Mood state monitoring method, device and storage medium

Abstract

The present application provides a mood state monitoring method, device, and storage medium. The method obtains autonomic nervous activity signals from a target individual; the autonomic nervous activity signals include multiple time-segmented autonomic nervous activity signals; extracts target sequences from the multiple time-segmented autonomic nervous activity signals to obtain multiple time-segmented physiological data; and calculates and compares the multiple time-segmented physiological data based on a preset feature space to determine the target individual's mood state. The preset feature space includes a preset stress level assessment feature space, a preset parasympathetic nervous system forced regulation state detection feature space, and a preset mood state assessment feature space. This method enables all-day mood state monitoring of a population, thereby effectively and completely inferring the population's mood state.

Description

Technical Field

[0001] The present application relates to the technical field of mood state monitoring, and in particular to a mood state monitoring method, device and storage medium. Background Art

[0002] A healthy state of mind, characterized by a balanced and harmonious relationship between the mind and body, corresponds to a healthy state of mind. A sub-healthy state, characterized by an unbalanced state between the mind and body, corresponds to a negative state of mind. Therefore, mood is closely related to human health. Furthermore, existing literature clearly indicates that short-term emotions can alter the activity patterns of the sympathetic and parasympathetic nervous systems. Therefore, monitoring a person's mood has the potential to enhance the application of research into human health.

[0003] Currently, emotional state detection technologies mainly focus on short-term and intense emotions, such as the research results of binary and tertiary classification of stress levels. There are also technologies for stress scoring. However, the stress levels and scoring results given by the above existing research and technologies lack empirical validation criteria, resulting in a lack of horizontal comparability of stress levels and scores. In addition, it is impossible to detect all-weather and diffuse mood states. Therefore, it is impossible to effectively and completely infer the mood state of the population based on the above stress levels and scoring results. Summary of the Invention

[0004] The purpose of the embodiments of the present invention is to provide a mood state monitoring method, device and storage medium, which can accurately infer the mood state of a person by calculating and comparing sequences extracted from the autonomic nervous activity signals of a person at multiple time periods throughout the day using multiple preset feature spaces.

[0005] In a first aspect, an embodiment of the present application provides a method for monitoring a mood state, the method comprising:

[0006] Acquire an autonomic nervous activity signal of a target individual; wherein the autonomic nervous activity signal includes a plurality of autonomic nervous activity signals in different time periods;

[0007] Extracting target sequences from the multiple time-divided autonomic nerve activity signals respectively to obtain multiple time-divided physiological data;

[0008] The physiological data of the multiple time periods are calculated and compared according to a preset feature space to determine the mood state of the target individual; wherein the preset feature space includes: a preset stress level assessment feature space, a preset parasympathetic nerve forced regulation state detection feature space, and a preset mood state assessment feature space.

[0009] The above-described mood state monitoring method determines the target individual's mood state by calculating and comparing target sequences extracted from the target individual's autonomic nervous system activity signals at multiple time intervals using a preset stress level assessment feature space, a preset parasympathetic nervous system forced regulation state detection feature space, and a preset mood state assessment feature space. Because the physiological data in the preset feature space is universally applicable and horizontally comparable, and distributed joint detection of physiological data from multiple time intervals ensures the integrity of mood state detection results, it enables effective and complete detection of an individual's mood state when monitoring an individual's physiological data collected in real time.

[0010] Optionally, obtaining the autonomic nervous activity signal of the target individual includes:

[0011] Acquiring physiological signals of the target individual collected by a medical sensor; wherein the physiological signals include autonomic nerve activity signals and three-dimensional acceleration signals of the human body;

[0012] Segmenting and calculating the autonomic nerve activity signal and the human body three-dimensional acceleration signal to obtain the human body three-dimensional acceleration signal and the autonomic nerve activity signal in multiple time periods;

[0013] Determining whether the multiple time-sharing three-dimensional human body acceleration signals are greater than a preset three-dimensional human body acceleration signal;

[0014] If yes, determining the autonomic nervous activity signals corresponding to the human body three-dimensional acceleration signals in the multiple time periods as the autonomic nervous activity signals of the non-target individual;

[0015] If not, the autonomic nervous activity signals corresponding to the three-dimensional acceleration signals of the human body in the multiple time periods are determined as the autonomic nervous activity signals of the target individual.

[0016] The above-mentioned mood state monitoring method segments and processes the target individual's autonomic nervous activity signals and three-dimensional human acceleration signals collected by a medical sensor to obtain multiple time-segmented three-dimensional human acceleration signals and autonomic nervous activity signals. A preset three-dimensional human acceleration signal is then used to distinguish whether physical activity (three-dimensional human acceleration signals) in the multiple time-segmented periods affects autonomic nervous activity. Time-segmented three-dimensional human acceleration signals and autonomic nervous activity signals greater than the preset three-dimensional human acceleration signal are then eliminated to obtain the target autonomic nervous activity signal. Because time-segmented three-dimensional human acceleration signals greater than the preset three-dimensional human acceleration signal can interfere with mood state detection, eliminating these time-segmented three-dimensional human acceleration signals further ensures the validity of mood state detection results.

[0017] Optionally, calculating and comparing the physiological data of the multiple time periods according to a preset feature space to determine the mood state of the target individual includes:

[0018] According to the preset stress level assessment feature space, respectively assessing the stress levels of the physiological data of the multiple time periods to obtain the physiological data of the multiple time periods after the stress level assessment, and calculating the physiological data of the multiple time periods according to a preset stress scoring method to obtain stress scores of the physiological data of the multiple time periods;

[0019] performing parasympathetic nerve forced regulation detection on the physiological data of the multiple time periods according to the preset parasympathetic nerve forced regulation state detection feature space, and obtaining parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods;

[0020] According to the preset mood state assessment feature space, the stress scores and parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods are analyzed and compared to determine the mood state of the target individual.

[0021] The above-mentioned mood state monitoring method determines an individual's mood state by using a preset stress level assessment feature space to assign stress scores to physiological data from multiple time periods, a preset parasympathetic nerve forced regulation state detection feature space to detect the time distribution and frequency of parasympathetic nerve forced regulation from physiological data from multiple time periods, and a preset mood state assessment feature space to analyze and compare the stress scores and parasympathetic nerve forced regulation time and frequency from physiological data from multiple time periods. Because the physiological data samples in the preset stress level assessment space, the preset parasympathetic nerve forced regulation state detection feature space, and the preset mood state assessment feature space are calibrated using multiple effective empirical criteria and selecting multiple physiological indicators that can reveal key characteristics of mental stress and parasympathetic nerve forced regulation, the universal applicability and cross-sectional comparability of the physiological data samples are ensured. Furthermore, by combining stress scores and parasympathetic nerve forced regulation in physiological data from multiple time periods to monitor mood state, the integrity of the mood state detection results is ensured, thereby enabling the effective and complete detection of an individual's mood state when monitoring physiological data collected in real time.

[0022] Optionally, the preset pressure level assessment feature space is constructed in the following manner:

[0023] According to the preset pressure level experience standard, the preset physiological data is calibrated at the pressure level to obtain the preset physiological data of multiple pressure levels;

[0024] According to the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms, the preset physiological data of the multiple pressure levels are calculated to obtain the calculation results of the preset physiological data of the multiple pressure levels, and the preset pressure level assessment feature space is constructed according to the calculation results.

[0025] The above-mentioned mood state monitoring method calibrates the preset physiological data to a pressure level using a preset stress level empirical criterion, and then represents the preset physiological data after the pressure level calibration using a periodic function of a target sequence rhythm and an average fluctuation function of two target sequence rhythms that reflect the key characteristics of the law of autonomic nervous activity, thereby obtaining a regular distribution of the preset physiological data in a feature space, which is used as a preset stress level assessment feature space. Since an effective preset stress level empirical criterion is used to calibrate the preset physiological data to a pressure level, preset physiological data of multiple pressure levels are obtained. Secondly, the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms are used to support the scatter distribution of the preset physiological data of multiple pressure levels in the preset stress level assessment space, thereby ensuring the effectiveness, general applicability and horizontal comparability of the pressure level calibration of the preset physiological data, so that the constructed preset stress level assessment feature space can effectively assess the stress level of the physiological data collected in real time.

[0026] Optionally, the preset stress scoring method includes Formula 1: Score is the stress score of the physiological data of the time period; x is the physiological data of the time period; x1 is the class center of the preset physiological data corresponding to the stress level of the physiological data of the time period; s is the standard deviation within the class of the preset physiological data; j is the stress level of the physiological data of the time period;

[0027] The step of performing stress level assessment on the physiological data of the multiple time periods according to the preset stress level assessment feature space to obtain the physiological data of the multiple time periods after the stress level assessment, and calculating the physiological data of the multiple time periods according to a preset stress scoring method to obtain stress scores for the physiological data of the multiple time periods includes:

[0028] Calculating the physiological data of the multiple time periods according to the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms to obtain calculation results of the physiological data of the multiple time periods;

[0029] Mapping the calculation results of the physiological data of the multiple time periods to the preset pressure level assessment feature space, and performing distance calculations between the cluster centers of the calculation results of the preset physiological data of the multiple pressure levels and the calculation results, to obtain distances between the calculation results of the physiological data of the multiple time periods and the cluster centers of the calculation results of the preset physiological data of the multiple pressure levels;

[0030] Determining a plurality of minimum distances from the distances between the calculated results of the physiological data of the plurality of time periods and the cluster centers of the calculated results of the preset physiological data of the plurality of pressure levels, and using the pressure levels of the preset physiological data corresponding to the plurality of minimum distances as the pressure levels of the physiological data of the plurality of time periods;

[0031] The stress scores of the physiological data of the multiple time periods are obtained according to Formula 1, the physiological data of the multiple time periods, the class centers of the preset physiological data of the stress levels corresponding to the physiological data of the multiple time periods, the class standard deviations of the preset physiological data and the stress levels of the physiological data of the multiple time periods.

[0032] The above-mentioned mood state monitoring method uses a periodic function of a target sequence rhythm and an average fluctuation function of two target sequence rhythms to represent the scatter distribution in a preset stress level assessment feature space using physiological data from multiple time periods. The method then observes the proximity of the physiological data from the multiple time periods to the preset physiological data for multiple stress levels, and uses the stress level of the preset physiological data closest to the physiological data from the multiple time periods as the stress level of the physiological data for the multiple time periods. Finally, the physiological data from the multiple time periods for which the stress level has been assessed are scored according to a preset scoring method to obtain stress scores for the physiological data from the multiple time periods. Because the preset stress level assessment feature space can effectively assess the stress level of physiological data, the raw stress score calculated when the stress score of the physiological data for which the stress level has been assessed is also valid, thereby providing an effective basis for stress scoring in mood state detection.

[0033] Optionally, the preset parasympathetic nerve forced regulation state detection feature space is constructed in the following manner:

[0034] Calibrate the preset physiological data according to the preset stress level empirical effect standard, the preset fatigue theoretical model, and the preset work and rest and eating irregularity effect standard to obtain a plurality of first physiological data and second physiological data; wherein the first physiological data is calibrated as parasympathetic nervous system forced regulation;

[0035] Performing continuous wavelet transform on the plurality of first physiological data and the second physiological data according to the wavelet basis function of the preset transformation scale interval, and obtaining a geometric morphology approximation degree measurement index of the plurality of first physiological data and the second physiological data and the physiological data of the plurality of time periods;

[0036] Calculating the average difference between the plurality of first physiological data and the second physiological data and the adjacent first physiological data and the second physiological data to obtain the average difference between the plurality of first physiological data and the second physiological data;

[0037] Acquire power values of the plurality of first physiological data and second physiological data at a preset frequency;

[0038] Obtaining calculation results of the plurality of first physiological data and the second physiological data based on the maximum transformation scale, the average difference, and the power value at the preset frequency of the plurality of first physiological data and the second physiological data;

[0039] The parasympathetic nerve forced regulation state detection feature space is constructed according to calculation results of the plurality of first physiological data and the second physiological data.

[0040] The above-mentioned mood state monitoring method evaluates the preset physiological data through a preset stress level empirical criterion and a preset fatigue theoretical model to obtain first physiological data and second physiological data, and then represents the first physiological data and the second physiological data using a wavelet approximation degree measurement index of a preset transformation scale interval that reflects the key characteristics of the law of autonomic nervous activity, an index of target sequence variation, and a relative activation degree index within a preset frequency, thereby obtaining a regular distribution of the preset physiological data in a feature space, which serves as a preset parasympathetic nervous system forced regulation state detection feature space. Since the preset physiological data are calibrated by using effective preset pressure level empirical criterion, preset fatigue theoretical model and other preset special events that lead to parasympathetic nerve forced regulation, physiological data with parasympathetic nerve forced regulation and without parasympathetic nerve forced regulation are obtained. Secondly, according to the wavelet approximation degree measurement index of the preset transformation scale interval, the target sequence variation index and the relative activation degree index within the preset frequency, the scatter distribution of physiological data with parasympathetic nerve forced regulation and without parasympathetic nerve forced regulation in the preset parasympathetic nerve forced regulation state detection feature space is supported, which ensures the effectiveness, universal applicability and horizontal comparability of the preset physiological data parasympathetic nerve forced regulation detection, so that the constructed preset parasympathetic nerve forced regulation state detection feature space can effectively perform parasympathetic nerve forced regulation detection on physiological data collected in real time.

[0041] Optionally, performing parasympathetic nerve forced regulation detection on the physiological data of the multiple time periods according to the preset parasympathetic nerve forced regulation state detection feature space to obtain the parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods includes:

[0042] Performing wavelet transform on the physiological data of the multiple time periods to obtain geometric morphology approximation degree measurement indicators of the physiological data of the multiple time periods;

[0043] Calculating the average difference between the physiological data of the multiple time periods and the physiological data of the adjacent time periods to obtain the average difference of the physiological data of the multiple time periods;

[0044] Acquire power values of the physiological data of the plurality of time periods at the preset frequency;

[0045] Obtaining calculation results of the physiological data of the multiple time periods based on the geometric morphology approximation degree measurement index, the average difference and the power value at the preset frequency of the multiple time periods;

[0046] Mapping the calculation results of the multiple time-divided physiological data to the preset parasympathetic nerve forced regulation state detection feature space, and performing distance calculations between the cluster centers of the calculation results of the multiple first physiological data and the second physiological data and the calculation results, respectively, to obtain distances between the calculation results of the multiple time-divided physiological data and the cluster centers of the calculation results of the multiple first physiological data and the second physiological data;

[0047] Determining a plurality of minimum distances respectively from the distances between the calculation results of the physiological data of the plurality of time periods and the cluster centers of the calculation results of the plurality of first physiological data and second physiological data;

[0048] If the preset physiological data corresponding to the multiple minimum distances is the first physiological data, the multiple time-divided physiological data corresponding to the multiple minimum distances are marked as parasympathetic nerve forced regulation, and the number of parasympathetic nerve forced regulation of the physiological data of the multiple time-divided time periods is recorded to obtain the forced regulation frequency of the physiological data of the multiple time-divided time periods.

[0049] The above-mentioned mood state monitoring method uses a wavelet basis function with a preset transformation scale interval, an index of target sequence variation, and a relative activation degree index within a preset frequency range to represent the scatter point distribution in a preset parasympathetic nerve forced modulation state detection feature space based on physiological data from multiple time periods. The method then observes the proximity of the physiological data from the multiple time periods to the preset physiological data with and without parasympathetic nerve forced modulation, labels the physiological data from the multiple time periods closest to the parasympathetic nerve forced modulation as parasympathetic nerve forced modulation, then records the number of parasympathetic nerve forced modulations, and finally determines the parasympathetic nerve forced modulation frequency of the physiological data from the multiple time periods based on the duration of the time periods and the number of parasympathetic nerve forced modulations. Since the preset parasympathetic nerve forced modulation state detection feature space can effectively detect parasympathetic nerve forced modulation in physiological data, it provides an effective basis for mood state detection in terms of parasympathetic nerve forced modulation frequency.

[0050] Optionally, the preset mood state assessment feature space is constructed in the following manner:

[0051] According to a preset self-assessment form, the preset physiological data is calibrated for the mood state to obtain the preset physiological data of the calibrated mood state;

[0052] Calculating the stress score and parasympathetic nerve forced regulation time distribution and frequency of the preset physiological data of the calibrated mood state, and obtaining the preset stress score, preset duration ratio, and preset parasympathetic nerve forced regulation time distribution and frequency for a good mood state and a poor mood state;

[0053] The preset mood state assessment feature space is constructed according to the preset stress scores, preset duration ratios, and preset parasympathetic nervous system forced regulation time distribution and frequency of good mood state and bad mood state.

[0054] The above-mentioned mood state monitoring method calibrates the preset physiological data for mood state using a preset self-assessment form, and through the stress score and parasympathetic nerve forced regulation detection of the physiological data of the determined mood state, obtains the preset stress score, preset duration ratio, and preset parasympathetic nerve forced regulation time distribution and frequency for good and poor mood states, and constructs a preset mood state feature space based on the above stress score and parasympathetic nerve forced regulation time distribution and frequency. Because the preset physiological data is calibrated for mood state using a preset self-assessment form, the stress score and parasympathetic nerve forced regulation of the physiological data of the calibrated mood state are analyzed to obtain the preset stress score, preset duration ratio, and preset forced regulation frequency for good and poor mood states, the validity of the various mood state indicators of the preset physiological data is ensured, so that the preset mood state feature space can effectively detect the mood state of the physiological data collected in real time.

[0055] Optionally, analyzing and comparing the stress scores and forced adjustment frequencies of the physiological data of the multiple time periods according to the preset mood state assessment feature space to determine the mood state of the target individual includes:

[0056] Grouping the stress scores of the physiological data of the multiple time periods according to the work and rest time of the target individual to obtain stress scores of the physiological data of the multiple wakefulness states and the sleep states;

[0057] The parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods are grouped according to a preset stress level empirical criterion and a preset fatigue theoretical model to obtain the parasympathetic nerve forced regulation time distribution and frequency of multiple physiological fatigue prone times and physiological fatigue non-prone times;

[0058] Determining, from the stress scores of the physiological data of the multiple wakefulness states, a stress score of the physiological data of the wakefulness state that is less than the preset stress score of the good mood state, to obtain a stress score of the physiological data of the first wakefulness state;

[0059] Determining the stress score of the physiological data of the sleep state having a stress score of zero from the stress scores of the physiological data of the multiple sleep states, to obtain the stress score of the physiological data of the first sleep state;

[0060] If the ratio of the duration corresponding to the stress score of the physiological data of the first wakefulness state to the duration of the multiple time periods is greater than the preset ratio of the duration of the good mood state, the ratio of the duration corresponding to the stress score of the physiological data of the first sleep state to the duration of the multiple time periods is greater than the preset ratio of the duration of the good mood state, and the forced adjustment frequency of the multiple physiological fatigue times is less than the preset adjustment frequency of the good mood state, then the mood state is determined to be good.

[0061] The above-mentioned mood state monitoring method groups the stress scores and parasympathetic nervous system forced modulation of physiological data from multiple time periods according to the target individual's sleep and work schedule, a preset stress level empirical criterion, and a preset fatigue theoretical model. This method then determines the stress scores of the physiological data for the wakefulness and sleep states, as well as the parasympathetic nervous system forced modulation frequencies of the physiological data for the physiologically fatigued state and the physiologically non-fatigue period. The method then determines the individual's mood state as being good if the stress score of the physiological data for the wakefulness state is less than the preset stress score for a good mood and is greater than the preset proportion of the good mood, the stress score of the sleep state is zero for a greater than the preset proportion of the good mood, and the parasympathetic nervous system forced modulation frequencies of the physiological data for the physiologically fatigued state are less than the preset forced modulation frequencies for a good mood. Because the stress scores of the physiological data for the wakefulness and sleep states and the parasympathetic nervous system forced modulation frequencies for the physiologically non-fatigue period are determined using the preset mood state feature space, the detection result indicating that the target individual's mood state is good is both valid and complete.

[0062] Optionally, analyzing and comparing the stress scores and forced adjustment frequencies of the physiological data of the multiple time periods according to the feature space of the preset mood state assessment to determine the mood state includes:

[0063] Determining, from the stress scores of the physiological data of the multiple wakefulness states, a stress score of the physiological data of the wakefulness state that is greater than the preset stress score of the poor mood state, to obtain a stress score of the physiological data of the second wakefulness state;

[0064] Determining, from the stress scores of the physiological data of the plurality of sleep states, a stress score of the physiological data of the sleep state that is greater than the preset stress score of the poor mood state, to obtain a stress score of the physiological data of the second sleep state;

[0065] If the ratio of the duration corresponding to the stress score of the physiological data of the second wakefulness state to the duration of the multiple time periods is greater than the preset ratio of the duration of the poor mood state, the ratio of the duration corresponding to the stress score of the physiological data of the second sleep state to the duration of the multiple time periods is greater than the preset ratio of the duration of the poor mood state, and the forced adjustment frequency of the multiple physiological fatigue-prone times is greater than the preset adjustment frequency of the poor mood state, then the mood state is determined to be poor.

[0066] The above-mentioned mood state monitoring method determines that the individual's mood state is poor when the stress score of the physiological data in the awake state is greater than the preset stress score for poor mood, the proportion of time that the stress score of the physiological data in the sleep state is greater than the preset stress score for poor mood, and the proportion of time that the stress score of the physiological data in the sleep state is greater than the preset stress score for poor mood, and the parasympathetic nervous system forced modulation frequency of the physiological data during the physiological fatigue-free period is greater than the preset forced modulation frequency for poor mood. Because the stress score of the physiological data in the awake state and the sleep state and the parasympathetic nervous system forced modulation frequency during the physiological fatigue-free period are judged using the preset mood state feature space, the detection result that the target individual's mood state is poor is valid and complete.

[0067] In a second aspect, an embodiment of the present application further provides a mood state monitoring device, the device comprising:

[0068] An acquisition module, configured to acquire an autonomic nervous activity signal of a target individual; wherein the autonomic nervous activity signal includes a plurality of autonomic nervous activity signals in different time periods;

[0069] An extraction module, configured to extract target sequences from the plurality of time-divided autonomic nerve activity signals, respectively, to obtain physiological data of the plurality of time-divided periods;

[0070] An analysis module is configured to calculate and compare the physiological data of the multiple time periods according to a preset feature space to determine the mood state of the target individual; wherein the preset feature space includes: a preset stress level assessment feature space, a preset parasympathetic nervous system forced regulation state detection feature space, and a preset mood state assessment feature space.

[0071] The mood state monitoring device provided in the above embodiment has the same beneficial effects as the mood state monitoring method provided in the above first aspect or any optional implementation manner of the first aspect, and will not be described in detail here.

[0072] In a third aspect, an embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the method described above is executed.

[0073] The computer-readable storage pool medium provided in the above embodiment has the same beneficial effects as the mood state monitoring method provided in the above first aspect, or any optional implementation of the first aspect, and is not described in detail here.

[0074] In summary, the present application provides a mood state monitoring method, device, and storage medium. This method acquires autonomic nervous activity signals from a target individual; the autonomic nervous activity signals include multiple time-segmented autonomic nervous activity signals; extracts target sequences from each of the multiple time-segmented autonomic nervous activity signals to obtain multiple time-segmented physiological data; and calculates and compares the multiple time-segmented physiological data based on a preset feature space to determine the target individual's mood state. The preset feature space includes a preset stress level assessment feature space, a preset parasympathetic nervous system forced regulation state detection feature space, and a preset mood state assessment feature space. This method enables all-day mood state monitoring of a population, thereby effectively and completely inferring the population's mood state. BRIEF DESCRIPTION OF THE DRAWINGS

[0075] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the embodiments of the present application. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without creative work.

[0076] Figure 1 A block diagram of an electronic device provided in an embodiment of the present application;

[0077] Figure 2 A schematic diagram of a first flow chart of a mood state monitoring method provided in an embodiment of the present application;

[0078] Figure 3 A second flow chart of the mood state monitoring method provided in an embodiment of the present application;

[0079] Figure 4 A schematic diagram of the waveform of a three-dimensional acceleration signal of a human body provided in an embodiment of the present application;

[0080] Figure 5 A schematic diagram of a waveform showing significant changes in autonomic nervous system activity caused by a three-dimensional acceleration signal of the human body provided in an embodiment of the present application;

[0081] Figure 6 A schematic diagram of a preset pressure level assessment feature space provided in an embodiment of the present application;

[0082] Figure 7 A schematic diagram of the stress score, heart rate sequence, parasympathetic nerve forced regulation marker, and human body three-dimensional acceleration signal of a subject in a good mood around the clock provided in the embodiments of the present application;

[0083] Figure 8A schematic diagram of the stress score, heart rate sequence, parasympathetic nerve forced regulation marker, and human body three-dimensional acceleration signal of a subject in a bad mood around the clock provided in the embodiments of the present application;

[0084] Figure 9 This is a schematic diagram of the structure of the mood state monitoring device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0085] The following embodiments of the technical solution of the present application will be described in detail with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present application and are therefore only examples and are not intended to limit the scope of protection of the present application.

[0086] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0087] In the description of the embodiments of this application, the technical terms "first," "second," etc. are used only to distinguish different objects and should not be understood to indicate or imply relative importance or to implicitly indicate the quantity, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, the meaning of "plurality" is more than two, unless otherwise specifically defined.

[0088] To facilitate understanding of this embodiment, the electronic device that executes the mood state monitoring method disclosed in the embodiment of this application is first introduced in detail.

[0089] like Figure 1 , which is a block diagram of an embodiment of an electronic device. The electronic device 100 may include a memory 111, a storage controller 112, a processor 113, a peripheral interface 114, an input and output unit 115, and a display unit 116. It will be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the electronic device 100. For example, the electronic device 100 may further include Figure 1 More or fewer components than shown, or with Figure 1 Different configurations shown.

[0090] The aforementioned memory 111, storage controller 112, processor 113, peripheral interface 114, input / output unit 115, and display unit 116 are electrically connected to each other, directly or indirectly, to enable data transmission or interaction. For example, these components may be electrically connected to each other via one or more communication buses or signal lines. The aforementioned processor 113 is used to execute the executable modules stored in the memory.

[0091] The memory 111 may be, but is not limited to, a random access memory (RAM), a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), etc. The memory 111 is used to store programs, and the processor 113 executes the programs after receiving an execution instruction. The method executed by the electronic device 100 defined by the process disclosed in any embodiment of the present application can be applied to the processor 113 or implemented by the processor 113.

[0092] The processor 113 may be an integrated circuit chip with signal processing capabilities. The processor 113 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. The methods, steps, and logic block diagrams disclosed in the embodiments of the present application may be implemented or executed. The general-purpose processor may be a microprocessor or any conventional processor.

[0093] The peripheral interface 114 couples various input / output devices to the processor 113 and the memory 111. In some embodiments, the peripheral interface 114, the processor 113, and the memory controller 112 can be implemented in a single chip. In other embodiments, they can be implemented in separate chips.

[0094] The input / output unit 115 is used to provide input data to the user and can be, but not limited to, a mouse and a keyboard.

[0095] The display unit 116 provides an interactive interface (e.g., a user operation interface) between the electronic device 100 and the user or is used to display image data for the user's reference. In this embodiment, the display unit can be a liquid crystal display or a touch display. If it is a touch display, it can be a capacitive touch screen or a resistive touch screen that supports single-point and multi-point touch operations. Supporting single-point and multi-point touch operations means that the touch display can sense touch operations generated simultaneously from one or more locations on the touch display, and pass the sensed touch operations to the processor for calculation and processing.

[0096] The electronic device 100 in this embodiment can be used to execute each step in each method provided in the embodiments of the present application. The following describes in detail the implementation process of the mood state monitoring method through several embodiments.

[0097] The principles of the mood state monitoring method provided in the embodiments of this application are as follows:

[0098] From a biological perspective, the autonomic nervous system is mainly divided into the sympathetic and parasympathetic nervous systems. They restrict each other and maintain the balance of the body's regulatory functions. That is, when people encounter special events (that is, when their mood state changes), the sympathetic and parasympathetic nerves will antagonize to maintain the body's balance (that is, the parasympathetic nerves are forced to regulate), and the forced regulation of the parasympathetic nerves is reflected in the changing patterns of autonomic nervous activity signals.

[0099] From a psychological perspective, when people encounter special events (that is, when their mood state changes), they will show different physiological reactions, behavioral manifestations under stress, psychological experiences of stress responses, etc., so mental stress needs to be calibrated. At different levels of mental stress, the corresponding changes in autonomic nervous activity signals will also be different.

[0100] In summary, a person's mood is determined by multiple factors, including their stress score, duration, and onset time, as well as the frequency and onset time of parasympathetic nervous system forced modulation. Therefore, to determine a person's mood, it's necessary to determine their stress score and parasympathetic nervous system forced modulation frequency. The parasympathetic nervous system forced modulation frequency and stress level correspond to the changing patterns of autonomic nervous system activity signals.

[0101] The mood state monitoring method provided in the embodiments of this application precisely calibrates the mental stress level and parasympathetic nervous system forced modulation state of a population beforehand. It then identifies the corresponding patterns of change in the population's autonomic nervous system activity signals. This clarifies the patterns of change in autonomic nervous system activity signals under different mental stress levels and under parasympathetic nervous system forced modulation. Therefore, in subsequent practical applications, the mental stress level and the presence of parasympathetic nervous system forced modulation can be determined simply by observing whether the autonomic nervous system activity signals meet these patterns.

[0102] See Figure 2 The figure shows a first flow chart of the mood state monitoring method provided by an embodiment of the present application.

[0103] Step S100: Acquire an autonomic nervous activity signal of a target individual; wherein the autonomic nervous activity signal includes a plurality of autonomic nervous activity signals in different time periods;

[0104] The target individual is the person whose mood state is to be monitored.

[0105] Autonomic nervous activity signals include but are not limited to electrocardiogram, pulse wave, heart rate, pulse rate and heart rate, etc., which are not limited in the embodiments of the present application.

[0106] It should be noted that the embodiment of the present application is to monitor the mood state of the target individual around the clock, so the autonomic nervous activity signals are collected for 24 hours, and 24 hours are divided into multiple time periods. Therefore, the autonomic nervous signal of the target individual is divided into autonomic nervous activity signals of multiple time periods.

[0107] Step S200: extracting target sequences from the autonomic nerve activity signals of the multiple time periods to obtain physiological data of the multiple time periods;

[0108] The physiological data of multiple time periods refers to a collection of target sequences of multiple time periods.

[0109] The target sequence refers to the rhythm time series extracted from the autonomic nervous activity signal. For example, when extracting a sequence from the electrocardiogram, the result is the electrocardiogram rhythm time series; when extracting a sequence from the heart rate, the result is the heart rate rhythm time series.

[0110] In addition, since different autonomic nervous activity signals have different display states, different autonomic nervous activity signals require corresponding signal processing before sequences can be extracted from the autonomic nervous activity signals. The embodiments of the present application do not specifically limit the signal processing methods for different autonomic nervous activities, and the corresponding signal processing method is selected according to the selected autonomic nervous activity signal.

[0111] Step S300: Calculate and compare physiological data of multiple time periods according to a preset feature space to determine the mood state of the target individual; wherein the preset feature space includes: a preset stress level assessment feature space, a preset parasympathetic nerve forced regulation state detection feature space, and a preset mood state assessment feature space.

[0112] The preset stress level assessment feature space is used to perform stress level determination and stress scoring on physiological data of multiple time periods, so as to obtain stress scores of the physiological data of multiple time periods.

[0113] The preset parasympathetic nerve forced regulation state monitoring feature space is used to detect the parasympathetic nerve forced regulation frequency of the physiological data of multiple time periods to obtain the parasympathetic nerve forced regulation frequency of the physiological data of multiple time periods.

[0114] The preset mood state feature space is used to analyze the physiological data stress scores and parasympathetic nerve forced regulation frequencies of multiple time periods to determine the mood state of the target individual.

[0115] Specifically, after obtaining the autonomic nervous activity signals of the target individual in multiple time periods, the target sequence is extracted from the autonomic nervous activity signals in multiple time periods, and then the target is calculated and compared according to the preset stress level assessment feature space, the preset parasympathetic nervous forced regulation state detection feature space, and the preset mood state assessment feature space to determine the mood state of the target individual.

[0116] In this embodiment, the target individual's mood state is determined by calculating and comparing target sequences extracted from the target individual's autonomic nervous system activity signals at multiple time intervals using a preset stress level assessment feature space, a preset parasympathetic nervous system forced regulation state detection feature space, and a preset mood state assessment feature space. Because the physiological data in the preset feature space is universally applicable and horizontally comparable, and distributed joint detection of physiological data from multiple time intervals ensures the integrity of the mood state detection results, enabling effective and complete detection of the individual's mood state when testing the individual's physiological data collected in real time.

[0117] Optionally, the above step S100 may specifically include: steps S110-S150.

[0118] See Figure 3 1 is a second flow chart of the mood state monitoring method provided in an embodiment of the present application.

[0119] Step S110: Acquire physiological signals of the target individual collected by a medical sensor; wherein the physiological signals include autonomic nerve activity signals and three-dimensional acceleration signals of the human body;

[0120] See Figure 4 The figure shows a schematic diagram of the waveform of a three-dimensional acceleration signal of a human body provided by an embodiment of the present application.

[0121] See Figure 5 A waveform diagram showing significant changes in autonomic nervous activity caused by a three-dimensional acceleration signal of the human body provided by an embodiment of the present application is shown.

[0122] It should be noted that when the human body's three-dimensional acceleration exceeds the preset three-dimensional human body acceleration, it causes a significant change in the autonomic nervous activity signal. This significant change is abnormal, and the autonomic nervous activity signal corresponds to the antagonism between the sympathetic and parasympathetic nerves. Therefore, if the autonomic nervous activity signal under abnormal conditions is subjected to forced parasympathetic nerve regulation, the detection result will inevitably be inaccurate. Inaccurate parasympathetic nerve regulation detection will inevitably interfere with the judgment of mood state monitoring. Therefore, when the human body three-dimensional acceleration signal of a certain period in the collected time period exceeds the preset three-dimensional human body acceleration, the target autonomic nervous activity signal of this period should be discarded to ensure the accuracy of mood state monitoring.

[0123] It is understandable that the medical sensor is set on the person whose mood state is to be detected, and the carrier of the medical sensor can be a device that can carry it, such as a smart watch. The embodiment of this application does not limit the carrier of the medical sensor, and it can be set according to actual needs.

[0124] The 3D human acceleration signal refers to the acceleration of a person in the X, Y, and Z dimensions, and is used to determine physical activity. When the 3D human acceleration signal exceeds a preset 3D human acceleration, it indicates a significant change in autonomic nervous system activity, meaning that the collected autonomic nervous system activity signal cannot represent the antagonistic state between the sympathetic and parasympathetic nervous systems.

[0125] In one embodiment, since a three-dimensional human acceleration signal that has not been detrended contains multiple high and low frequencies, it is necessary to detrend (denoise) the three-dimensional human acceleration signal. Therefore, under the premise of a sampling rate of 512 Hz, the low-frequency coefficients of a six-layer wavelet decomposition are set to zero, and then the signal is reconstructed to obtain a three-dimensional human acceleration signal with high-frequency fluctuations.

[0126] Step S120: performing segmentation calculation on the autonomic nerve activity signal and the human body three-dimensional acceleration signal to obtain the human body three-dimensional acceleration signal and the autonomic nerve activity signal in multiple time periods;

[0127] The segmentation calculation method can be a preset sliding time window, or other methods that can segment time, which is not limited in the embodiments of the present application.

[0128] Step S130: determining whether the multiple time-sharing three-dimensional human body acceleration signals are greater than a preset three-dimensional human body acceleration signal;

[0129] Step S140: If yes, then determining the autonomic nervous activity signals corresponding to the human body three-dimensional acceleration signals in the multiple time periods as the autonomic nervous activity signals of the non-target individual;

[0130] Step S150: If not, the autonomic nervous activity signals corresponding to the human body three-dimensional acceleration signals in multiple time periods are determined as the autonomic nervous activity signals of the target individual.

[0131] The preset three-dimensional acceleration signal refers to a threshold at which the three-dimensional acceleration signal is sufficient to cause significant changes in autonomic nervous activity. That is, when the preset three-dimensional acceleration signal is exceeded, the autonomic nervous activity signal will cause significant changes in autonomic nervous activity.

[0132] The autonomic nervous system activity signal of a non-target individual is an autonomic nervous system activity signal that cannot normally represent the antagonism between the sympathetic and parasympathetic nerves. Conversely, the autonomic nervous system activity signal of a target individual is an autonomic nervous system activity signal that can normally represent the antagonism between the sympathetic and parasympathetic nerves.

[0133] Specifically, by acquiring the three-dimensional acceleration signal and autonomic nervous activity signal of the human body collected by the medical sensor, and then segmenting and processing the three-dimensional acceleration signal and the autonomic nervous activity signal of the human body, the three-dimensional acceleration signal and the autonomic nervous activity signal of the human body in multiple time periods are obtained. Finally, the preset three-dimensional acceleration signal of the human body with the three-dimensional acceleration detrended fluctuation signal is used to distinguish whether the physical activity significantly affects the autonomic nervous activity, thereby obtaining the autonomic nervous activity signal that can represent the antagonistic state between the sympathetic and parasympathetic nerves, thus completing the preprocessing and collection of the autonomic nervous activity.

[0134] In this embodiment, the above-mentioned mood state monitoring method segments and processes the target individual's autonomic nervous activity signals and 3D human acceleration signals collected by a medical sensor to obtain multiple time-segmented 3D human acceleration signals and autonomic nervous activity signals. A preset 3D human acceleration signal is then used to distinguish whether physical activity (3D human acceleration signals) in the multiple time-segmented periods affects autonomic nervous activity. Time-segmented 3D human acceleration signals and autonomic nervous activity signals greater than the preset 3D human acceleration signal are eliminated to obtain the target autonomic nervous activity signal. Because time-segmented 3D human acceleration signals greater than the preset 3D human acceleration signal can interfere with mood state detection, eliminating these time-segmented 3D human acceleration signals further ensures the validity of mood state detection results.

[0135] Optionally, the above step S300 may specifically include: steps S310-S330.

[0136] Step S310: performing stress level assessments on the physiological data of the multiple time periods according to a preset stress level assessment feature space to obtain the physiological data of the multiple time periods after the stress level assessments, and calculating the physiological data of the multiple time periods according to a preset stress scoring method to obtain stress scores for the physiological data of the multiple time periods;

[0137] Step S320: performing parasympathetic nerve forced regulation detection on the physiological data of multiple time periods according to a preset parasympathetic nerve forced regulation state detection feature space, and obtaining parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods;

[0138] Step S330: Analyze and compare the stress scores and parasympathetic nerve forced regulation time distribution and frequency of physiological data of multiple time periods according to the preset mood state assessment feature space to determine the mood state of the target individual.

[0139] In this embodiment, a stress score is calculated for physiological data from multiple time periods using a preset stress level assessment feature space, parasympathetic nervous system forced regulation time distribution and frequency detection is performed on physiological data from multiple time periods using a preset parasympathetic nervous system forced regulation state detection feature space, and a mood state analysis and comparison is performed on the stress scores and parasympathetic nervous system forced regulation time distribution and frequency of physiological data from multiple time periods using a preset mood state assessment feature space to determine an individual's mood state. Because the calibration of physiological data samples in the preset stress level assessment space, the preset parasympathetic nervous system forced regulation state detection feature space, and the preset mood state assessment feature space utilizes multiple effective empirical validation criteria and selects multiple physiological indicators that can reveal key characteristics of mental stress and parasympathetic nervous system forced regulation, the universal applicability and cross-sectional comparability of the physiological data samples are ensured. Furthermore, by jointly monitoring the stress scores and parasympathetic nervous system forced regulation of physiological data from multiple time periods, the integrity of the mood state detection results is ensured, thereby enabling effective and complete detection of the individual's mood state when monitoring the individual's physiological data collected in real time.

[0140] Optionally, the preset pressure level assessment feature space is constructed in the following manner:

[0141] See Figure 6 Schematic diagram of the preset pressure level assessment feature space provided by the embodiment of the present application

[0142] 1) Calibrate the preset physiological data according to the preset pressure level empirical standard to obtain preset physiological data of multiple pressure levels;

[0143] The preset stress level experience criterion includes but is not limited to prior knowledge of stress sources, psychological experience of stress response, physiological reaction and behavioral performance, which are not limited in the embodiments of the present application.

[0144] The multiple pressure levels include but are not limited to four pressure levels: no pressure, weak pressure, medium pressure, and strong pressure. Other classification levels are also possible, and this application embodiment does not make specific limitations here.

[0145] 2) Based on the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms, the preset physiological data of multiple pressure levels are calculated to obtain calculation results of the preset physiological data of multiple pressure levels, and a preset pressure level assessment feature space is constructed based on the calculation results.

[0146] The periodic function of the target sequence rhythm is a measure of the balance between sympathetic and parasympathetic nerve activities.

[0147] The average fluctuation function of the two target sequence rhythms is a measure of the competition between sympathetic and parasympathetic nerve activities.

[0148] The following are specific examples:

[0149] Example 1: Under the influence of a strong stressor during a master's thesis defense (prior knowledge of the stressor), the subjects reported their stress level as strong (the psychological experience of a stress response), their heartbeats were significantly faster than usual (a physiological response to stress), and they were unable to correctly answer questions from experts or even pass the defense (a behavioral manifestation of stress). In this case, the physiological data of this group of people in the thesis defense scenario is marked as strong stress.

[0150] Example 2: During a nap, the subjects are physically and mentally relaxed (prior knowledge of stressors), report feeling stress-free (psychological experience), and their heartbeat is significantly slower than before the nap (physiological response). Their body posture and facial expressions indicate relaxation during the nap (behavioral manifestations). Therefore, the physiological data for this group of people during the nap is marked as stress-free.

[0151] The stress labeling method in other real-life situations also combines the above-mentioned prior knowledge of stress sources, psychological experience of stress response, physiological reaction and behavioral performance to jointly calibrate the level label of stress physiological data.

[0152] Specifically, through a variety of stress level empirical standards, the physiological data in different real-life situations can be effectively calibrated for stress levels. In addition, the above two physiological indicators are key features reflecting the forced regulation of the parasympathetic nervous system. Therefore, the periodic function of the target sequence rhythm and the average fluctuation function of the two target sequence rhythms can effectively reflect the changes in the autonomic nervous activity signals of the physiological data of different stress levels. For the physiological data actually collected subsequently, after mapping the collected physiological data to the preset stress level assessment space, the stress level of the collected physiological data can be directly determined based on the autonomic nervous activity signal, which also ensures the effectiveness of the physiological data stress level assessment.

[0153] In this embodiment, the preset physiological data is pressure-calibrated using a preset pressure-level empirical criterion. The pressure-calibrated preset physiological data is then represented using a periodic function of a target sequence rhythm and an average fluctuation function of two target sequence rhythms, which are key features of the law of autonomic nervous system activity. This results in a regular distribution of the preset physiological data in a feature space, which serves as a preset pressure-level assessment feature space. Because the preset physiological data is pressure-calibrated using an effective preset pressure-level empirical criterion, preset physiological data of multiple pressure levels are obtained. Furthermore, the scatter distribution of the preset physiological data of multiple pressure levels in the preset pressure-level assessment space is supported based on the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms. This ensures the effectiveness, universal applicability, and horizontal comparability of the pressure-level calibration of the preset physiological data, enabling the constructed preset pressure-level assessment feature space to effectively assess the pressure level of physiological data collected in real time.

[0154] Optionally, the preset stress scoring method includes Formula 1: Score is the stress score of the physiological data of the time period; x is the physiological data of the time period; x1 is the class center of the preset physiological data corresponding to the stress level of the physiological data of the time period; s is the standard deviation within the class of the preset physiological data; j is the stress level of the physiological data of the time period. The above step 310 may specifically include: steps S311-S314.

[0155] It should be noted that the essence of the preset stress scoring formula is to evenly map the physiological data that has been subjected to the preset stress level empirical standard into the interval of [0-100]. Since the preset stress level assessment feature space only contains physiological data of four stress levels, the stress scoring interval for each stress level is: no stress: [0-25], weak stress: [25-50], medium stress [50-75], and strong stress [75-100]. That is to say, when there are five stress levels in the physiological data, the corresponding stress scoring interval becomes: no stress: [0-20], weak stress: [20-40], medium stress [40-60], strong stress [60-80], and extremely strong stress [80-100]. Therefore, the embodiment of the present application does not specifically limit the division of stress levels, and can be set according to actual needs.

[0156] In addition, the value 25 in Formula 1 is due to the four pressure levels adopted in the embodiment of the present application, and then the four pressure levels are mapped to the interval of [0-100] and divided equally, that is, 100 / 4=25. If the pressure levels are divided into five types, then it is 100 / 5=20, and so on.

[0157] In addition, the j in Formula 1 does not refer to the four levels of no pressure, weak pressure, medium pressure and strong pressure, but refers to the specific value after sorting the four levels of no pressure, weak pressure, medium pressure and strong pressure. The larger j is, the stronger the pressure is. Since the pressure score interval of no pressure is [0-25], which is the smallest, then the no pressure level is 0, the pressure score interval of weak pressure is [25-50], which is second, then the weak pressure level is 1, and so on, the medium pressure level is 2, and the strong pressure level is 3.

[0158] In one embodiment, when constructing a preset stress level assessment feature space, after the physiological data is represented by two indicators, namely, the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms, the physiological data is not in the no-stress scatter point distribution or the strong-stress scatter point distribution, resulting in a score less than 0 or greater than 100 when calculating the stress score. In this case, the value should be selected as 0 or 100, and should not be less than 0 or greater than 100.

[0159] Step S311: Calculating the physiological data of multiple time periods based on the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms to obtain calculation results of the physiological data of the multiple time periods;

[0160] Step S312: Mapping the calculated results of the physiological data of the multiple time periods to the preset stress level assessment feature space, and performing distance calculations between the cluster centers of the calculated results of the preset physiological data of the multiple stress levels and the calculated results, to obtain the distances between the calculated results of the physiological data of the multiple time periods and the cluster centers of the calculated results of the preset physiological data of the multiple stress levels;

[0161] Step S313: determining a plurality of minimum distances from the distances between the calculated physiological data of the plurality of time periods and the cluster centers of the calculated physiological data of the plurality of preset pressure levels, and using the pressure levels of the preset physiological data corresponding to the plurality of minimum distances as the pressure levels of the physiological data of the plurality of time periods;

[0162] Step S314: Obtain stress scores for the physiological data of the multiple time periods according to Formula 1, the physiological data of the multiple time periods, the class centers of the preset physiological data corresponding to the stress levels of the physiological data of the multiple time periods, the standard deviation within the class of the preset physiological data, and the stress levels of the physiological data of the multiple time periods.

[0163] It is understandable that the preset stress level assessment feature space includes physiological data of multiple stress levels. When the physiological data of multiple time periods are represented by the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms to form a scatter point distribution in the preset stress level assessment feature space, the distribution relationship between the scatter point distribution of the physiological data of multiple time periods and the scatter point distribution of the physiological data of multiple pressure levels can be observed, and then the pressure level of the physiological data of the pressure level closest to the physiological data of the time period is used as the pressure level of the physiological data of the time period. Finally, according to Formula 1, the physiological data of the determined pressure level is calculated to obtain the stress score of the physiological data.

[0164] In this embodiment, physiological data from multiple time periods is used to represent the scatter distribution in a preset stress level assessment feature space using a periodic function of a target sequence rhythm and an average fluctuation function of two target sequence rhythms. The proximity of the physiological data from the multiple time periods to the preset physiological data for the multiple stress levels is then observed, and the stress level of the preset physiological data closest to the physiological data from the multiple time periods is used as the stress level of the physiological data for the multiple time periods. Finally, the physiological data from the multiple time periods for which stress levels have been assessed are scored according to a preset scoring method to obtain stress scores for the physiological data from the multiple time periods. Because the preset stress level assessment feature space can effectively assess the stress level of physiological data, the resulting stress score calculated when the stress score is calculated for the physiological data for which stress levels have been assessed is also valid, thereby providing an effective basis for stress scoring in mood state detection.

[0165] Optionally, the preset parasympathetic nerve forced regulation state detection feature space is constructed in the following manner:

[0166] It should be noted that the embodiment of the present application adopts three types of indicators, of which the first type is a key indicator for describing the forced regulation of the parasympathetic nerves. The other two types of indicators are classic indicators for analyzing the laws of autonomic nervous activity using visceral rhythm time series. When the first type of indicator forms a high-dimensional feature space together with the other two types of indicators, the complementary information between the features better describes the distribution law of the two types of data samples in the feature space of whether the parasympathetic nerves are forced to be regulated. Therefore, the present invention detects whether forced regulation of the parasympathetic nerves occurs by the distribution of the first physiological data and the second physiological data in the preset parasympathetic nerve forced regulation state monitoring feature space.

[0167] 1) Calibrate the preset physiological data according to a preset stress level empirical effect standard, a preset fatigue theoretical model, a preset work and rest effect standard, and a preset dietary irregularity effect standard to obtain a plurality of first physiological data and second physiological data; wherein the first physiological data is calibrated as parasympathetic nervous system forced regulation;

[0168] The first physiological data refers to the physiological signal containing autonomic nervous activity information when the mental alertness is high during the time when the physical fatigue is not easy, and the mind and body are in a state of relaxation, which is marked as parasympathetic nervous system forced regulation.

[0169] The second physiological data refers to physiological signals containing information on autonomic nervous system activity when a person is under strong pressure or has low mental alertness during a period of physiological fatigue, during special life events that deviate from the equilibrium state.

[0170] Physiological fatigue-free time refers to the time after a normal night's sleep, starting from the time of getting up in the morning, and the following 3 hours, obtained based on the preset fatigue theory model.

[0171] In one embodiment, the fatigue level of physiological data is divided into fatigue levels (0-10, 0 means no fatigue at all, higher scores mean more fatigue, and 10 means exhaustion), and then the level of mental alertness is judged based on the fatigue level.

[0172] Physiological fatigue prone time refers to the physiological fatigue prone time 6 hours after waking up from a normal sleep at night, which is obtained based on a preset fatigue theory model.

[0173] Whether strong stress is strong is determined based on four empirical criteria: prior knowledge of the stress source, psychological experience, physiological reaction, and behavioral performance.

[0174] 2) performing continuous wavelet transform on the plurality of first physiological data and the second physiological data according to the wavelet basis function of the preset transformation scale interval, and obtaining a geometric morphological approximation degree measurement index of the plurality of first physiological data and the second physiological data and the physiological data of the plurality of time periods;

[0175] The preset transformation scale interval refers to the numerical range of the wavelet coefficients of the wavelet basis function, and is not limited in this embodiment of the present application.

[0176] In one embodiment, a key indicator reflecting forced regulation of the parasympathetic nerves is used: a specific wavelet basis function type is used to perform scaling within a preset transformation scale range to approximate the geometric forms of multiple first physiological data and second physiological data, and multiple transformation scales that best reflect the first physiological data and the second physiological data are found. Then, a maximum transformation scale is selected from the multiple transformation scales that best approximate the first physiological data and the second physiological data. The reason for selecting the maximum transformation scale is that the maximum transformation scale can represent the forced regulation of the parasympathetic nerves.

[0177] 3) calculating the average difference between the plurality of first physiological data and the second physiological data and the adjacent first physiological data and the second physiological data, to obtain the average difference between the plurality of first physiological data and the second physiological data;

[0178] In one embodiment, a first-class indicator reflecting the patterns of autonomic nervous system activity is used: the indicator Tn, which describes small-scale heart rate variability in the time domain. The subscript n indicates that more than one such indicator is available, such as the mean of adjacent heart rate differences. This indicator is used to reveal the relative degree of competition between the sympathetic and parasympathetic nervous systems. A high value for the indicator indicates a more intense competition between the sympathetic and parasympathetic nervous systems, while a low value indicates inhibition of one of the autonomic nervous system branches.

[0179] 4) obtaining power values of a plurality of first physiological data and second physiological data at a preset frequency;

[0180] In one embodiment, the first type of classical indicators reflecting the patterns of autonomic nervous system activity is used: 3. An indicator Fn is taken in the frequency domain to describe the relative activation of the parasympathetic nervous system. The subscript n indicates that there is more than one such indicator, such as the total power of the heart rate sequence in the 0.15Hz-0.4Hz subband.

[0181] 5) obtaining calculation results of the plurality of first physiological data and the second physiological data based on the maximum transformation scale, the average difference, and the power value at a preset frequency of the plurality of first physiological data and the second physiological data;

[0182] 6) Constructing a parasympathetic nerve forced regulation state detection feature space based on calculation results of the plurality of first physiological data and the second physiological data.

[0183] In this embodiment, the preset physiological data are evaluated by using a preset stress level empirical criterion and a preset fatigue theoretical model, a preset work and rest criterion, and an irregular diet criterion to obtain first physiological data and second physiological data. The first physiological data and the second physiological data are then represented by a wavelet approximation degree measurement index of a preset transformation scale interval that reflects the key characteristics of the law of autonomic nervous system activity, an index of target sequence variation, and a relative activation degree index within a preset frequency, thereby obtaining a regular distribution of the preset physiological data in the feature space as a preset parasympathetic nervous system forced regulation state detection feature space. Since the preset physiological data are calibrated by using effective preset pressure level empirical criterion, preset fatigue theoretical model and other preset special events that lead to parasympathetic nerve forced regulation, physiological data with parasympathetic nerve forced regulation and without parasympathetic nerve forced regulation are obtained. Secondly, according to the wavelet approximation degree measurement index of the preset transformation scale interval, the target sequence variation index and the relative activation degree index within the preset frequency, the scatter distribution of physiological data with parasympathetic nerve forced regulation and without parasympathetic nerve forced regulation in the preset parasympathetic nerve forced regulation state detection feature space is supported, which ensures the effectiveness, universal applicability and horizontal comparability of the preset physiological data parasympathetic nerve forced regulation detection, so that the constructed preset parasympathetic nerve forced regulation state detection feature space can effectively perform parasympathetic nerve forced regulation detection on physiological data collected in real time.

[0184] Optionally, step S320 may specifically include: steps S321-S327.

[0185] Step S321: performing wavelet transform on the physiological data of the multiple time periods to obtain geometric morphology approximation degree measurement indicators of the physiological data of the multiple time periods;

[0186] Step S322: calculating the average difference between the physiological data of the multiple time periods and the physiological data of the adjacent time periods to obtain the average difference of the physiological data of the multiple time periods;

[0187] Step S323: obtaining power values of physiological data of multiple time periods at a preset frequency;

[0188] Step S324: obtaining calculation results of the physiological data of the multiple time periods based on the geometric approximation degree metric, the average difference and the power value at the preset frequency of the physiological data of the multiple time periods;

[0189] Step S325: Mapping the calculation results of the physiological data of the multiple time periods to a preset parasympathetic nerve forced regulation state detection feature space, and performing distance calculations between the cluster centers of the calculation results of the multiple first physiological data and the second physiological data and the calculation results, respectively, to obtain distances between the calculation results of the physiological data of the multiple time periods and the cluster centers of the calculation results of the multiple first physiological data and the second physiological data;

[0190] It should be noted that the physiological data of multiple time periods refer to the target sequence of the autonomic nervous activity signals of multiple time periods, and the target sequence is a wavelength, which is similar to the first physiological data and the second physiological data, so it can be directly compared with the transformation scale of the first physiological data and the second physiological data without any calculation.

[0191] It is understood that because the first physiological data is pre-determined to indicate the presence of parasympathetic nervous system activity using a preset stress level empirical criterion and a preset fatigue model, the first physiological data in the preset parasympathetic nervous system forced modulation state detection feature space satisfies the aforementioned three indicators, indicating the presence of parasympathetic nervous system forced modulation. Therefore, when the calculated results of the physiological data for each time period are similar to the indicators of the first physiological data, it indicates that parasympathetic nervous system forced modulation has occurred in the physiological data for that time period.

[0192] Step S326: determining a plurality of minimum distances from the distances between the calculation results of the physiological data of the plurality of time periods and the cluster centers of the calculation results of the plurality of first physiological data and the second physiological data;

[0193] Step S327: If the preset physiological data corresponding to the multiple minimum distances is the first physiological data, the multiple time-divided physiological data corresponding to the multiple minimum distances are marked as parasympathetic nerve forced regulation, and the number of parasympathetic nerve forced regulation of the physiological data of the multiple time-divided time periods is recorded to obtain the parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time-divided time periods.

[0194] Specifically, the physiological data, average difference and power at a preset frequency of multiple time periods are mapped to a preset parasympathetic nerve forced regulation state detection feature space, and then the relationship between the physiological data of the time period and the scatter distribution of the first physiological data and the scatter distribution of the second physiological data is observed, and the physiological data of the time period close to the scatter distribution of the first physiological data is marked as parasympathetic nerve forced regulation. Finally, the number of parasympathetic nerve forced regulation marks is recorded, and the regulation frequency within the time period is calculated based on the number of adjustments.

[0195] In this embodiment, the scatter distribution of the physiological data of multiple time periods in a preset parasympathetic nerve forced modulation state detection feature space is represented using a wavelet basis function with a preset transformation scale interval, an index of target sequence variation, and a relative activation degree index within a preset frequency. The physiological data of the multiple time periods are then observed to determine their proximity to the preset physiological data with and without parasympathetic nerve forced modulation. The physiological data of the multiple time periods closest to the physiological data with parasympathetic nerve forced modulation are respectively labeled as parasympathetic nerve forced modulation. The number of parasympathetic nerve forced modulation events is then recorded. Finally, the parasympathetic nerve forced modulation frequency of the physiological data of the multiple time periods is determined based on the duration of the time periods and the number of occurrences of parasympathetic nerve forced modulation. Since the preset parasympathetic nerve forced modulation state detection feature space can effectively detect parasympathetic nerve forced modulation in physiological data, it provides an effective basis for mood state detection in terms of parasympathetic nerve forced modulation frequency.

[0196] Optionally, the preset mood state assessment feature space is constructed in the following manner:

[0197] 1) According to a preset self-assessment form, the preset physiological data is calibrated for the mood state to obtain the preset physiological data of the calibrated mood state;

[0198] 2) calculating the stress score and parasympathetic nerve forced modulation time distribution and frequency of the preset physiological data of the calibrated mood state, and obtaining the preset stress score, preset parasympathetic nerve forced modulation time distribution and frequency for a good mood state and a poor mood state;

[0199] The preset self-assessment scales include, but are not limited to, self-assessment of fatigue by time period during the day, self-assessment of sleep duration and quality at night, self-assessment of anxiety, self-assessment of stress, and self-assessment of depression. Any scale that can score the mood state of a group of people is within the scope of protection of the embodiments of the present application, and the embodiments of the present application do not make specific limitations here.

[0200] The following is a detailed explanation of the self-assessment form: for example, the subject participated in a graduation thesis defense within a certain period of time on the same day, and self-evaluated that he was very anxious and stressed before and during the defense. He also felt anxious and stressed while waiting for the results after the defense. He reported his fatigue level every hour during the waking hours of the whole day; the Self-Rating Anxiety Scale SAS, the Self-Rating Depression Scale SDS, the Life Events Scale LES, the Perceived Stress Scale PSS, etc. were used to measure stress, anxiety and depression in the past week or even longer. The subjects self-reported the time they fell asleep, the time they got up, whether they woke up in the middle of the night, and their self-perceived sleep quality on the previous day and the same day.

[0201] It is understandable that the preset stress scores, preset duration ratios and preset forced adjustment frequencies for people with good moods and bad moods are obtained by conducting multiple preset self-assessment tests on people with good moods and people with bad moods, and combining the preset stress level assessment feature space and the preset parasympathetic nerve forced adjustment state detection feature space, and then obtaining the stress scores, duration ratios, occurrence times and parasympathetic nerve forced adjustment and occurrence times of the two types of people in multiple time periods throughout the day.

[0202] 3) Construct a preset mood state assessment feature space based on the preset stress scores of good mood state and bad mood state, and the preset parasympathetic nervous system forced regulation time distribution and frequency.

[0203] Specifically, self-assessment reports of the population are collected based on multiple preset self-assessment forms, and the fatigue levels and work and rest time of multiple time periods are analyzed. Then, the stress scores and the number of parasympathetic nerve forced adjustments of multiple time periods are detected by combining the preset stress level assessment feature space and the preset parasympathetic nerve forced regulation state detection feature space. Finally, the corresponding stress scores, duration ratios, and parasympathetic nerve forced regulation frequencies under good and bad mood states are obtained.

[0204] In this embodiment, the preset physiological data is calibrated for mood states using a preset self-assessment form. The stress score and parasympathetic force modulation detection of the physiological data for the determined mood state are used to obtain the preset stress scores, preset duration percentages, and preset parasympathetic force modulation time distribution and frequency for good and poor mood states. A preset mood state feature space is then constructed based on the aforementioned stress scores and parasympathetic force modulation time distribution and frequency. By using the preset self-assessment form to calibrate the preset physiological data for mood states, analyzing the stress scores and parasympathetic force modulation of the calibrated mood state physiological data, and obtaining the preset stress scores, preset duration percentages, and preset force modulation frequency for good and poor mood states, the validity of the various mood state indicators of the preset physiological data is ensured, enabling the preset mood state feature space to effectively detect mood states in the physiological data collected in real time.

[0205] Optionally, the above step S330 may specifically include: steps: 331-S335.

[0206] For example, the indicators for determining mood states in the embodiments of the present application include: a stress score above 50, the duration of which is a sensitive indicator of poor mood, and the time of occurrence (whether it occurs during sleep) is a sensitive and specific indicator of poor mood. The frequency and time distribution of parasympathetic nervous system forced modulation are sensitive and specific indicators of poor mood. The combination of these stress and parasympathetic nervous system forced modulation indicators is a key technical indicator for detecting poor mood.

[0207] Step S331: grouping the stress scores of the physiological data of multiple time periods according to the target individual's work and rest time to obtain stress scores of the physiological data of multiple wakefulness states and sleep states;

[0208] The target individual's work and rest time refers to the time when the group is sleeping and not sleeping. Since everyone's work and rest time is different, the physiological data of different time periods can be grouped according to actual needs.

[0209] Step S332: Grouping the parasympathetic nervous system forced modulation time distribution and frequency of physiological data in multiple time periods according to a preset stress level empirical criterion and a preset fatigue theoretical model to obtain the parasympathetic nervous system forced modulation time distribution and frequency of multiple physiological fatigue prone times and physiological fatigue non-fatality times;

[0210] Step S333: determining a stress score of physiological data of an awake state that is less than a preset stress score of a good mood state from the stress scores of physiological data of the multiple awake states, thereby obtaining a stress score of physiological data of the first awake state;

[0211] Step S334: determining the stress score of the physiological data of the sleep state having a stress score of zero from the stress scores of the physiological data of the multiple sleep states, to obtain the stress score of the physiological data of the first sleep state;

[0212] Step S335: If the ratio of the duration corresponding to the stress score of the physiological data of the first awake state to the duration of multiple time periods is greater than the preset ratio of the duration of a good mood, the ratio of the duration corresponding to the stress score of the physiological data of the first sleep state to the duration of multiple time periods is greater than the preset ratio of the duration of a good mood, and the forced adjustment frequency of multiple physiological fatigue-prone times is less than the preset adjustment frequency of a good mood, then the mood state is determined to be good.

[0213] Specifically, if the proportion of time during which the mental stress scores in the wakeful state are lower than the good mood experience threshold throughout the day is higher than the good mood experience threshold, the proportion of time during which the mental stress scores in the sleep state are zero is higher than the good mood experience threshold, the number of forced regulation of the parasympathetic nervous system in the wakeful state and the sleep state is lower than the good mood experience threshold, and most of them are distributed within the physiologically fatigued time, then the mood state is judged to be good.

[0214] The following are examples:

[0215] See Figure 7 The embodiment of the present application provides a schematic diagram of the stress score, heart rate sequence, parasympathetic nerve forced regulation mark and human body three-dimensional acceleration signal of the subject in a good mood around the clock.

[0216] like Figure 7 The following are the stress scores and parasympathetic nervous system forced regulation markers of the subjects in a good mood. This data sample has been confirmed to be in a good mood by multiple empirical criteria: there were no stressors or behavioral triggers that induced a bad mood on the day of data collection and within the week before and after. Using the method of the present invention, the mood state of the subject on that day was detected by physiological signal processing. Figure 7 As shown in the figure, the stress score of 7 indicates that the subjects were at the lowest stress level for the majority of the day, with scores below 25 for the vast majority of the time. Parasympathetic nervous system over-regulation occurred only occasionally during physiologically fatigue-prone periods. In the absence of daytime naps, fatigue-induced parasympathetic nervous system over-regulation occurred during physiologically fatigue-prone periods, approximately 6:00 PM to 9:00 PM. Therefore, the positive mood detected from the frequency and temporal distribution of stress and parasympathetic nervous system over-regulation is consistent with the true positive mood label.

[0217] In this embodiment, the stress scores and parasympathetic nervous system forced modulation of physiological data from multiple time periods are grouped according to the target individual's sleep and wakefulness schedule, a preset stress level empirical criterion, and a preset fatigue theoretical model. The stress scores of the physiological data for the wakefulness and sleep states, as well as the parasympathetic nervous system forced modulation frequencies of the physiological data for the physiologically fatigued state and the physiologically non-fatigued state, are obtained. Then, if the stress score of the physiological data for the wakefulness state is less than the preset stress score for the duration of a good mood and greater than the preset duration of a good mood, the stress score of the sleep state is zero for a duration greater than the preset duration of a good mood, and the parasympathetic nervous system forced modulation frequency of the physiological data for the physiologically fatigued state is less than the preset forced modulation frequency for a good mood, then the individual's mood state is determined to be good. Because the stress scores of the physiological data for the wakefulness and sleep states and the parasympathetic nervous system forced modulation frequencies for the physiologically non-fatigued state are determined using the preset mood state feature space, the detection result indicating that the target individual's mood state is good is valid and complete.

[0218] Optionally, the above step S330 may further specifically include: steps: 331'-S333'.

[0219] Step S331′: determining a stress score of physiological data of an awake state that is greater than a preset stress score for a poor mood state from the stress scores of physiological data of multiple awake states, to obtain a stress score of physiological data of a second awake state;

[0220] Step S332′: determining the stress score of the physiological data of the sleep state that is greater than the preset stress score of the poor mood state from the stress scores of the physiological data of the multiple sleep states, and obtaining the stress score of the physiological data of the second sleep state;

[0221] Step S333': If the ratio of the duration corresponding to the stress score of the physiological data of the second awake state to the duration of multiple time periods is greater than the preset ratio of the duration of the bad mood state, the ratio of the duration corresponding to the stress score of the physiological data of the second sleep state to the duration of multiple time periods is greater than the preset ratio of the duration of the bad mood state, and the forced adjustment frequency of multiple physiological fatigue-prone times is greater than the preset adjustment frequency of the bad mood state, then it is determined that the mood state is bad.

[0222] Specifically, if the proportion of time during which the mental stress scores in the wakeful state are higher than the bad mood experience threshold throughout the day is higher than the bad mood experience threshold, the proportion of time during which the mental stress scores in the sleep state are higher than the bad mood experience threshold is higher than the bad mood experience threshold, the number of forced regulation of the parasympathetic nervous system in the wakeful state and the sleep state is higher than the bad mood experience threshold, and is not only distributed during the time when physiological fatigue is easy, but also more distributed during the time when physiological fatigue is not easy, then it is judged that the mood state is not good.

[0223] The following are examples:

[0224] See Figure 8 A schematic diagram of the stress score, heart rate sequence, parasympathetic nerve forced regulation mark and human body three-dimensional acceleration signal of a subject in a bad mood around the clock provided by an embodiment of the present application is shown.

[0225] like Figure 8 The figure shows the all-day mental stress score and parasympathetic nervous system forced regulation marker of a subject who was in a bad mood due to graduation defense. This data sample has confirmed the bad mood state and the causes of this state through a variety of empirical criteria: the subject participated in the graduation defense from 14:00 to 16:00 that afternoon. Before the defense, he had an uneasy expectation of the unknown result, and he was uneasy about facing the expert's presentation during the defense. After the defense, he had a rumination-like recollection of the event and a large consumption of physical and mental resources, which caused the subject to be in an uneasy mood all day. The mood state monitoring method of the embodiment of the present application is used to detect the mood state of the subject on that day through physiological signal processing. For example Figure 8 As shown. Figure 8The subject's stress score indicates that, from before the defense until the start of morning data collection (approximately 8:50 AM), the subject spent most of their time in a state of stress (a score greater than 25 indicates moderate stress, and a score greater than 50 indicates strong stress), accompanied by intensive parasympathetic force modulation, indicating the persistent occurrence of events that could easily lead to a bad mood. Furthermore, the force activation marker appeared intensively not only during the physiologically fatigued period 6 hours after waking up in the morning, but also during the physiologically fatigued period 3 hours after waking up in the morning. After the defense, the subject's stress level decreased somewhat, but the subject remained in a weak stress state (a score greater than 25 and less than 50), still accompanied by intensive parasympathetic force modulation, indicating the persistent occurrence of events that could easily lead to a bad mood. During sleep that night, the subject's stress score rarely reached the 0-point range typical of a good sleep state. Therefore, the poor mood detected from the frequency and temporal distribution of stress and parasympathetic force modulation matched the true poor mood label, indicating that the subject's mood state was determined to be poor.

[0226] In one embodiment, if the mental stress score, duration ratio, and parasympathetic nervous system forced regulation frequency of each time period during the whole day's wakefulness state are between the above two judgment indicators, the mood state is determined to be normal.

[0227] In this embodiment, the individual's mood state is determined to be poor when the stress score of the physiological data in the awake state is greater than the preset stress score for poor mood, the percentage of time the stress score of the physiological data in the sleep state is greater than the preset stress score for poor mood, and the percentage of time the stress score of the physiological data in the sleep state is greater than the preset stress score for poor mood, and the parasympathetic nervous system forced modulation frequency of the physiological data during the physiological fatigue-free period is greater than the preset forced modulation frequency for poor mood. Because the stress scores of the physiological data in the awake state and the sleep state and the parasympathetic nervous system forced modulation frequency during the physiological fatigue-free period are determined using the preset mood state feature space, the detection result indicating that the target individual's mood state is poor is valid and complete.

[0228] See Figure 9 The figure shows a schematic diagram of the structure of the mood state monitoring device provided in an embodiment of the present application.

[0229] The embodiment of the present application provides a mood state monitoring device 200, comprising:

[0230] An acquisition module 210 is configured to acquire an autonomic nervous activity signal of a target individual; wherein the autonomic nervous activity signal includes a plurality of autonomic nervous activity signals in different time periods;

[0231] An extraction module 220 is configured to extract target sequences from the multiple time-divided autonomic nerve activity signals to obtain physiological data of the multiple time-divided periods;

[0232] The analysis module 230 is used to calculate and compare the physiological data of the multiple time periods according to a preset feature space to determine the mood state of the target individual; wherein the preset feature space includes: a preset stress level assessment feature space, a preset parasympathetic nerve forced regulation state detection feature space, and a preset mood state assessment feature space.

[0233] Optionally, the acquisition module 210 is further used to: acquire physiological signals of the target individual collected by the medical sensor; wherein the physiological signals include autonomic nervous activity signals and three-dimensional human body acceleration signals; perform segmentation calculation on the autonomic nervous activity signals and the three-dimensional human body acceleration signals to obtain a plurality of time-divided three-dimensional human body acceleration signals and autonomic nervous activity signals; determine whether the plurality of time-divided three-dimensional human body acceleration signals are greater than a preset three-dimensional human body acceleration signal; if so, determine the autonomic nervous activity signals corresponding to the plurality of time-divided three-dimensional human body acceleration signals as the autonomic nervous activity signals of the non-target individual; if not, determine the autonomic nervous activity signals corresponding to the plurality of time-divided three-dimensional human body acceleration signals as the autonomic nervous activity signals of the target individual.

[0234] Optionally, the analysis module 230 is further used to: perform stress level assessment on the physiological data of the multiple time periods according to the preset stress level assessment feature space, obtain the physiological data of the multiple time periods after stress level assessment, and calculate the physiological data of the multiple time periods according to the preset stress scoring method to obtain the stress scores of the physiological data of the multiple time periods; perform parasympathetic nerve forced regulation detection on the physiological data of the multiple time periods according to the preset parasympathetic nerve forced regulation state detection feature space, and obtain the parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods; analyze and compare the stress scores and parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods according to the preset mood state assessment feature space to determine the mood state of the target individual.

[0235] Optionally, the analysis module 230 is also used to: calibrate the preset physiological data for pressure levels according to the preset pressure level empirical standard to obtain preset physiological data of multiple pressure levels; calculate the preset physiological data of the multiple pressure levels according to the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms to obtain the calculation results of the preset physiological data of the multiple pressure levels, and construct the preset pressure level assessment feature space according to the calculation results.

[0236] Optionally, the analysis module 230 is further configured to: calculate the physiological data of the multiple time periods according to the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms to obtain calculation results of the physiological data of the multiple time periods; map the calculation results of the physiological data of the multiple time periods to the preset stress level assessment feature space, and perform distance calculations between the cluster centers of the calculation results of the preset physiological data of the multiple stress levels and the calculation results to obtain distances between the calculation results of the physiological data of the multiple time periods and the cluster centers of the calculation results of the preset physiological data of the multiple stress levels; determine multiple minimum distances from the distances between the calculation results of the physiological data of the multiple time periods and the cluster centers of the calculation results of the preset physiological data of the multiple stress levels, and use the stress levels of the preset physiological data corresponding to the multiple minimum distances as the stress levels of the physiological data of the multiple time periods; and obtain stress scores of the physiological data of the multiple time periods according to Formula 1, the physiological data of the multiple time periods, the cluster centers of the preset physiological data corresponding to the pressure levels of the physiological data of the multiple time periods, the cluster standard deviations of the preset physiological data, and the stress levels of the physiological data of the multiple time periods.

[0237] Optionally, the analysis module 230 is also used to: calibrate the preset physiological data according to the preset stress level experience standard and the preset fatigue theoretical model, the preset work and rest standard and the irregular diet standard to obtain multiple first physiological data and second physiological data; wherein, the first physiological data is calibrated as parasympathetic nerve forced regulation; according to the wavelet basis function of the preset transformation scale interval, perform continuous wavelet transform on the multiple first physiological data and the second physiological data respectively to obtain a geometric morphological approximation degree measurement index of the multiple first physiological data and the second physiological data and the physiological data of the multiple time periods; calculate the average difference between the multiple first physiological data and the second physiological data and the adjacent first physiological data and the second physiological data respectively to obtain the average difference of the multiple first physiological data and the second physiological data; obtain the power value of the multiple first physiological data and the second physiological data at the preset frequency; obtain the calculation results of the multiple first physiological data and the second physiological data based on the maximum transformation scale, the average difference and the power value at the preset frequency of the multiple first physiological data and the second physiological data; construct the parasympathetic nerve forced regulation state detection feature space according to the calculation results of the multiple first physiological data and the second physiological data.

[0238] Optionally, the analysis module 230 is further used to: perform wavelet transform on the physiological data of the multiple time periods to obtain geometric approximation degree measurement indicators of the physiological data of the multiple time periods; perform average difference calculation on the physiological data of the multiple time periods and the physiological data of the adjacent time periods to obtain the average difference of the physiological data of the multiple time periods; obtain the power value of the physiological data of the multiple time periods at the preset frequency; obtain the calculation results of the physiological data of the multiple time periods based on the geometric approximation degree measurement indicators, average difference and power value at the preset frequency of the physiological data of the multiple time periods; map the calculation results of the physiological data of the multiple time periods to the preset parasympathetic nerve forced regulation state detection feature space, and map the multiple first physiological data and second physiological data to the preset parasympathetic nerve forced regulation state detection feature space. The class centers of the calculation results of the physiological data are respectively calculated with the calculation results to obtain the distances between the calculation results of the physiological data of the multiple time periods and the class centers of the calculation results of the multiple first physiological data and the second physiological data; multiple minimum distances are respectively determined from the distances between the calculation results of the physiological data of the multiple time periods and the class centers of the calculation results of the multiple first physiological data and the second physiological data; if the preset physiological data corresponding to the multiple minimum distances is the first physiological data, the multiple time-divided physiological data corresponding to the multiple minimum differences are marked as parasympathetic nerve forced regulation, and the number of parasympathetic nerve forced regulation of the physiological data of the multiple time periods is recorded to obtain the parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods.

[0239] Optionally, the analysis module 230 is further used to: calibrate the preset physiological data for the mood state according to a preset self-assessment form to obtain the preset physiological data of the calibrated mood state; calculate the stress score and the parasympathetic nerve forced regulation time distribution and frequency of the preset physiological data of the calibrated mood state to obtain the preset stress score, preset duration ratio and preset forced regulation frequency of a good mood state and a bad mood state; and construct the preset mood state assessment feature space according to the preset stress score, preset parasympathetic nerve forced regulation time distribution and frequency of a good mood state and a bad mood state.

[0240] Optionally, the analysis module 230 is further used to: group the stress scores of the physiological data of the multiple time periods according to the work and rest time of the target individual, and obtain the stress scores of the physiological data of the multiple wakefulness states and the sleep state; group the parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods according to the preset stress level empirical criterion and the preset fatigue theoretical model, and obtain the parasympathetic nerve forced regulation time distribution and frequency of multiple physiological fatigue prone times and physiological fatigue non-prone times; determine the stress score of the physiological data of the wakefulness state that is less than the preset stress score of the good mood state from the stress scores of the physiological data of the multiple wakefulness states, and obtain the first wakefulness state the stress score of the physiological data of the first sleep state; determining the stress score of the physiological data of the sleep state with a stress score of zero from the stress scores of the physiological data of the multiple sleep states, and obtaining the stress score of the physiological data of the first sleep state; if the ratio of the duration corresponding to the stress score of the physiological data of the first wakeful state to the duration of the multiple divided time periods is greater than the preset ratio of the duration of the good mood state, the ratio of the duration corresponding to the stress score of the physiological data of the first sleep state to the duration of the multiple divided time periods is greater than the preset ratio of the duration of the good mood state, and the forced adjustment frequency of the multiple physiological fatigable time states is less than the preset adjustment frequency of the good mood state, then it is determined that the mood state is good.

[0241] Optionally, the analysis module 230 is further used to: determine, from the stress scores of the physiological data of the multiple wakeful states, a stress score of the physiological data of the wakeful state that is greater than the preset stress score of the poor mood state, to obtain the stress score of the physiological data of the second wakeful state; determine, from the stress scores of the physiological data of the multiple sleep states, a stress score of the physiological data of the sleep state that is greater than the preset stress score of the poor mood state, to obtain the stress score of the physiological data of the second sleep state; if the ratio of the duration corresponding to the stress score of the physiological data of the second wakeful state to the duration of the multiple time periods is greater than the preset ratio of the poor mood state, the ratio of the duration corresponding to the stress score of the physiological data of the second sleep state to the duration of the multiple time periods is greater than the preset ratio of the poor mood state, and the forced adjustment frequency of the multiple physiological fatigue-prone times is greater than the preset adjustment frequency of the poor mood state, then it is determined that the mood state is poor.

[0242] It should be understood that the device corresponds to the aforementioned embodiment of the mood state monitoring method and is capable of executing each of the steps involved in the aforementioned embodiment of the method. The specific functions of the device can be found in the description above, and a detailed description is omitted here to avoid repetition. The device includes at least one software functional module that can be stored in a memory in the form of software or firmware or embedded in the device's operating system (OS).

[0243] An embodiment of the present application further provides a storage medium, on which a computer program is stored. When the computer program is run by a processor, the above method is executed.

[0244] Among them, the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0245] In the several embodiments provided in the embodiments of the present application, it should be understood that the disclosed devices and methods can also be implemented in other ways. The device embodiments described above are merely illustrative. For example, the flowcharts and block diagrams in the accompanying drawings show the possible architectures, functions, and operations of the devices, methods, and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment, or a portion of code, and the module, program segment, or a portion of code contains one or more executable instructions for implementing the specified logical functions. It should also be noted that in some alternative implementations, the functions marked in the boxes can also occur in a different order than the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of boxes in the block diagram and / or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or action, or can be implemented using a combination of dedicated hardware and computer instructions.

[0246] In addition, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0247] The above description is only an optional implementation method of the embodiment of the present application, but the protection scope of the embodiment of the present application is not limited to this. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed in the embodiment of the present application, and they should all be covered by the protection scope of the embodiment of the present application.

Claims

1. A method for monitoring mood state, characterized in that: The method comprises: Acquire an autonomic nervous activity signal of a target individual; wherein the autonomic nervous activity signal includes a plurality of autonomic nervous activity signals in different time periods; Extracting target sequences from the multiple time-divided autonomic nerve activity signals respectively to obtain multiple time-divided physiological data; Calculating and comparing the physiological data of the multiple time periods according to a preset feature space to determine the mood state of the target individual; wherein the preset feature space includes: a preset stress level assessment feature space, a preset parasympathetic nerve forced regulation state detection feature space, and a preset mood state assessment feature space; The step of calculating and comparing the physiological data of the plurality of time periods according to a preset feature space to determine the mood state of the target individual includes: According to the preset stress level assessment feature space, respectively assessing the stress levels of the physiological data of the multiple time periods to obtain the physiological data of the multiple time periods after the stress level assessment, and calculating the physiological data of the multiple time periods according to a preset stress scoring method to obtain stress scores of the physiological data of the multiple time periods; performing parasympathetic nerve forced regulation detection on the physiological data of the multiple time periods according to the preset parasympathetic nerve forced regulation state detection feature space, and obtaining parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods; According to the preset mood state assessment feature space, the stress scores and parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods are analyzed and compared to determine the mood state of the target individual.

2. The method for monitoring mood state according to claim 1, characterized in that: The obtaining of the target individual's autonomic nervous activity signal comprises: Acquiring physiological signals of the target individual collected by a medical sensor; wherein the physiological signals include autonomic nerve activity signals and three-dimensional acceleration signals of the human body; Segmenting and calculating the autonomic nerve activity signal and the human body three-dimensional acceleration signal to obtain the human body three-dimensional acceleration signal and the autonomic nerve activity signal in multiple time periods; Determining whether the multiple time-sharing three-dimensional human body acceleration signals are greater than a preset three-dimensional human body acceleration signal; If yes, determining the autonomic nervous activity signals corresponding to the human body three-dimensional acceleration signals in the multiple time periods as the autonomic nervous activity signals of the non-target individual; If not, the autonomic nervous activity signals corresponding to the three-dimensional acceleration signals of the human body in the multiple time periods are determined as the autonomic nervous activity signals of the target individual.

3. The method for monitoring mood state according to claim 1, characterized in that: The preset pressure level assessment feature space is constructed in the following manner: According to the preset pressure level experience standard, the preset physiological data is calibrated at the pressure level to obtain the preset physiological data of multiple pressure levels; According to the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms, the preset physiological data of the multiple pressure levels are calculated to obtain the calculation results of the preset physiological data of the multiple pressure levels, and the preset pressure level assessment feature space is constructed according to the calculation results.

4. The method for monitoring mood state according to claim 3, characterized in that: in, The preset stress scoring method includes Formula 1: Score is the stress score of the physiological data of the time period; x is the physiological data of the time period; x1 is the class center of the preset physiological data corresponding to the stress level of the physiological data of the time period; s is the class standard deviation of the preset physiological data; j is the pressure level of the physiological data in the time period; The step of performing stress level assessment on the physiological data of the multiple time periods according to the preset stress level assessment feature space to obtain the physiological data of the multiple time periods after the stress level assessment, and calculating the physiological data of the multiple time periods according to a preset stress scoring method to obtain stress scores for the physiological data of the multiple time periods includes: Calculating the physiological data of the multiple time periods according to the periodic function of the target sequence rhythm and the average fluctuation function of two target sequence rhythms to obtain calculation results of the physiological data of the multiple time periods; Mapping the calculation results of the physiological data of the multiple time periods to the preset pressure level assessment feature space, and performing distance calculations between the cluster centers of the calculation results of the preset physiological data of the multiple pressure levels and the calculation results, to obtain distances between the calculation results of the physiological data of the multiple time periods and the cluster centers of the calculation results of the preset physiological data of the multiple pressure levels; Determining a plurality of minimum distances from the distances between the calculated results of the physiological data of the plurality of time periods and the cluster centers of the calculated results of the preset physiological data of the plurality of pressure levels, and using the pressure levels of the preset physiological data corresponding to the plurality of minimum distances as the pressure levels of the physiological data of the plurality of time periods; The stress scores of the physiological data of the multiple time periods are obtained according to Formula 1, the physiological data of the multiple time periods, the class centers of the preset physiological data of the stress levels corresponding to the physiological data of the multiple time periods, the class standard deviations of the preset physiological data and the stress levels of the physiological data of the multiple time periods.

5. The method for monitoring mood state according to claim 1, characterized in that: The preset parasympathetic nerve forced regulation state detection feature space is constructed in the following way: Calibrate the preset physiological data according to the preset stress level empirical effect standard, the preset fatigue theoretical model, and the preset work and rest and eating irregularity effect standard to obtain a plurality of first physiological data and second physiological data; wherein the first physiological data is calibrated as parasympathetic nervous system forced regulation; Performing continuous wavelet transform on the plurality of first physiological data and the second physiological data according to the wavelet basis function of the preset transformation scale interval, and obtaining a geometric morphology approximation degree measurement index of the plurality of first physiological data and the second physiological data and the physiological data of the plurality of time periods; Calculating the average difference between the plurality of first physiological data and the second physiological data and the adjacent first physiological data and the second physiological data to obtain the average difference between the plurality of first physiological data and the second physiological data; Acquire power values of the plurality of first physiological data and second physiological data at a preset frequency; Obtaining calculation results of the plurality of first physiological data and the second physiological data based on the maximum transformation scale, the average difference, and the power value at the preset frequency of the plurality of first physiological data and the second physiological data; The parasympathetic nerve forced regulation state detection feature space is constructed according to calculation results of the plurality of first physiological data and the second physiological data.

6. The method for monitoring mood state according to claim 5, characterized in that: The method of performing parasympathetic nerve forced regulation detection on the physiological data of the multiple time periods according to the preset parasympathetic nerve forced regulation state detection feature space to obtain the parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods includes: Performing wavelet transform on the physiological data of the multiple time periods to obtain geometric morphology approximation degree measurement indicators of the physiological data of the multiple time periods; Calculating the average difference between the physiological data of the multiple time periods and the physiological data of the adjacent time periods to obtain the average difference of the physiological data of the multiple time periods; Acquire power values of the physiological data of the plurality of time periods at the preset frequency; Obtaining calculation results of the physiological data of the multiple time periods based on the geometric approximation degree metric, the average difference, and the power value at the preset frequency of the physiological data of the multiple time periods; Mapping the calculation results of the multiple time-divided physiological data to the preset parasympathetic nerve forced regulation state detection feature space, and performing distance calculations between the cluster centers of the calculation results of the multiple first physiological data and the second physiological data and the calculation results, respectively, to obtain distances between the calculation results of the multiple time-divided physiological data and the cluster centers of the calculation results of the multiple first physiological data and the second physiological data; Determining a plurality of minimum distances respectively from the distances between the calculation results of the physiological data of the plurality of time periods and the cluster centers of the calculation results of the plurality of first physiological data and second physiological data; If the preset physiological data corresponding to the multiple minimum distances is the first physiological data, the multiple time-divided physiological data corresponding to the multiple minimum distances are marked as parasympathetic nerve forced regulation, and the number of parasympathetic nerve forced regulation of the physiological data of the multiple time-divided time periods is recorded to obtain the parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time-divided time periods.

7. The method for monitoring mood state according to claim 1, characterized in that: The preset mood state assessment feature space is constructed in the following manner: According to a preset self-assessment form, the preset physiological data is calibrated for the mood state to obtain the preset physiological data of the calibrated mood state; Calculating the stress score and parasympathetic nerve forced regulation time distribution and frequency of preset physiological data of the calibrated mood state, and obtaining the preset stress score, preset parasympathetic nerve forced regulation time distribution and frequency for a good mood state and a bad mood state; The preset mood state assessment feature space is constructed according to the preset stress scores of a good mood state and a bad mood state, and the preset parasympathetic nerve forced regulation time distribution and frequency.

8. The method for monitoring mood state according to claim 7, characterized in that: The analyzing and comparing the stress scores and the forced adjustment time distribution and frequency of the physiological data of the multiple time periods according to the preset mood state assessment feature space to determine the mood state of the target individual includes: Grouping the stress scores of the physiological data of the multiple time periods according to the work and rest time of the target individual to obtain stress scores of the physiological data of the multiple wakefulness states and the sleep states; The parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods are grouped according to a preset stress level empirical criterion and a preset fatigue theoretical model to obtain the parasympathetic nerve forced regulation time distribution and frequency of multiple physiological fatigue prone times and physiological fatigue non-prone times; Determining, from the stress scores of the physiological data of the multiple wakefulness states, a stress score of the physiological data of the wakefulness state that is less than the preset stress score of the good mood state, to obtain a stress score of the physiological data of the first wakefulness state; Determining the stress score of the physiological data of the sleep state having a stress score of zero from the stress scores of the physiological data of the multiple sleep states, to obtain the stress score of the physiological data of the first sleep state; If the ratio of the duration corresponding to the stress score of the physiological data of the first wakefulness state to the duration of the multiple time periods is greater than the preset ratio of the duration of the good mood state, the ratio of the duration corresponding to the stress score of the physiological data of the first sleep state to the duration of the multiple time periods is greater than the preset ratio of the duration of the good mood state, and the forced adjustment frequency of the multiple physiological fatigue times is less than the preset adjustment frequency of the good mood state, then the mood state is determined to be good.

9. The method for monitoring mood state according to claim 8, characterized in that: The step of analyzing and comparing the stress scores and forced adjustment frequencies of the physiological data of the plurality of time periods according to the feature space of the preset mood state assessment to determine the mood state includes: Determining, from the stress scores of the physiological data of the multiple wakefulness states, a stress score of the physiological data of the wakefulness state that is greater than the preset stress score of the poor mood state, to obtain a stress score of the physiological data of the second wakefulness state; Determining, from the stress scores of the physiological data of the plurality of sleep states, a stress score of the physiological data of the sleep state that is greater than the preset stress score of the poor mood state, to obtain a stress score of the physiological data of the second sleep state; If the ratio of the duration corresponding to the stress score of the physiological data of the second wakefulness state to the duration of the multiple time periods is greater than the preset ratio of the duration of the poor mood state, the ratio of the duration corresponding to the stress score of the physiological data of the second sleep state to the duration of the multiple time periods is greater than the preset ratio of the duration of the poor mood state, and the forced adjustment frequency of the multiple physiological fatigue-prone times is greater than the preset adjustment frequency of the poor mood state, then the mood state is determined to be poor.

10. A mood state monitoring device, characterized in that: The device comprises: An acquisition module, configured to acquire an autonomic nervous activity signal of a target individual; wherein the autonomic nervous activity signal includes a plurality of autonomic nervous activity signals in different time periods; An extraction module, configured to extract target sequences from the plurality of time-divided autonomic nerve activity signals, respectively, to obtain physiological data of the plurality of time-divided periods; an analysis module, configured to calculate and compare the physiological data of the plurality of time periods according to a preset feature space to determine the mood state of the target individual; wherein the preset feature space includes: a preset stress level assessment feature space, a preset parasympathetic nervous system forced regulation state detection feature space, and a preset mood state assessment feature space; The analysis module is specifically used to: According to the preset stress level assessment feature space, respectively assessing the stress levels of the physiological data of the multiple time periods to obtain the physiological data of the multiple time periods after the stress level assessment, and calculating the physiological data of the multiple time periods according to a preset stress scoring method to obtain stress scores of the physiological data of the multiple time periods; performing parasympathetic nerve forced regulation detection on the physiological data of the multiple time periods according to the preset parasympathetic nerve forced regulation state detection feature space, and obtaining parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods; According to the preset mood state assessment feature space, the stress scores and parasympathetic nerve forced regulation time distribution and frequency of the physiological data of the multiple time periods are analyzed and compared to determine the mood state of the target individual.

11. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 9 is executed.

Images