CN106419937A - Mental stress analysis system based on heart sound HRV theory - Google Patents

Abstract

The invention discloses a mental stress analysis system based on the heart sound HRV theory, which uses heart rate variability as an evaluation index of stress level, and based on the influence of heart sound HRV analysis on mental stress and fatigue, combines heart rate variability and heart sound analysis, and proposes A mental stress analysis system that can accurately measure human mental stress. The system uses a wireless heart sound collector to achieve long-term collection of heart sounds, which lasts for 3-5 minutes; in order to quickly and effectively extract heart rate variability from heart sound signals, an automatic HRV extraction method is proposed. Adaptive algorithm; through time domain and frequency domain analysis of HRV, twelve indicators in time domain and frequency domain are obtained, and 8 main characteristic indicators are graphically represented. Compared with the electrocardiogram and mental stress analyzer, the present invention is convenient to use, simple to operate, complete in functions, especially low in cost, and is suitable for popularization and use in grassroots medical units, families, schools and individuals.

Description

Mental Stress Analysis System Based on Heart Sound HRV Theory

technical field

The invention belongs to the field of diagnosis of mental diseases, in particular to a mental stress analysis method based on heart sound HRV theory.

Background technique

In modern society, as the pace of work and life is getting faster and faster, the psychological pressure that modern people are facing is also increasing. According to the statistical report of the National Health and Family Planning Commission, as of the end of 2015, the number of people suffering from mental illness in China has surpassed that of cardiovascular disease, ranking first in the number of patients with diseases in my country.

Psychological stress is the main cause of sub-health. Too much stress can damage your health. If the psychological pressure cannot be released in time, it will cause mental illness. Excessive stress will lead to sub-health of the body, and in severe cases it will lead to direct disease. Stress is the biggest health killer around us.

However, more than 40% of the people in the survey did not take any measures. Sociologists point out that people are under a lot of pressure nowadays, and many people don't know how to resolve it. For the current situation, there is an urgent need for an instrument that can intuitively detect and analyze the mental state and fatigue of the human body. It can detect the mental stress of the subject, so that the user can have a deep and clear understanding of the patient's mental state. And according to the patient's mental state to give corresponding advice.

Contents of the invention

The purpose of this invention is to propose a device that can intuitively detect and analyze the mental state and fatigue degree of the human body, which can detect the mental stress of the subject, so that the user can have a deep and clear understanding of the mental state of the patient based on Psychological stress analysis system based on heart sound HRV theory.

For this reason, the present invention a kind of mental stress analysis system based on heart sound HRV theory, system is made up of heart sound signal acquisition module, wireless communication module and PC end system platform.

The heart sound signal acquisition module is connected with the PC terminal system platform, and the heart sound fluctuation signal transmitted from the user's chest wall is directly transmitted to the transducer element through the pressure sensitive element by the electric sensor, and then the collected heart sound signal is converted from analog to digital , and transmitted to the PC terminal system platform for processing;

The wireless communication module is connected with the PC terminal system platform, and transmits data between the heart sound collection module and the PC terminal system platform in a wireless manner;

The PC-end system platform is used to analyze the data input by the heart sound acquisition module, restore the heart sound signal, use the heart sound extraction HRV adaptive algorithm to obtain HRV, and process the HRV with a variety of algorithms, and analyze it in the frequency domain and time domain at the same time to obtain Leading indicator of mental stress.

The above-mentioned adaptive algorithm for extracting HRV using heart sound to obtain HRV specifically includes the following steps

①Resample the heart sound signal, and take a point every certain number of points. In the case of ensuring no aliasing, the amount of data is reduced, which is conducive to subsequent processing;

② Use Chebyshev type I low-pass filter to filter out low-frequency components and set the cut-off frequency;

③Use a high-pass filter to eliminate the base drift caused by the low-pass filter, and set it at the cut-off frequency;

④ Perform differential operations on the above signals so that the amplitudes of the positive and negative semi-axes are nearly equal;

⑤Squaring and windowing the signal to take the average operation, the squaring aims to change the signal with a negative amplitude into a signal with a positive amplitude, and the window to take the average is to smooth the squared signal, and its transfer function is :

In the formula, 32 is the number of points for windowing;

⑥ Locate the S1 peak. The heart sound signal has two obvious peaks, the S1 peak and the S2 peak. Since the S1 peak amplitude is higher, the S1 peak is used as the peak to be detected, and the location of the S1 peak is drawn as a graph;

⑦Threshold setting and judgment, the threshold setting is based on the currently detected peak attribute, using the adaptive iterative method, when the currently detected peak is positioned as peak(i), update T(i):

In the formula, N is the total number of peaks located;

⑧ According to the final determined S1 peak value, determine the cardiac cycle, draw it as an IBI (Inter-Beat Interval) file, and store the tester's cardiac cycle sequence.

Further, the certain number of points mentioned in the above step ① is preferably 15.

Further, the cut-off frequency in step ② is 150 Hz, and the cut-off frequency in step ③ is 630 Hz.

In order to ensure the collection effect, the heart sound collection lasts for 3-5 minutes.

The main indicators of mental stress mentioned above include time-domain indicators and frequency-domain indicators. Time-domain indicators include cardiac cycle, overall standard deviation, average heart rate, heart rate standard deviation, and continuous interval standard deviation; frequency-domain indicators include total energy TP, low-frequency components LF, High frequency component HF and low frequency/high frequency ratio, low frequency power to high frequency power ratio LF/HF, high frequency energy density value and low frequency energy density value.

According to the time-domain indicators and frequency-domain indicators, the user's mental stress state is analyzed from five aspects: the stability of the autonomic nervous system, the activity of the autonomic nervous system, the ability to withstand stress, the stress index, and the fatigue index, and a final analysis report is generated.

Compared with prior art, the beneficial effect of the present invention is:

① There is currently no algorithm for extracting heart rate variability from heart sound signals. This invention proposes an adaptive algorithm for extracting HRV from heart sounds. This algorithm has high calculation accuracy and good robustness, and the actual use effect meets the expected requirements.

②The results of the present invention are basically the same as those of professional testing equipment, and the cost is only one percent of that of similar foreign products. Compared with the electrocardiogram and mental stress analyzer, the present invention is convenient to use, simple to operate, complete in functions, especially low in cost, and is suitable for popularization and use in grassroots medical units, families, schools and individuals.

Description of drawings

Figure 1 is a block diagram of the system.

Fig. 2 is a flowchart of an adaptive algorithm for heart sound extraction HRV.

Figure 3 is the differentiated heart sound signal.

Fig. 4 is the averaged heart sound signal after square and windowing.

Figure 5 shows the location of the S1 peak.

Figure 6 shows the location of the S1 peak after thresholding.

Fig. 7 is a spectrum diagram of HRV energy.

detailed description

The specific embodiment of the present invention will be further described in detail in conjunction with the accompanying drawings.

A mental stress analysis system based on heart sound HRV theory proposed by the present invention is mainly composed of a heart sound signal acquisition module, a wireless communication module and a PC terminal system platform. The system block diagram is shown in Figure 1

The heart sound signal acquisition module adopts a shoulder-mounted wireless heart sound collector, which is connected to the PC terminal system platform. The heart sound fluctuation signal from the chest wall is directly transmitted to the transducer element through the pressure sensitive element by using the electric sensor, and then the collected heart sound signal is converted from analog to digital, and transmitted to the PC terminal system platform for processing. The signal acquisition time is 3 minutes.

The wireless communication module is connected with the PC terminal, and transmits data between the heart sound collection module and the PC terminal system platform in a wireless manner.

The PC-side system platform is connected with the wireless communication module to analyze the data input by the heart sound acquisition module, restore the heart sound signal, and extract HRV from the heart sound signal, and process the HRV with a variety of algorithms, simultaneously in the frequency domain and time domain. Perform analysis, obtain 11 main indicators, and graphically represent 8 of the main characteristic indicators, and analyze the user's performance from five aspects: the stability of the autonomic nervous system, the activity of the autonomic nervous system, the ability to resist stress, the stress index, and the fatigue index. Mental stress state, get the final analysis report.

The mental stress analysis method based on heart sound HRV theory is as follows:

(1) Analyze the data transmitted by the heart sound collector, restore the heart sound signal, filter and amplify the signal, and obtain the heart sound signal.

(2) Utilize the heart sound to extract the adaptive algorithm of HRV to obtain HRV, algorithm flowchart is shown in accompanying drawing 2 as shown in accompanying drawing 2, and algorithm comprises the following steps

①Resample the heart sound signal, take a point every 15 points, and reduce the amount of data while ensuring no aliasing, which is conducive to subsequent processing;

② Use Chebyshev type I low-pass filter to filter out low-frequency components, and set the cut-off frequency at 150Hz

③Use a high-pass filter to eliminate the base drift caused by the low-pass filter, and set the cut-off frequency at 630Hz

④ The differential operation is performed on the signal so that the amplitudes of the positive and negative semi-axes are almost equal, and the obtained signal after heart sound processing is shown in Figure 3.

⑤Squaring and windowing the signal to take the average operation. Squaring is designed to turn a negative-magnitude signal into a positive-magnitude signal. The purpose of windowing and averaging is to smooth the squared signal. The processed heart sound signal is as shown in Figure 4. Therefore, its transfer function is:

In the formula, 32 is the number of points for windowing;

⑥ Locate the S1 peak. There are two obvious peaks in the heart sound signal, namely the S1 peak and the S2 peak. Since the S1 peak amplitude is higher, the S1 peak is taken as the peak to be detected here. And the location of the S1 peak is plotted as shown in Fig. 5 .

⑦Threshold setting and judgment. The threshold setting adopts an adaptive iterative method, according to the currently detected peak properties. When the currently detected peak is positioned as peak(i), update T(i):

In the formula, N is the total number of peaks located, and the updated peak is located above the red line in Figure 6;

⑧ According to the final determined S1 peak value, determine the cardiac cycle, draw it as an IBI file, and store the tester's cardiac cycle sequence.

(3) Analyze HRV in frequency domain and time domain to obtain time domain index and frequency domain index.

①Analyze the change trend of heart rate variability by statistical methods, and obtain time-domain indicators, which can generally evaluate the regulation and control effect of autonomic nervous system on heart rate. Time-domain metrics include cardiac cycle, population standard deviation SDNN, mean heart rate, heart rate standard deviation, and serial interval standard deviation RMSSD.

The formula for calculating the average cardiac cycle is

The standard deviation SDNN calculation formula is

The formula for calculating the average heart rate is HR=60*1000/meanSS

The formula for calculating the standard deviation of heart rate is

The calculation formula of continuous interval standard deviation RMSSD is

Among them, N is the total number of heartbeats; SS _i is the ith S1 peak interval; meanSS is the average value of N S1 peak intervals

② The frequency domain analysis method is to use a classic power spectrum estimation theory - Welch periodogram method to obtain the frequency domain index. Including total energy TP, low frequency component LF, high frequency component HF and ultra-low frequency component VLF, ratio of low frequency power to high frequency power LF/HF, high frequency energy density value HF(n.u.) and low frequency energy density value LF(n.u.).

The modern spectral estimation of the autoregressive AR model is performed on the SS interval series of HRV, and the energy distribution of HRV is quantitatively described. The AR power spectral density curve is obtained through the Welch algorithm, as shown in Figure 7.

Total power TP: refers to the sum of each power component in all frequency ranges is the overall active state of the autonomic nervous system, indicating the ability of the autonomic nervous system to regulate the body. Its calculation expression is:

High-frequency power HF: mainly reflects the activity of the vagus nerve, which is related to respiratory movement, and generally increases excessively when breathing slowly or deeply. Its calculation expression is:

Low-frequency power LF: The efferent nerve of the autonomic nervous system is affected by both the sympathetic and vagus nerves, and mainly reflects the activity of the sympathetic nerve. Its calculation expression is:

Ultra-low frequency power VLF: It mainly represents sympathetic tension, which is closely related to the thermoregulatory system, and is also related to cardiopulmonary function related to vasomotion and hormones. Its calculation expression is:

The ratio of low-frequency power to high-frequency power LF/HF: Reflects the complexity of the autonomic nervous system sympathetic and vagus nerves

High-frequency energy density value HF(n.u.): represents the sympathetic nervous system of the autonomic nervous system, and its calculation expression is LF(n.u.)=LF/(LF+HF)×100%

Low-frequency energy density value HF(n.u.): represents the parasympathetic nervous system of the autonomic nervous system, and its calculation expression is HF(n.u.)=HF/(LF+HF)×100%

(4) Analyze the mental stress state of the testers. Including the following five indicators: the stability of the autonomic nervous system, the activity of the autonomic nervous system, the ability to resist stress, the stress index, and the fatigue index.

The stability of the autonomic nervous system is determined by the ratio of sympathetic and parasympathetic nerves. When the ratio of sympathetic and parasympathetic nerves is out of balance, the balance of the autonomic nervous system is temporarily disrupted, which will cause a series of mental problems.

The activity of the autonomic nervous system is determined by the low-frequency energy LF and high-frequency energy HF, which refers to the level of the tester's current body adjustment ability.

Stress resistance refers to the ability of the body to adapt to environmental changes or various pressure stimuli and maintain a stable internal environment. The significance of HRV reduction is that the complexity of heart rate dynamic variation is reduced, indicating that the body's ability to adapt to environmental changes is reduced, that is, the body's ability to resist stress is reduced.

The stress index is determined by the deviation of the cardiac cycle and heart rate, and refers to the mental stress state of the current tester.

The fatigue index is determined by the activity of the autonomic nervous system. The lower the activity of the autonomic nervous system, the heavier the pressure, the higher the fatigue index, which will lead to physical and mental fatigue.

(5) According to the mental stress state of the testers obtained from the analysis, the final analysis report is obtained. And has a printing function, the tester can print out the analysis report. Stress analysis reports and tester information, as well as IBI files containing the tester's cardiac cycle sequence, will be stored in the database for easy viewing at any time.

Claims (8)

1. a kind of stress analysis system theoretical based on heart sound HRV it is characterised in that：Described system is by heart sound signal acquisition Module, wireless communication module and PC end system platform composition；

Heart sound signal acquisition module is connected with PC end system platform, using electric-type sensor, the wall of the chest of user is spread out of the heart coming Sound fluctuation signal is directly delivered on inverting element by pressure cell, and then the cardiechema signals collecting are simulated with-number Word is changed, and transmits to PC end system platform, is processed；

Wireless communication module is connected with PC end system platform, and wirelessly in heart sound acquisition module and PC end system platform Between transmission data；

PC end system platform is used for analyzing the data of heart sound acquisition module input, reduces cardiechema signals, extracts HRV's using heart sound Adaptive algorithm obtains HRV, and HRV is processed using many algorithms, is analyzed in frequency domain and time domain simultaneously, obtains essence The leading indicator of refreshing pressure.

2. a kind of stress analysis system theoretical based on heart sound HRV according to claim 1 is it is characterised in that institute State the adaptive algorithm using heart sound extracts HRV and obtain HRV, comprise the steps of：

1. resampling is carried out to cardiechema signals, take a point every certain points, in the case of guarantee not aliasing, reduce Data volume, is conducive to follow-up process；

2. using Chebyshev I type low pass filter, filter low frequency component, cut-off frequency is set；

3. eliminated using high-pass filter and floated by the base that low pass filter brings, be arranged on cut-off frequency；

4. derivation operation is carried out to above-mentioned signal so that the amplitude of positive and negative semiaxis is almost equal；

5. signal is carried out square and adding window takes average operation, square be intended to the signal being negative by amplitude to be changed into amplitude being positive letter Number, adding window make even be intended to square after signal be smoothed, its transmission function is：

$H_{window}\left(Z\right)=\frac{1}{32}\·\frac{1-z^{-32}}{1-z^{-1}}$

In formula, 32 is the points of adding window；

6. position S1 crest, cardiechema signals have 2 obvious peak values to be respectively S1 peak value and S2 peak value, due to S1 peak amplitude more Height, using S1 peak as peak value to be detected, and the positioning drafting pattern by S1 peak value；

7. threshold value sets and judges, threshold value sets according to currently detected peak value attribute, using adaptive iteration method, current inspection When measuring peak value and orientating peak (i) as, update T (i):

$T\left(i\right)=\frac{\underset{i=1}{\overset{N}{\Σ}}peak\left(i\right)}{N}$

In formula, N is the peak value sum of positioning；

8. according to the final S1 peak value determining, determine cardiac cycle, be plotted as IBI file, deposit tester's cardiac cycle sequence.

3. a kind of stress analysis system theoretical based on heart sound HRV according to claim 2 it is characterised in that：Step Rapid 1. described in certain points be 15.

4. a kind of stress analysis system theoretical based on heart sound HRV according to claim 2 it is characterised in that：Step Rapid 2. in cut-off frequency be 150Hz.

5. a kind of stress analysis system theoretical based on heart sound HRV according to claim 2 it is characterised in that：Step Rapid 3. in cut-off frequency be 630Hz.

6. a kind of stress analysis system theoretical based on heart sound HRV according to claim 1 it is characterised in that：Adopt The time of collection heart sound continues 3-5 minute.

7. a kind of stress analysis system theoretical based on heart sound HRV according to claim 1 it is characterised in that：Institute The leading indicator stating stress comprises time domain index and frequency-domain index, and time domain index includes cardiac cycle, standard deviation in population, Average heart rate, heart rate criteria difference and consecutive intervals standard deviation；Frequency-domain index includes gross energy TP, low-frequency component LF, high frequency one-tenth Divide the ratio LF/HF of HF and low frequency/high frequency ratio, low frequency power and high frequency power, high-frequency energy density value and low frequency energy close Angle value.

8. a kind of stress analysis system theoretical based on heart sound HRV according to claim 7 it is characterised in that：Root According to described time domain index and frequency-domain index, from autonomic stability, autonomic activity, anti-pressure ability, Pressure index, 5 aspects of fatigue exponent analyze the stress state of user, generate final analysis report.