WO2021035924A1 - 一种心肺和谐系列指标评测方法、装置及系统 - Google Patents

一种心肺和谐系列指标评测方法、装置及系统 Download PDF

Info

Publication number
WO2021035924A1
WO2021035924A1 PCT/CN2019/113393 CN2019113393W WO2021035924A1 WO 2021035924 A1 WO2021035924 A1 WO 2021035924A1 CN 2019113393 W CN2019113393 W CN 2019113393W WO 2021035924 A1 WO2021035924 A1 WO 2021035924A1
Authority
WO
WIPO (PCT)
Prior art keywords
cardiopulmonary
signal
harmony
interval sequence
resonance
Prior art date
Application number
PCT/CN2019/113393
Other languages
English (en)
French (fr)
Inventor
吴健康
崔佳佳
Original Assignee
中科宁心电子科技(南京)有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 中科宁心电子科技(南京)有限公司 filed Critical 中科宁心电子科技(南京)有限公司
Publication of WO2021035924A1 publication Critical patent/WO2021035924A1/zh

Links

Images

Classifications

    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/02—Detecting, measuring or recording pulse, heart rate, blood pressure or blood flow; Combined pulse/heart-rate/blood pressure determination; Evaluating a cardiovascular condition not otherwise provided for, e.g. using combinations of techniques provided for in this group with electrocardiography or electroauscultation; Heart catheters for measuring blood pressure
    • A61B5/0205—Simultaneously evaluating both cardiovascular conditions and different types of body conditions, e.g. heart and respiratory condition
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A—HUMAN NECESSITIES
    • A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
    • A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
    • A61B5/00—Measuring for diagnostic purposes; Identification of persons
    • A61B5/72—Signal processing specially adapted for physiological signals or for diagnostic purposes
    • A61B5/7235—Details of waveform analysis
    • A61B5/7246—Details of waveform analysis using correlation, e.g. template matching or determination of similarity

Definitions

  • the present disclosure relates to the field of information processing technology, and in particular, to a method, device and system for evaluating a series of indicators of heart-lung harmony.
  • the human cardiopulmonary metabolism system provides nutrients to various organs and cells to ensure the normal operation of life bodies.
  • the internationally recognized respiratory heart rate modulation (Respiration Sinus Arrhythmia, RSA) is a physiological mechanism for the coordination of the heart and lungs to ensure efficient delivery of oxygen and nutrients, and recovery of metabolites.
  • RSA respiratory Sinus Arrhythmia
  • a large number of studies have proved that the level of RSA is directly related to the physical health and mental health of the human body and represents the activity level of the parasympathetic nerve. In this way, giving quantitative and accurate evaluation indicators of RSA has become a key technology and method for related scientific research and clinical applications.
  • the human cardiopulmonary metabolic system is regulated by the autonomic nervous system, and is affected by various factors such as external environment, internal infection, exercise, psychology, etc., making the research of RSA evaluation indicators face great difficulties.
  • One of the objectives of the present disclosure is to provide a cardiopulmonary harmony series index evaluation method, device and system, which can obtain a relatively accurate and complete digital index series of respiratory heart rate modulation (RSA) to realize the evaluation of cardiopulmonary function.
  • RSA respiratory heart rate modulation
  • the present disclosure provides a series of index evaluation methods for cardiopulmonary harmony, including:
  • testee Perform filtering and denoising preprocessing on the testee’s ECG signal, respiration signal and motion signal, determine the testee’s posture according to the motion signal, and determine whether the testee is in a resting state according to the posture, and obtain the State the ECG signal and breathing signal of the subject in a resting state;
  • the series of indicators of cardiopulmonary harmony include cardiopulmonary harmony curve, cardiopulmonary harmony degree, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary resonance factor, cardiopulmonary harmony curve bandwidth, and cardiopulmonary resonance quality Factor;
  • the cardiopulmonary harmony curve is used to characterize the influence of respiration on different frequency points in the frequency domain of the modulation of the RR interval sequence;
  • the cardiopulmonary harmony is the maximum influence value in the cardiopulmonary harmony curve;
  • the cardiopulmonary resonance Frequency is the resonance frequency corresponding to the maximum influence value;
  • the cardiopulmonary resonance factor is the mean value of the square of the influence value;
  • the cardiopulmonary resonance factor is the mean value of the square of the influence value;
  • the cardiopulmonary resonance quality factor It is the ratio of the cardiopulmonary resonance frequency to the bandwidth of the cardiopulmonary harmony curve.
  • the present disclosure proposes the concept of the cardiopulmonary harmony series indicators and provides a method for obtaining the cardiopulmonary harmony series indicators, which can provide a reliable reference basis for the evaluation of the cardiopulmonary metabolism system.
  • constructing a binary regression mathematical model according to the RR interval sequence and the respiratory signal includes:
  • a binary regression mathematical model is constructed according to the processed RR interval sequence and the breathing signal in a resting state.
  • a more accurate RR interval sequence can be obtained, thereby improving the accuracy of the index of the cardiopulmonary harmony series.
  • obtaining abnormal points in the RR interval sequence includes:
  • the value in the RR interval series does not satisfy the preset formula, the value is an abnormal point
  • the preset formula is: RRI i is the sequence value of the i-th RR interval, Is the average value of the RR interval sequence, RRI i-1 is the value of the i-1th RR interval sequence, Std(RRI) is the standard deviation of the RR interval sequence, and i is a positive integer.
  • the abnormal point can be accurately obtained through the above-mentioned preset formula.
  • obtaining abnormal points in the RR interval sequence, and replacing the abnormal points with a cubic spline interpolation method to obtain the processed RR interval sequence including:
  • the present disclosure makes the sampling rate of the RR interval sequence and the respiratory signal the same through resampling, so that the sequences of the two are synchronized.
  • constructing a binary regression mathematical model according to the RR interval sequence and the respiratory signal includes:
  • the RR interval sequence and the respiratory signal are respectively normalized, and the following binary regression mathematical model is constructed according to the normalized RR interval sequence and the normalized respiratory signal:
  • a 11,j , A 12,j , A 21,j and A 22,j are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p is a binary regression mathematical model
  • the present disclosure can accurately obtain a series of indicators of cardiopulmonary harmony by constructing a binary regression data model.
  • transforming the parameters in the binary regression mathematical model to the frequency domain to obtain a transformation matrix includes:
  • the parameters in the binary regression mathematical model are Fourier transformed to obtain a transformation matrix.
  • the present disclosure obtains parameter information in the frequency domain by performing Fourier transform.
  • transforming the parameters in the binary regression mathematical model to the frequency domain to obtain a transformation matrix includes:
  • X 1 (f) is the frequency domain representation of the RR interval sequence
  • X 2 (f) is the frequency domain representation of the respiratory signal
  • E 1 (f) and E 2 (f) are the Fourier transforms of the regression residuals
  • I is a plural unit.
  • analyzing the transformation matrix to obtain a cardiopulmonary harmony curve includes:
  • the Granger causality analysis method is used to analyze the transformation matrix to obtain the influence values of respiration on different frequency points of the RR interval sequence in the frequency domain, and obtain the cardiopulmonary harmony curve according to the influence values of different frequency points.
  • an accurate modulation of breathing to heart rate can be obtained through the Granger causality analysis method.
  • analyzing the transformation matrix to obtain a cardiopulmonary harmony curve includes:
  • the transformation matrix is analyzed by the following formula to obtain the influence value of respiration on different frequency points of the RR interval sequence in the frequency domain:
  • H(f) is the pair of matrix blocks The result of the inversion; S(f) is obtained by matrix transformation of X(f); G y ⁇ x (f) is the influence value of respiration on different frequency points of the RR interval sequence in the frequency domain; x represents the RR interval sequence, and y represents the respiratory signal;
  • the cardiopulmonary harmony curve is obtained according to the influence values of different frequency points.
  • a heart-lung harmony series indicator evaluation device including:
  • the preprocessing module is configured to perform filtering and denoising preprocessing on the subject’s ECG signal, breathing signal and motion signal, determine the subject’s posture according to the motion signal, and determine whether the subject is in Resting state, obtaining the ECG signal and breathing signal of the subject in the resting state;
  • a model construction module configured to extract an RR interval sequence from the ECG signal, and construct a binary regression mathematical model according to the RR interval sequence and the breathing signal in a resting state;
  • a transformation module configured to transform the parameters in the binary regression mathematical model to the frequency domain to obtain a transformation matrix
  • the index obtaining module is configured to analyze the transformation matrix to obtain cardiopulmonary harmony series indicators; wherein, the cardiopulmonary harmony series indicators include cardiopulmonary harmony curve, cardiopulmonary harmony degree, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary harmony curve bandwidth, and cardiopulmonary harmony Resonance quality factor; the cardiopulmonary harmony curve is used to characterize the influence of respiration on different frequency points in the frequency domain of the modulation of the RR interval sequence; the cardiopulmonary harmony is the maximum influence value in the cardiopulmonary harmony curve; The cardiopulmonary resonance frequency is the resonance frequency corresponding to the maximum influence value; the cardiopulmonary resonance factor is the mean value of the square of the influence value; the cardiopulmonary resonance quality factor is the difference between the cardiopulmonary resonance frequency and the bandwidth of the cardiopulmonary harmony curve ratio.
  • the cardiopulmonary harmony series indicators include cardiopulmonary harmony curve, cardiopulmonary harmony degree, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary harmony curve bandwidth, and cardiopulmonary harmony Resonance quality factor
  • the cardiopulmonary harmony curve is used to characterize the influence of respiration on different frequency points in the frequency domain of the modulation of the RR interval
  • model building module is specifically configured as:
  • a binary regression mathematical model is constructed according to the processed RR interval sequence and the respiratory signal.
  • model building module is specifically configured as:
  • the value in the RR interval series does not satisfy the preset formula, the value is an abnormal point
  • the preset formula is: RRI i is the sequence value of the i-th RR interval, Is the average value of the RR interval sequence, RRI i-1 is the value of the i-1th RR interval sequence, Std(RRI) is the standard deviation of the RR interval sequence, and i is a positive integer.
  • model building module is specifically configured as:
  • model building module is also configured to:
  • the RR interval sequence and the respiratory signal are respectively normalized, and the following binary regression mathematical model is constructed according to the normalized RR interval sequence and the normalized respiratory signal:
  • a 11,j , A 12,j , A 21,j and A 22,j are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p is a binary regression mathematical model
  • the transformation module is specifically configured as:
  • the parameters in the binary regression mathematical model are Fourier transformed to obtain a transformation matrix.
  • the transformation module is specifically configured as:
  • X 1 (f) is the frequency domain representation of the RR interval sequence
  • X 2 (f) is the frequency domain representation of the respiratory signal
  • E 1 (f) and E 2 (f) are the Fourier transforms of the regression residuals
  • I is a plural unit.
  • the indicator obtaining module is specifically configured as:
  • the Granger causality analysis method is used to analyze the transformation matrix to obtain the influence values of respiration on different frequency points of the RR interval sequence in the frequency domain, and obtain the cardiopulmonary harmony curve according to the influence values of different frequency points.
  • the indicator obtaining module is specifically configured as:
  • the transformation matrix is analyzed by the following formula to obtain the influence value of respiration on different frequency points of the RR interval sequence in the frequency domain:
  • H(f) is the pair of matrix blocks The result of the inversion;
  • S(f) is obtained by matrix transformation of X(f);
  • Gy ⁇ x(f) is the influence value of respiration on different frequency points of the RR interval sequence in the frequency domain;
  • x represents the RR interval sequence, and
  • y represents the respiratory signal;
  • the cardiopulmonary harmony curve is obtained according to the influence values of different frequency points.
  • the present disclosure provides a cardiopulmonary harmony series index evaluation system, including: a wearable electronic device, the cardiopulmonary harmony series index evaluation device described in the second aspect, and a user and data management subsystem;
  • the wearable electronic device is configured to collect the ECG signal, breathing signal and exercise signal of the subject, and send the ECG signal, the breathing signal and the exercise signal to the cardiopulmonary harmony series index evaluation device via Bluetooth. signal;
  • the cardiopulmonary harmony series index evaluation device runs on a smart phone or a portable computing device PAD, receives the subject’s ECG signal, respiration signal, and exercise signal from the wearable electronic device via Bluetooth, and uses it according to the test subject’s
  • the patient’s ECG signal, breathing signal and exercise signal are analyzed and processed to obtain a series of indicators of cardiopulmonary harmony, and upload the obtained series of indicators of cardiopulmonary harmony to the user and the data management subsystem.
  • the wearable electronic device is a miniature electronic device worn on the chest, and the wearable electronic device includes:
  • ECG electrodes configured to collect single-lead ECG signals
  • Respiration sensor configured to collect respiratory signals according to the principle of measuring chest impedance or chest movement caused by respiration
  • a motion sensor configured to measure the three-dimensional acceleration of the subject's torso relative to the vertical angle
  • the synchronization acquisition unit is configured to amplify and AD convert the ECG signal and respiration signal, receive controller instructions, and synchronously collect the ECG, respiration and motion signals according to the controller instruction, pack them into data packets, and send them to the Bluetooth transmission unit;
  • the Bluetooth transmission unit sends the data packets received from the synchronization acquisition unit to the processing and analysis subsystem.
  • the user and data management subsystem includes a central database and a doctor terminal that are communicatively connected;
  • the central database receives testee data from a cardiopulmonary harmony series indicator evaluation device, wherein the testee data includes the testee's cardiopulmonary harmony series indicators;
  • the central database receives operating instructions from the doctor's terminal, and generates a detection report according to the cardiopulmonary harmony series indicators and the operating instructions.
  • the present disclosure provides an electronic device, including: a processor, a memory, and a bus, where:
  • the processor and the memory complete communication with each other through the bus;
  • the memory stores program instructions executable by the processor, and the processor invokes the program instructions to execute the method steps of the first aspect.
  • the present disclosure provides a non-transitory computer-readable storage medium, including:
  • the non-transitory computer-readable storage medium stores computer instructions that cause the computer to execute the method steps of the first aspect.
  • FIG. 1 is a schematic diagram of a method for evaluating a series of indicators of cardiopulmonary harmony provided by the present disclosure
  • Figure 2 is a schematic flow chart of another method for evaluating a series of indicators of cardiopulmonary harmony provided by the present disclosure
  • FIG. 3 is a schematic diagram of the ECG signal and the respiratory signal provided by the present disclosure
  • Figure 4 is a schematic diagram of a series of indicators for evaluating cardiopulmonary harmony provided by the present disclosure
  • FIG. 5 is a schematic diagram of the cardiopulmonary harmony curve corresponding to the testee provided by the present disclosure
  • Fig. 6 is a schematic structural diagram of a heart-lung harmony series index evaluation device provided by the present disclosure
  • FIG. 7 is a schematic diagram of the structure of a heart-lung harmony series index evaluation system provided by the present disclosure.
  • Respiration modulates the heart rate (respiratory sinus arrhythmia, RSA): when the human body does inhalation, the thorax expands. This activity excites the lung stretch receptors in the smooth muscle layer of the bronchus and bronchioles, sends impulses and impulses Passed to the respiratory center.
  • the respiratory center integrates the incoming information, inhibits the activity of the vagus nerve, and enhances the activity of the sympathetic nerves, thereby increasing the heart rate and increasing blood pressure. After the blood pressure rises, the arterial baroreceptors located under the adventitia of the carotid sinus and the aortic arch are excited, emit impulses and transfer them to the respiratory center.
  • the respiratory center integrates the information of the incoming impulses again to enhance the activity of the vagus nerve and inhibit
  • the activity of the sympathetic nerve slows down the heart rate and lowers blood pressure.
  • the increase in blood pressure causes the blood volume of the atria and ventricles to increase, which excites the cardiopulmonary receptors located in the walls of the atria, ventricles, and pulmonary circulation.
  • they further reduce sympathetic nerve activity and enhance vagus nerve activity.
  • the heart rate slows down and blood pressure drops.
  • a slower heart rate causes a decrease in the partial pressure of oxygen in the blood, and the carotid body and aortic body chemoreceptors feel changes, emit impulses and pass into the respiratory center, which speeds up breathing and increases heart rate.
  • the RSA metric is expressed in the following manner:
  • the RMSSD root mean square value of the ECG in the expiratory and inspiratory intervals is used to represent the RSA measurement index. This method does not consider the physiological mechanism of RSA at all, and of course it is not effective.
  • CPC CardioPulmonary Coupling
  • the degree of cross-correlation between heart rate sequence and respiratory signal in the transform domain is used to express the degree of cardiopulmonary interaction and harmony.
  • the frequency domain power of NN sequence and respiratory signal is used.
  • the similarity of the spectrum indicates the degree of cardiopulmonary interaction and harmony.
  • the heart rate variation part is similar to the power spectrum of the respiratory signal, but is not caused by respiratory modulation at all. Therefore, this representation is inaccurate and may be wrong.
  • CPC is generally considered to better characterize the state of autonomic nervous regulation in different sleep periods, but our experiments on the MIT standard database have proved that its effect is very limited.
  • this embodiment proposes a new evaluation digital indicator series for evaluating the human cardiopulmonary metabolic system, Cardiopulmonary Resonance Indicators (CRI).
  • CRI Cardiopulmonary Resonance Indicators
  • Cardiopulmonary harmony series indicators include cardiopulmonary harmony curve G(f), cardiopulmonary resonance Amplitude (CRA), cardiopulmonary resonance frequency, cardiopulmonary resonance factor (CRF), cardiopulmonary harmony curve bandwidth, and cardiopulmonary resonance quality factor (Cardiopulmonary) Resonance Qualityfactor, CRQ).
  • the cardiopulmonary harmony curve G(f) is used to characterize the influence of respiration on the modulation of the RR interval sequence at different frequency points in the frequency domain;
  • the cardiopulmonary harmony is the maximum influence value in the cardiopulmonary harmony curve;
  • the cardiopulmonary resonance frequency is the maximum The resonance frequency corresponding to the influence value;
  • the cardiopulmonary resonance factor is the mean value of the square of the influence value of different frequency points;
  • the cardiopulmonary resonance quality factor is the ratio of the cardiopulmonary resonance frequency to the bandwidth of the cardiopulmonary harmony curve.
  • the cardiopulmonary harmony series indicators proposed in this embodiment can completely describe the personalized RSA characteristics of the human body.
  • the subject In the resting state, the subject’s physical and psychological state has an impact on breathing and heart rate, which is further manifested in the resonance quality factor of the cardiopulmonary harmony curve G(f) and the cardiopulmonary harmony: the cardiopulmonary function system is good (physical and psychological Good condition), the breathing is stable, the cardiopulmonary resonance factor is high, and the cardiopulmonary harmony is great; on the other hand, G(f) also changes with the breathing rate: if the breathing rate is from the usual 20 breaths per minute (about 0.3 Hz) to deep breathing When the method drops to 6 times per minute (about 0.1 Hz), the cardiopulmonary harmony increases, and the maximum respiratory rate is about 0.1 Hz.
  • everyone has a maximum resonant frequency, and everyone’s maximum resonant frequency is different, but they are all around 0.1 Hz.
  • the main body that executes the method can be an electronic device, which can be a desktop computer, a tablet computer, a server, and other devices with data processing functions.
  • This method Including the following processing:
  • Processing 101 Perform filtering and denoising preprocessing on the subject’s ECG signal, breathing signal and motion signal, determine the subject’s posture based on the preprocessed motion signal, and determine whether the subject is based on the posture In the resting state, the ECG signal and the breathing signal of the subject in the resting state are obtained.
  • the electronic device receives or collects the ECG signal, respiration signal and motion signal of the subject, and performs filtering and denoising preprocessing on the ECG signal, respiration information and motion signal.
  • the ECG signal is a comprehensive reflection of the electrical activities of countless cardiomyocytes of the heart. It records the depolarization and repolarization process of heart cells from a macroscopic perspective, and to a certain extent objectively reflects the physiological conditions of various parts of the heart.
  • the respiration signal is a time series used to characterize the breath of the subject.
  • the motion signal is used to characterize the current state of the subject, and the current state of the subject includes the resting state and the motion state.
  • the motion signal may be a measured three-dimensional acceleration signal of the subject.
  • the resting state includes sleeping state, sitting state and bed rest state, and the motion state includes walking, running, etc.
  • the motion signal can determine the posture of the examinee, and then determine whether the examinee is in a resting state according to the posture. It should be noted that the above-mentioned method for determining the resting state of the subject is only an example. In addition to determining the motion signal, it may also be determined by other methods, which is not specifically limited in the embodiment of the present application.
  • the electronic device when it determines that the subject is in a resting state, it can obtain the heartbeat of the subject in a resting state for a preset period of time from the pre-processed ECG signal and respiratory signal, respectively.
  • the preset time period can be set to, for example, 3 minutes, 5 minutes, or other time lengths. Taking the preset time period of 3 minutes as an example, when the electronic device determines that the subject is at rest, it can obtain 3 minutes of heartbeat from the collected ECG and breathing signals of the subject at rest. Electrical signals and breathing signals. It should be noted that the acquired 3-minute ECG signal and respiration signal should be in the same time period.
  • the exercise signal it is known that the subject is in a resting state between 1:30 PM and 3:00 PM on a certain day. Therefore, the heart during the time period of 1:40 PM-1:43 PM on that day can be obtained. Electrical signals and breathing signals.
  • Processing 102 extracting the RR interval sequence from the ECG signal in the resting state, and constructing a binary regression mathematical model according to the RR interval sequence and the breathing signal in the resting state.
  • the ECG signal includes a variety of complexes, such as: P wave, QRS complex, PR interval, T wave, RR interval, etc.
  • the electronic device can extract the RR interval sequence from the ECG signal of the subject in the resting state, and construct a binary regression mathematical model based on the RR interval sequence and the resting state of the respiratory signal.
  • Processing 103 Use a binary regression model to describe the modulation of heart rate by breathing.
  • the parameters in the binary regression mathematical model are transformed into the frequency domain to obtain a transformation matrix.
  • the constructed binary regression data model is also in the time domain.
  • the Fourier transform can be used Transform the parameters in the binary regression mathematical model to the frequency domain.
  • Processing 104 Analyze the transformation matrix to obtain a series of indicators of cardiopulmonary harmony.
  • Granger causality analysis can be used to analyze the transformation matrix to obtain the influence value of the modulation of the RR interval sequence in the preset time period on the different frequency points in the frequency domain.
  • the influence value of can obtain the cardiopulmonary harmony curve, the cardiopulmonary harmony degree, the cardiopulmonary resonance frequency, the cardiopulmonary resonance factor, the cardiopulmonary harmony curve bandwidth and the cardiopulmonary resonance quality factor, namely the cardiopulmonary harmony series indicators.
  • a binary regression mathematical model is constructed based on the ECG signal and the respiratory signal to obtain a series of indicators of the cardiopulmonary harmony.
  • the series of indicators of the cardiopulmonary harmony start from the resonance state of the cardiopulmonary system and calculate the relationship to the heart rate through the respiratory signal and the ECG signal.
  • the CRI proposed in this embodiment starts from the mathematical modeling of the physiological process of RSA, and accurately describes the physiological phenomenon of respiratory modulation of the heart rate. Therefore, it can be further used to describe the work efficiency and health status of the cardiopulmonary metabolic system and to evaluate the efficiency of the cardiopulmonary metabolic system. Provide accurate parameter data.
  • the accuracy of staging the deep sleep period of CRI proposed in this embodiment is about 11% higher than that of CPC.
  • Figure 2 is a schematic flow chart of another method for evaluating a series of indicators of cardiopulmonary harmony provided by the present disclosure. As shown in Figure 2, the method includes the following processing:
  • Processing 201 monitor the subject with an electrocardiogram and respiration monitor.
  • an ECG and respiration monitor with both ECG signal acquisition and respiratory signal acquisition can be selected, and the ECG and respiration monitor can be set on the subject. . In this way, the obtained ECG signal and respiration signal are synchronized in time.
  • Processing 202 Obtain an ECG signal and a breathing signal.
  • the electronic device can obtain the ECG signal, the respiration signal, and the motion signal from the ECG and respiration monitor.
  • FIG. 3 exemplarily shows a schematic diagram of the ECG signal and the respiration signal provided by an embodiment of the present application.
  • the electronic device can learn the state information of the subject according to the motion signal.
  • the state information of the subject indicates that the subject is in a resting state (such as sleeping, deep breathing, etc.)
  • the sliding window is used to obtain the state information.
  • the unit length of the sliding window may be the aforementioned preset time period (for example, 3-5 minutes).
  • Process 203 Extract the RR interval sequence.
  • the electronic device extracts the RR interval sequence from the ECG signal.
  • the electronic device obtains the abnormal point from the RR interval sequence, where the abnormal point can be determined by the following formula:
  • RRI i is the sequence value of the i-th RR interval, Is the average value of the RR interval series, RRI i-1 is the value of the i-1th RR interval series, and Std(RRI) is the standard deviation of the RR interval series. Therefore, if the value of the i-th RR interval sequence does not satisfy the above formula, it means that the i-th RR interval sequence is an abnormal point, and i is a positive integer.
  • the electronic device After acquiring the abnormal points, the electronic device removes the abnormal points and performs cubic spline interpolation processing on the removed abnormal points to replace the removed abnormal points with corresponding interpolation values to obtain the processed RR interval sequence.
  • the processed RR interval sequence has basically the same sampling rate as the respiratory signal.
  • Process 205 Normalize the RR interval sequence and the respiratory signal.
  • the electronic device resamples the processed RR interval sequence to obtain a processed RR interval sequence that is basically the same as the sampling rate of the respiratory signal, and normalizes the processed RR interval sequence and the respiratory signal to make the processed RR interval sequence
  • the mean value of the RR interval sequence and the respiratory signal are both 0, and the variance is both 1.
  • Processing 206 Construct a binary regression mathematical model to characterize the modulation of heart rate by breathing.
  • the electronic device can construct the following binary regression mathematical model according to the normalized RR interval sequence and the normalized respiratory signal:
  • a 11,j , A 12,j , A 21,j and A 22,j are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p is a binary regression mathematical model
  • the parameter p is selected based on the principle of maximizing duration and minimizing residuals, that is, calculating the Akaike Information Criterion (AIC) and Bayes Information Criterion corresponding to different p values (Bayesian Information Criterion, BIC) coefficient, when the AIC coefficient and the BIC coefficient take the minimum value, the corresponding p value is the delay p of the binary regression mathematical model.
  • AIC Akaike Information Criterion
  • BIC Bayes Information Criterion corresponding to different p values
  • T is the selection duration of the calculated regression
  • n is the calculated regression sample size.
  • X 1 (f) is the frequency domain representation of the RR interval sequence
  • X 2 (f) is the frequency domain representation of the respiratory signal
  • E 1 (f) and E 2 (f) are the Fourier transforms of the regression residuals
  • I is a plural unit.
  • Processing 208 Analyze the transformed matrix.
  • the electronic device can use the Granger causality analysis method to calculate the influence value G of the respiration on the change of the RR interval sequence in the frequency domain at different frequency points in the preset time period.
  • the calculation of the influence value G is derived from Granger causality analysis theory, through Measure the reduction of binary regression residuals relative to unary regression residuals, and characterize and measure the influence relationship between two sequences.
  • S(f) is obtained by matrix transformation of X(f);
  • G y ⁇ x (f) is the influence value of respiration on different frequency points of the RR interval sequence in the frequency domain;
  • x represents the RR interval sequence, and
  • y represents the respiratory signal.
  • the cardiopulmonary harmony curve G(f) can be obtained, and further, the cardiopulmonary harmony curve G(f) can be obtained cardiopulmonary harmony CRA, cardiopulmonary resonance frequency, cardiopulmonary resonance factor CRF, cardiopulmonary harmony curve bandwidth and cardiopulmonary resonance Quality factor CRQ.
  • Fig. 4 is a schematic diagram of a series of indicators for evaluating cardiopulmonary harmony provided by the present disclosure.
  • the abscissa represents the frequency
  • the ordinate represents the influence value at different frequency points.
  • CRA maxG(f)
  • the frequency corresponding to CRA is the cardiopulmonary resonance frequency f A
  • the cardiopulmonary resonance factor CRF mean(G2(f)), which is a measure of the resonance energy of the cardiopulmonary system
  • the cardiopulmonary resonance quality factor CRQ is the cardiopulmonary system
  • the measurement of the resonance characteristic is the ratio of the resonance frequency f A to the bandwidth of the cardiopulmonary harmony curve ⁇ f when the value of the harmony curve G(f) drops to 0.707, namely:
  • the cardiopulmonary harmony represents the maximum depth of the coupling resonance of the cardiopulmonary system in the current measurement state, and reflects the degree of the testee's cardiopulmonary coupling. Among them, the greater the cardiopulmonary harmony, the better the current harmony state of the subject.
  • the cardiopulmonary resonance factor expresses the coupling state of the heart and lungs at the energy level. The higher the CRF, the greater the modulation of the heart by the respiratory system, and the better the resonance state of the heart and lungs.
  • the cardiopulmonary resonance quality factor CRQ is high, and the cardiopulmonary harmony curve bandwidth ⁇ f is narrow, indicating that the cardiopulmonary metabolism system is highly efficient.
  • Processing 209 slide the sliding window, and calculate the CRI in real time.
  • the unit length of a sliding window may be the aforementioned preset time period. After sliding the sliding window, the ECG signal and the breathing signal in the next sliding window (ie, the sliding window after sliding) are acquired, And according to processing 201-processing 208, the ECG signal and respiration signal in the next sliding window are analyzed to realize long-term monitoring of the subject.
  • the present disclosure obtains the digital measurement of the modulation of the heart rate by breathing by acquiring the ECG signal and the respiratory signal of the test subject, using the binary regression mathematical model and Granger causality analysis to obtain the digital measure of the heart rate modulation of the heart, that is, the cardiopulmonary harmony series indicators.
  • the cardiopulmonary harmony series indicators can be used as a pair A series of more accurate evaluation digital indicators for the evaluation of the cardiopulmonary metabolism system.
  • FIG. 5 is a schematic diagram of the cardiopulmonary harmony curve corresponding to the testee provided by the present disclosure, which shows the influence of the modulation of the heart rate by the testee on the time domain and the frequency domain under different respiratory frequencies and different physiological states.
  • the value, from top to bottom in Figure 5, is the transition from the resting state to the adjusted state that can reduce the breathing rate.
  • Fig. 6 is a schematic structural diagram of a heart-lung harmony series index evaluation device provided by the present disclosure.
  • the device can be a module, program segment or code on an electronic device. It should be understood that this device corresponds to the above-mentioned method embodiment of FIG. 1 and can perform various steps involved in the method embodiment of FIG. 1. For specific functions of the device, refer to the above description. To avoid repetition, detailed descriptions are appropriately omitted here.
  • the device can include a preprocessing module 601, a model construction module 602, a transformation module 603, and an index obtaining module 604, where:
  • the preprocessing module 601 is configured to perform filtering and denoising preprocessing on the subject’s ECG signal, respiration signal and motion signal, determine the subject’s posture based on the preprocessed motion signal, and determine the subject’s posture according to the posture. Whether the examinee is in a resting state, obtain the ECG signal and breathing signal of the examinee in the resting state.
  • the ECG signal, respiration signal and motion signal of the testee are received, and the ECG signal, respiration information and motion signal are filtered and denoised preprocessing.
  • the ECG signal is a comprehensive reflection of the electrical activities of countless cardiomyocytes of the heart. It records the depolarization and repolarization process of heart cells from a macroscopic perspective, and to a certain extent objectively reflects the physiological conditions of various parts of the heart.
  • the breathing signal is used to characterize the time series of the subject's breathing.
  • the motion signal is used to characterize the current state of the subject, for example: the subject is in a resting state, a sleeping state, a deep breathing state, a slow breathing modulation state, or a motion state, etc.
  • the motion signal may be a measured three-dimensional acceleration signal of the subject.
  • the resting state can include sleeping state, sitting state, bedridden state, etc.
  • the posture of the subject can be determined by the motion signal, and whether the subject is in the resting state can be determined according to the posture. If so, from the pre-processed ECG
  • the ECG signal of the examinee in the resting state is obtained from the signal, and the breathing signal of the examinee in the resting state is obtained from the pre-processed breathing signal.
  • the model construction module 602 is configured to extract an RR interval sequence from the ECG signal in the resting state, and construct a binary regression mathematical model according to the RR interval sequence and the breathing signal in the resting state.
  • the ECG signal includes a variety of complexes, such as: P wave, QRS complex, PR interval, T wave, RR interval, etc. Therefore, the model construction module 602 can extract the RR interval sequence from the ECG signal, and construct a binary regression mathematical model based on the RR interval sequence and the respiratory signal. It should be noted that the respiratory signal is the respiratory time sequence of the subject in a preset time period.
  • the transformation module 603 is configured to transform the parameters in the binary regression mathematical model to the frequency domain to obtain a transformation matrix.
  • both the RR interval sequence and the respiratory signal are signals in the time domain. Therefore, the constructed binary regression data model is also in the time domain.
  • the transformation module 603 can pass Fourier Leaf transform transforms the parameters in the binary regression mathematical model to the frequency domain.
  • the index obtaining module 604 is configured to analyze the transformation matrix to obtain a series of indicators of cardiopulmonary harmony; wherein, the series of indicators of cardiopulmonary harmony include cardiopulmonary harmony curve, cardiopulmonary harmony degree, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary harmony curve bandwidth, and cardiopulmonary harmony Resonance quality factor; the cardiopulmonary harmony curve is used to characterize the influence of respiration on different frequency points in the frequency domain of the modulation of the RR interval sequence; the cardiopulmonary harmony is the maximum influence value in the cardiopulmonary harmony curve; The cardiopulmonary resonance frequency is the resonance frequency corresponding to the maximum influence value; the cardiopulmonary resonance factor is the mean value of the square of the influence value; the cardiopulmonary resonance quality factor is the difference between the cardiopulmonary resonance frequency and the bandwidth of the cardiopulmonary harmony curve ratio.
  • the index obtaining module 604 can use Granger causality analysis to analyze the transformation matrix, and obtain the influence value of the modulation of the RR interval sequence on the frequency domain by the respiration in the preset time period.
  • the cardiopulmonary harmony curve, cardiopulmonary harmony degree, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary harmony curve bandwidth and cardiopulmonary resonance quality factor can be obtained through the influence value of each frequency point, that is, the cardiopulmonary harmony series indicators.
  • the present disclosure constructs a binary regression mathematical model based on the ECG signal and the respiratory signal, and then obtains a series of indicators of the cardiopulmonary harmony.
  • the series of indicators of the cardiopulmonary harmony starts from the resonance state of the cardiopulmonary system and calculates the modulation of the heart rate through the respiratory signal and the ECG signal.
  • the measurement parameters of intensity and effect provide accurate parameter data for evaluating the efficiency of the cardiopulmonary metabolic system.
  • model construction module 602 may be specifically configured as:
  • model construction module 602 may be specifically configured as:
  • the preset formula is: RRI i is the sequence value of the i-th RR interval, Is the average value of the RR interval sequence, RRI i-1 is the value of the i-1th RR interval sequence, Std(RRI) is the standard deviation of the RR interval sequence, and i is a positive integer.
  • model construction module 602 may be specifically configured as:
  • model construction module 602 may also be configured as:
  • the RR interval sequence and the respiratory signal are respectively normalized, and the following binary regression mathematical model is constructed according to the normalized RR interval sequence and the normalized respiratory signal:
  • a 11,j , A 12,j , A 21,j and A 22,j are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p are regression coefficients of a binary regression mathematical model
  • X 1 is the RR interval sequence
  • X 2 is the respiratory signal
  • ⁇ 1 (t) and ⁇ 2 (t) are both regression residuals
  • p represents the length of the selected regression sequence
  • t represents the time at which the RR interval sequence or the respiratory signal is located
  • j is a positive integer
  • 1 ⁇ j ⁇ p is a binary regression mathematical model
  • the transformation module 603 may be specifically configured as:
  • the transformation module may be specifically configured as:
  • X 1 (f) is the frequency domain representation of the RR interval sequence
  • X 2 (f) is the frequency domain representation of the respiratory signal
  • E 1 (f) and E 2 (f) are the Fourier transforms of the regression residuals
  • I is a plural unit.
  • the index obtaining module 604 may be specifically configured as:
  • the Granger causality analysis method is used to analyze the transformation matrix to obtain the influence values of respiration on different frequency points of the RR interval sequence in the frequency domain, and obtain the cardiopulmonary harmony curve according to the influence values of different frequency points.
  • the index obtaining module 604 may be specifically configured as:
  • the transformation matrix is analyzed by the following formula to obtain the influence value of respiration on different frequency points of the RR interval sequence in the frequency domain:
  • H(f) is the pair of matrix blocks The result of the inversion; S(f) is obtained by matrix transformation of X(f); G y ⁇ x (f) is the influence value of respiration on different frequency points of the RR interval sequence in the frequency domain; x represents the RR interval sequence, and y represents the respiratory signal;
  • the cardiopulmonary harmony curve is obtained according to the influence values of different frequency points.
  • Fig. 7 is a series of index evaluation system for cardiopulmonary harmony provided by the present disclosure.
  • the system may include: wearable electronic device 701, cardiopulmonary harmony series index evaluation device 702, and user and data management subsystem 703, in which:
  • the wearable electronic device 701 is configured to collect the ECG signal, respiration signal, and motion signal of the subject, and send the ECG signal, the respiration signal, and all the signals to the cardiopulmonary harmony series indicator evaluation device 702 via Bluetooth.
  • the motion signal ;
  • the cardiopulmonary harmony series indicator evaluation device 702 runs on a smart phone or a portable computing device PAD, receives the subject’s ECG signal, respiration signal, and exercise signal from the wearable electronic device 701 via Bluetooth, and performs according to the The subject’s ECG signal, breathing signal and exercise signal are analyzed and processed to obtain a series of indicators of cardiopulmonary harmony, and upload the obtained series of indicators of cardiopulmonary harmony to the user and data management subsystem 703.
  • the wearable electronic device 701 may be a miniature electronic device worn on the chest, and the wearable electronic device may include an ECG electrode 7011, a respiration sensor 7012, a motion sensor 7013, and a synchronous collection Unit 7014 and Bluetooth transmission unit 7015.
  • the ECG electrode 7011, the respiration sensor 7012, and the motion sensor 7013 are respectively connected to the synchronization collection unit 7014, and the synchronization collection unit 7014 is connected to the Bluetooth transmission unit 7015.
  • the ECG electrode 7011 is configured to collect a single-lead ECG signal.
  • the wearable electronic device 7015 may include two ECG electrodes 7011. In use, the two ECG electrodes 7011 may be attached to the standard ECG lead position on the chest of the subject.
  • the respiration sensor 7012 is configured to collect respiration signals according to the principle of measuring chest impedance or chest movement caused by respiration.
  • the motion sensor 7013 is configured to measure the three-dimensional acceleration of the subject’s torso relative to the vertical angle; when collecting the motion signal of the subject, the motion sensor can be worn on the subject’s chest, so that the subject can be measured The three-dimensional acceleration signal of the angle of the torso relative to the vertical direction.
  • the synchronization acquisition unit 7014 is configured to amplify and AD convert ECG signals and respiratory signals, receive controller instructions, and synchronously collect ECG signals, respiratory signals, and exercise signals according to the controller instructions, and convert the synchronized ECG signals and respiratory signals. Pack it with the motion signal into a data packet and send it to the Bluetooth transmission unit;
  • the Bluetooth transmission unit 7015 is configured to send the data packets received from the synchronization acquisition unit 7014 to the processing and analysis subsystem.
  • the wearable electronic device 701 may further include a controller 7016, and the synchronization acquisition unit 7014 and the Bluetooth transmission unit 7015 are respectively connected to the controller 7016.
  • the controller 7016 may be configured to control the synchronization acquisition unit 7014 and the Bluetooth transmission unit 7015.
  • the user and data management subsystem 703 may include a central database 7031 and a doctor terminal 7032 that are communicatively connected to each other.
  • the central database 7031 can receive testee data from the cardiopulmonary harmony series index evaluation device 702, wherein the testee data may include the testee's cardiopulmonary harmony series indicators.
  • the central database 7031 receives the operation instructions of the doctor terminal 7032, and generates a detection report according to the cardiopulmonary harmony series indicators and the operation instructions.
  • the present disclosure also provides an electronic device, including: a processor, a memory, and a bus; wherein the processor and the memory communicate with each other through the bus.
  • the processor is configured to call the program instructions in the memory to execute the methods provided in the foregoing embodiments, for example, including: acquiring the ECG signal, respiration signal, and motion signal of the subject, and according to the motion
  • the signal determines the state information of the subject; obtains the ECG signal and the breathing signal in the resting state from the state information of the subject; Extracts the RR interval sequence from the ECG signal, and according to The RR interval sequence and the respiratory signal construct a binary regression mathematical model; transform the parameters in the binary regression mathematical model to the frequency domain to obtain a transformation matrix; analyze the transformation matrix to obtain a series of indicators of cardiopulmonary harmony
  • the cardiopulmonary harmony series indicators include cardiopulmonary harmony curve, cardiopulmonary harmony degree, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary harmony curve bandwidth and cardiopulmonary resonance quality factor;
  • the cardiopulmonary harmony curve is used to characterize the sequence of respiratory versus RR intervals
  • the modulation of the influence values of different frequency points in the frequency domain; the cardiopulmonary harmony is the maximum influence value in the cardiopulmonary harmony curve;
  • the embodiment also discloses a computer program product.
  • the computer program product includes a computer program stored on a non-transitory computer-readable storage medium.
  • the computer program includes program instructions. When the program instructions are executed by a computer, The computer can implement the methods provided in the foregoing embodiments.
  • the method includes, for example, acquiring the ECG signal, respiration signal, and motion signal of the subject, and determining the state information of the subject according to the motion signal; Obtain the ECG signal and the breathing signal in the resting state from the state information of the subject; extract the RR interval sequence from the ECG signal, and construct it based on the RR interval sequence and the breathing signal Binary regression mathematical model; transform the parameters in the binary regression mathematical model to the frequency domain to obtain a transformation matrix; analyze the transformation matrix to obtain a series of indicators of cardiopulmonary harmony; wherein, the series of indicators of cardiopulmonary harmony include cardiopulmonary harmony Curve, cardiopulmonary harmony, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary harmony curve bandwidth and cardiopulmonary resonance quality factor; the cardiopulmonary harmony curve is used to characterize the influence of respiration on the modulation of the RR interval sequence at different frequency points in the frequency domain
  • the cardiopulmonary harmony is the maximum influence value in the cardiopulmonary harmony curve; the cardiopulmonary resonance frequency is the resonance frequency corresponding to the maximum influence value; the cardiopulmonary resonance factor is the mean value of the
  • This embodiment also provides a non-transitory computer-readable storage medium, the non-transitory computer-readable storage medium storing computer instructions that cause the computer to execute the methods provided in the foregoing method embodiments, such as Including: acquiring the ECG signal, respiration signal and motion signal of the subject, and determining the state information of the subject according to the motion signal; obtaining the state information of the subject in a resting state Extract the RR interval sequence from the ECG signal, and construct a binary regression mathematical model based on the RR interval sequence and the respiratory signal; combine the binary regression mathematical model Transform the parameters of to the frequency domain to obtain a transformation matrix; analyze the transformation matrix to obtain a series of indicators of cardiopulmonary harmony; wherein, the series of indicators of cardiopulmonary harmony include cardiopulmonary harmony curve, cardiopulmonary harmony degree, cardiopulmonary resonance frequency, cardiopulmonary resonance factor, cardiopulmonary harmony The bandwidth of the harmony curve and the quality factor of the cardiopulmonary resonance; the cardiopulmonary harmony curve is used to characterize the influence of respiration on the modulation of the RR interval sequence at different frequency points in the
  • the disclosed apparatus, system, and method may be implemented in other ways.
  • the above-described embodiments are merely illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation.
  • multiple units or components may be combined or may be Integrate into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be through some communication interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separate, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • the functional modules in the various embodiments of the present disclosure may be integrated together to form an independent part, or each module may exist alone, or two or more modules may be integrated to form an independent part.
  • the cardiopulmonary harmony series index evaluation method, device and system provided in the present disclosure provide a measurement parameter for the modulation intensity and effect of heart rate calculated from the resonant state of the cardiopulmonary system based on the respiratory signal and the ECG signal, which is the cardiopulmonary metabolism system
  • the evaluation of efficiency provides reliable and accurate parameters.

Landscapes

  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Engineering & Computer Science (AREA)
  • Physiology (AREA)
  • Surgery (AREA)
  • Medical Informatics (AREA)
  • Veterinary Medicine (AREA)
  • Physics & Mathematics (AREA)
  • Public Health (AREA)
  • Biophysics (AREA)
  • Pathology (AREA)
  • Biomedical Technology (AREA)
  • Heart & Thoracic Surgery (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Animal Behavior & Ethology (AREA)
  • Artificial Intelligence (AREA)
  • Psychiatry (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Signal Processing (AREA)
  • Cardiology (AREA)
  • Pulmonology (AREA)
  • Measurement Of The Respiration, Hearing Ability, Form, And Blood Characteristics Of Living Organisms (AREA)

Abstract

一种心肺和谐系列指标评测方法、装置及系统。该方法包括：对被测者的心电信号、呼吸信号和运动信号进行滤波和去噪预处理（101），从预处理后的心电信号中提取RR间期序列，并根据RR间期序列和呼吸信号构建二元回归数学模型（102）；将二元回归数学模型中的参数变换到频域获得变换矩阵（103）；对变换矩阵进行分析获得心肺和谐系列指标（S104）。心肺和谐系列指标用于量化心肺代谢系统的评估，包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子，其中心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值。

Description

一种心肺和谐系列指标评测方法、装置及系统
相关申请的交叉引用
本申请要求于2019年08月27日提交中国专利局的申请号为201910799044.8，名称为“一种心肺和谐系列指标评测方法、装置及系统”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本公开涉及信息处理技术领域，具体而言，涉及一种心肺和谐系列指标评测方法、装置及系统。
背景技术
人体心肺代谢系统为各器官和细胞提供营养，保证生命体的正常运行。国际上公认的呼吸心率调制(RespirationSinusArrhythmia，RSA)是心肺协调工作，以保障高效输送氧气和营养，回收代谢物的生理机制。大量研究证明，RSA的水平直接与人体的生理健康和心理健康相关、表征副交感神经的活动程度。这样，定量、准确地给出RSA的评测指标，成为相关科学研究和临床应用的关键技术和手段。然而，人体心肺代谢系统受自主神经调控，受外部环境、内部感染、运动、心理、等多种因素的影响，使得RSA评测指标研究面临很大困难。
发明内容
本公开的目的之一在于提供一种心肺和谐系列指标评测方法、装置及系统，可以获得较为准确、完整的呼吸心率调制(RSA)的数字指标系列，以实现对心肺功能的评估。
第一方面，本公开提供一种心肺和谐系列指标评测方法，包括：
对被测者的心电信号、呼吸信号和运动信号进行滤波和去噪预处理，根据运动信号，判定被测者的姿态，并根据姿态确定所述被测者是否处于静息状态，获得所述被测者在静息状态下的心电信号和呼吸信号；
从处于静息状态下的所述心电信号中提取RR间期序列，并根据所述RR间期序列和处于静息状态下的所述呼吸信号构建二元回归数学模型；
将所述二元回归数学模型中的参数变换到频域，获得变换矩阵；
对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述的影响值的平方的均值；所述心肺谐振因子为所述的影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
本公开提出了心肺和谐系列指标这一概念，并提供了心肺和谐系列指标获得的方法，可以为心肺代 谢系统的评估提供可靠的参考依据。
可选地，根据所述RR间期序列和所述呼吸信号构建二元回归数学模型，包括：
获取所述RR间期序列中的异常点，并用三次样条插值方法替换所述异常点，获得处理后RR间期序列；
根据所述处理后RR间期序列和处于静息状态下的所述呼吸信号构建二元回归数学模型。
本公开通过将RR间期序列中的异常点去除，并用三次样条插值法进行插值，可以获得较为准确的RR间期序列，进而提高心肺和谐系列指标的准确度。
可选地，获取所述RR间期序列中的异常点，包括：
若所述RR间期序列中的值不满足预设公式，则该值为异常点；
其中，所述预设公式为：
Figure PCTCN2019113393-appb-000001
RRI i为第i个RR间期序列值，
Figure PCTCN2019113393-appb-000002
为所述RR间期序列的平均值，RRI i-1为第i-1个RR间期序列值，Std(RRI)为所述RR间期序列的标准偏差，i为正整数。
本公开通过上述预设公式可以准确地获得异常点。
可选地，获取所述RR间期序列中的异常点，并用三次样条插值方法替换所述异常点，获得处理后RR间期序列，包括：
获取所述RR间期序列中的异常点，用三次样条插值方法替换所述异常点，并对插值后的RR间期序列进行重采样，获得所述处理后RR间期序列；其中，所述处理后RR间期序列的采样率与所述呼吸信号的采样率相同。
本公开通过重采样使得RR间期序列与呼吸信号的采样率相同，使得二者的序列是同步的。
可选地，根据所述RR间期序列和所述呼吸信号构建二元回归数学模型，包括：
分别将所述RR间期序列和所述呼吸信号进行归一化处理，并根据归一化后的RR间期序列和归一化后的呼吸信号构建如下二元回归数学模型：
Figure PCTCN2019113393-appb-000003
Figure PCTCN2019113393-appb-000004
其中，A 11,j、A 12,j、A 21,j和A 22,j均为二元回归数学模型的回归系数；X 1为所述RR间期序列；X 2为所述呼吸信号；ξ 1(t)和ξ 2(t)均为回归残差；p表示选取的回归序列的长度，t表示所述RR间期序列或所述呼吸信号所处的时刻，j为正整数，且1≤j≤p。
本公开通过构建二元回归数据模型可以准确地获得心肺和谐系列指标。
可选地，将所述二元回归数学模型中的参数变换到频域，获得变换矩阵，包括：
将所述二元回归数学模型中的参数进行傅里叶变换，获得变换矩阵。
本公开通过进行傅里叶变换，从而获得频域上的参数信息。
可选地，将所述二元回归数学模型中的参数变换到频域，获得变换矩阵，包括：
将所述二元回归数学模型中的参数变换到频域，获得如下变换矩阵：
Figure PCTCN2019113393-appb-000005
其中，
Figure PCTCN2019113393-appb-000006
X 1(f)为RR间期序列的频域表示，X 2(f)为呼吸信号的频域表示，E 1(f)和E 2(f)均为回归残差量的傅里叶变换，i为复数单位。
可选地，对所述变换矩阵进行分析获得心肺和谐曲线，包括：
利用格兰杰因果关系分析方法对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值，根据不同频点的影响值获得所述心肺和谐曲线。
本公开通过格兰杰因果关系分析方法可以获得准确的呼吸对心率的调制情况。
可选地，对所述变换矩阵进行分析获得心肺和谐曲线，包括：
通过以下公式对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值：
S(f)＝<X(f)X *(f)>＝<H(f)∑H *(f)>；
Figure PCTCN2019113393-appb-000007
Figure PCTCN2019113393-appb-000008
其中，
Figure PCTCN2019113393-appb-000009
H(f)为对矩阵块
Figure PCTCN2019113393-appb-000010
求逆的结果；S(f)为经过X(f)进行矩阵变换获得；
Figure PCTCN2019113393-appb-000011
G y→x(f)为呼吸对RR间期序列在频域上不同频点的影响值；x表示RR间期序列，y表示呼吸信号；
根据不同频点的影响值获得所述心肺和谐曲线。
第二方面，本公开提供一种心肺和谐系列指标评测装置，包括：
预处理模块，配置成对被测者的心电信号、呼吸信号和运动信号进行滤波和去噪预处理，根据运动信号，判定被测者的姿态，并根据姿态确定所述被测者是否处于静息状态，获得所述被测者在静息状态下的心电信号和呼吸信号；
模型构建模块，配置成从所述心电信号中提取RR间期序列，并根据所述RR间期序列和静息状态下的呼吸信号构建二元回归数学模型；
变换模块，配置成将所述二元回归数学模型中的参数变换到频域，获得变换矩阵；
指标获得模块，配置成对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列 指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述的影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
可选地，模型构建模块具体配置成：
获取所述RR间期序列中的异常点，并用三次样条插值方法替换所述异常点，获得处理后RR间期序列；
根据所述处理后RR间期序列和所述呼吸信号构建二元回归数学模型。
可选地，模型构建模块具体配置成：
若所述RR间期序列中的值不满足预设公式，则该值为异常点；
其中，所述预设公式为：
Figure PCTCN2019113393-appb-000012
RRI i为第i个RR间期序列值，
Figure PCTCN2019113393-appb-000013
为所述RR间期序列的平均值，RRI i-1为第i-1个RR间期序列值，Std(RRI)为所述RR间期序列的标准偏差，i为正整数。
可选地，模型构建模块具体配置成：
获取所述RR间期序列中的异常点，用三次样条插值方法替换所述异常点，并对插值后的RR间期序列进行重采样，获得所述处理后RR间期序列；其中，所述处理后RR间期序列的采样率与所述呼吸信号的采样率相同。
可选地，模型构建模块还配置成：
分别将所述RR间期序列和所述呼吸信号进行归一化处理，并根据归一化后的RR间期序列和归一化后的呼吸信号构建如下二元回归数学模型：
Figure PCTCN2019113393-appb-000014
Figure PCTCN2019113393-appb-000015
其中，A 11,j、A 12,j、A 21,j和A 22,j均为二元回归数学模型的回归系数；X 1为所述RR间期序列；X 2为所述呼吸信号；ξ 1(t)和ξ 2(t)均为回归残差；p表示选取的回归序列的长度，t表示所述RR间期序列或所述呼吸信号所处的时刻，j为正整数，且1≤j≤p。
可选地，变换模块具体配置成：
将所述二元回归数学模型中的参数进行傅里叶变换，获得变换矩阵。
可选地，变换模块具体配置成：
将所述二元回归数学模型中的参数变换到频域，获得如下变换矩阵：
Figure PCTCN2019113393-appb-000016
其中，
Figure PCTCN2019113393-appb-000017
X 1(f)为RR间期序列的频域表示，X 2(f)为呼吸信号的频域表示，E 1(f)和E 2(f)均为回归残差量的傅里叶变换，i为复数单位。
可选地，指标获得模块具体配置成：
利用格兰杰因果关系分析方法对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值，根据不同频点的影响值获得所述心肺和谐曲线。
可选地，指标获得模块具体配置成：
通过以下公式对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值：
S(f)＝<X(f)X *(f)>＝<H(f)∑H *(f)>；
Figure PCTCN2019113393-appb-000018
Figure PCTCN2019113393-appb-000019
其中，
Figure PCTCN2019113393-appb-000020
H(f)为对矩阵块
Figure PCTCN2019113393-appb-000021
求逆的结果；S(f)为经过X(f)进行矩阵变换获得；
Figure PCTCN2019113393-appb-000022
Gy→x(f)为呼吸对RR间期序列在频域上不同频点的影响值；x表示RR间期序列，y表示呼吸信号；
根据不同频点的影响值获得所述心肺和谐曲线。
第三方面，本公开提供一种心肺和谐系列指标评测系统，包括：穿戴式电子设备、第二方面所述心肺和谐系列指标评测装置，以及用户和数据管理子系统；
所述穿戴式电子设备配置成采集被测者的心电信号、呼吸信号和运动信号，并通过蓝牙向所述心肺和谐系列指标评测装置发送所述心电信号、所述呼吸信号和所述运动信号；
所述心肺和谐系列指标评测装置运行在智能手机或便携式计算设备PAD上，通过蓝牙接收来自穿戴式电子设备的所述被测者的心电信号、呼吸信号和运动信号，并根据所述被测者的心电信号、呼吸信号和运动信号进行分析处理，获得心肺和谐系列指标，并将获得的心肺和谐系列指标上传至用户和数据管理子系统。
可选地，所述穿戴式电子设备为微型、佩戴于胸前的电子设备，且所述穿戴式电子设备包括：
心电电极，配置成采集单导联心电信号；
呼吸传感器，配置成根据测量胸阻抗原理或由呼吸引起的胸部运动采集呼吸信号；
运动传感器，配置成测量所述被测者的躯干相对于垂直方向角度的三维加速度；
同步采集单元，配置成对心电信号和呼吸信号进行放大和AD转换，接收控制器指令，并根据控制器指令同步采集心电、呼吸和运动信号，打包成数据包，送往蓝牙传输单元；
蓝牙传输单元，将从同步采集单元接收到的数据包发往处理和分析子系统。
可选地，所述用户和数据管理子系统包括通信连接的中心数据库和医生终端；
所述中心数据库接收来自心肺和谐系列指标评测装置的被测者数据，其中所述被测者数据包括被测者的心肺和谐系列指标；
所述中心数据库接收医生终端的操作指令，根据所述心肺和谐系列指标和所述操作指令生成检测报告。
第四方面，本公开提供一种电子设备，包括：处理器、存储器和总线，其中，
所述处理器和所述存储器通过所述总线完成相互间的通信；
所述存储器存储有可被所述处理器执行的程序指令，所述处理器调用所述程序指令能够执行第一方面的方法步骤。
第五方面，本公开提供一种非暂态计算机可读存储介质，包括：
所述非暂态计算机可读存储介质存储计算机指令，所述计算机指令使所述计算机执行第一方面的方法步骤。
本公开的其他特征和优点将在随后的说明书阐述，并且，部分地从说明书中变得显而易见，或者通过实施本公开的实施例了解。本公开的目的和其他优点可通过在所写的说明书、权利要求书、以及附图中所特别指出的结构来实现和获得。
附图说明
为了更清楚地说明本公开的技术方案，下面将对本公开中所需要使用的附图作简单地介绍，应当理解，以下附图仅示出了本公开的某些实施例，因此不应被看作是对范围的限定，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他相关的附图。
图1为本公开提供的一种心肺和谐系列指标的评测方法流程示意图；
图2为本公开提供的另一种心肺和谐系列指标评测方法流程示意图；
图3为本公开提供的心电信号和呼吸信号的示意图；
图4为本公开提供的评价心肺和谐系列指标的示意图；
图5为本公开提供的被测者对应的心肺和谐曲线示意图；
图6为本公开提供的一种心肺和谐系列指标评测装置结构示意图；
图7为本公开提供的一种心肺和谐系列指标评测系统结构示意图。
具体实施方式
下面将结合本实施例中的附图，对本实施例中的技术方案进行描述。
呼吸对心率的调制(respiratory sinus arrhythmia，RSA)表现为：人体在做吸气运动时，胸廓扩张，这一活动使支气管和细支气管的平滑肌层中的肺牵张感受器兴奋，发送冲动并将冲动传入到呼吸中枢。呼吸中枢将传入的信息进行整合，抑制迷走神经的活动，增强交感神经活性，从而使心率加快，血压升高。血压升高后，位于颈动脉窦和主动脉弓血管外膜下的动脉压力感受器兴奋，发放冲动并将冲动传入到呼吸中枢，呼吸中枢再次将传入的冲动的信息整合后，增强迷走神经活性，抑制交感神经的活性，使心率减慢，血压降低。同时，血压升高引起心房和心室的血容量增多，使位于心房、心室和肺循环大血管壁的心肺感受器兴奋，它们发放的冲动经过呼吸中枢的整合后，进一步降低交感神经活性，增强迷走神经活性，使心率减慢，血压下降。心率减慢引起血液中氧分压降低，颈动脉体和主动脉体化学感受器感受到变化，发放冲动并传入呼吸中枢，使呼吸加快，心率加快。
基于以上分析，获取能够体现呼吸对心率的调制的本质的评测数字指标，对人体心肺代谢系统性能的评估非常重要。
一些实施例，采用如下方式来表示RSA度量指标：
例如，使用呼气和吸气间隔中的心电的RMSSD均方根值表示RSA度量指标，这种方式完全没有考虑RSA的生理学机制，当然谈不上有效性。
又如，国际上提出心肺耦合系数(CardioPulmonaryCoupling，CPC)的方法，用心率序列和呼吸信号在变换域内的互相关程度来表示心肺交互与和谐程度，如，采用NN序列和呼吸信号的频率域功率谱的相似度来表示心肺交互与和谐程度。然而，心率序列中可能存在一心率变化部分，该心率变化部分与呼吸信号的功率谱相似，但完全不是由呼吸调制引起的。因此，这种表示方式是不准确的，可能出错。
CPC通常被认为比较好地表征了不同睡眠期内的自主神经调控状态，但我们在MIT标准数据库上的实验证明，其作用非常有限。
又如，使用呼吸频率区间内的心率变异性功率谱相对于总功率谱的比值(称为“和谐指数Coh”)，相对较好地表征了心肺和谐水平。
但是，经研究发现，上述表示方式都未能从呼吸对心率调制的本质出发，从而无法获得较为准确、完整的评测指标，以实现对人体心肺代谢系统效率的评估。
因此，本实施例提出一种用于评估人体心肺代谢系统的新的评测数字指标系列，心肺和谐系列指标(Cardiopulmonary Resonance Indices，CRI)。
心肺和谐系列指标包括心肺和谐曲线G(f)、心肺和谐度(Cardiopulmonary Resonance Amplitude，CRA)、心肺谐振频率、心肺谐振因子(Cardiopulmonary Resonance Factor,CRF)、心肺和谐曲线带宽和心肺谐振品质因子(Cardiopulmonary Resonance Qualityfactor，CRQ)。
其中，心肺和谐曲线G(f)用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；心肺和谐度为心肺和谐曲线中的最大影响值；心肺谐振频率为该最大影响值对应的谐振频率；心肺谐振因子 为不同频点的影响值的平方的均值；心肺谐振品质因子为心肺谐振频率与心肺和谐曲线带宽的比值。
本实施例提出的心肺和谐系列指标可以完整地描述人体的个性化RSA特性。在静息状态下，被测者的生理和心理状态对呼吸和心率产生影响，该影响进一步表现为心肺和谐曲线G(f)的谐振品质因子和心肺和谐度：心肺功能系统好(生理和心理状态好)，则呼吸平稳，心肺谐振因子高，心肺和谐度大；另一方面，G(f)也随呼吸频率变化：如果呼吸频率从通常的每分钟20次(约0.3Hz)以深呼吸的方式降到每分钟6次(约0.1Hz)，则心肺和谐度增大，达到最大值的呼吸频率在0.1Hz左右。每个人都有一个达到最大值的谐振频率，每个人的最大谐振频率都不一样，但都在0.1Hz附近。
下面将给出心肺和谐系列指标的评测方法，如图1所示，执行该方法的主体可以是电子设备，该电子设备可以为台式电脑、平板电脑、服务器等具有数据处理功能的设备，该方法包括以下处理：
处理101：对被测者的心电信号、呼吸信号和运动信号进行滤波和去噪预处理，根据预处理后的运动信号，判定被测者的姿态，并根据姿态确定所述被测者是否处于静息状态，获得所述被测者处于静息状态的心电信号和呼吸信号。
在实施过程中，电子设备接收或采集被测者的心电信号、呼吸信号和运动信号，并对心电信号、呼吸信息和运动信号进行滤波和去噪预处理。
其中，心电信号是心脏的无数心肌细胞电活动的综合反映，从宏观上记录心脏细胞的除极和复极过程，在一定程度上客观反映了心脏各部位的生理状况。呼吸信号是用于表征被测者呼吸的时间序列。
运动信号用于表征被测者当前的状态，被测者当前的状态包括静息状态和运动状态。本实施例中，运动信号可以是测得被测者的三维加速度信号。静息状态包括睡眠状态、静坐状态和卧床状态等，运动状态包括走、跑等。通过运动信号可以确定被测者的姿态，进而根据该姿态确定被测者是否处于静息状态。应当说明的是，上述关于被测者的静息状态的确定方式仅为举例，除了通过运动信号确定外，还可以通过其他方式确定，本申请实施例对此不做具体限定。
可选地，电子设备在确定被测者处于静息状态时，可以分别从经过预处理的心电信号和呼吸信号中，获取被测者处于静息状态下的、预设时间段内的心电信号和呼吸信号，预设时间段例如可以设定为3分钟、5分钟，或者其他时长。以预设时间段是3分钟为例，电子设备确定被测者处于静息状态时，可以从采集的被测者处于静息状态下的心电信号和呼吸信号中，分别获取3分钟的心电信号和呼吸信号。应当说明的是，获取到的3分钟的心电信号和呼吸信号应当是同一时间段的。例如：根据运动信号获知在某一天的下午1点半到3点之间被测者处于静息状态，因此，可以获取这天下午1点40分-1点43分这一时间段内的心电信号和呼吸信号。
处理102：从处于静息状态的心电信号中提取RR间期序列，并根据所述RR间期序列和处于静息状态的呼吸信号构建二元回归数学模型。
心电信号包括多种波群，例如：P波、QRS波群、PR间期、T波、RR间期等。在实施过程中，电 子设备可以从被测者处于静息状态下的心电信号中提取出RR间期序列，并根据RR间期序列和处于静息状态的呼吸信号构建二元回归数学模型。
处理103：使用二元回归模型描述呼吸对心率的调制。为了求解需要，将所述二元回归数学模型中的参数变换到频域，获得变换矩阵。
在实施过程中，由于RR间期序列和呼吸信号都是时域上的信号，因此，构建的二元回归数据模型也是时域上的，为了获得呼吸对心率的调制，可以通过傅里叶变换将二元回归数学模型中的参数变换到频域。
处理104：对所述变换矩阵进行分析获得心肺和谐系列指标。
在实施过程中，可以利用格兰杰因果关系分析法对变换矩阵进行分析，获得在预设时间段内呼吸对RR间期序列的调制在频域上不同频点的影响值，通过各个频点的影响值可以获得心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子，即心肺和谐系列指标。
本公开中根据心电信号和呼吸信号构建二元回归数学模型，进而获得心肺和谐系列指标，该心肺和谐系列指标是从心肺系统的共振状态出发，通过呼吸信号和心电信号计算出对于心率的调制强度和效果的度量参数。本实施例提出的CRI从RSA的生理过程的数学建模出发，准确地描述了呼吸调制心率的生理现象，因而可以进一步用于描述心肺代谢系统的工作效率和健康状况，为评估心肺代谢系统效率提供了准确的参数数据。
值得说明的是，相较于CPC，本实施例提出的CRI对深睡期的分期准确率比CPC提高了11％左右。
图2为本公开提供的另一种心肺和谐系列指标评测方法流程示意图，如图2所示，该方法包括以下处理：
处理201：利用心电和呼吸监测仪监测被测者。
其中，为了同步采集被测者的心电信号和呼吸信号，可以选择同时具备心电信号采集和呼吸信号采集的心电和呼吸监测仪，并将心电和呼吸监测仪设置在被测者身上。如此，获得的心电信号和呼吸信号在时间上是同步的。
处理202：获取心电信号和呼吸信号。
电子设备可以从心电和呼吸监测仪获取心电信号、呼吸信号和运动信号，图3示例性地示出了本申请实施例提供的心电信号和呼吸信号的示意图。实施过程中，电子设备可以根据运动信号可以获知被测者的状态信息，当被测者的状态信息指示该被测者处于静息状态(如，睡眠、深呼吸等)时，采用滑动窗口从获得的心电信号和呼吸信号中获取静息状态下的心电信号和呼吸信号。其中，滑动窗口的单位长度可以为上述的预设时间段(如，3-5分钟)。
处理203：提取RR间期序列。
电子设备从心电信号中提取RR间期序列。
处理204：剔除异常点。
电子设备从RR间期序列中获取异常点，其中，异常点可以通过如下公式进行确定：
Figure PCTCN2019113393-appb-000023
RRI i为第i个RR间期序列值，
Figure PCTCN2019113393-appb-000024
为RR间期序列的平均值，RRI i-1为第i-1个RR间期序列值，Std(RRI)为RR间期序列的标准偏差。因此，如果第i个RR间期序列的值不满足上述公式，则说明第i个RR间期序列为异常点，i为正整数。
电子设备在获取到异常点之后，将异常点剔除，并对被剔除的异常点进行三次样条插值处理，以将被剔除的异常点替换为对应的插值，获得处理后RR间期序列。其中，处理后RR间期序列与呼吸信号具有基本相同的采样率。
处理205：归一化RR间期序列与呼吸信号。
电子设备对处理后RR间期序列进行重采样，获得与呼吸信号的采样率基本相同的处理后RR间期序列，并分别将处理后RR间期序列和呼吸信号进行归一化，使得处理后RR间期序列和呼吸信号的均值均为0，方差均为1。
处理206：构建二元回归数学模型，表征呼吸对心率的调制。
详细地，电子设备可以根据归一化后的RR间期序列和归一化后的呼吸信号构建如下二元回归数学模型：
Figure PCTCN2019113393-appb-000025
Figure PCTCN2019113393-appb-000026
其中，A 11,j、A 12,j、A 21,j和A 22,j均为二元回归数学模型的回归系数；X 1为所述RR间期序列；X 2为所述呼吸信号；ξ 1(t)和ξ 2(t)均为回归残差；p表示选取的回归序列的长度，t表示所述RR间期序列或所述呼吸信号所处的时刻，j为正整数，且1≤j≤p。
在上述二元回归数学模型中，参数p是基于最大化时长、最小化残差的原则来选取的，即计算不同p值对应的赤池信息准则(Akaike Information Criterion，AIC)和贝叶斯信息准则(Bayesian Information Criterion，BIC)系数，当AIC系数和BIC系数取到最小值时，对应的p值即为二元回归数学模型的延迟p。其中，AIC系数和BIC系数可以通过如下计算式获得：
Figure PCTCN2019113393-appb-000027
其中，T为计算的回归的选取时长，n为计算的回归样本量。
处理207：傅里叶变换。
对二元回归数学模型中的参数进行傅里叶变换。其中，傅里叶变换公式为：
Figure PCTCN2019113393-appb-000028
其中，
Figure PCTCN2019113393-appb-000029
X 1(f)为RR间期序列的频域表示，X 2(f)为呼吸信号的频域表示，E 1(f)和E 2(f)均为回归残差量的傅里叶变换，i为复数单位。
为了便于后续的计算，可以对上述矩阵做如下变换：
Figure PCTCN2019113393-appb-000030
其中，
Figure PCTCN2019113393-appb-000031
为对矩阵块
Figure PCTCN2019113393-appb-000032
求逆的结果。
处理208：对变换后矩阵进行分析。
电子设备在获得变换后矩阵之后，可以利用格兰杰因果关系分析法，计算出所述预设时间段内呼吸对RR间期序列的变化在频域上的不同频点的影响值G。
其中，影响值G的计算源于格兰杰因果关系分析理论，通过
Figure PCTCN2019113393-appb-000033
度量二元回归残差相对于一元回归残差的减少，表征和度量两个序列之间的影响关系。
将这一理论应用到RR间期序列和呼吸信号的频域的度量上，其中频率的影响值可以通过下列公式计算获得：
S(f)＝<X(f)X *(f)>＝<H(f)∑H *(f)>；
在计算过程中，将二元回归模型在频域上表示出来如下：
Figure PCTCN2019113393-appb-000034
通过矩阵变换技巧，得到影响值G：
Figure PCTCN2019113393-appb-000035
其中，
Figure PCTCN2019113393-appb-000036
S(f)为经过X(f)进行矩阵变换获得；
Figure PCTCN2019113393-appb-000037
G y→x(f)为呼吸对RR间期序列在频域上不同频点的影响值；x表示RR间期序列，y表示呼吸信号。
根据不同频点的影响值可以获得心肺和谐曲线G(f)，进一步，根据心肺和谐曲线G(f)可以获得心肺和谐度CRA、心肺谐振频率、心肺谐振因子CRF、心肺和谐曲线带宽和心肺谐振品质因子CRQ。
图4为本公开提供的评价心肺和谐系列指标的示意图，图4中，横坐标表示频率，纵坐标表示不同频点下的影响值。
其中，CRA＝maxG(f)，CRA对应的频率为心肺谐振频率f A；心肺谐振因子CRF＝mean(G2(f))，是对心肺系统谐振能量的度量；心肺谐振品质因子CRQ为心肺系统的谐振特性的度量，是谐振频率f A与和谐曲线G(f)的值下降到0.707时的心肺和谐曲线带宽△f的比值，即：
Figure PCTCN2019113393-appb-000038
心肺和谐度表征当前测量状态下的心肺系统的耦合共振的最大深度，反映了被测者的心肺耦合的程度。其中，心肺和谐度越大，被测者当前的和谐状态越好。心肺谐振因子在能量层面表述了心肺的耦合状态，CRF越高，表明心脏受呼吸系统的调制作用越大，心肺的谐振状态越佳。另外，心肺谐振品质因子CRQ高、心肺和谐曲线带宽△f窄，说明心肺代谢系统工作效率高。
处理209：对滑动窗口进行滑动，实时计算CRI。
本实施例中，一个滑动窗口的单位长度可以为上述的预设时间段，对滑动窗口进行滑动后，获取下一个滑动窗口(即，滑动后的滑动窗口)内的心电信号和呼吸信号，并按照处理201-处理208对下一个滑动窗口内的心电信号和呼吸信号进行分析，以实现对被测者的长期监测。
本公开通过获取被测者的心电信号和呼吸信号，利用二元回归数学模型和格兰杰因果关系分析获得呼吸对心率调制的数字度量，即心肺和谐系列指标，心肺和谐系列指标可以作为对心肺代谢系统进行评估的较为准确的评测数字指标系列。
图5为本公开提供的被测者对应的心肺和谐曲线示意图，其中示出了被测者在不同的呼吸频率和不同生理状态下，呼吸对心率的调制作用在时域和频域上的影响值，图5中从上到下为被测者从静息状态过渡到可以降低呼吸率的调整状态。
图6为本公开提供的一种心肺和谐系列指标评测装置结构示意图，该装置可以是电子设备上的模块、程序段或代码。应理解，该装置与上述图1方法实施例对应，能够执行图1方法实施例涉及的各个步骤，该装置具体的功能可以参见上文中的描述，为避免重复，此处适当省略详细描述。从功能上划分，该装置可以包括预处理模块601、模型构建模块602、变换模块603和指标获得模块604，其中：
预处理模块601配置成对被测者的心电信号、呼吸信号和运动信号进行滤波和去噪预处理，根据预处理后的运动信号，判定被测者的姿态，并根据姿态确定所述被测者是否处于静息状态，获得所述被测者在静息状态下的心电信号和呼吸信号。
在实施过程中，接收被测者的心电信号、呼吸信号和运动信号，并对心电信号、呼吸信息和运动信号进行滤波和去噪预处理。其中，心电信号是心脏的无数心肌细胞电活动的综合反映，从宏观上记录心脏细胞的除极和复极过程，在一定程度上客观反映了心脏各部位的生理状况。呼吸信号用于表征被测者呼吸的时间序列。运动信号用于表征被测者当前的状态，例如：被测者处于静息状态、睡眠状态、深呼吸状态、缓慢呼吸的调制状态或运动状态等。其中，运动信号可以是测得被测者的三维加速度信号。
静息状态可以包括睡眠状态、静坐状态、卧床状态等，通过运动信号可以确定被测者所处的姿态，根据姿态确定被测者是否处于静息状态，若是，则从预处理后的心电信号中获得被测者处于静息状态的心电信号，从预处理后的呼吸信号中获得被测者处于静息状态的呼吸信号。
模型构建模块602配置成从所述处于静息状态的心电信号中提取RR间期序列，并根据所述RR间期序列和所述处于静息状态的呼吸信号构建二元回归数学模型。
心电信号中包括多种波群，例如：P波、QRS波群、PR间期、T波、RR间期等。因此，模型构建模块602可以从心电信号中提取出RR间期序列，并根据RR间期序列和呼吸信号构建二元回归数学模型。应当说明的是，呼吸信号是被测者在预设时间段内的呼吸时间序列。
变换模块603配置成将所述二元回归数学模型中的参数变换到频域，获得变换矩阵。
在实施过程中，RR间期序列和呼吸信号都是时域上的信号，因此，构建的二元回归数据模型也是时域上的，为了获得呼吸对心率的调制，变换模块603可以通过傅里叶变换将二元回归数学模型中的参数变换到频域。
指标获得模块604配置成对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述的影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
在实施过程中，指标获得模块604可以利用格兰杰因果关系分析法对变换矩阵进行分析，获得在预设时间段内呼吸对RR间期序列的调制在频域上不同频点的影响值，通过各个频点的影响值可以获得心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子，即心肺和谐系列指标。
本公开根据心电信号和呼吸信号构建二元回归数学模型，进而获得心肺和谐系列指标，该心肺和谐系列指标是从心肺系统的共振状态出发，通过呼吸信号和心电信号计算出对于心率的调制强度和效果的度量参数，为评估心肺代谢系统效率提供了准确的参数数据。
在上述实施例的基础上，模型构建模块602具体可以配置成：
确定所述RR间期序列中的异常点，剔除所述异常点，并对剔除的异常点进行三次样条插值处理，以将所述剔除的异常点替换为插值，获得处理后RR间期序列；根据所述处理后RR间期序列和所述处于静息状态的呼吸信号构建二元回归数学模型。
在上述实施例的基础上，所述模型构建模块602具体可以配置成：
将所述RR间期序列中不满足预设公式的值确定为异常点；
其中，所述预设公式为：
Figure PCTCN2019113393-appb-000039
RRI i为第i个RR间期序列值，
Figure PCTCN2019113393-appb-000040
为所述RR间期序列的平均值，RRI i-1为第i-1个RR间期序列值，Std(RRI)为所述RR间期序列的标准偏差，i为正整数。
在上述实施例的基础上，模型构建模块602具体可以配置成：
获取所述RR间期序列中的异常点，用三次样条插值方法替换所述异常点，并对插值后的RR间期序列进行重采样，获得所述处理后RR间期序列；其中，所述处理后RR间期序列的采样率与所述呼吸信号的采样率相同。
在上述实施例的基础上，模型构建模块602还可以配置成：
分别对所述RR间期序列和所述呼吸信号进行归一化处理，并根据归一化后的RR间期序列和归一化后的呼吸信号构建如下二元回归数学模型：
Figure PCTCN2019113393-appb-000041
Figure PCTCN2019113393-appb-000042
其中，A 11,j、A 12,j、A 21,j和A 22,j均为二元回归数学模型的回归系数；X 1为所述RR间期序列；X 2为所述呼吸信号；ξ 1(t)和ξ 2(t)均为回归残差；p表示选取的回归序列的长度，t表示所述RR间期序列或所述呼吸信号所处的时刻，j为正整数，且1≤j≤p。
在上述实施例的基础上，变换模块603具体可以配置成：
对所述二元回归数学模型中的参数进行傅里叶变换，获得变换矩阵。
在上述实施例的基础上，所述变换模块具体可以配置成：
将所述二元回归数学模型中的参数变换到频域，获得如下变换矩阵：
Figure PCTCN2019113393-appb-000043
其中，
Figure PCTCN2019113393-appb-000044
X 1(f)为RR间期序列的频域表示，X 2(f)为呼吸信号的频域表示，E 1(f)和E 2(f)均为回归残差量的傅里叶变换，i为复数单位。
在上述实施例的基础上，指标获得模块604具体可以配置成：
利用格兰杰因果关系分析方法对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值，根据不同频点的影响值获得所述心肺和谐曲线。
在上述实施例的基础上，指标获得模块604具体可以配置成：
通过以下公式对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值：
S(f)＝<X(f)X *(f)>＝<H(f)∑H *(f)>；
Figure PCTCN2019113393-appb-000045
Figure PCTCN2019113393-appb-000046
其中，
Figure PCTCN2019113393-appb-000047
H(f)为对矩阵块
Figure PCTCN2019113393-appb-000048
求逆的结果；S(f)为经过X(f)进行矩阵变换获 得；
Figure PCTCN2019113393-appb-000049
G y→x(f)为呼吸对RR间期序列在频域上不同频点的影响值；x表示RR间期序列，y表示呼吸信号；
根据不同频点的影响值获得所述心肺和谐曲线。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的装置的具体工作过程，可以参考前述方法中的对应过程，在此不再过多赘述。
图7为本公开提供的一种心肺和谐系列指标评测系统，该系统可以包括：穿戴式电子设备701、心肺和谐系列指标评测装置702和用户和数据管理子系统703，其中：
所述穿戴式电子设备701配置成采集被测者的心电信号、呼吸信号和运动信号，并通过蓝牙向所述心肺和谐系列指标评测装置702发送所述心电信号、所述呼吸信号和所述运动信号；
所述心肺和谐系列指标评测装置702运行在智能手机或便携式计算设备PAD上，通过蓝牙接收来自穿戴式电子设备701的所述被测者的心电信号、呼吸信号和运动信号，并根据所述被测者的心电信号、呼吸信号和运动信号进行分析处理，获得心肺和谐系列指标，并将获得的心肺和谐系列指标上传至用户和数据管理子系统703。
在上述实施例的基础上，穿戴式电子设备701可以为微型的、佩戴于胸前的电子设备，且所述穿戴式电子设备可以包括心电电极7011、呼吸传感器7012、运动传感器7013、同步采集单元7014和蓝牙传输单元7015。心电电极7011、呼吸传感器7012和运动传感器7013分别与同步采集单元7014连接，同步采集单元7014与蓝牙传输单元7015连接。
其中，心电电极7011配置成采集单导联心电信号。本实施例中，穿戴式电子设备7015可以包括两个心电电极7011，在使用中，将两个心电电极7011可以贴在被测者的胸前标准心电导联位置。
呼吸传感器7012配置成根据测量胸阻抗原理或由呼吸引起的胸部运动采集呼吸信号。
运动传感器7013配置成测量被测者的躯干相对于垂直方向角度的三维加速度；在采集被测者的运动信号时，可以将运动传感器佩带在被测者的胸前，从而可以测量到被测者的躯干相对于垂直方向角度的三维加速度信号。
同步采集单元7014配置成对心电信号和呼吸信号进行放大和AD转换，接收控制器指令，并根据控制器指令同步采集心电信号、呼吸信号和运动信号，将同步的心电信号、呼吸信号和运动信号打包成数据包，送往蓝牙传输单元；
蓝牙传输单元7015配置成将从同步采集单元7014接收到的数据包发往处理和分析子系统。
另外，穿戴式电子设备701还可以包括控制器7016，同步采集单元7014和蓝牙传输单元7015分别与控制器7016连接。控制器7016可以配置成对同步采集单元7014和蓝牙传输单元7015进行控制。
用户和数据管理子系统703可以包括相互通信连接的中心数据库7031和医生终端7032。
所述中心数据库7031可以接收来自心肺和谐系列指标评测装置702的被测者数据，其中所述被测者数据可以包括被测者的心肺和谐系列指标。
所述中心数据库7031接收医生终端7032的操作指令，根据所述心肺和谐系列指标和所述操作指令生成检测报告。
本公开还提供一种电子设备，包括：处理器(processor)、存储器(memory)和总线；其中，所述处理器和存储器通过所述总线完成相互间的通信。
所述处理器配置成调用所述存储器中的程序指令，以执行上述各实施例所提供的方法，例如包括：获取所述被测者的心电信号、呼吸信号和运动信号，根据所述运动信号确定所述被测者的状态信息；从所述被测者的状态信息中获取处于静息状态下的心电信号和呼吸信号；从所述心电信号中提取RR间期序列，并根据所述RR间期序列和所述呼吸信号构建二元回归数学模型；将所述二元回归数学模型中的参数变换到频域，获得变换矩阵；对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
本实施例还公开一种计算机程序产品，所述计算机程序产品包括存储在非暂态计算机可读存储介质上的计算机程序，所述计算机程序包括程序指令，当所述程序指令被计算机执行时，计算机可以实现上述各实施例所提供的方法，该方法例如包括：获取所述被测者的心电信号、呼吸信号和运动信号，根据所述运动信号确定所述被测者的状态信息；从所述被测者的状态信息中获取处于静息状态下的心电信号和呼吸信号；从所述心电信号中提取RR间期序列，并根据所述RR间期序列和所述呼吸信号构建二元回归数学模型；将所述二元回归数学模型中的参数变换到频域，获得变换矩阵；对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
本实施例还提供一种非暂态计算机可读存储介质，所述非暂态计算机可读存储介质存储计算机指令，所述计算机指令使所述计算机执行上述各方法实施例所提供的方法，例如包括：获取所述被测者的心电信号、呼吸信号和运动信号，根据所述运动信号确定所述被测者的状态信息；从所述被测者的状态信息中获取处于静息状态下的心电信号和呼吸信号；从所述心电信号中提取RR间期序列，并根据所述RR 间期序列和所述呼吸信号构建二元回归数学模型；将所述二元回归数学模型中的参数变换到频域，获得变换矩阵；对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
在本公开的实施例中，应该理解到，所揭露装置、系统和方法，可以通过其它的方式实现。以上所描述的实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，又例如，多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些通信接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
另外，作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
再者，在本公开各个实施例中的各功能模块可以集成在一起形成一个独立的部分，也可以是各个模块单独存在，也可以两个或两个以上模块集成形成一个独立的部分。
在本文中，诸如第一和第二等之类的关系术语仅仅用来将一个实体或者操作与另一个实体或操作区分开来，而不一定要求或者暗示这些实体或操作之间存在任何这种实际的关系或者顺序。
以上所述仅为本公开的实施例而已，并不用于限制本公开的保护范围，对于本领域的技术人员来说，本公开可以有各种更改和变化。凡在本公开的精神和原则之内，所作的任何修改、等同替换、改进等，均应包含在本公开的保护范围之内。
工业实用性
本公开提供的心肺和谐系列指标评测方法、装置及系统，提供了从心肺系统的共振状态出发，根据呼吸信号和心电信号计算出的针对心率的调制强度和效果的度量参数，为心肺代谢系统效率的评估提供了可靠、准确的参数。

Claims (16)

  1. 一种心肺和谐系列指标评测方法，其特征在于，包括：
    对被测者的心电信号、呼吸信号和运动信号进行滤波和去噪预处理；
    根据预处理后的运动信号，判定被测者的姿态，并根据所述姿态确定所述被测者是否处于静息状态；
    获得所述被测者在静息状态下的心电信号和呼吸信号；
    从处于静息状态下的心电信号中提取RR间期序列，并根据所述RR间期序列和处于静息状态下的呼吸信号构建二元回归数学模型；
    将所述二元回归数学模型中的参数变换到频域，获得变换矩阵；
    对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
  2. 根据权利要求1所述的方法，其特征在于，所述获得所述被测者在静息状态下的心电信号和呼吸信号，包括：
    若根据所述姿态确定所述被测者处于静息状态，则从预处理后的心电信号获得所述处于静息状态的心电信号，从所述预处理后的呼吸信号中获得所述处于静息状态的呼吸信号。
  3. 一种心肺和谐系列指标评测装置，其特征在于，包括：
    预处理模块，配置成对被测者的心电信号、呼吸信号和运动信号进行滤波和去噪预处理，根据预处理后的运动信号，判定被测者的姿态，并根据所述姿态确定所述被测者是否处于静息状态，获得所述被测者在静息状态下的心电信号和呼吸信号；
    模型构建模块，配置成从处于静息状态的心电信号中提取RR间期序列，并根据所述RR间期序列和处于静息状态的呼吸信号构建二元回归数学模型；
    变换模块，配置成将所述二元回归数学模型中的参数变换到频域，获得变换矩阵；
    指标获得模块，配置成对所述变换矩阵进行分析获得心肺和谐系列指标；其中，所述心肺和谐系列指标包括心肺和谐曲线、心肺和谐度、心肺谐振频率、心肺谐振因子、心肺和谐曲线带宽和心肺谐振品质因子；所述心肺和谐曲线用于表征呼吸对RR间期序列的调制在频域上不同频点的影响值；所述心肺和谐度为所述心肺和谐曲线中的最大影响值；所述心肺谐振频率为所述最大影响值对应的谐振频率；所述心肺谐振因子为所述的影响值的平方的均值；所述心肺谐振品质因子为所述心肺谐振频率与所述心肺和谐曲线带宽的比值。
  4. 根据权利要求3所述的装置，其特征在于，所述预处理模块获得所述被测者在静息状态下的心电 信号和呼吸信号的方式为：
    若根据所述姿态确定所述被测者处于静息状态，从预处理后的心电信号中获得所述处于静息状态的心电信号，以及从预处理后的呼吸信号中获得所述处于静息状态的呼吸信号。
  5. 根据权利要求3或4所述的装置，其特征在于，所述模型构建模块具体配置成：
    确定所述RR间期序列中的异常点，并用三次样条插值方法替换所述异常点，获得处理后RR间期序列；
    根据所述处理后RR间期序列和所述呼吸信号构建二元回归数学模型。
  6. 根据权利要求3或4所述的装置，其特征在于，所述模型构建模块具体配置成：
    确定所述RR间期序列中的异常点，从所述RR间期序列中剔除所述异常点；
    对被剔除的异常点进行三次样条插值处理，以将所述被剔除的异常点替换成对应的插值，得到处理后RR间期序列，其中，处理后RR间期序列与所述呼吸信号具有基本相同的采样率；
    根据所述处理后RR间期序列和所述呼吸信号构建二元回归数学模型。
  7. 根据权利要求5或6所述的装置，其特征在于，模型构建模块确定所述RR间期序列中的异常点的方式为：
    将所述RR间期序列中不满足预设公式的值确定为所述异常点；
    其中，所述预设公式为：
    Figure PCTCN2019113393-appb-100001
    RRI i为第i个RR间期序列值，
    Figure PCTCN2019113393-appb-100002
    为所述RR间期序列的平均值，RRI i-1为第i-1个RR间期序列值，Std(RRI)为所述RR间期序列的标准偏差，i为正整数。
  8. 根据权利要求5所述的装置，其特征在于，模型构建模块具体配置成：
    获取所述RR间期序列中的异常点，用三次样条插值方法替换所述异常点，并对插值后的RR间期序列进行重采样，获得所述处理后RR间期序列；其中，所述处理后RR间期序列的采样率与所述呼吸信号的采样率相同。
  9. 根据权利要求3-8中任意一项所述的装置，其特征在于，模型构建模块还配置成：
    分别对所述RR间期序列和所述呼吸信号进行归一化处理，并根据归一化后的RR间期序列和归一化后的呼吸信号构建如下二元回归数学模型：
    Figure PCTCN2019113393-appb-100003
    Figure PCTCN2019113393-appb-100004
    其中，A 11,j、A 12,j、A 21,j和A 22,j均为二元回归数学模型的回归系数；X 1为所述RR间期序列；X 2为所 述呼吸信号；ξ 1(t)和ξ 2(t)均为回归残差；p表示选取的回归序列的长度，t表示所述RR间期序列或所述呼吸信号所处的时刻，j为正整数，且1≤j≤p。
  10. 根据权利要求3-9中任意一项所述的装置，其特征在于，变换模块具体配置成：
    对所述二元回归数学模型中的参数进行傅里叶变换，获得变换矩阵。
  11. 根据权利要求10所述的装置，其特征在于，变换模块对所述二元回归数学模型中的参数进行傅里叶变换的方式：
    将所述二元回归数学模型中的参数变换到频域，获得如下变换矩阵：
    Figure PCTCN2019113393-appb-100005
    其中，
    Figure PCTCN2019113393-appb-100006
    X 1(f)为RR间期序列的频域表示，X 2(f)为呼吸信号的频域表示，E 1(f)和E 2(f)均为回归残差量的傅里叶变换，i为复数单位。
  12. 根据权利要求3-11中任意一项所述的装置，其特征在于，指标获得模块具体配置成：
    利用格兰杰因果关系分析方法对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值，根据不同频点的影响值获得所述心肺和谐曲线。
  13. 根据权利要求12所述的装置，其特征在于，指标获得模块具体配置成：
    通过以下公式对所述变换矩阵进行分析，获得呼吸对RR间期序列在频域上不同频点的影响值：
    S(f)＝<X(f)X *(f)>＝<H(f)∑H *(f)>；
    Figure PCTCN2019113393-appb-100007
    Figure PCTCN2019113393-appb-100008
    其中，
    Figure PCTCN2019113393-appb-100009
    H(f)为对矩阵块
    Figure PCTCN2019113393-appb-100010
    求逆的结果；S(f)为经过X(f)进行矩阵变换获得；
    Figure PCTCN2019113393-appb-100011
    G y→x(f)为呼吸对RR间期序列在频域上不同频点的影响值；x表示RR间期序列，y表示呼吸信号；
    根据不同频点的影响值获得所述心肺和谐曲线。
  14. 一种心肺和谐系列指标评测系统，其特征在于，包括：穿戴式电子设备、权利要求3-13中任意一项所述心肺和谐系列指标评测装置，以及用户和数据管理子系统；
    所述穿戴式电子设备配置成采集被测者的心电信号、呼吸信号和运动信号，并通过蓝牙向所述心肺和谐系列指标评测装置发送所述心电信号、所述呼吸信号和所述运动信号；
    所述心肺和谐系列指标评测装置运行在智能手机或便携式计算设备PAD上，通过蓝牙接收来自穿戴式电子设备的所述被测者的心电信号、呼吸信号和运动信号，并根据所述被测者的心电信号、呼吸信号和运动信号进行分析处理，获得心肺和谐系列指标，并将获得的心肺和谐系列指标上传至用户和数据 管理子系统。
  15. 根据权利要求14所述的心肺和谐系列指标评测系统，其特征在于，所述穿戴式电子设备为微型的、佩戴于胸前的电子设备，且所述穿戴式电子设备包括：
    心电电极，配置成采集单导联心电信号；
    呼吸传感器，配置成根据测量胸阻抗原理或由呼吸引起的胸部运动采集呼吸信号；
    运动传感器，配置成测量被测者躯干的三维加速度；
    同步采集单元，配置成对心电信号和呼吸信号进行放大和AD转换，接收控制器指令，并根据控制器指令同步采集心电、呼吸和运动信号，打包成数据包，送往蓝牙传输单元；
    蓝牙传输单元，配置成将从同步采集单元接收到的数据包发往处理和分析子系统。
  16. 根据权利要求14或15所述的心肺和谐系列指标评测系统，其特征在于，所述用户和数据管理子系统包括通信连接的中心数据库和医生终端；
    所述中心数据库配置成接收来自心肺和谐系列指标评测装置的被测者数据，其中所述被测者数据包括被测者的心肺和谐系列指标；
    所述中心数据库还配置成接收医生终端的操作指令，根据所述心肺和谐系列指标和所述操作指令生成检测报告。
PCT/CN2019/113393 2019-08-27 2019-10-25 一种心肺和谐系列指标评测方法、装置及系统 WO2021035924A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910799044.8A CN110495862A (zh) 2019-08-27 2019-08-27 一种心肺和谐系列指标评测方法、装置及系统
CN201910799044.8 2019-08-27

Publications (1)

Publication Number Publication Date
WO2021035924A1 true WO2021035924A1 (zh) 2021-03-04

Family

ID=68590135

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/113393 WO2021035924A1 (zh) 2019-08-27 2019-10-25 一种心肺和谐系列指标评测方法、装置及系统

Country Status (2)

Country Link
CN (1) CN110495862A (zh)
WO (1) WO2021035924A1 (zh)

Families Citing this family (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110801221B (zh) * 2019-12-09 2021-04-27 中山大学 基于无监督特征学习的睡眠呼吸暂停片段检测设备
CN111000541A (zh) * 2019-12-25 2020-04-14 南京宁康中科医疗技术有限公司 个性化深呼吸训练处方的制定和实施方法、系统及装置
CN112674737A (zh) * 2020-11-13 2021-04-20 河北省药品医疗器械检验研究院 一种心肺功能监测系统
CN113143271A (zh) * 2020-12-28 2021-07-23 中科宁心电子科技(南京)有限公司 一种孕产妇心理健康评测和深呼吸情绪调理系统
CN112914536B (zh) * 2021-03-24 2023-08-15 平安科技（深圳）有限公司 运动状态的检测方法、装置、计算机设备和存储介质
CN114098756A (zh) * 2021-11-08 2022-03-01 杭州菲诗奥医疗科技有限公司 一种基于单通道ecg信号的心肺耦合分析方法
CN115299887B (zh) * 2022-10-10 2023-01-03 安徽星辰智跃科技有限责任公司 一种动态代谢功能的检测量化方法和系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050267362A1 (en) * 2004-05-17 2005-12-01 Beth Israel Deaconess Medical Center Assessment of sleep quality and sleep disordered breathing based on cardiopulmonary coupling
CN103876733A (zh) * 2014-03-12 2014-06-25 西安交通大学 用于心肺系统相位同步分析的系统和方法
CN105496377A (zh) * 2014-10-08 2016-04-20 吴健康 一种心率变异生物反馈锻炼系统方法和设备
CN109620179A (zh) * 2018-12-26 2019-04-16 南京茂森电子技术有限公司 一种呼吸对心率调制的定量分析方法及相关系统

Family Cites Families (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5964788A (en) * 1997-10-28 1999-10-12 Pacesetter, Inc. Method and apparatus for controlling a pacemaker using respiration
WO2015189304A1 (en) * 2014-06-12 2015-12-17 Koninklijke Philips N.V. Resting heart rate monitor system
CN104840191A (zh) * 2015-04-30 2015-08-19 吴健康 测试心脏运动功能的设备、系统和方法
CN105147268A (zh) * 2015-06-15 2015-12-16 辛勤 一种健康监测方法及装置
CN106166065B (zh) * 2016-06-22 2019-12-17 天津理工大学 一种基于社交网络的可穿戴心电健康交互平台及其实现方法
CN109620209B (zh) * 2018-12-31 2023-12-19 南京茂森电子技术有限公司 一种动态心电、呼吸和运动监测系统和方法
CN109846694B (zh) * 2019-04-10 2023-10-13 军事科学院系统工程研究院卫勤保障技术研究所 在人工通气间期进行心电节律辨识的心肺复苏一体化装置及方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20050267362A1 (en) * 2004-05-17 2005-12-01 Beth Israel Deaconess Medical Center Assessment of sleep quality and sleep disordered breathing based on cardiopulmonary coupling
CN103876733A (zh) * 2014-03-12 2014-06-25 西安交通大学 用于心肺系统相位同步分析的系统和方法
CN105496377A (zh) * 2014-10-08 2016-04-20 吴健康 一种心率变异生物反馈锻炼系统方法和设备
CN109620179A (zh) * 2018-12-26 2019-04-16 南京茂森电子技术有限公司 一种呼吸对心率调制的定量分析方法及相关系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
WU JIANKANG: "Cardiac rehabilitation assessment and personalized intervention based on autonomic balance", 7 July 2019 (2019-07-07), pages 1 - 7, XP055786595, Retrieved from the Internet <URL:http://www.365heart.com/show/136730.shtml> *

Also Published As

Publication number Publication date
CN110495862A (zh) 2019-11-26

Similar Documents

Publication Publication Date Title
WO2021035924A1 (zh) 一种心肺和谐系列指标评测方法、装置及系统
US7862515B2 (en) Apparatus for detecting sleep apnea using electrocardiogram signals
US9265430B2 (en) Method, system and software product for the measurement of heart rate variability
KR101656611B1 (ko) 무구속적으로 측정한 생체신호를 이용하여 산소탈포화지수를 획득하는 방법
EP2654560B1 (en) Heart failure detection with a sequential classifier
US8700137B2 (en) Cardiac performance monitoring system for use with mobile communications devices
WO2021129067A1 (zh) 个性化深呼吸训练处方的制定和实施方法、系统及装置
US11395634B2 (en) Estimating physiological states based on changes in CRI
US20150057512A1 (en) Wearable heart failure monitor patch
WO2020006845A1 (zh) 一种健康储备评估方法、装置及其应用
CN105455797B (zh) 自主神经心脏调控功能测量方法和设备
CN109475311A (zh) 确定受试者周期性生理过程频率的方法以及确定受试者周期性生理过程频率的装置和系统
JP3190388U (ja) 生理信号を収集する改良装置
CN103892812B (zh) 匹配度分析技术
WO2016061545A1 (en) Assessing effectiveness of cpr
WO2020133536A1 (zh) 一种睡眠状态判断的方法及装置
TWI225394B (en) Method and device for analysis of heart rate variability (HRV)
CN112568876A (zh) 基于lstm多生理参数的老年人健康预测系统及预测方法
GB2469547A (en) Measurement of heart rate variability
CN105266764B (zh) 一种中医宗气测评装置
Surrel et al. Low-power wearable system for real-time screening of obstructive sleep apnea
Gavriel et al. Smartphone as an ultra-low cost medical tricorder for real-time cardiological measurements via ballistocardiography
CN113143271A (zh) 一种孕产妇心理健康评测和深呼吸情绪调理系统
Dong et al. An integrated framework for evaluation on typical ECG-derived respiration waveform extraction and respiration
CN115024716A (zh) 基于心率标签生成的心冲击图信号重建方法

Legal Events

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

Ref document number: 19942637

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19942637

Country of ref document: EP

Kind code of ref document: A1