A portable device, method, equipment and program product for evaluating autonomic nervous function

Abstract

The present invention discloses a portable autonomic nervous function assessment device, method, equipment, and program product. Specifically, the device utilizes an intelligent respiratory monitoring and guidance program to individually adjust a user's breathing to a specific frequency, utilizes a heart rate acquisition device to record the user's heart rate fluctuations during this process, quantitatively analyzes the user's cardiopulmonary coupling strength based on a feature extraction algorithm, and constructs a mapping model between physiological characteristics and clinical indicators to achieve quantitative assessment and portable monitoring of autonomic nervous function. Based on accurate decoding of the cardiopulmonary coupling process, the present invention realizes a portable autonomic nervous function assessment device, method, equipment, and program product that is fast, inexpensive, easy to operate, and has widespread application. The device requires a simple measurement device and can be used for personalized home health assessments.

Description

Technical Field

[0001] The present invention relates to a portable evaluation technology for autonomic nervous system function, and belongs to the technical field of signal processing. Background Art

[0002] The autonomic nervous system is closely linked to the regulation of heart rate, blood pressure, digestion, urination, and body temperature, and is crucial for maintaining internal balance. In our fast-paced society, as individuals' sub-health conditions worsen, the negative impact of autonomic nervous system imbalances or disorders on quality of life becomes increasingly significant. Therefore, monitoring autonomic nervous system function plays a guiding role in assessing health levels and adjusting lifestyles.

[0003] The measurement principle of heart rate variability (HRV) is simple and is often used to monitor autonomic nervous function, but this analysis method is greatly affected by individual differences. HRV analysis in the evaluation of autonomic nervous system function, especially in the evaluation of sympathetic nervous function, often relies on a longer signal monitoring time. Generally, in order to achieve more accurate quantitative analysis of autonomic nervous function, researchers will conduct a comprehensive analysis based on indicators such as blood pressure variability (BPV) or pre-ejection period (PEP). However, BPV analysis requires the use of a continuous dynamic blood pressure monitor, and PEP analysis requires the use of an impedance electrocardiogram. Both of these are complex in principle and expensive, making them difficult to use for individualized daily health management. Therefore, a sensitive, accurate, and economical method for evaluating autonomic nervous function is needed. Summary of the Invention

[0004] Purpose of the invention: In view of the above-mentioned deficiencies in the prior art, the purpose of the present invention is to provide a portable assessment device, method, equipment and program product for autonomic nervous function, so as to realize quantitative assessment and portable monitoring of autonomic nervous function.

[0005] Technical solution: To achieve the above-mentioned purpose, the present invention adopts the following technical solution:

[0006] A portable device for evaluating autonomic nervous system function, comprising:

[0007] The heart rate extraction and preprocessing module is used to collect physiological signals and obtain the user's heart rate fluctuations through preprocessing operations;

[0008] Intelligent breathing test and guidance module, which is used to adaptively measure the user's breathing ability, determine a personalized breathing rhythm, and guide the user to follow the breathing rhythm;

[0009] The cardiopulmonary coupling feature extraction module is used to perform spectral analysis and causal analysis on the user's heart rate signal during breathing training, and calculate cardiopulmonary coupling indicators, including the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during long inhalation and short exhalation, and the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during short inhalation and long exhalation;

[0010] The autonomic nervous function analysis module is used to integrate physiological characteristics and determine the user's autonomic nervous system status through an autonomic nervous balance analysis model and an autonomic nervous activity analysis model; the autonomic nervous balance is represented by the ratio of the cardiopulmonary follow-up response index in different processes, and the autonomic nervous activity is represented by the cardiopulmonary coupling strength.

[0011] Furthermore, in the heart rate extraction and preprocessing module, the preprocessing operation steps specifically include:

[0012] Find the peak points in the physiological signal and obtain a peak point sequence {r(n), n=1, 2, ..., N} derived from the physiological signal, where N is the total number of points in the sequence;

[0013] Performing a first-order difference operation on the peak point sequence to calculate the peak point interval sequence {rr(n), n=1, 2, ..., N-1};

[0014] The user's beat-to-beat heart rate is calculated based on the peak point interval sequence:

[0015] Furthermore, in the intelligent breathing test and guidance module, the steps of measuring breathing capacity and formulating and guiding breathing rhythm specifically include:

[0016] Introducing the designated breathing method and rhythm through voice and / or video, guiding the user to inhale continuously and evenly until no more air can be inhaled, and then to exhale continuously and evenly until no more air can be exhaled;

[0017] The first-order difference of the user's heart rate during the calculation process is: hr′(n)=hr(n)-hr(n-1), n=2,...,N-1, where N is the total number of peak points in the physiological signal sequence, hr i The user's beat-to-beat heart rate;

[0018] The first point in time with hr′(n) ≥ 2 bmp is defined as the starting inhalation key frame n1; the first point in time after the inhalation key frame with hr′(n) ≤ 0.5 bmp is defined as the transition key frame n2; the first point in time after the transition key frame with hr′(n) ≤ -2 bmp is defined as the starting exhalation key frame n2; the first point in time after the starting exhalation key frame with hr′(n) ≥ -0.5 bmp is defined as the ending key frame n4;

[0019] Calculate the user's maximum constant inspiratory time: and maximum constant exhalation time: in to is the time corresponding to the corresponding key frame;

[0020] Based on the measured maximum uniform inhalation and exhalation time of the user, the user's breathing rhythm is personalized and adaptively set, and the user is guided to breathe according to the set rhythm through voice, vibration and / or video, including long inhalation and short exhalation training and short inhalation and long exhalation training.

[0021] Furthermore, in the cardiopulmonary coupling feature extraction module, the steps of calculating the cardiopulmonary follow-up response index and the cardiopulmonary coupling strength specifically include:

[0022] The heart rate fluctuation is converted from the time domain to the frequency domain through discrete Fourier transform:

[0023]

[0024] Where i is the imaginary unit, N is the number of sampling points, hr(n) is the heart rate signal, and HR[k] is the frequency domain distribution of the heart rate signal;

[0025] Calculate the power of each point in the discrete frequency domain to obtain the power spectrum density of the heart rate signal: where f s is the sampling frequency of the original physiological signal.

[0026] Calculate the cardiopulmonary response index during long inhalation and short exhalation: Where f1 is the fixed respiratory frequency during long inhalation and short exhalation, peak s1 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C1 is The integer part of ;

[0027] Calculate the cardiopulmonary response index during short inhalation and long exhalation: Where f2 is the fixed respiratory frequency during short inhalation and long exhalation, peak s2 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C2 is The integer part of ;

[0028] The heart rate signal hr(n) is decomposed into multiple components HR with physical meaning through EEMD j (n), j = 1, 2, ..., m;

[0029] The respiratory timing signal is converted into Disassemble into multiple components with physical meaning RESP j (n), j = 1, 2, ..., m;

[0030] HR is obtained by Hilbert transform j Phase φ of (n) x (n), and RESP j Phase φ of (n)y (n);

[0031] Calculate HR j (n) and RESP j (n) coupling strength: where i is the imaginary unit;

[0032] The cardiopulmonary coupling strength of the user is calculated as: CCS = max{Coh1, Coh2, ..., Coh m}.

[0033] Furthermore, in the autonomic nervous function analysis module, the step of determining the state of the user's autonomic nervous system specifically includes:

[0034] Calculate the user's autonomic balance: CRI1 is the cardiopulmonary response index during long inhalation and short exhalation, and CRI2 is the cardiopulmonary response index during short inhalation and long exhalation.

[0035] Based on the statistical analysis results, determine whether the user's autonomic nervous system state deviates from the normal range;

[0036] If the user's autonomic nervous system status deviates from the normal range, the autonomic nervous system activity is calculated: CCS1 is the cardiopulmonary coupling strength during long inhalation and short exhalation, and CCS2 is the cardiopulmonary coupling strength during short inhalation and long exhalation; SNA represents the sympathetic nerve activity, and PNA represents the parasympathetic nerve activity;

[0037] A feature vector [SNA, SNA] is constructed, and based on the feature vector, a distance between the user's autonomic nervous activity and each type of autonomic nervous dysfunction obtained by clustering is calculated, thereby determining the specific type of the user's autonomic nervous dysfunction.

[0038] Furthermore, the autonomic dysfunction type is obtained by performing K-means clustering on the collected sample data, and the number of clusters is set to 6, corresponding to 6 types of autonomic nervous system imbalance:

[0039] a. Sympathetic side: sympathetic nerves are strong and parasympathetic nerves are normal;

[0040] b. Sympathetic side: Sympathetic nerves are normal, parasympathetic nerves are weak;

[0041] c. Sympathetic side: strong sympathetic nerves and weak parasympathetic nerves;

[0042] d. Biased towards the parasympathetic side: sympathetic nerves are normal, parasympathetic nerves are strong;

[0043] e. Biased towards the parasympathetic side: sympathetic nerves are weak and parasympathetic nerves are normal;

[0044] f. Biased towards the parasympathetic nerve side: the sympathetic nerves are weak and the parasympathetic nerves are strong.

[0045] A portable method for assessing autonomic nervous function comprises the following steps:

[0046] Collect physiological signals and obtain the user's heart rate fluctuations through preprocessing operations;

[0047] Adaptively measures the user's breathing ability, determines a personalized breathing rhythm, and guides the user to follow that breathing rhythm;

[0048] Perform spectrum analysis and causal analysis on the user's heart rate signal during breathing training to calculate cardiopulmonary coupling indicators, including the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during long inhalation and short exhalation, and the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during short inhalation and long exhalation.

[0049] Physiological characteristics are integrated and the user's autonomic nervous system state is determined through an autonomic nervous balance analysis model and an autonomic nervous activity analysis model; the autonomic nervous balance is represented by the ratio of the cardiopulmonary follow-up response index in different processes, and the autonomic nervous activity is represented by the cardiopulmonary coupling strength.

[0050] An electronic device comprises a memory, a processor and a computer program / instruction stored in the memory and executable on the processor, wherein the computer program / instruction, when executed by the processor, implements the steps of the portable autonomic nervous function assessment method.

[0051] Furthermore, the electronic device further includes a physiological signal acquisition device, or acquires the collected physiological signals or the user's heart rate signal by communicating with the physiological signal acquisition device.

[0052] A computer program product comprises a computer program / instruction, which implements the steps of the portable autonomic nervous function assessment method when executed by a processor.

[0053] Beneficial effects: Compared with the prior art, the present invention has the following beneficial effects:

[0054] As an improvement and innovation, the present invention (1) proposes a portable autonomic nervous function assessment device based on the cardiopulmonary coupling principle, which only needs to extract and analyze the user's heart rate signal during the dynamic process, has the characteristics of simplicity and lightweight, and can be transplanted to various types of wearable health devices or other terminals with heart rate monitoring functions that are currently widely used; (2) proposes an intelligent breathing test and guidance method, which adaptively measures the user's breathing ability through heart rate signals, sets a personalized breathing frequency, and guides the user to grasp the breathing rhythm through multiple sensory media such as vision, hearing, and touch; (3) proposes a cardiopulmonary coupling analysis framework, which integrates spectrum and causal analysis to accurately and quantitatively extract the user's cardiopulmonary coupling characteristics, and realize objective and detailed evaluation of the user's autonomic nervous system function.

[0055] The beneficial effects of the present invention are: based on the precise decoding of the cardiopulmonary coupling process, a fast, low-cost, easy-to-operate, and popularizable portable autonomic nervous function assessment device, method, equipment, and program product are realized. The required measurement device is simple and can be used for individualized home health assessment. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a schematic diagram of the technical solution of the present invention.

[0057] Figure 2 Schematic diagram of the system structure of the present invention.

[0058] Figure 3 The figure is a schematic diagram of a specific implementation process of a certain embodiment of the present invention. DETAILED DESCRIPTION

[0059] The technical details of the technical solution of the present invention are further described below with reference to the accompanying drawings and specific embodiments.

[0060] Rhythmic breathing can induce a significant cardiopulmonary coupling phenomenon, that is, a high degree of follow-up response between breathing and heart rate, which is closely related to the autonomic nervous system. When the autonomic nervous system loses balance or is damaged, the cardiopulmonary coupling phenomenon related to autonomic nervous function will also weaken. Therefore, the present invention proposes a new idea, that is, measuring the user's heart rate under a specific breathing rhythm set adaptively, combining a feature extraction algorithm to quantitatively construct a correlation between the cardiopulmonary coupling intensity and the autonomic nervous function, in the hope of realizing a fast, accurate, low-cost, easy to operate, and popularizable autonomic nervous function assessment method. Relevant embodiments are as follows:

[0061] Example 1: Figure 1As shown, an embodiment of the present invention discloses a portable autonomic nervous function assessment device, which mainly includes: a heart rate extraction and preprocessing module, an intelligent breathing test and guidance module, a cardiopulmonary coupling feature extraction module, and an autonomic nervous function analysis module. The heart rate extraction and preprocessing module is used to collect physiological signals and obtain the user's heart rate fluctuations through preprocessing operations. The intelligent breathing test and guidance module is used to adaptively measure the user's breathing capacity, determine a personalized breathing rhythm, and guide the user to follow this breathing rhythm. The cardiopulmonary coupling feature extraction module is used to perform spectral analysis and causal analysis on the user's heart rate signal during breathing training, and calculate cardiopulmonary coupling indicators, including the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during long inhalation and short exhalation, and the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during short inhalation and long exhalation. The autonomic nervous function analysis module is used to integrate physiological features and determine the user's autonomic nervous system status through an autonomic nervous balance analysis model and an autonomic nervous activity analysis model. The autonomic nervous balance is characterized by the ratio of the cardiopulmonary follow-up response index during different processes, and the autonomic nervous activity is characterized by the cardiopulmonary coupling strength. The central controller is used to control each module.

[0062] The specific processing idea of ​​the heart rate extraction and preprocessing module can be: collect original physiological signals containing heart rate information such as PPG or ECG through the device, use filtering processing to remove baseline and high-frequency noise, and calculate the heart rate information during the detection process based on the peak feature points in the original physiological signal.

[0063] Specifically, in some embodiments, the preprocessing steps in the heart rate extraction and preprocessing module specifically include:

[0064] Find the peak points in the physiological signal and obtain the peak point sequence {r(n), n = 1, 2, ..., N} derived from the physiological signal, where N is the total number of points in the sequence;

[0065] Perform a first-order difference operation on the peak point sequence to calculate the peak point interval sequence {rr(n), i=1, 2, ..., N-1};

[0066] The user's beat-to-beat heart rate is calculated based on the peak point interval sequence:

[0067] The specific processing idea of ​​the intelligent breathing test and guidance module can be: analyze the user's heart rate signal during the process of uniform deep breathing, calculate a comprehensive indicator to evaluate the user's breathing ability, set a suitable breathing rhythm based on the indicator, and guide the user to follow the breathing rhythm based on sensory media such as vision, hearing, and touch.

[0068] Specifically, in some embodiments, the steps of measuring breathing capacity and formulating and guiding breathing rhythm in the intelligent breathing test and guidance module specifically include:

[0069] Introducing the designated breathing method and rhythm through voice and / or video, guiding the user to inhale continuously and evenly until no more air can be inhaled, and then to exhale continuously and evenly until no more air can be exhaled;

[0070] The first-order difference of the user's heart rate during the calculation process is: hr′(n)=hr(n)-hr(n-1), n=2,...,N-1, where N is the total number of peak points in the physiological signal sequence, hr i The user's beat-to-beat heart rate;

[0071] The first point in time with hr′(n) ≥ 2 bmp is defined as the starting inspiration key frame n1; the first point in time after the inspiration key frame with hr′(n) ≤ 0.5 bmp is defined as the transition key frame n2; the first point in time after the transition key frame with hr′(n) ≤ -2 bmp is defined as the starting exhalation key frame n3; the first point in time after the starting exhalation key frame with hr′(n) ≥ -0.5 bmp is defined as the ending key frame n4;

[0072] Calculate the user's maximum constant inspiratory time: and maximum constant exhalation time: in to is the time corresponding to the corresponding key frame;

[0073] Based on the measured maximum uniform inhalation and exhalation time of the user, the user's breathing rhythm is personalized and adaptively set, and the user is guided to breathe according to the set rhythm through voice, vibration and / or video, including long inhalation and short exhalation training and short inhalation and long exhalation training.

[0074] The specific processing idea of ​​the cardiopulmonary coupling feature extraction module can be: based on methods such as spectrum analysis and causal analysis, extract the cardiopulmonary coupling features related to the autonomic nervous system.

[0075] Specifically, in some embodiments, the steps of calculating the cardiopulmonary follow-up response index and the cardiopulmonary coupling strength in the cardiopulmonary coupling feature extraction module specifically include:

[0076] The heart rate fluctuation is converted from the time domain to the frequency domain through discrete Fourier transform:

[0077]

[0078] Where i is the imaginary unit, N is the number of sampling points, hr(n) is the heart rate signal, and HR[k] is the frequency domain distribution of the heart rate signal;

[0079] Calculate the power of each point in the discrete frequency domain to obtain the power spectrum density of the heart rate signal: where f s is the sampling frequency of the original physiological signal.

[0080] Calculate the cardiopulmonary response index during long inhalation and short exhalation: Where f1 is the fixed respiratory frequency during long inhalation and short exhalation, peak s1 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C1 is The integer part of ;

[0081] Calculate the cardiopulmonary response index during short inhalation and long exhalation: Where f2 is the fixed respiratory frequency during short inhalation and long exhalation, peak s2 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C2 is The integer part of ;

[0082] The heart rate signal hr(n) is decomposed into multiple components HR with physical meaning through EEMD j (n), j = 1, 2, ..., m;

[0083] The respiratory timing signal is converted into Disassemble into multiple components with physical meaning RESP j (n), j = 1, 2, ..., m;

[0084] HR is obtained by Hilbert transform j Phase φ of (n) x (n), and RESP j Phase φ of (n) y (n);

[0085] Calculate HR j (n) and RESP j (n) coupling strength: where i is the imaginary unit;

[0086] The cardiopulmonary coupling strength of the user is calculated as: CCS = max{Coh1, Coh2, ..., Coh m}.

[0087] The specific processing ideas of the autonomic nervous function analysis module can be: based on the statistical analysis results, a multidimensional comprehensive model is constructed to clarify the mapping relationship between cardiopulmonary coupling characteristics and autonomic nervous state, and to achieve quantitative analysis of physiological indicators including but not limited to autonomic nervous balance, sympathetic nervous activity, parasympathetic nervous activity, etc.

[0088] Specifically, in some embodiments, the step of determining the state of the user's autonomic nervous system in the autonomic nervous function analysis module specifically includes:

[0089] Calculate the user's autonomic balance: CRI1 is the cardiopulmonary response index during long inhalation and short exhalation, and CRI2 is the cardiopulmonary response index during short inhalation and long exhalation.

[0090] Based on the statistical analysis results, determine whether the user's autonomic nervous system state deviates from the normal range;

[0091] If the user's autonomic nervous system status deviates from the normal range, the autonomic nervous system activity is calculated: CCS1 is the cardiopulmonary coupling strength during long inhalation and short exhalation, and CCS2 is the cardiopulmonary coupling strength during short inhalation and long exhalation; SNA represents the sympathetic nerve activity, and PNA represents the parasympathetic nerve activity;

[0092] A feature vector [SNA, SNA] is constructed, and based on the feature vector, a distance between the user's autonomic nervous activity and each type of autonomic nervous dysfunction obtained by clustering is calculated, thereby determining the specific type of the user's autonomic nervous dysfunction.

[0093] Example 2: Figure 2 As shown, an embodiment of the present invention discloses a portable method for evaluating autonomic nervous function, comprising the following steps:

[0094] (1) Heart rate signal extraction: collect physiological signals and obtain the user's heart rate fluctuation through preprocessing operations;

[0095] (2) Intelligent breathing test and guidance: Adaptively measure the user's breathing ability, determine the personalized breathing rhythm, and guide the user to follow the breathing rhythm;

[0096] (3) Cardiopulmonary coupling feature extraction: Spectral analysis and causal analysis are performed on the user's heart rate signal during breathing training to calculate the cardiopulmonary coupling indicators, including the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during long inhalation and short exhalation, and the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during short inhalation and long exhalation;

[0097] (4) Construction of an autonomic nervous function analysis model: Integrating physiological characteristics, the user's autonomic nervous system status is determined through an autonomic nervous balance analysis model and an autonomic nervous activity analysis model; the autonomic nervous balance is represented by the ratio of the cardiopulmonary follow-up response index in different processes, and the autonomic nervous activity is represented by the cardiopulmonary coupling strength.

[0098] The execution subject of the heart rate signal extraction method in step (1) can be a terminal such as a wearable device, including but not limited to any terminal or system with the ability to collect physiological signals such as PPG signals, ECG signals, millimeter wave radar, etc. that can be used to calculate heart rate information.

[0099] This embodiment can record raw physiological signals including heart rate information using a physiological signal acquisition device that integrates a PPG photoelectric sensor or an ECG sensor. The PPG photoelectric sensor can include visible light, such as green and red light, as well as invisible light of various wavelengths, such as infrared light. The ECG sensor can include a single lead or multiple lead electrodes.

[0100] The influence of baseline drift, power frequency interference and other noises in the original physiological signal on heart rate extraction can be reduced by filtering processing. The filtering processing can be bandpass filtering. The filtering can use a zero-phase shift digital filter to bandpass filter the signal. The signal can be processed by bandpass filtering to obtain a physiological signal with baseline drift and high-frequency noise removed.

[0101] The findpeaks function provided in MATLAB software can be used to find the peak points in the physiological signal and obtain the peak point sequence {r(n), n=1, 2, ..., N} derived from the physiological signal.

[0102] The peak point interval sequence {rr(n), i=1, 2, ..., N-1} can be calculated by performing a first-order difference operation on the peak point sequence.

[0103] The user's beat-to-beat heart rate can be calculated based on the peak point interval sequence:

[0104] The execution subject of the implementation method of the intelligent breathing test and guidance program in step (2) can be a mobile phone, tablet computer, wearable device and other devices, or a laptop computer, desktop computer and other devices with interaction, storage and data processing capabilities, or other intelligent terminal devices.

[0105] In this embodiment, the method in step (2) can be used to execute a physical key or a virtual key to trigger the intelligent respiratory monitoring and guidance program. The program can include the following modules: a test guidance module, a collection module, a test analysis module, and a training guidance module.

[0106] The test guidance module can introduce the specified breathing method and rhythm through voice (or video) and other means. The breathing method and rhythm specified in this module can be: continuous and even inhalation until no more gas can be inhaled, and then continuous and even exhalation until no more gas can be exhaled.

[0107] The function of the acquisition module can be to capture the user's heart rate fluctuations during the respiratory monitoring process. This module can be implemented based on the heart rate extraction method in step (1).

[0108] The test analysis module can be used to convert heart rate fluctuations captured by the acquisition module into respiratory-related features based on the principle of cardiopulmonary coupling. Respiratory-related features may include maximum uniform inhalation time and maximum uniform exhalation time. The analysis module can perform a first-order difference calculation on the user's heart rate fluctuations during the test to obtain the heart rate variability: hr′(n) = hr(n) - hr(n-1), where n = 2, ..., N-1.

[0109] The first point in time with hr′(n) ≥ 2 bmp can be defined as the starting inhalation key frame n1; the first point in time after the inhalation key frame with hr′(n) ≤ 0.5 bmp can be defined as the transition key frame n2; the first point in time after the transition key frame with hr′(n) ≤ -2 bmp can be defined as the starting exhalation key frame n3; the first point in time after the starting exhalation key frame with hr′(n) ≥ -0.5 bmp can be defined as the ending key frame n4.

[0110] The measured maximum uniform inspiratory time can be expressed as:

[0111] The measured maximum uniform exhalation time can be expressed as:

[0112] The training guidance module can be used to customize and adaptively set an appropriate breathing rhythm based on the user's maximum uniform inhalation and exhalation times measured by the test analysis module, and guide the user to follow the set rhythm through voice, vibration, video, and other means. To prevent individual breathing habits from affecting subsequent analysis results, the training guidance module's breathing exercises can include: 3-minute long inhalation and short exhalation training, and 3-minute short inhalation and long exhalation training.

[0113] Among them, the inhalation time in long inhalation and short exhalation training can be set to (0.75×Δt1) seconds, and the exhalation time can be set to (0.55×Δt2) seconds; the inhalation time in short inhalation and long exhalation training can be set to (0.55×Δt1) seconds, and the exhalation time can be set to (0.75×Δt2) seconds.

[0114] The execution subject of the cardiopulmonary coupling feature extraction method in step (3) can be a device such as a mobile phone, a tablet computer, a wearable device, or a device with data processing capabilities such as a laptop computer and a desktop computer.

[0115] This embodiment can extract the user's cardiopulmonary coupling characteristics through power spectrum analysis and causal analysis.

[0116] Power spectrum analysis can convert heart rate fluctuations from the time domain to the frequency domain through discrete Fourier transform:

[0117] The power spectrum density of the heart rate signal can be obtained by calculating the power of each point in the discrete frequency domain:

[0118] Among them, the power spectrum density corresponding to the i-th peak point of the heart rate signal power spectrum can be expressed as: PSD (peak i ).

[0119] For the long inhalation and short exhalation process, the fixed respiratory frequency can be expressed as:

[0120] From the heart rate-respiration tracking response, we can know that the cardiopulmonary tracking response index during long inhalation and short exhalation can be expressed as:

[0121]

[0122] For the short inhalation and long exhalation process, the fixed respiratory frequency can be expressed as:

[0123] From the heart rate-respiration tracking response, we can know that the cardiopulmonary tracking response index during short inhalation and long exhalation can be expressed as:

[0124]

[0125] Causal analysis can decompose the heart rate signal into multiple components with physical meaning through EEMD j (n), j = 1, 2, ..., m.

[0126] According to the set breathing rhythm, the state value of the inhalation process can be set to "1", and the state value of the exhalation process can be set to "0" to obtain the timing signal of the specified breathing rhythm Similarly, the respiratory timing signal can be decomposed into multiple components RESP with physical meaning through EEMD j (n), j = 1, 2, ..., m.

[0127] HR can be solved by Hilbert transform j Phase φ of (t) x (n), and RESP j Phase φ of (t) y (n).

[0128] HR j (t) and RESP j The coupling strength between (t) can be expressed as: where i is the imaginary unit.

[0129] From the heart rate-respiration following response, it can be seen that the cardiopulmonary coupling strength during breathing training can be expressed as: CCS = max{Coh1, Coh2, ..., Coh m}.

[0130] The cardiopulmonary coupling strength during long inhalation and short exhalation can be recorded as CCS1, and the cardiopulmonary coupling strength during short inhalation and long exhalation can be recorded as CCS2.

[0131] The execution subject of the method for constructing the autonomic nervous function analysis model in step (4) can be a device such as a mobile phone, a tablet computer, a wearable device, or a device with data processing capabilities such as a laptop computer and a desktop computer.

[0132] The autonomic nervous system balance analysis model can be constructed by integrating the cardiopulmonary response index in different processes:

[0133] The normal range for ANB can be selected based on the statistical distribution of large sample sizes. A high value can be set as ANB ≥ 1.25, while a low value can be set as ANB ≤ 0.8. A higher ANB indicates a sympathetic balance in the autonomic nervous system, while a lower ANB indicates a parasympathetic balance.

[0134] After preliminary judgment through power spectrum analysis, the cardiopulmonary response index of individuals with autonomic nervous system imbalance in different processes can be integrated to construct an autonomic nervous system activity analysis model:

[0135] Among them, SNA represents the activity of sympathetic nerves, and PNA represents the activity of parasympathetic nerves.

[0136] In order to achieve a more comprehensive representation of the autonomic nervous system and provide targeted personalized recommendations, it is necessary to classify individuals with different autonomic nervous system imbalances. i , its autonomic nervous function characteristics can be represented by the feature vector ANS(U i )=[SNA i , PNA i ] indicates that the set {ANS(U1), ANS(U2), ..., ANS(U N )} for clustering, the number of clusters k can be selected as 6. Cluster analysis obtains the set {K n (x n ,y n ), n = 1, 2, ..., 6}, corresponding to the following 6 types of autonomic nervous system imbalance:

[0137] a. Sympathetic side: sympathetic nerves are strong and parasympathetic nerves are normal;

[0138] b. Sympathetic side: Sympathetic nerves are normal, parasympathetic nerves are weak;

[0139] c. Sympathetic side: strong sympathetic nerves and weak parasympathetic nerves;

[0140] d. Biased towards the parasympathetic side: sympathetic nerves are normal, parasympathetic nerves are strong;

[0141] e. Biased towards the parasympathetic side: sympathetic nerves are weak and parasympathetic nerves are normal;

[0142] f. Biased towards the parasympathetic side: sympathetic nerves are weak and parasympathetic nerves are strong;

[0143] For U i , you can calculate the Euclidean distance from each set Then D k ={D n} min , the corresponding index k is the user's autonomic nervous system imbalance category. Based on the classification results, we can further automatically select personalized adjustment solutions from the constructed knowledge graph and push them to the specified user. The specific analysis results are shown in the following example. Figure 3 .

[0144] As shown in the above implementation process, this method can first determine the balance of the autonomic nervous system based on spectrum analysis, and then solve the autonomic nervous system activity of individuals with autonomic nervous system imbalance / disorder through causal analysis, and determine the type of disorder. Among them, the power spectrum analysis principle is simple, the operation is fast, and it occupies less processor and storage resources; while the causal analysis is more complicated to calculate due to the introduction of EEMD and Hilbert transform, and occupies more processor and storage resources. Therefore, the diversion step processing strategy introduced in this embodiment, that is, first make a rough judgment through power spectrum analysis with less computing power consumption. If the user's autonomic nervous system is found to be unbalanced, more computing power will be further called to solve the type of imbalance, which can save onboard computing power to a certain extent.

[0145] Example 3: An electronic device disclosed in an embodiment of the present invention includes a memory, a processor, and a computer program / instructions stored in the memory and executable on the processor. When executed by the processor, the computer program / instructions implement the steps of the portable autonomic nervous function assessment method. The electronic device can be a mobile phone, tablet computer, wearable device, or the like, including a physiological signal acquisition device; it can also be a mobile phone, tablet computer, laptop computer, desktop computer, etc., which acquires collected physiological signals or processed heart rate signals by communicating with the physiological signal acquisition device.

[0146] Example 4: A computer program product disclosed in an embodiment of the present invention comprises a computer program / instruction, which implements the steps of the portable method for assessing autonomic nervous function when executed by a processor. The program / instruction code for implementing the method of the present invention can be written in any combination of one or more programming languages. These program / instruction codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that the program / instruction code enables the steps of the method of the present invention to be implemented when executed by the processor or controller. The program / instruction code can be executed entirely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine, or entirely on a remote machine or server. Anything not described in detail in the present invention is a common knowledge to those skilled in the art.

Claims

1. A portable device for evaluating autonomic nervous function, characterized in that: The device comprises: The heart rate extraction and preprocessing module is used to collect physiological signals and obtain the user's heart rate fluctuations through preprocessing operations; Intelligent breathing test and guidance module, which is used to adaptively measure the user's breathing ability, determine a personalized breathing rhythm, and guide the user to follow the breathing rhythm; The cardiopulmonary coupling feature extraction module is used to perform spectral analysis and causal analysis on the user's heart rate signal during breathing training, and calculate cardiopulmonary coupling indicators, including the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during long inhalation and short exhalation, and the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during short inhalation and long exhalation. The calculation steps of the cardiopulmonary follow-up response index and cardiopulmonary coupling strength include: The heart rate fluctuation is converted from the time domain to the frequency domain through discrete Fourier transform: Where i is the imaginary unit, N is the total number of sequence points, hr(n) represents the beat-to-beat heart rate signal, and HR[k] represents the frequency domain distribution of the heart rate signal; Calculate the power of each point in the discrete frequency domain to obtain the power spectrum density of the heart rate signal: where f s is the sampling frequency of the original physiological signal; Calculate the cardiopulmonary response index during long inhalation and short exhalation: Where f1 is the fixed respiratory frequency during long inhalation and short exhalation, peak s1 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C1 is The integer part of ; Calculate the cardiopulmonary response index during short inhalation and long exhalation: Where f2 is the fixed respiratory frequency during short inhalation and long exhalation, peak s2 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C2 is The integer part of ; The beat-to-beat heart rate signal hr(n) is decomposed into multiple physically meaningful components HR by EEMD. j (n),j=1,2,…,m; The respiratory timing signal is converted into Disassemble into multiple components with physical meaning RESP j (n),j=1,2,…,m; HR is obtained by Hilbert transform j Phase φ of (n) x (n), and RESP j Phase φ of (n) y (n); Calculate HR j (n) and RESP j (n) coupling strength: The cardiopulmonary coupling strength of the user is calculated as: CCS = max{Coh1, Coh2,…, Coh m }; The autonomic nervous function analysis module is used to integrate physiological characteristics and determine the user's autonomic nervous system status through an autonomic nervous balance analysis model and an autonomic nervous activity analysis model; the autonomic nervous balance is represented by the ratio of the cardiopulmonary follow-up response index in different processes, and the autonomic nervous activity is represented by the cardiopulmonary coupling strength.

2. The portable device for evaluating autonomic nervous function according to claim 1, characterized in that: In the heart rate extraction and preprocessing module, the preprocessing operation steps specifically include: Find the peak points in the physiological signal and obtain the peak point sequence {r(n), n = 1, 2, ..., N} derived from the physiological signal, where N is the total number of points in the sequence; Perform first-order difference operation on the peak point sequence to calculate the peak point interval sequence {rr(n), i = 1, 2, ..., N-1}; The user's beat-to-beat heart rate signal is calculated based on the peak point interval sequence:

3. The portable device for evaluating autonomic nervous function according to claim 1, characterized in that: In the intelligent breathing test and guidance module, the steps of measuring breathing capacity and formulating and guiding breathing rhythm specifically include: Introducing the designated breathing method and rhythm through voice and / or video, guiding the user to inhale continuously and evenly until no more air can be inhaled, and then to exhale continuously and evenly until no more air can be exhaled; The first-order difference of the user's heart rate during calculation is: hr′(n)=hr(n)-hr(n-1), n=2,…,N-1; The first point in time with hr′(n) ≥ 2 bmp is defined as the starting inspiration key frame n1; the first point in time after the inspiration key frame with hr′(n) ≤ 0.5 bmp is defined as the transition key frame n2; the first point in time after the transition key frame with hr′(n) ≤ -2 bmp is defined as the starting exhalation key frame n3; the first point in time after the starting exhalation key frame with hr′(n) ≥ -0.5 bmp is defined as the ending key frame n4; Calculate the user's maximum constant inspiratory time: and maximum constant exhalation time: in to is the time corresponding to the corresponding key frame; Based on the measured maximum uniform inhalation and exhalation time of the user, the user's breathing rhythm is personalized and adaptively set, and the user is guided to breathe according to the set rhythm through voice, vibration and / or video, including long inhalation and short exhalation training and short inhalation and long exhalation training.

4. The portable device for evaluating autonomic nervous function according to claim 1, characterized in that: In the autonomic nervous function analysis module, the step of determining the state of the user's autonomic nervous system specifically includes: Calculate the user's autonomic balance: CRI1 is the cardiopulmonary response index during long inhalation and short exhalation, and CRI2 is the cardiopulmonary response index during short inhalation and long exhalation. Based on the statistical analysis results, determine whether the user's autonomic nervous system state deviates from the normal range; If the user's autonomic nervous system status deviates from the normal range, the autonomic nervous system activity is calculated: CCS1 is the cardiopulmonary coupling strength during long inhalation and short exhalation, and CCS2 is the cardiopulmonary coupling strength during short inhalation and long exhalation; SNA represents the sympathetic nerve activity, and PNA represents the parasympathetic nerve activity; A feature vector [SNA, PNA] is constructed, and based on the feature vector, a distance between the user's autonomic nervous activity and each type of autonomic nervous dysfunction obtained by clustering is calculated, thereby determining the specific type of the user's autonomic nervous dysfunction.

5. The portable device for evaluating autonomic nervous function according to claim 4, characterized in that: The autonomic dysfunction type is obtained by performing K-means clustering on the collected sample data. The number of clusters is set to 6, corresponding to 6 types of autonomic dysfunction: a. Sympathetic side: sympathetic nerves are strong and parasympathetic nerves are normal; b. Sympathetic side: Sympathetic nerves are normal, parasympathetic nerves are weak; c. Sympathetic side: strong sympathetic nerves and weak parasympathetic nerves; d. Biased towards the parasympathetic side: sympathetic nerves are normal, parasympathetic nerves are strong; e. Biased towards the parasympathetic side: sympathetic nerves are weak and parasympathetic nerves are normal; f. Biased towards the parasympathetic nerve side: the sympathetic nerves are weak and the parasympathetic nerves are strong.

6. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instructions are executed by a processor, the following steps are implemented: Collect physiological signals and obtain the user's heart rate fluctuations through preprocessing operations; Adaptively measures the user's breathing ability, determines a personalized breathing rhythm, and guides the user to follow that breathing rhythm; Perform spectrum analysis and causal analysis on the user's heart rate signal during breathing training to calculate cardiopulmonary coupling indicators, including the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during long inhalation and short exhalation, and the cardiopulmonary follow-up response index and cardiopulmonary coupling strength during short inhalation and long exhalation. The calculation steps of the cardiopulmonary follow-up response index and the cardiopulmonary coupling strength include: The heart rate fluctuation is converted from the time domain to the frequency domain through discrete Fourier transform: Where i is the imaginary unit, N is the total number of sequence points, hr(n) represents the beat-to-beat heart rate signal, and HR[k] represents the frequency domain distribution of the heart rate signal; Calculate the power of each point in the discrete frequency domain to obtain the power spectrum density of the heart rate signal: where f s is the sampling frequency of the original physiological signal; Calculate the cardiopulmonary response index during long inhalation and short exhalation: Where f1 is the fixed respiratory frequency during long inhalation and short exhalation, peak s1 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C1 is The integer part of ; Calculate the cardiopulmonary response index during short inhalation and long exhalation: Where f2 is the fixed respiratory frequency during short inhalation and long exhalation, peak s2 Indicates the maximum frequency of the peak point of the heart rate signal spectrum, C2 is The integer part of ; The beat-to-beat heart rate signal hr(n) is decomposed into multiple physically meaningful components HR by EEMD. j (n),j=1,2,…,m; The respiratory timing signal is converted into Disassemble into multiple components with physical meaning RESP j (n),j=1,2,…,m; HR is obtained by Hilbert transform j Phase φ of (n) x (n), and RESP j Phase φ of (n) y (n); Calculate HR j (n) and RESP j (n) coupling strength: The cardiopulmonary coupling strength of the user is calculated as: CCS = max{Coh1, Coh2,…, Coh m }; Physiological characteristics are integrated and the user's autonomic nervous system state is determined through an autonomic nervous balance analysis model and an autonomic nervous activity analysis model; the autonomic nervous balance is represented by the ratio of the cardiopulmonary follow-up response index in different processes, and the autonomic nervous activity is represented by the cardiopulmonary coupling strength.

Images