flow chart of steps of another embodiment of a method for determining a pressure value according to an embodiment of the present application
  • FIG. 6 is a schematic structural diagram of an embodiment of an apparatus for determining a pressure value according to an embodiment of the present application
  • FIG. 7 is a schematic structural diagram of another embodiment of an apparatus for determining a pressure value according to an embodiment of the present application.
  • FIG. 8 is a schematic structural diagram of another embodiment of an apparatus for determining a pressure value according to an embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of another embodiment of an apparatus for determining a pressure value according to an embodiment of the present application.
  • Embodiments of the present application provide a method and apparatus for determining a pressure value for improving the accuracy of determining a pressure value.
  • Physiological signal It is a bioelectrical signal generated by the combination of the human endocrine system and the autonomic nervous system.
  • the physiological signals include but are not limited to EEG signals, EMG signals, electrocardiography (ECG) signals, and pulse (photoplethysmography, PPG). ) Signal, skin conductance (SC), respiration (RSP), and heart rate.
  • EEG signals EMG signals
  • ECG signals electrocardiography
  • PPG photoplethysmography
  • PPG photoplethysmography
  • SC skin conductance
  • RSP respiration
  • heart rate heart rate
  • the physiological signal may be exemplified by taking an electrocardiogram signal and a pulse as an example.
  • ECG signal based on the bioelectrical changes of cardiomyocytes, it reflects the changes in the overall potential of the whole heart during the excitatory process.
  • the ECG signal is not subject to the subjective consciousness of the human being, and can more objectively reflect the human's Emotional state.
  • Pulse signal The pulse wave is generated from the periodic contraction and relaxation of the heart. It is a shock wave of blood and blood vessel wall formed by the blood circulation of the heart rhythm. This shock wave starts at the root end of the aorta and then travels along the main and branch arteries to the surrounding blood vessels.
  • the pulse wave sensor is used to record the shock wave on the body surface to form a pulse signal.
  • a method for determining a pressure value is provided in the embodiment of the present application.
  • the pressure value in the embodiment of the present application is a mental pressure value, and the method is applied to a device for determining a pressure value, and the device may include but is not limited to a mobile phone and wearable. Equipment (for example, watches, bracelets, headphones, smart clothes, etc.).
  • the device may be described by taking a wearable device as an example.
  • the device includes two models for predicting pressure values.
  • the two models are a classification model and a calibration model.
  • the classification model and the calibration model are obtained by acquiring a large number of physiological signals and then extracting feature vectors in the physiological signals. Learning and training a large number of feature vectors in the feature vector data set.
  • NB naive bayes
  • LR logistic regression
  • DT decision tree
  • SVM support vector machine
  • KNN K-nearest_neighbors
  • neural networks random forests, etc.
  • the classification model includes a mapping relationship between the input feature vector and its corresponding pressure category.
  • the input of the classification model is a feature vector of the physiological signal
  • the output is a pressure value corresponding to the pressure category.
  • the classification model can be classified into two categories.
  • the model is described as an example, and the pressure category includes, but is not limited to, normal pressure (also referred to as "normal pressure") and high pressure.
  • the classification model may also be a three-category, a four-category model, etc., for example, the pressure category that the classification model can output may be a first pressure category, a second pressure category, and a third pressure category, etc.
  • the pressure corresponding to each pressure category can be gradually increased, and the pressure of each type of column can be preset to correspond to a specific value, and the preset specific value is used as the output pressure value corresponding to the input feature vector.
  • the classification model may be described by taking a two-category model as an example, that is, the classification model outputs pressure corresponding to two categories.
  • the pressure value for example, the preset pressure value corresponding to the atmospheric pressure category is 50, and the preset pressure value corresponding to the high pressure category is 70.
  • the specific numerical values of the preset pressure values in the embodiments of the present application are only for illustrative purposes, and do not result in a limited description in the embodiments of the present application.
  • the calibration model includes a mapping relationship between the input feature vector and its corresponding specific pressure value.
  • the pressure value output by the calibration model may be a specific discrete pressure value such as 51, 62, etc., and the calibration model predicts the calibrated pressure based on the calibration information. value.
  • Calibration information used to calibrate the output of the pressure value used for the reference information
  • the calibration information includes the calibration pressure value and the calibration feature vector
  • the calibration pressure value can be after the actual collection of the user's physiological signal for a period of time, according to the self-evaluation score or scale
  • the mode gives the pressure value at this time
  • the calibration feature vector is a feature vector corresponding to the calibration pressure value.
  • the calibration information is related to each individual user, and the calibration information may be different.
  • a classification model that triggers updates to calibration information A classification model that triggers updates to calibration information.
  • a calibration model that outputs the pressure value after calibration.
  • the pressure value output by the classification model is a category pressure value (also referred to as “first pressure value” in the embodiment of the present application), and the calibration model outputs
  • the pressure value is a target pressure value (also referred to as a "second pressure value” in the embodiment of the present application).
  • the pre-trained model is used to predict the current pressure value of the user according to the feature vector.
  • the mental state directly from the physiological signal by the model. For example, the pressure value predicted according to the physiological signal of the user after the exercise may be too high, which requires calibration of the pressure value.
  • the device collects the physiological signal of the user, extracts the target feature vector in the physiological signal, and then determines whether there is calibration information, and the calibration information can be understood as the reference information for calibrating the output pressure value, if there is calibration information And determining, by using the classification model, the first pressure value corresponding to the pressure category according to the target feature vector, and determining, by using the calibration model, the second pressure value after the calibration according to the calibration information and the target feature vector; and according to the first pressure value and the second pressure value The difference between the parameters updates the calibration information.
  • the first pressure value is a preset pressure value corresponding to the pressure category.
  • the classification model outputs a pressure value
  • the preset pressure value output by the classification model actually represents a range, for example, within a normal pressure range.
  • a plurality of discrete pressure values are classified into a normal pressure category by a classification model, and the normal pressure category is output only corresponding to a preset pressure value, so as to avoid differences between physiological signals of different individuals and the same individual under different states.
  • the calibration information after which the device outputs a relatively accurate second pressure value after calibration.
  • the active calibration is not required, and the calibration information is automatically triggered, that is, the output pressure value is calibrated, and the accuracy of the output pressure value is improved, and in this process, the user has no feeling and the experience is good.
  • the device is described as an executive body.
  • One embodiment of a method for determining a pressure value is provided in the embodiment of the present application, including:
  • Step 101 Collect a physiological signal of the user.
  • the physiological signal of the user may be collected by the user.
  • the trigger button is set on the wearable device.
  • the sensor starts to collect the user.
  • Physiological signal is set on the wearable device.
  • the physiological signal is passively collected.
  • the wearable device collects the physiological signal of the user through the sensor, and the process of collecting the physiological signal by the user is not perceived.
  • the physiological signal includes an electrocardiographic signal of a pulse signal.
  • Pre-processing the collected physiological signals for example, filtering and denoising the acquired physiological signals.
  • Step 102 Acquire a target feature vector according to the physiological signal.
  • the feature information in the pre-processed physiological signal is extracted, and the target feature vector of the feature information is calculated according to the feature information, and the target feature vector can be used as an input of the classification model and the calibration model.
  • FIG. 2 is a schematic diagram of the periodic waveform of the electrocardiogram signal.
  • the characteristic information included in the ECG signal may be:
  • P wave The first wave recorded on the electrocardiogram, the waveform of the P wave is small and smooth.
  • QRS complex Completely reflects the depolarization process of the ventricle. The group wave contains three closely connected potential changes, namely a downward Q wave, an upward R wave and an S wave following the R wave downward. 3. ST segment: From the end of the QRS complex to the start of the T wave, it represents that the myocardial cells in the ventricle are in a depolarized state, which is the period before the end of the ventricular depolarization to the repolarization. 4, T wave: represents the repolarization process of the ventricle.
  • FIG. 3 is a schematic diagram of the periodic waveform of the pulse signal.
  • the characteristic information included in the pulse signal may be:
  • the main wave rise branch AB represents the rapid ejection period of the ventricle.
  • Wave-by-wave descending branch BD This band represents the process from the end of ejection to the next cardiac cycle.
  • the starting point of the ascending branch A The time at which the AB segment begins at the waveform is the time when the aortic valve is open, and the ventricle quickly starts to emit blood. This point is usually considered as the boundary between two adjacent cardiac cycles.
  • Main wave peak B indicates the maximum intra-arterial pressure.
  • heavy beat wave C represents the beginning of ventricular relaxation, the beginning of blood flow in the aorta.
  • Step 103 Determine, by using a classification model, a first pressure value corresponding to the pressure category according to the target feature vector.
  • the target feature vector is used as an input of the classification model, and the classification model outputs a first pressure value corresponding to the pressure category.
  • the first pressure value corresponding to the normal pressure of the classification model output is 50.
  • the first pressure value corresponding to the high pressure output by the classification model is 70.
  • Step 104 Determine whether calibration information exists; if there is calibration information, execute step 105; if there is no calibration information, perform step 108.
  • the calibration information includes a calibration pressure value, or the calibration information includes a calibration pressure value and a calibration feature vector.
  • Step 105 If calibration information exists, the calibration model is used to determine the second pressure value after calibration according to the calibration information and the target feature vector.
  • the specific manner of determining the calibrated second pressure value according to the calibration information and the target feature vector by using the calibration model may be:
  • determining a difference feature vector between the target feature vector and the calibration feature vector using a calibration model to determine a differential pressure value corresponding to the difference feature vector according to the difference vector; determining the first according to the calibration pressure value and the differential pressure value Two pressure values.
  • the target feature vector is a
  • the calibration feature vector is b
  • the difference feature vector between the target feature vector and the calibration feature vector is (ab). If the calibration pressure value is 50 and the differential pressure value is 5, the calibrated first The second pressure value is 55.
  • the error value of the calibration pressure value and the pressure value output by the calibration model is determined, and then the pressure value output by the calibration model is calibrated according to the error value, and finally the second pressure value after calibration is obtained.
  • the calibration pressure value is 50
  • the target feature vector is used as an input
  • the target feature vector is predicted by using a calibration model
  • the obtained pressure value is 60, indicating that the error value between the calibration model and the calibration pressure value is 10, and the obtained
  • the pressure value is calibrated
  • the calibration model outputs a pressure value of 65 according to the feature vector, and subtracts 10 from the obtained pressure value to obtain a second pressure value of 55 after calibration.
  • Step 106 Update calibration information according to a difference parameter between the first pressure value and the second pressure value.
  • the difference parameter has a plurality of representations, and the difference parameter may be a difference, and the difference parameter may also be between the first pressure value and the second pressure value
  • the difference parameter, or the difference parameter may also be a ratio between the first pressure value and the second pressure value multiplied by a coefficient or the like.
  • the difference parameter is described by taking the difference as an example.
  • An absolute difference ⁇ P between the first pressure value (P1) and the second pressure value (P2), ⁇ P
  • a threshold eg, the threshold is 50
  • the difference parameter is less than or equal to the threshold, it is determined whether the duration between the current time and the time when the calibration information was last updated is greater than a second preset value (for example, two weeks), and if so, the calibration information is deleted, and the counter is recorded.
  • the value is set to zero. Since the human body's physical condition is gradual and not transient, a calibration can only reflect the physiological signal-to-pressure mapping relationship of the user over a period of time. This relationship changes as the user's physical condition changes, so the setting is made.
  • a calibration interval mechanism to achieve automatic recalibration enables the device to capture the latest state of the user and update the calibration information without the user having to actively intervene.
  • Step 107 Output a second pressure value.
  • the second pressure value after the calibration is output, and the wearable device can display the second pressure value, and the user can know the current mental stress state according to the second pressure value.
  • Step 108 If the calibration information does not exist, save the first pressure value and the target feature vector as calibration information.
  • the first pressure value and the target feature vector are saved, and the first pressure value is used as a calibration pressure value, and the target feature vector is used as a calibration feature vector.
  • Step 109 Output a first pressure value.
  • the wearable device outputs a first pressure value (eg, 50), for example, the wearable device displays a pressure value of 50.
  • the device collects the physiological signal of the user, and extracts the target feature vector in the physiological signal; and then determines whether there is calibration information, and the calibration information can be understood as the reference information for calibrating the output pressure value, if there is calibration Information, the classification model is used to determine the first pressure value corresponding to the pressure category according to the target feature vector, and the calibration model is used to determine the second pressure value after calibration according to the calibration information and the target feature vector; and the first pressure value and the second pressure may be determined according to the first pressure value and the second pressure The difference parameter between the values updates the calibration information.
  • the preset pressure value of the classification model output is actually A range, for example, a plurality of discrete pressure values belonging to the normal pressure range are classified into a normal pressure category by a classification model, and the normal pressure category is output only corresponding to a preset pressure value, thereby avoiding different individuals and the same individual Differences in physiological signals in different states, improve The accuracy of the calibration information, after which the device outputs a relatively accurate second pressure value after calibration. If there is no calibration information, the first pressure value is saved as the calibration pressure value, and the target feature vector is saved as the calibration feature vector, and the first pressure value is output.
  • the active participation of the user is not required, and the difference parameter between the first pressure value and the second pressure value automatically triggers the update of the calibration information, that is, the output pressure value is calibrated, and the accuracy of the output pressure value is improved. And in this process, the user has no feeling and the experience is good.
  • FIG. 4 Another embodiment of the method for determining the pressure value is also provided in the embodiment of the present application.
  • the device is further provided with an acceleration sensor for detecting the state of the user.
  • an acceleration sensor for detecting the state of the user.
  • the difference between the embodiment and the above embodiment is that when the acceleration detects that the state of the user is a resting state, the classification model is used according to the resting state.
  • the pressure category is determined to be a normal pressure, and a first pressure value corresponding to the normal pressure.
  • Step 401 Collect a physiological signal of the user.
  • Step 402 Acquire a target feature vector according to the physiological signal.
  • Step 401 and step 402 can be understood in conjunction with step 101 and step 102 in the corresponding embodiment of FIG. 1, and details are not described herein.
  • Step 403 The acceleration sensor is used to detect the state of the user, and the state includes a resting state, where the resting state refers to a state when the user is in a quiet rest.
  • step 403 may be performed before step 401, or may be performed after step 401 and before step 402.
  • the specific timing of step 403 is not limited.
  • Step 404 When the acceleration sensor detects that the state of the user is a resting state, the classification model is used to determine, according to the resting state, that the pressure category is a normal pressure and a first pressure value corresponding to the normal pressure.
  • Step 405 Determine whether calibration information exists; if there is calibration information, execute step 406; if there is no calibration information, perform step 409.
  • Step 406 If calibration information exists, the calibration model is used to determine the second pressure value after calibration according to the calibration information and the target feature vector.
  • Step 407 Update calibration information according to a difference parameter between the first pressure value and the second pressure value.
  • Step 408 Output a second pressure value.
  • Steps 407 to 408 can be understood in conjunction with steps 106 and 107 in the embodiment corresponding to FIG. 1, and details are not described herein.
  • Step 409 If there is no calibration information, and the acceleration sensor detects that the user is in a resting state, the first pressure value corresponding to the normal pressure is saved as the calibration pressure value, and the target feature vector is used as the calibration pressure vector.
  • Step 410 Output a first pressure value.
  • the acceleration sensor detects that the user is in a non-rest state (for example, a motion state)
  • the method steps in the embodiment corresponding to FIG. 1 are performed.
  • the pressure type output by the classification model is normal pressure. If the calibration information is not accurate, the calibration count is accumulated, several times. After restarting the new calibration (update calibration information), the user is not required to participate in the automatic startup calibration process. And by identifying the static state of the user, an auxiliary means for identifying the pressure of the classification model is added. When the user is detected to be in a resting state, the classification model is used to directly output the pressure value corresponding to the normal pressure, thereby improving the identification of the classification model. effectiveness.
  • FIG. 5 is a schematic flowchart of a step of determining a pressure value in an embodiment of the present application.
  • the user wears a smart watch, which has the function of measuring the pressure value, the user can actively activate the function of measuring the pressure value in the watch, the watch acquires the physiological signal of the user, or the user can also automatically collect the physiological signal in the watch menu. That is, the user does not need to actively trigger, and the watch automatically collects the physiological signals of the user.
  • the classification model outputs a first pressure value corresponding to the pressure category according to the acquired physiological signal.
  • the watch After acquiring the physiological signal through the sensor, the watch acquires the target feature vector in the physiological signal, and uses the target feature vector as the input of the classification model, and the classification model outputs the pressure value corresponding to the pressure category according to the target feature vector, for example, the first pressure
  • the value is 50 (indicating that the pressure category is normal pressure).
  • step A31 Determine whether there is calibration information. If not, step A31 is performed, and if yes, step A34 is performed.
  • the first pressure value (50) and the target feature vector are saved as calibration information.
  • the watch shows that the current pressure value is 50.
  • the calibration model is used to obtain a second pressure value according to the calibration information and the target feature vector, that is, the difference feature vector according to the acquired target feature vector and the calibration feature vector, the difference feature vector includes an increased feature vector and a reduced feature vector, if the difference The feature vector is an increased feature vector, and the corresponding pressure value is added based on the calibration pressure value; if the difference feature vector is the reduced feature vector, the corresponding pressure value is reduced based on the calibration pressure value.
  • the calibration model outputs a differential pressure value corresponding to the difference vector (eg, 5), and then determines the second pressure value according to the calibration pressure value and the differential pressure value, and the second pressure value is a calibration pressure value and The sum of the differential pressure values, if the calibration pressure value is 50 and the differential pressure value is 5, the second pressure value after calibration is 55.
  • A352 Determine whether the calibration counter meets the first preset value (such as 5). If execution A3521 is satisfied, A3522 is executed if it is not satisfied.
  • A3521 delete the calibration information, and set the counter to zero to trigger the calibration.
  • A3522 displays a second pressure value (such as 55).
  • A350 Determine whether the time interval between the current time and the time when the calibration information was last deleted is greater than a second preset value (2 weeks). If yes, go to step A3521; if no, go to step A3522.
  • step A31 is executed to save the first pressure value and the currently acquired feature vector as calibration information, and the calibration information is updated, and the output is output.
  • the pressure value is the first pressure value (such as 70) output by the classification model, and the next time the watch judges whether calibration information exists (step A3), at this time, since the first pressure value (70) and the corresponding feature vector have been saved as For calibration information, proceed to step A34 to obtain a second pressure value using the calibration model.
  • the calibration information is updated in the process of determining the pressure value, and then the calibrated pressure value can be output.
  • the user's active participation, automatic calibration, and acquisition are not required.
  • the physiological signal to determine the corresponding pressure value and display that is, to ensure the user's good experience and improve the accuracy of the output pressure value, the user can adjust his mental state as soon as possible according to the output pressure value, to avoid excessive stress due to mental stress Affects physical health.
  • an embodiment of an apparatus for determining a pressure value is provided in an embodiment of the present application, where the apparatus is configured to perform the steps actually performed by the apparatus in the foregoing method embodiment.
  • the apparatus includes :
  • the collecting module 601 is configured to collect a physiological signal of the user
  • the obtaining module 602 is configured to obtain a target feature vector according to the physiological signal collected by the collecting module 601;
  • the category pressure value determining module 603 is configured to determine, according to the target feature vector acquired by the obtaining module 602, the category pressure value corresponding to the pressure category by using the classification model,
  • the determining module 604 is configured to determine whether calibration information exists
  • the calibration module 606 is configured to update the calibration information according to the difference parameter between the category pressure value determined by the category pressure value determining module 603 and the target pressure value determined by the target pressure value determining module 605;
  • the output module 607 is configured to output a target pressure value determined by the target pressure value determining module 605.
  • the device further includes a storage module 608;
  • the storage module 608 is configured to save the category pressure value and the target feature vector as calibration information when the determining module 604 determines that the calibration information does not exist;
  • the output module 607 is further configured to output the category pressure value determined by the category pressure value determining module 603.
  • the calibration information includes a calibration pressure value and a calibration feature vector; the target pressure value determination module 605 is further specifically configured to:
  • the difference pressure value corresponding to the difference feature vector is determined according to the difference vector
  • the target pressure value is determined based on the calibration pressure value and the differential pressure value.
  • the calibration module 606 is further specifically configured to:
  • the calibration information is deleted, and the value recorded by the counter is set to zero.
  • FIG. 7 another embodiment of the apparatus 700 is further provided by the embodiment of the present application, and the apparatus further includes an update module 609;
  • the determining module 604 is further configured to: determine, when the difference parameter is less than or equal to the threshold, whether the duration between the current time and the time when the calibration information was last updated is greater than a second preset value;
  • the update module 609 when the duration between the current time and the time when the calibration information was last updated is greater than the second preset value, deletes the calibration information in the storage module 608 and sets the value of the counter record to zero.
  • another embodiment of the apparatus 800 is further provided by the embodiment of the present application, and further includes a detecting module.
  • the detecting module 610 is configured to detect a state of the user by using an acceleration sensor
  • the category pressure value determining module 603 is further configured to: when the detecting module 610 determines that the state of the user is a resting state, determine, by using the classification model, the pressure category is a normal pressure according to a resting state determined by the detecting module, and a category pressure value corresponding to the normal pressure. .
  • FIG. 6-8 the devices in Figures 6-8 are presented in the form of functional modules.
  • a “module” herein may refer to an application-specific integrated circuit (ASIC), circuitry, a processor and memory that executes one or more software or firmware programs, integrated logic circuitry, and/or other functions that provide the functionality described above. Device.
  • ASIC application-specific integrated circuit
  • FIG. 9 the apparatus of Figures 6-8 can take the form shown in Figure 9.
  • the embodiment of the present application further provides another device for determining the pressure value.
  • FIG. 9 for the convenience of description, only parts related to the embodiment of the present invention are shown. Without specific details, please refer to the present invention. Example method section.
  • the device is described by taking a wearable device as an example.
  • FIG. 9 is a block diagram showing a partial structure of an apparatus related to a terminal provided by an embodiment of the present application.
  • the wearable device includes a memory 920, an input unit 930, a display unit 940, a sensor 950, an audio circuit 960, a processor 980, and the like.
  • the specific components of the device will be specifically described below with reference to FIG. 9:
  • the memory 920 can be used to store software programs and modules, and the processor 980 executes various functional applications and data processing of the wearable device by running software programs and modules stored in the memory 920.
  • the memory 920 may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function, and the like; the storage data area may store data created according to use of the wearable device, and the like.
  • memory 920 can include high speed random access memory, and can also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other volatile solid state storage device.
  • the input unit 930 can be configured to receive input numeric or character information and to generate key signal inputs related to user settings and function controls of the wearable device.
  • the input unit 930 may include a touch panel 931 and other input devices 932.
  • the touch panel 931 also referred to as a touch screen, can collect touch operations on or near the user (such as a user using a finger, a stylus, or the like on the touch panel 931 or near the touch panel 931. operating).
  • Other input devices 932 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons, etc.).
  • the display unit 940 can be used to display information input by the user or information provided to the user and pressure values of various menus and outputs of the wearable device.
  • the wearable device can also include at least one type of sensor 950, such as a light sensor, a motion sensor, and other sensors, acceleration sensors.
  • Various sensors can collect the physiological signals of the user.
  • the accelerometer sensor can detect the magnitude of acceleration in all directions (usually three axes). When stationary, the magnitude and direction of gravity can be detected, which can be used for identification.
  • wearable device posture detecting the user's resting state
  • vibration recognition related functions such as pedometer, tapping
  • thermometers thermometers
  • infrared sensors that can be configured for wearable devices Other sensors, etc.
  • An audio circuit 960, a speaker 961, and a microphone 962 can provide an audio interface between the user and the wearable device.
  • the audio circuit 960 can transmit the converted electrical data of the received audio data to the speaker 961, and convert it into a sound signal output by the speaker 961.
  • the microphone 962 converts the collected sound signal into an electrical signal, and the audio circuit 960 After receiving, it is converted to audio data, and then processed by the audio data output processor 980, transmitted to the, for example, another wearable device via the RF circuit 910, or outputted to the memory 920 for further processing.
  • Processor 980 is a control center for the wearable device that connects various portions of the entire wearable device with various interfaces and lines, by running or executing software programs and/or modules stored in memory 920, and for recalling storage in memory 920. Data, perform various functions of the wearable device and process data to monitor the wearable device as a whole.
  • processor 980 can include one or more processing units.
  • the processor 980 included in the wearable device causes the wearable device to perform the method steps actually performed by the device for determining the pressure value in the method embodiment described above.
  • a computer storage medium for storing computer software instructions for use in the above apparatus is provided in the embodiment of the present application, and includes method steps for performing the apparatus in the foregoing method embodiment.
  • the chip comprises: a processing unit and a communication unit
  • the processing unit may be, for example, a processor
  • the communication unit may be, for example, an input/output interface, Pin or circuit, etc.
  • the processing unit may execute computer execution instructions stored by the storage unit to cause the chip within the terminal to perform the wireless communication method of any of the above aspects.
  • the storage unit is a storage unit in the chip, such as a register, a cache, etc., and the storage unit may also be a storage unit located outside the chip in the terminal, such as a read-only memory (read) -only memory, ROM) or other types of static storage devices, random access memory (RAM), etc. that can store static information and instructions.
  • the processor mentioned in any of the above may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more for controlling the above.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • the integrated circuit of the program execution of the first aspect wireless communication method may be a general-purpose central processing unit (CPU), a microprocessor, an application-specific integrated circuit (ASIC), or one or more for controlling the above.
  • CPU central processing unit
  • ASIC application-specific integrated circuit
  • the disclosed system, apparatus, and method may be implemented in other manners.
  • the device embodiments described above are merely illustrative.
  • the division of the unit is only a logical function division.
  • there may be another division manner for example, multiple units or components may be combined or Can be integrated into another system, or some features can be ignored or not executed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in an electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated unit if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium.
  • a computer readable storage medium A number of instructions are included to cause a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present application.
  • the foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, and the like. .

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Hospice & Palliative Care (AREA)
  • Pathology (AREA)
  • Developmental Disabilities (AREA)
  • Psychiatry (AREA)
  • Psychology (AREA)
  • Social Psychology (AREA)
  • Physics & Mathematics (AREA)
  • Child & Adolescent Psychology (AREA)
  • Biophysics (AREA)
  • Educational Technology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • Medical Informatics (AREA)
  • Molecular Biology (AREA)
  • Surgery (AREA)
  • Animal Behavior & Ethology (AREA)
  • General Health & Medical Sciences (AREA)
  • Public Health (AREA)
  • Veterinary Medicine (AREA)
  • Measuring Pulse, Heart Rate, Blood Pressure Or Blood Flow (AREA)

Abstract

一种确定压力值的方法，包括：采集用户的生理信号，根据生理信号获取目标特征向量，判断是否存在校准信息；若存在校准信息，采用校准模型根据校准信息和目标特征向量确定校准后的目标压力值，所述校准模型包括特征向量与压力值的对应关系，所述校准信息用于所述校准模型输出所述目标压力值所使用的基准信息；输出所述目标压力值。还提供了一种确定压力值的装置，用于提高测量的压力值的准确性。

Description

一种确定压力值的方法及装置 技术领域
本申请涉及智能终端领域，尤其涉及一种确定压力值的方法及装置。
背景技术
随着可穿戴设备的兴起，越来越多的设备可以在日常的情景下采集人们的生理数据，如心率、皮肤温度等。通过长时程采集生理数据，能够更加有效地对人们的生理健康情况做出统计和预测，尽管此类产品较多，但在精神压力的评测上面仍然关注较少。精神压力是现代社会人们所面临的主要心理健康问题之一，高精神压力对人们的工作生活效率、生活品质等方面都有重要影响。长期处于有压力状态会诱使各类疾病的发生，因此定性定量地评测用户的精神压力是非常有价值的。
当前测试精神压力的产品涉及两大类：有校准类和无校准类。
其中，有校准评测需要用户主动参与，通过获得用户的主观压力自评(如用户做文字上的压力测试、用户手动选取当前心情状态等)，再结合精神压力测试模型来实现用户的精神压力评测，此种测评方式，需要用户的主动参与，适用性较差，用户产生心情描述是偶发的，具有主观性且可能是不真实的。无校准测评不需要用户的主动参与，但是受限于用户的生理信号与人体之间的个体依赖性，精度较低。
发明内容
本申请实施例提供了一种确定压力值的方法及装置，该方法可以应用于一种确定压力值的装置，该装置可以包括但不限定于手机和可穿戴设备(例如，手表、手环、耳机、智能服饰等等)，本申请实施例中的压力值只是精神层面的心理的压力值，本申请实施例中可以通过装置测量用户的心理的压力值，提高测量压力值的准确性。
第一方面，本申请实施例提供了一种确定压力值的方法，包括：装置采集用户的生理信号，该生理信号是由人体内泌系统和自主神经系统共同控制产生的生物电信号，该生理信号可以为脑电信号，肌电信号，心电信号、脉搏信号等，该装置可以根据生理信号获取目标特征向量，判断是否存在校准信息，该校准信息为用于校准输出的压力值所使用的基准信息，若存在校准信息，可以采用校准模型根据校准信息和目标特征向量确定校准后的目标压力值，校准模型包括特征向量与压力值的对应关系，校准信息用于校准模型输出目标压力值所使用的基准信息；然后，输出目标压力值；本申请实施例中，校准信息为用于校准输出的压力值所使用的基准信息，例如，该校准信息可以为实际采集一段时间用户的生理信号后，根据用户自评打分或量表的方式给出此时的压力值，该校准信息与每个用户个体具有关联性，人体不同，校准信息可能不同，将目标特征向量作为校准模型的输入，通过校准模型输出校准后的目标压力值，不需要用户的主动参与，即对输出的压力值进行校准，提高了输出压力值的准确性，且在此过程中，用户无感，体验佳。
在一种可能的实现方式中，将目标特征向量为分类模型的输入，采用分类模型根据目 标特征向量确定压力类别对应的类别压力值，分类模型包括特征向量与压力类别的映射关系；根据类别压力值和目标压力值之间的差异参数更新校准信息。本申请实施例中，该校准信息不断更新，且更新过程不需要用户主动参与，而是根据分类模型的输出值和校准模型的输出值之间的差异参数来不断的更新校准信息，提高校准信息的准确性，从而提高输出的目标压力值的准确性。
在一种可能的实现方式中，若不存在校准信息，采用分类模型根据目标特征向量确定压力类别对应的类别压力值，分类模型包括特征向量与压力类别的对应关系；将类别压力值和目标特征向量作为校准信息进行保存；然后，输出类别压力值，本申请实施例中，若不存在校准信息的情况下，该装置显示分类模型输出的类别压力值。
在一种可能的实现方式中，校准信息包括校准压力值和校准特征向量，采用校准模型根据校准信息和目标特征向量确定校准后的目标压力值具体可以为：确定目标特征向量与校准特征向量之间的差异特征向量，采用校准模型根据差异向量确定差异特征向量对应的差异压力值，然后根据校准压力值和差异压力值确定目标压力值；本申请实施例中，可以根据目标特征向量在校准特征向量的基础上是增加了(增量)还是减少(减量)了，并根据增量或减量来确定差异压力值，若目标特征向量在校准特征向量的基础上是增加了，在该校准压力值的基础上增加该差异压力值，若目标特征向量在校准特征向量的基础上是减少了，在该校准压力值的基础上减少该差异压力值，从而得到校准后的目标压力值。
在一种可能的实现方式中，根据类别压力值和目标压力值之间的差异参数更新校准信息具体的方式可以为：确定类别压力值和目标压力值之间的差异参数，判断差异参数是否大于阈值，若差异参数大于阈值，则将计数器记录的数值增加1，当计数器记录的数值达到第一预设值时，删除校准信息，并将计数器记录的数值置零；本申请实施例中，若该差值大于阈值，则表明这个模型输出的压力值差异过大，校准信息可能不准确，需要对校准信息进行更新，但是为了避免类别压力值与目标压力值之间的差值大于阈值为偶然情况，则将计数器记录的数值加1，当计数器记录的数值达到第一预设值时，触发删除校准信息，以对校准信息进行更新，并将计数器记录的数值置零，提高校准信息的准确性。
在一种可能的实现方式中，若差异参数小于或者等于阈值，则判断当前时刻与上次更新校准信息的时刻之间的时长是否大于第二预设值；若当前时刻与上次更新校准信息的时刻之间的时长大于第二预设值，则删除校准信息，并将计数器记录的数值置零，由于人的身体情况是渐变的，并不是瞬变的，所以一次校准仅能反映用户一段时间内的生理信号到压力的映射关系，这个关系随着用户的身体情况的改变会改变，所以设定一种校准间隔机制来实现自动重新校准，就能够在用户不需要主动介入的情况下，装置自己捕捉用户的最新状态，更新该校准信息，以提高校准信息的准确性。
在一种可能的实现方式中，该装置包括加速度传感器，采用加速度传感器检测用户的状态，该状态包括静息状态，当用户的状态为静息状态时，采用分类模型根据静息状态确定压力类别为正常压力，及正常压力对应的类别压力值，本申请实施例中，若通过加速度传感器确定用户处于静息状态，则装置显示正常压力对应的类别压力值，本实施例中，增加了当用户为休息状态下的场景，若确定用户为静息状态，分类模型输出的压力类别为正 常压力，这时候若校准信息不准，则会累加校准计数，若干次后重新启动新校准(更新校准信息)，不需要用户参与自动启动校准流程。且通过对用户静状态的识别，增加了对分类模型识别压力的一种辅助手段，当检测到用户处于静息状态时，采用分类模型直接输出正常压力对应的压力值，提高了分类模型的识别效率。
第二方面，本发明实施例提供了一种计算机存储介质，用于储存上述装置所用的计算机软件指令，其包含用于执行上述方面所设计的程序。
第三方面，本发明实施例提供了一种确定压力值的装置，具有实现上述方法中实际中装置所执行的功能。该功能可以通过硬件实现，也可以通过硬件执行相应的软件实现。该硬件或软件包括一个或多个与上述功能相对应的模块。
第四方面，确定压力值的装置的结构中包括存储器和处理器。其中存储器用于存储计算机可执行程序代码。该程序代码包括指令，当该处理器执行该指令时，该指令使该装置执行上述方法中所涉及的信息或者指令。
附图说明
图1为本申请实施例中一种确定压力值的方法的一个实施例的步骤流程示意图；
图2为本申请实施例中的心电信号周期波形示意图；
图3为本申请实施例中的脉搏信号的周期波形示意图；
图4为本申请实施例中一种确定压力值的方法的另一个实施例的步骤流程示意图；
图5为本申请实施例中一种确定压力值的方法的另一个实施例的步骤流程示意图；
图6为本申请实施例中一种确定压力值的装置的一个实施例的结构示意图；
图7为本申请实施例中一种确定压力值的装置的另一个实施例的结构示意图；
图8为本申请实施例中一种确定压力值的装置的另一个实施例的结构示意图；
图9为本申请实施例中一种确定压力值的装置的另一个实施例的结构示意图。
具体实施方式
本申请实施例提供了一种确定压力值的方法及设备，用于提高确定压力值的准确度。
为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分的实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本申请保护的范围。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
为了方便理解，首先对本申请中涉及的词语进行解释说明。
生理信号：是由人体内泌系统和自主神经系统共同控制产生的生物电信号，生理信号包括但不限定于脑电信号，肌电信号、心电(electrocardiography，ECG)信号、脉搏(photoplethysmography，PPG)信号、皮肤电导(skin conductance，SC)、呼吸(respiration，RSP)以及心率。本申请实施例中，该生理信号可以以心电信号和脉搏为例进行举例说明。
心电信号：是以心肌细胞的生物电变化为基础的，它反映了整个心脏在兴奋过程中的综合电位变化，心电信号不受人的主观意识的支配，可较为客观地反映出人的情绪状态。
脉搏信号：脉搏波的产生源于心脏周期性的收缩和舒张运动，它是由心脏节律射血而形成的一种血液和血管壁的震荡波。这种震荡波开始于主动脉根端，随后沿主干及分支动脉向周围血管传导，采用脉搏传感器将震荡波在体表记录下来，就形成了脉搏信号。
本申请实施例中提供了一种确定压力值的方法，本申请实施例中的压力值为精神压力值，该方法应用于确定压力值的装置，该装置可以包括但不限定于手机和可穿戴设备(例如，手表、手环、耳机、智能服饰等等)。本申请实施例中，该装置可以以可穿戴设备为例进行说明。该装置中包括用于预测压力值的两个模型，这两个模型为分类模型和校准模型，该分类模型和校准模型均是通过获取大量的生理信号，然后提取生理信号中的特征向量，通过对特征向量数据集中的大量特征向量进行学习和训练得到的。
这两个模型的类型可以为：朴素贝叶斯(naive bayes,NB)、逻辑回归(logistic regression,LR)、决策树(decision tree,DT)、支持向量机(support vector machine,SVM)、证据推理(evidential reasoning)、K近邻(k_nearest_neighbors,KNN)、神经网络、随机森林等。
其中，分类模型包含了输入的特征向量到其对应的压力类别的映射关系，该分类模型的输入为生理信号的特征向量，输出为压力类别对应的压力值，例如，该分类模型可以以二分类模型为例进行说明，该压力类别包括但不限定于正常压力(也称为“常压”)和高压。在实际应用中，该分类模型也可以为三分类、四分类模型等等，比如，该分类模型可以输出的压力类别可以为第一压力类别，第二压力类别和第三压力类别等，这三个压力类别对应的压力可以逐渐增大，每一类列别的压力可以预设对应一个特定数值，该预设的特定数值作为输入的特征向量对应的输出压力值。本申请实施例中，对于分类模型具体可以输出几个压力类别并不限定，在本申请实施例中，该分类模型可以以二分类模型为例进行说明，即分类模型输出两个类别的压力对应的压力值，例如，常压类别对应的预设压力值为50，高压类别对应的预设压力值为70。需要说明的是，本申请实施例中对于预设压力值的具体数值只为举例说明，并不造成对本申请实施例中的限定性说明。
校准模型包含输入的特征向量到其对应的具体压力值的映射关系，例如，校准模型输出的压力值可以为51、62等这样具体的离散压力值，该校准模型根据校准信息预测校准后的压力值。
可以理解的是：
校准信息，用于校准输出的压力值所使用的基准信息，校准信息包括校准压力值和校 准特征向量，校准压力值可以为实际采集一段时间用户的生理信号后，根据自评打分或量表的方式给出此时的压力值，该校准特征向量为该校准压力值对应的特征向量。该校准信息与每个用户个体具有关联性，人体不同，校准信息可能不同。
分类模型，用于触发对校准信息的更新。
校准模型，用于输出校准之后的压力值。
需要说明的是，本申请实施例中，为了区分分类模型和校准模型，分类模型输出的压力值为类别压力值(本申请实施例中也称为“第一压力值”)，校准模型输出的压力值为目标压力值(本申请实施例中也称为“第二压力值”)。
由于生理信号与用户的精神状态具有关联性，也就是说，不同的精神状态下人体所产生的生理信号中的特征向量是不同的，采用预先训练好的模型根据特征向量预测用户当前的压力值，但是，直接通过模型根据生理信号预测精神状态可能会出现不准确的情况，例如，根据用户在运动完之后的生理信号预测的压力值可能会偏高，这就需要对该压力值进行校准。本申请实施例中，装置采集用户的生理信号，提取生理信号中的目标特征向量；然后判断是否存在校准信息，该校准信息可以理解为对输出的压力值进行校准的基准信息，若存在校准信息，则采用分类模型根据目标特征向量确定压力类别对应的第一压力值，并采用校准模型根据校准信息和目标特征向量确定校准后的第二压力值；可以根据第一压力值和第二压力值之间的差异参数更新校准信息。第一压力值为压力类别对应的预设压力值，虽然该分类模型输出的是一个压力值，但是实际上分类模型输出的预设压力值表示的是一个范围，例如，属于正常压力范围内的多个离散的压力值均通过分类模型分类到正常压力类别，该正常压力类别只对应一个预设压力值输出，这样可以避免不同的个体及同一个个体在不同状态下的生理信号带来的差异，提高校准信息的准确性，之后该装置输出校准之后的相对准确的第二压力值。本申请实施例中，不需要用户的主动参与，自动触发更新校准信息，即对输出的压力值进行校准，提高了输出压力值的准确性，且在此过程中，用户无感，体验佳。
请参阅图1，下面以该装置为执行主体进行描述，本申请实施例中提供了一种确定压力值的方法的一个实施例包括：
步骤101、采集用户的生理信号。
装置中设置有各类传感器，通过传感器采集用户的生理信号。在一种可以实现的方式中，采集用户的生理信号可以是用户主动采集的，例如，可穿戴设备上设置有触发按键，当用户对该触发按键进行按压时或点击时，传感器开始采集用户的生理信号。在另一种可能实现的方式中，被动的采集生理信号，当用户戴上可穿戴设备后，可穿戴设备通过传感器采集用户的生理信号，用户采集生理信号的过程无感知。
例如该生理信号包括脉搏信号的心电信号。对采集的生理信号进行预处理，例如，对采集得到生理信号进行滤波去噪等。
步骤102、根据生理信号中获取目标特征向量。
提取预处理后的生理信号中的特征信息，根据特征信息计算该特征信息的目标特征向量，该目标特征向量可以作为分类模型和校准模型的输入。
例如，请参阅图2进行理解特征信息，图2为心电信号周期波形示意图。其中，心电信号所包括的特征信息可以为：
1、P波:心电图上记录的第一个波，P波的波形小而光滑。2、QRS波群：完整的反映了心室的除极过程。该群波包含三个紧密相连的电位变化，即向下的Q波，向上的R波和紧接着R波向下的S波。3、ST段：从QRS波群的末端到T波的起点，它代表心室中的心肌细胞均处于去极化状态，是心室除极结束到复极之前的时段。4、T波：代表心室的复极化过程。
请参阅图3进行理解脉搏信号的特征信息，图3为脉搏信号的周期波形示意图。其中，脉搏信号所包括的特征信息可以为：
1、主波上升支AB；代表心室的快速射血期。2、逐波下降支BD：该波段代表射血末期至下一心动周期的过程。3、升支起始点A：在波形上表现为AB段开始的时刻，是主动脉瓣开放的时刻，心室快速开始射血的标志。通常将该点看作两个相邻心动周期的分界。4、主波波峰B：表示最大动脉内压力。5、重搏波C：代表心室开始舒张，主动脉内血液返流的开始。
步骤103、采用分类模型根据目标特征向量确定压力类别对应的第一压力值。
将目标特征向量作为分类模型的输入，分类模型输出压力类别对应的第一压力值，例如，分类模型输出正常压力对应的第一压力值为50。或者，分类模型输出的高压对应的第一压力值为70。
步骤104、判断是否存在校准信息；若存在校准信息则执行步骤105；若不存在校准信息则执行步骤108。
判断是否存储有校准信息，该校准信息包括校准压力值，或者，该校准信息包括校准压力值和校准特征向量。
步骤105、若存在校准信息，则采用校准模型根据校准信息和目标特征向量确定校准后的第二压力值。
若存在该校准信息，采用校准模型根据校准信息和目标特征向量确定校准后的第二压力值的具体方式可以为：
在一种可能的实现方式中，确定目标特征向量与校准特征向量之间的差异特征向量，采用校准模型根据差异向量确定差异特征向量对应的差异压力值；根据校准压力值和差异压力值确定第二压力值。
例如，目标特征向量为a，校准特征向量为b，目标特征向量与校准特征向量之间的差异特征向量为(a-b)，若校准压力值为50，差异压力值为5，则校准后的第二压力值为55。
在另一种可能的实现方式中，确定校准压力值与校准模型输出的压力值的误差值，然后根据该误差值对校准模型输出的压力值进行校准，最后得到校准后的第二压力值。
例如，校准压力值为50，目标特征向量作为输入，采用校准模型对该目标特征向量进行预测，得到的压力值为60，表明校准模型与校准压力值之间的误差值为10，对得到的压力值进行校准，输出的第二压力值为校准模型得到的压力值与误差值之间的差值，即60-10＝50，后续在确定输出压力值时，每次都将得到的压力值减去10，得到校准后的第二 压力值，例如，下一次，校准模型根据特征向量输出的压力值为65，将得到的压力值减去10，得到校准后的第二压力值为55。
步骤106、根据第一压力值和第二压力值之间的差异参数更新校准信息。
确定第一压力值和第二压力值之间的差异参数，该差异参数有多种表现形式，该差异参数可以为差值，该差异参数也可以为第一压力值与第二压力值之间比值，或者，该差异参数也可以为第一压力值与第二压力值之间的比值乘以一个系数等等，本申请实施例中，该差异参数以差值为例进行说明。
第一压力值(P1)与第二压力值(P2)之间的绝对差ΔP，ΔP＝|P1-P2|；判断差异参数是否大于阈值(例如，该阈值为50)；若差异参数大于阈值，则将计数器记录的数值增加1，计数器用于记录差异参数大于阈值的次数。若该差值大于阈值，则表明这个模型输出的压力值差异过大，校准信息可能不准确，需要对校准信息进行更新，但是为了避免第一压力值与第二压力值之间的差值大于阈值为偶然情况，则将计数器记录的数值加1，当计数器记录的数值达到第一预设值(例如5)时，触发删除校准信息，以对校准信息进行更新，并将计数器记录的数值置零。
若差异参数小于或者等于阈值，则判断当前时刻与上次更新校准信息的时刻之间的时长是否大于第二预设值(例如，二周)，若是，则删除校准信息，并将计数器记录的数值置零。由于人的身体情况是渐变的，并不是瞬变的，所以一次校准仅能反映用户一段时间内的生理信号到压力的映射关系，这个关系随着用户的身体情况的改变会改变，所以设定一种校准间隔机制来实现自动重新校准，就能够在用户不需要主动介入的情况下，装置自己捕捉用户的最新状态，更新该校准信息。
步骤107、输出第二压力值。
输出校准后的第二压力值，可穿戴设备可以显示该第二压力值，用户可以根据该第二压力值得知自身当前的精神压力状况。
步骤108、若不存在该校准信息，将第一压力值和目标特征向量作为校准信息进行保存。
保存第一压力值和该目标特征向量，将该第一压力值作为校准压力值，将该目标特征向量作为校准特征向量。
步骤109、输出第一压力值。
可穿戴设备输出第一压力值(如50)，例如，可穿戴设备显示的压力值为50。
本申请实施例中，装置采集用户的生理信号，并提取生理信号中的目标特征向量；然后判断是否存在校准信息，该校准信息可以理解为对输出的压力值进行校准的基准信息，若存在校准信息，则采用分类模型根据目标特征向量确定压力类别对应的第一压力值，并采用校准模型根据校准信息和目标特征向量确定校准后的第二压力值；可以根据第一压力值和第二压力值之间的差异参数更新校准信息，由于第一压力值为压力类别对应的预设压力值，虽然该分类模型输出的是一个压力值，但是实际上分类模型输出的预设压力值表示的是一个范围，例如，属于正常压力范围内的多个离散的压力值均通过分类模型分类到正常压力类别，该正常压力类别只对应一个预设压力值输出，这样可以避免不同的个体及同 一个个体在不同状态下的生理信号带来的差异，提高校准信息的准确性，之后该装置输出校准之后的相对准确的第二压力值。如不存在校准信息，则保存第一压力值作为校准压力值，保存目标特征向量作为校准特征向量，输出第一压力值。本申请实施例中，不需要用户的主动参与，第一压力值和第二压力值之间的差异参数自动触发更新校准信息，即对输出的压力值进行校准，提高了输出压力值的准确性，且在此过程中，用户无感，体验佳。
在上述实施例的基础上，请参阅图4进行理解，本申请实施例中还提供了确定压力值的方法的另一个实施例。
该装置中还设置有加速度传感器，该加速度传感器用于检测用户的状态，本实施例与上述实施例的区别在于，当加速度检测到用户的状态为静息状态时，采用分类模型根据静息状态确定压力类别为正常压力，及正常压力对应的第一压力值。
步骤401、采集用户的生理信号。
步骤402、根据生理信号中获取目标特征向量。
步骤401和步骤402可以结合图1对应的实施例中的步骤101和步骤102进行理解，此处不赘述。
步骤403、采用加速度传感器检测用户的状态，该状态包括静息状态，该静息状态是指用户在安静休息时的状态。
需要说明的是，步骤403可以在步骤401之前执行，也可以在步骤401之后，步骤402之前执行，步骤403的具体时序并不限定。
步骤404、当加速传感器检测用户的状态为静息状态时，采用分类模型根据静息状态确定压力类别为正常压力及正常压力对应的第一压力值。
步骤405、判断是否存在校准信息；若存在校准信息则执行步骤406；若不存在校准信息则执行步骤409。
步骤406、若存在校准信息，则采用校准模型根据校准信息和目标特征向量确定校准后的第二压力值。
步骤407、根据第一压力值和第二压力值之间的差异参数更新校准信息。
步骤408、输出第二压力值。
步骤407至步骤408可以结合图1对应的实施例中的步骤106和步骤107进行理解，此处不赘述。
步骤409、若不存在校准信息，且加速度传感器检测到用户处于静息状态时，则保存正常压力对应的第一压力值作为校准压力值，目标特征向量作为校准压力向量。
步骤410、输出第一压力值。
需要说明的是，在示例中，若加速度传感器检测到用户处于非静息状态(例如运动状态)时，则执行图1对应的实施例中的方法步骤。
本实施例中，增加了当用户为休息状态下的场景，若确定用户为静息状态，分类模型输出的压力类别为正常压力，这时候若校准信息不准，则会累加校准计数，若干次后重新启动新校准(更新校准信息)，不需要用户参与自动启动校准流程。且通过对用户静状态的识别，增加了对分类模型识别压力的一种辅助手段，当检测到用户处于静息状态时，采用 分类模型直接输出正常压力对应的压力值，提高了分类模型的识别效率。
下面结合一个应用场景对本申请实施例进行说明，请结合图5进行理解，图5为本申请实施例中确定压力值的步骤流程示意图。
A1、用户开始进行压力值测量。
用户佩戴智能手表，该智能手表具有测量压力值的功能，用户可以主动启动手表中的测量压力值的功能，手表获取用户的生理信号，或者，用户也可以在手表菜单中设置自动采集生理信号，即不需要用户主动触发，手表自动采集用户的生理信号。
A2、分类模型根据获取到的生理信号输出压力类别对应的第一压力值。
手表通过传感器获取到生理信号后，获取生理信号中的目标特征向量，将该目标特征向量作为分类模型的输入，分类模型根据该目标特征向量输出压力类别对应的压力值，如，该第一压力值为50(表示压力类别为正常压力)。
A3、判断是否存在校准信息。若不存在则执行步骤A31，若存在则执行步骤A34。
A31、将第一压力值(50)和目标特征向量作为校准信息进行保存。
A32、将计数器置零。
A33、输出第一压力值。
手表显示当前的压力值为50。
A34、采用校准模型得到第二压力值。采用校准模型根据校准信息和目标特征向量得到第二压力值，即根据获取到的目标特征向量与校准特征向量的差异特征向量，该差异特征向量包括增加的特征向量和减少的特征向量，若差异特征向量是增加的特征向量，则在校准压力值的基础上增加对应的压力值；若差异特征向量是减少的特征向量，则在校准压力值的基础上减少对应的压力值。具体的，若校准信息中包含的校准压力值为50，校准特征向量为a,校准特征向量和目标特征向量(如b)之间的差异特征向量为(a-b)，然后将该差异向量(a-b)作为校准模型的输入，该校准模型输出该差异向量对应的差异压力值(如5)，然后，根据校准压力值和差异压力值确定该第二压力值，第二压力值为校准压力值与差异压力值之和，若校准压力值为50，差异压力值为5，则校准后的第二压力值为55。
A35、判断第一压力值和第二压力值之间的绝对差ΔP是否大于阈值，ΔP＝|P1-P2|，若是，则执行步骤A350，若否，则执行步骤A351。
A351、两个模型输出压力值偏差较大，校准计数器自动加1。
A352、判断校准计数器是否满足第一预设值(如5)。若满足执行A3521，若不满足则执行A3522。
A3521、删除校准信息，并将计数器置零，以触发校准。
A3522、显示第二压力值(如55)。
A350、判断当前时刻与上次删除校准信息时刻之间的时间间隔是否大于第二预设值(2周)。若是，执行步骤A3521；若否，则执行步骤A3522。
在应用场景中，当ΔP大于50，则表明此时校准信息可能已经不准确，当ΔP大于50的次数达到5次，就说明当前的校准信息已经不准确了，就可以触发重新校准了，当手表检测是否存在校准信息时，此时是不存在的(因为已经删除)，则执行步骤A31，保存第一压 力值和当前获取的特征向量作为校准信息，及校准信息被更新了，此时输出的压力值为分类模型输出的第一压力值(如70)，当下一次手表判断是否存在校准信息时(步骤A3)，此时因为已经保存了第一压力值(70)和对应的特征向量作为校准信息，则继续执行步骤A34，采用校准模型得到第二压力值。如此根据图5中的步骤进行循环往复不断的在确定压力值的过程中对校准信息进行更新，进而可以输出校准后的压力值，本实施中不需要用户的主动参与，自动校准，并通过获取到的生理信号确定对应的压力值并显示，即保证了用户的良好体验又提高了输出压力值的准确性，用户可以根据输出的压力值，尽快调整自己的精神状态，避免由于精神压力过大影响身体健康。
请参阅图6所示，本申请实施例中提供了一种确定压力值的装置的一个实施例，该装置用于执行如上述方法实施例中装置所实际执行的步骤，具体的，该装置包括：
采集模块601，用于采集用户的生理信号；
获取模块602，用于根据采集模块601采集的生理信号获取目标特征向量；
类别压力值确定模块603，用于采用分类模型根据获取模块602获取的目标特征向量确定压力类别对应的类别压力值，
判断模块604，用于判断是否存在校准信息；
目标压力值确定模块605，当判断模块604确定存在校准信息时，采用校准模型根据校准信息和目标特征向量确定校准后的目标压力值；
校准模块606，用于根据类别压力值确定模块603确定的类别压力值和目标压力值确定模块605确定的目标压力值之间的差异参数更新校准信息；
输出模块607，用于输出目标压力值确定模块605确定的目标压力值。
在一种可能的实现方式中，装置还包括存储模块608；
存储模块608，用于当判断模块604确定不存在校准信息时，将类别压力值和目标特征向量作为校准信息进行保存；
输出模块607，还用于输出类别压力值确定模块603确定的类别压力值。
在一种可能的实现方式中，校准信息包括校准压力值和校准特征向量；目标压力值确定模块605还具体用于：
确定目标特征向量与校准特征向量之间的差异特征向量；
采用校准模型根据差异向量确定差异特征向量对应的差异压力值；
根据校准压力值和差异压力值确定目标压力值。
在一种可能的实现方式中，校准模块606还具体用于：
确定类别压力值和目标压力值之间的差异参数；
判断差异参数是否大于阈值；
若差异参数大于阈值，则将计数器记录的数值增加1；
当计数器记录的数值达到第一预设值时，删除校准信息，并将计数器记录的数值置零。
参阅图7所示，在图6对应的实施例的基础上，本申请实施例还提供了该装置700的另一个实施例，该装置还包括更新模块609；
判断模块604，还用于当差异参数小于或者等于阈值时，判断当前时刻与上次更新校 准信息的时刻之间的时长是否大于第二预设值；
更新模块609，还用于当前时刻与上次更新校准信息的时刻之间的时长大于第二预设值时，则删除存储模块608中的校准信息，并将计数器记录的数值置零。
参阅图8所示，在图6对应的实施例的基础上，本申请实施例还提供了该装置800的另一个实施例，还包括检测模块；
检测模块610，用于采用加速度传感器检测用户的状态；
类别压力值确定模块603，还用于当检测模块610确定用户的状态为静息状态时，采用分类模型根据检测模块确定的静息状态确定压力类别为正常压力，及正常压力对应的类别压力值。
进一步的，图6-图8中的装置是以功能模块的形式来呈现。这里的“模块”可以指特定应用集成电路(application-specific integrated circuit，ASIC)，电路，执行一个或多个软件或固件程序的处理器和存储器，集成逻辑电路，和/或其他可以提供上述功能的器件。在一个简单的实施例中，本领域的技术人员可以想到图6-图8中的装置可以采用图9所示的形式。
本申请实施例还提供了另一种确定压力值的装置，如图9所示，为了便于说明，仅示出了与本发明实施例相关的部分，具体技术细节未揭示的，请参照本发明实施例方法部分。该装置以可穿戴设备为例进行说明。图9示出的是与本申请实施例提供的终端相关的装置的部分结构的框图。参考图9，可穿戴设备包括：存储器920、输入单元930、显示单元940、传感器950、音频电路960、处理器980等部件。下面结合图9对装置的各个构成部件进行具体的介绍：
存储器920可用于存储软件程序以及模块，处理器980通过运行存储在存储器920的软件程序以及模块，从而执行可穿戴设备的各种功能应用以及数据处理。存储器920可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序等；存储数据区可存储根据可穿戴设备的使用所创建的数据等。此外，存储器920可以包括高速随机存取存储器，还可以包括非易失性存储器，例如至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
输入单元930可用于接收输入的数字或字符信息，以及产生与可穿戴设备的用户设置以及功能控制有关的键信号输入。具体地，输入单元930可包括触控面板931以及其他输入设备932。触控面板931，也称为触摸屏，可收集用户在其上或附近的触摸操作(比如用户使用手指、触笔等任何适合的物体或附件在触控面板931上或在触控面板931附近的操作)。其他输入设备932可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)。
显示单元940可用于显示由用户输入的信息或提供给用户的信息以及可穿戴设备的各种菜单及输出的压力值。
可穿戴设备还可包括至少一种传感器950，比如光传感器、运动传感器以及其他传感器、加速度传感器。各种传感器可以采集用户的生理信号，作为运动传感器的一种，加速计传感器可检测各个方向上(一般为三轴)加速度的大小，静止时可检测出重力的大小及 方向，可用于识别可穿戴设备姿态的应用(检测用户的静息状态)、振动识别相关功能(比如计步器、敲击)等；至于可穿戴设备还可配置的陀螺仪、气压计、湿度计、温度计、红外线传感器等其他传感器，在此不再赘述。
音频电路960、扬声器961，传声器962可提供用户与可穿戴设备之间的音频接口。音频电路960可将接收到的音频数据转换后的电信号，传输到扬声器961，由扬声器961转换为声音信号输出；另一方面，传声器962将收集的声音信号转换为电信号，由音频电路960接收后转换为音频数据，再将音频数据输出处理器980处理后，经RF电路910以发送给比如另一可穿戴设备，或者将音频数据输出至存储器920以便进一步处理。
处理器980是可穿戴设备的控制中心，利用各种接口和线路连接整个可穿戴设备的各个部分，通过运行或执行存储在存储器920内的软件程序和/或模块，以及调用存储在存储器920内的数据，执行可穿戴设备的各种功能和处理数据，从而对可穿戴设备进行整体监控。可选的，处理器980可包括一个或多个处理单元。
在本发明实施例中，该可穿戴设备所包括的处理器980使得可穿戴设备执行如上述方法实施例中确定压力值的装置所实际执行的方法步骤。
本申请实施例中提供了一种计算机存储介质，用于储存上述装置所用的计算机软件指令，其包含用于执行上述方法实施例中装置所执行的方法步骤。
在另一种可能的设计中，当该装置为终端内的芯片时，芯片包括：处理单元和通信单元，所述处理单元例如可以是处理器，所述通信单元例如可以是输入/输出接口、管脚或电路等。该处理单元可执行存储单元存储的计算机执行指令，以使该终端内的芯片执行上述第一方面任意一项的无线通信方法。可选地，所述存储单元为所述芯片内的存储单元，如寄存器、缓存等，所述存储单元还可以是所述终端内的位于所述芯片外部的存储单元，如只读存储器(read-only memory，ROM)或可存储静态信息和指令的其他类型的静态存储设备，随机存取存储器(random access memory，RAM)等。
其中，上述任一处提到的处理器，可以是一个通用中央处理器(CPU)，微处理器，特定应用集成电路(application-specific integrated circuit，ASIC)，或一个或多个用于控制上述第一方面无线通信方法的程序执行的集成电路。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统，装置和单元的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统，装置和方法，可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码的介质。

Claims (16)

  1. 一种确定压力值的方法，其特征在于，包括：
    采集用户的生理信号；
    根据所述生理信号获取目标特征向量；
    判断是否存在校准信息；
    若存在所述校准信息，采用校准模型根据所述校准信息和所述目标特征向量确定校准后的目标压力值，所述校准模型包括特征向量与压力值的对应关系，所述校准信息为用于所述校准模型输出所述目标压力值所使用的基准信息；
    输出所述目标压力值。
  2. 根据权利要求1所述的方法，其特征在于,所述方法还包括：
    采用分类模型根据所述目标特征向量确定压力类别对应的类别压力值，所述分类模型包括特征向量与压力类别的映射关系；
    根据所述类别压力值和所述目标压力值之间的差异参数更新所述校准信息。
  3. 根据权利要求1所述的方法，其特征在于，所述方法还包括：
    若不存在所述校准信息，采用分类模型根据所述目标特征向量确定压力类别对应的类别压力值，所述分类模型包括特征向量与压力类别的对应关系；
    将所述类别压力值和所述目标特征向量作为校准信息进行保存；
    输出所述类别压力值。
  4. 根据权利要求1所述的方法，其特征在于，所述校准信息包括校准压力值和校准特征向量，所述采用校准模型根据所述校准信息和所述目标特征向量确定校准后的目标压力值，包括：
    确定所述目标特征向量与所述校准特征向量之间的差异特征向量；
    采用所述校准模型根据所述差异向量确定所述差异特征向量对应的差异压力值；
    根据所述校准压力值和所述差异压力值确定所述目标压力值。
  5. 根据权利要求2所述的方法，其特征在于，所述根据所述类别压力值和所述目标压力值之间的差异参数更新所述校准信息，包括：
    确定所述类别压力值和所述目标压力值之间的差异参数；
    判断所述差异参数是否大于阈值；
    若所述差异参数大于所述阈值，则将计数器记录的数值增加1；
    当所述计数器记录的数值达到第一预设值时，删除所述校准信息，并将所述计数器记录的数值置零。
  6. 根据权利要求5所述的方法，其特征在于，所述方法还包括：
    若所述差异参数小于或者等于所述阈值，则判断当前时刻与上次更新所述校准信息的时刻之间的时长是否大于第二预设值；
    若当前时刻与上次更新所述校准信息的时刻之间的时长大于所述第二预设值，则删除所述校准信息，并将所述计数器记录的数值置零。
  7. 根据权利要求1-6中任一项所述的方法，其特征在于，所述方法还包括：
    采用加速度传感器检测用户的状态；
    当所述用户的状态为静息状态时，采用所述分类模型根据所述静息状态确定压力类别为正常压力，及正常压力对应的类别压力值。
  8. 一种确定压力值的装置，其特征在于，包括：
    采集模块，用于采集用户的生理信号；
    获取模块，用于根据所述采集模块采集的生理信号获取目标特征向量；
    判断模块，用于判断是否存在校准信息；
    目标压力值确定模块，当判断模块确定存在所述校准信息时，采用校准模型根据所述校准信息和所述目标特征向量确定校准后的目标压力值，所述校准模型包括特征向量与压力值的对应关系，所述校准信息用于所述校准模型输出所述目标压力值所使用的基准信息；
    输出模块，用于输出所述目标压力值确定模块确定的所述目标压力值。
  9. 根据权利要求8所述的装置，其特征在于，还包括：类别压力值确定模块和校准模块；
    所述类别压力值确定模块，用于采用分类模型根据所述目标特征向量确定压力类别对应的类别压力值，所述分类模型包括特征向量与压力类别的映射关系；
    所述校准模块，用于根据所述类别压力值确定模块确定的所述类别压力值和所述目标压力值确定模块确定的所述目标压力值之间的差异参数更新所述校准信息。
  10. 根据权利要求8所述的装置，其特征在于，所述装置还包括存储模块；
    所述类别压力值确定模块，用于采用分类模型根据所述目标特征向量确定压力类别对应的类别压力值，所述分类模型包括特征向量与压力类别的映射关系；
    所述存储模块，用于当判断模块确定不存在所述校准信息时，将所述类别压力值确定模块确定的类别压力值和所述目标特征向量作为校准信息进行保存；
    所述输出模块，还用于输出所述类别压力值确定模块确定的所述类别压力值。
  11. 根据权利要求8所述的装置，其特征在于，所述校准信息包括校准压力值和校准特征向量；所述目标压力值确定模块还具体用于：
    确定所述目标特征向量与所述校准特征向量之间的差异特征向量；
    采用所述校准模型根据所述差异向量确定所述差异特征向量对应的差异压力值；
    根据所述校准压力值和所述差异压力值确定所述目标压力值。
  12. 根据权利要求9所述的装置，其特征在于，所述校准模块还具体用于：
    确定所述类别压力值和所述目标压力值之间的差异参数；
    判断所述差异参数是否大于阈值；
    若所述差异参数大于所述阈值，则将计数器记录的数值增加1；
    当所述计数器记录的数值达到第一预设值时，删除所述校准信息，并将所述计数器记录的数值置零。
  13. 根据权利要求12所述的装置，其特征在于，所述装置还包括更新模块；
    所述判断模块，还用于当所述差异参数小于或者等于所述阈值时，判断当前时刻与上次更新所述校准信息的时刻之间的时长是否大于第二预设值；
    所述更新模块，还用于当前时刻与上次更新所述校准信息的时刻之间的时长大于所述第二预设值时，则删除所述校准信息，并将所述计数器记录的数值置零。
  14. 根据权利要求8-13中任一项所述的装置，其特征在于，还包括检测模块；
    所述检测模块，用于采用加速度传感器检测用户的状态；
    所述类别压力值确定模块，还用于当所述用户的状态为静息状态时，采用所述分类模型根据所述检测模块确定的所述静息状态确定压力类别为正常压力，及正常压力对应的类别压力值。
  15. 一种确定压力值的装置，其特征在于，包括：
    存储器，用于存储计算机可执行程序代码；
    处理器，与所述存储器耦合；
    其中所述程序代码包括指令，当所述处理器执行所述指令时，所述指令使所述装置执行如权利要求1-7中任一项所述的方法。
  16. 一种计算机存储介质，其特征在于，用于储存上述注册服务器所用的计算机软件指令，其包含用于执行如权利要求1-7中任一项所述的方法。
PCT/CN2018/081571 2018-04-02 2018-04-02 一种确定压力值的方法及装置 WO2019191864A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
CN201880072336.5A CN111315296B (zh) 2018-04-02 2018-04-02 一种确定压力值的方法及装置
PCT/CN2018/081571 WO2019191864A1 (zh) 2018-04-02 2018-04-02 一种确定压力值的方法及装置

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2018/081571 WO2019191864A1 (zh) 2018-04-02 2018-04-02 一种确定压力值的方法及装置

Publications (1)

Publication Number Publication Date
WO2019191864A1 true WO2019191864A1 (zh) 2019-10-10

Family

ID=68099759

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/081571 WO2019191864A1 (zh) 2018-04-02 2018-04-02 一种确定压力值的方法及装置

Country Status (2)

Country Link
CN (1) CN111315296B (zh)
WO (1) WO2019191864A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114490374A (zh) * 2022-01-21 2022-05-13 阿里巴巴（中国）有限公司 应用的压力测试方法、电子设备、存储介质及程序产品

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103610456A (zh) * 2013-12-08 2014-03-05 季忠 基于生理信号特征参数的颅脑压力无创监测分析系统
CN104239288A (zh) * 2014-09-18 2014-12-24 清华大学 基于微博的青少年心理压力检测和疏导的方法及系统
CN104490407A (zh) * 2014-12-08 2015-04-08 清华大学 一种可穿戴式心理压力评测装置及方法
US20150120205A1 (en) * 2013-10-24 2015-04-30 Samsung Electronics Co., Ltd. System and method for managing stress
CN105559803A (zh) * 2015-12-17 2016-05-11 国家电网公司 一种电网企业工作人员心理压力状况评估方法
CN106934199A (zh) * 2015-12-31 2017-07-07 深圳先进技术研究院 一种心理压力评估方法及装置
CN106923846A (zh) * 2015-12-31 2017-07-07 深圳先进技术研究院 一种心理压力评估与反馈系统

Family Cites Families (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100580618B1 (ko) * 2002-01-23 2006-05-16 삼성전자주식회사 생리 신호의 단시간 모니터링을 통한 사용자 정서 인식장치 및 방법
EP1895890B1 (en) * 2005-03-03 2013-05-22 Ull Meter A/S Evaluation of sympathetic tone
US8157730B2 (en) * 2006-12-19 2012-04-17 Valencell, Inc. Physiological and environmental monitoring systems and methods
US20100295244A1 (en) * 2007-04-04 2010-11-25 Koninklijke Philips Electronics N.V. Determining stress level based on the performance of the person in a game or puzzle
US20090156886A1 (en) * 2007-12-12 2009-06-18 Synapse Research Company Method and apparatus for providing automatic eye focused therapy
US9173567B2 (en) * 2011-05-13 2015-11-03 Fujitsu Limited Triggering user queries based on sensor inputs
CN103674350B (zh) * 2012-09-17 2016-06-01 联想(北京)有限公司 一种压力传感器和检测压力值的方法
EP3328278A4 (en) * 2015-08-02 2019-04-03 G Medical Innovations Holdings Ltd DEVICE, SYSTEM AND METHOD FOR NONINVASIVE MONITORING OF PHYSIOLOGICAL PARAMETERS
CN106859618A (zh) * 2017-03-21 2017-06-20 中国计量大学 一种基于人体状态的健康分析系统
CN107233102A (zh) * 2017-05-26 2017-10-10 重庆邮电大学 基于bp神经网络算法的多参数心理压力评估方法

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150120205A1 (en) * 2013-10-24 2015-04-30 Samsung Electronics Co., Ltd. System and method for managing stress
CN103610456A (zh) * 2013-12-08 2014-03-05 季忠 基于生理信号特征参数的颅脑压力无创监测分析系统
CN104239288A (zh) * 2014-09-18 2014-12-24 清华大学 基于微博的青少年心理压力检测和疏导的方法及系统
CN104490407A (zh) * 2014-12-08 2015-04-08 清华大学 一种可穿戴式心理压力评测装置及方法
CN105559803A (zh) * 2015-12-17 2016-05-11 国家电网公司 一种电网企业工作人员心理压力状况评估方法
CN106934199A (zh) * 2015-12-31 2017-07-07 深圳先进技术研究院 一种心理压力评估方法及装置
CN106923846A (zh) * 2015-12-31 2017-07-07 深圳先进技术研究院 一种心理压力评估与反馈系统

Also Published As

Publication number Publication date
CN111315296A (zh) 2020-06-19
CN111315296B (zh) 2021-06-01

Similar Documents

Publication Publication Date Title
JP6659830B2 (ja) 生体情報分析装置、システム、及び、プログラム
US20210275104A1 (en) Apparatus and method for correcting error of bio-information sensor, and apparatus and method for estimating bio-information
Pierleoni et al. An android-based heart monitoring system for the elderly and for patients with heart disease
JP6676499B2 (ja) 疲労度判定装置、疲労度判定方法、疲労度判定プログラム、及び、生体情報測定装置
US11617545B2 (en) Methods and systems for adaptable presentation of sensor data
EP3801227A1 (en) Method and apparatus for estimating a trend in a blood pressure surrogate
JP6831382B2 (ja) 血圧測定の信頼性を評価する方法およびこれを実装する装置
WO2019191864A1 (zh) 一种确定压力值的方法及装置
CN109890276B (zh) 血压监测方法、装置和设备
CN112120715A (zh) 压力监测与缓解系统
RU2782412C2 (ru) Способ и устройство для количественной оценки тенденции в суррогатном показателе кровяного давления
WO2021141572A1 (en) Methods and systems for adaptable presentation of sensor data

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18913925

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18913925

Country of ref document: EP

Kind code of ref document: A1