WO2020024312A1

WO2020024312A1 - Breathing signal extraction method, apparatus, processing device and system

Info

Publication number: WO2020024312A1
Application number: PCT/CN2018/099254
Authority: WO
Inventors: 叶飞
Original assignee: 深圳市大耳马科技有限公司
Priority date: 2018-08-03
Filing date: 2018-08-07
Publication date: 2020-02-06
Also published as: CN109009023A

Abstract

The present invention is applicable to the field of medicine and provides a breathing signal extraction method, apparatus, processing device and system. The method comprises: acquiring a waveform of a monitoring signal for cardiac impulses; acquiring a time interval between the same feature events according to the waveform of the monitoring signal for cardiac impulses, and acquiring a breathing waveform according to the change in the time interval between the same feature events over time. The present invention can prevent a breathing signal from being affected or even distorted due to a weak breathing signal or external low-frequency disturbance under certain scenarios, and thus can acquire the breathing signal more accurately. Also, an exhaling process and an inhaling process can be determined directly from the rise and fall of a breathing waveform acquired by means of the present invention, and a breathing signal can be more conveniently combined with cardiac parameters for clinical analysis and calculation, meeting more clinical requirements.

Description

Extraction method, device, processing equipment and system for breathing signals

Technical field

The invention belongs to the medical field, and particularly relates to a method, a device, a processing device and a system for extracting respiratory signals.

Background technique

The sensor can sense and collect the vibration data signal of the body. The original vibration signal collected by the sensor usually includes the body's heart beat signal, breathing signal, environmental micro-vibration signal, interference signal caused by body movement, and the circuit's own noise signal. If the breathing signal is obtained from the original vibration signal, the original vibration signal needs to be pre-processed (such as filtering, etc.) to capture the breathing waveform.

Because the sensor is sensitive to pressure changes caused by changes in vibration displacement, the pressure changes caused by the exhalation and inhalation of the body are related to the measurement position of the sensor. The breathing waveforms that may be obtained at different positions are different, making it difficult to judge the breath from the breathing waveform. And inspiratory band. In addition, the breathing waveform may be very weak or distorted under the influence of external low-frequency disturbances in some scenarios.

Therefore, it is difficult to determine the actual expiratory and inspiratory process by using the above methods to obtain the respiratory waveform. On the other hand, in some clinical scenarios, for example, the ratio of expiratory and inspiratory time needs to be calculated, and the expiratory and inspiratory time needs to be analyzed. When it is necessary to analyze the characteristics of the systolic time of exhalation and inspiratory time, it is difficult to meet the actual analysis and calculation requirements.

technical problem

The purpose of the present invention is to provide a method, a device, a computer-readable storage medium, a processing device, and a system for extracting respiratory signals, which are intended to solve the difficulty in determining the actual exhalation and inhalation process in the prior art method for obtaining respiratory waveforms. The breathing waveform in the scene may be very weak or distorted due to the influence of external low-frequency disturbances.

Technical solutions

In a first aspect, the present invention provides a method for extracting a breathing signal, the method comprising:

Obtain the waveform of the heart beat monitoring signal;

Obtain the time interval of the same characteristic event according to the waveform of the heart beat monitoring signal, and obtain the breathing waveform according to the time interval of the same characteristic event.

In a second aspect, the present invention provides a device for extracting respiratory signals, the extraction device comprising:

An acquisition module for acquiring a waveform of a heart beat monitoring signal; and

The breathing waveform acquisition module is configured to obtain the time interval of the same characteristic event according to the waveform of the heart beat monitoring signal, and obtain the breathing waveform according to the time interval of the same characteristic event.

In a third aspect, the present invention provides a computer-readable storage medium that stores a computer program that, when executed by a processor, implements the steps of the breathing signal extraction method described above.

In a fourth aspect, the present invention provides a breathing signal extraction processing device, including:

One or more processors;

Memory; and

One or more computer programs, wherein the one or more computer programs are stored in the memory and configured to be executed by the one or more processors, which are implemented when the processors execute the computer programs The steps of the method for extracting a breathing signal as described above.

In a fifth aspect, the present invention provides a respiratory signal extraction system, the extraction system includes:

A generation module configured to generate a waveform of a heart beat monitoring signal; and

An extraction processing device connected to the generation module, such as the above-mentioned breathing signal.

Beneficial effect

In the present invention, the waveform of the heart beat monitoring signal is obtained, and then the time interval of the same characteristic event is obtained according to the waveform of the heart beat monitoring signal, and the breathing waveform is obtained according to the time interval of the same characteristic event. Therefore, it is possible to prevent the breathing signal from being affected or even distorted due to the weak breathing signal or the external low-frequency disturbance in some scenes. The present invention can more accurately obtain the breathing signal, and the rise or fall of the breathing waveform obtained by the invention can directly determine the call In the process of inhalation or inhalation, it is more convenient to combine the respiratory signal with the parameters related to the heart for clinical analysis and calculation to meet more clinical needs.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a flowchart of a method for extracting a breathing signal according to a first embodiment of the present invention.

FIG. 2 is a schematic diagram of generating a time-domain waveform of a BCG signal based on an original vibration signal, wherein FIG. 2 (a) is a schematic diagram of an original vibration signal waveform, and FIG. 2 (b) is a schematic diagram of a time-domain waveform of the BCG signal.

FIG. 3 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the BCG signal, and obtaining the breathing waveform according to the change of the time interval of the same characteristic event with time. FIG. 3 (a) is a diagram of the BCG signal. Schematic diagram of time domain waveforms, Fig. 3 (b) is a schematic diagram of time intervals of the same characteristic events, and Fig. 3 (c) is a schematic diagram of respiratory waveforms.

Figure 4 shows the time-domain waveform of the BCG signal generated from the original vibration signal. The waveform of the BCG signal obtained by the second-order differential transformation is a schematic diagram, where Figure 4 (a) is a schematic diagram of the original vibration signal waveform, and Figure 4 (b) is A schematic diagram of the time domain waveform of the BCG signal. FIG. 4 (c) is a schematic diagram of the waveform of the BCG signal obtained through the second-order differential transformation.

FIG. 5 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peaks of the waveform of the second-order differential transformation of the BCG signal, and obtaining the breathing waveform according to the time interval of the same characteristic event. FIG. 5 ( a) is a waveform diagram of the BCG signal obtained by second-order differential transformation. FIG. 5 (b) is a time interval diagram of the same characteristic event, and FIG. 5 (c) is a breathing waveform diagram.

FIG. 6 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the ECG signal, and obtaining the breathing waveform according to the change of the time interval of the same characteristic event with time. Figure 6 (b) shows the time interval of the same characteristic event, and Figure 6 (c) shows the respiratory waveform.

FIG. 7 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the PPG signal, and obtaining the breathing waveform according to the change of the time interval of the same characteristic event with time, wherein FIG. 7 (a) is the waveform of the PPG signal A schematic diagram, FIG. 7 (b) is a time interval diagram of the same characteristic event, and FIG. 7 (c) is a breathing waveform diagram.

FIG. 8 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the SCG signal, and obtaining the breathing waveform according to the change of the time interval of the same characteristic event with time, wherein FIG. 8 (a) is the waveform of the SCG signal 8 (b) is a time interval diagram of the same characteristic event, and FIG. 8 (c) is a breathing waveform diagram.

FIG. 9 is a functional block diagram of a breathing signal extraction device provided by Embodiment 2 of the present invention.

FIG. 10 is a specific structural block diagram of a breathing signal extraction processing device provided in Embodiment 4 of the present invention.

FIG. 11 is a specific structural block diagram of a breathing signal extraction system provided by Embodiment 5 of the present invention.

Best Mode of the Invention

In order to make the objectives, technical solutions, and beneficial effects of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

In order to explain the technical solution of the present invention, the following description is made through specific embodiments.

Embodiment one:

Please refer to FIG. 1. The breathing signal extraction method provided by the first embodiment of the present invention includes the following steps: It should be noted that if there is substantially the same result, the breathing signal extraction method of the present invention is not as shown in FIG. Process sequence is limited.

S101. Obtain a waveform of a heart beat monitoring signal.

In the first embodiment of the present invention, the heartbeat monitoring signals may be Ballistocardiogram (BCG) signals, Electrocardiogram (ECG) signals, Phonocardiogram (PCG) signals, and Seismocardiogram (SCG) signals. Photoelectric volume pulse wave (Photoplethysmograph, PPG).

When the heart beat monitoring signal is a BCG signal, a PCG signal or an SCG signal, the heart beat monitoring signal is obtained through a vibration sensor. When the heart beat monitoring signal is an ECG signal, the heart beat monitoring signal is obtained through an electrocardiograph; when the heart beat monitoring signal is a PPG signal, the heart beat monitoring signal is obtained through a PPG signal collector.

In the first embodiment of the present invention, the vibration sensor may be an acceleration sensor, a speed sensor, a displacement sensor, a pressure sensor, a strain sensor, or a sensor that converts the equivalent of a physical quantity equivalently based on acceleration, speed, pressure, or displacement (for example, Electrostatic charge sensitive sensors, inflatable micro-motion sensors, radar sensors, etc.). The strain sensor may be an optical fiber strain sensor.

The vibration sensor can be placed on a contact surface behind a human lying on his back, a contact surface behind a human lying on his back within a predetermined range of inclination angles, a contact surface behind a human lying on a wheelchair or other reclining object, and the like.

The living body may be a living body performing vital sign signal monitoring. For example, hospital patients, caregivers (such as elderly, imprisoned, etc.). Generally, in order to ensure the quality of the collected original vibration signals, the body needs to be measured in a quiet state.

The waveform of the heart beat monitoring signal represents a signal that is detected and recorded by a vibration sensor when the heart beat process causes the body to vibrate.

When the heartbeat monitoring signal is obtained through a vibration sensor, S101 may specifically be: filtering and scaling the original vibration signal obtained by the vibration sensor to generate a heartbeat monitoring signal waveform.

When filtering the original vibration signal, one or more of IIR filter, FIR filter, wavelet filter, zero-phase bidirectional filter, polynomial fitting smoothing filter, etc. can be used according to the requirements of the filtered signal characteristics. Combination to filter and denoise the original vibration signal.

When filtering the original vibration signal, the following steps may also be included:

Determine whether the original vibration signal carries a power frequency interference signal, and if so, filter the power frequency noise through a power frequency notch.

Fig. 2 is a schematic diagram of generating a time-domain waveform of a BCG signal based on an original vibration signal. Each waveform has obvious characteristics and good consistency, regular periodicity, clear outline, and stable baseline. The “J” peak of the BCG signal can be extracted from the data characteristics. The "J" peak has the following characteristics: a relatively narrow spike, and a sharper falling edge after the spike ends, gradually decreasing to the lowest point of the waveform of the heartbeat. The "J" peak represents the maximum of one of acceleration, pressure, and displacement that the vibration effect caused by ejection is sensitive to by the vibration sensor. The heartbeat is divided by the "J" peaks of the filtered BCG signal. This data includes a total of 30 "J" peaks from 1 to 30.

S102. Obtain the time interval of the same characteristic event according to the waveform of the heart beat monitoring signal, and obtain the breathing waveform according to the time interval of the same characteristic event.

In the first embodiment of the present invention, S102 may specifically be:

Obtain the time interval of the same characteristic event according to the same characteristic peak / valley of the waveform of the heart beat monitoring signal or the waveform of the same characteristic peak / valley after the waveform of the heart beat monitoring signal is transformed, and according to the time interval of the same characteristic event, The change in time acquires the breathing waveform.

The waveform of the heart beat monitoring signal after transformation may be: the time interval of the heart beat monitoring signal waveform that does not affect the same characteristic events on its time domain signal through integral transformation, differential transformation (such as second-order differential transformation), etc. The waveform of the transformation of the distribution characteristics.

Obtaining a breathing waveform according to the change of the time interval of the same characteristic event over time may be: based on the change of the time interval of the same characteristic event over time, extracting the breathing waveform using linear interpolation, cubic spline fitting, polynomial fitting, and the like.

The time interval for obtaining the same characteristic event according to the same characteristic peak / valley of the waveform of the heart beat monitoring signal, or the waveform of the same characteristic peak / valley of the transformed waveform according to the waveform of the heart beat monitoring signal may specifically be:

Based on the waveform of the heart beat monitoring signal or the waveform of the heart beat monitoring signal after transformation, each identical characteristic peak / valley is detected, and the time interval between each identical characteristic peak / valley and the same characteristic peak / valley of the adjacent preamble is calculated. The time interval is taken as the time interval of the same characteristic event corresponding to the same characteristic peak / valley.

FIG. 3 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the BCG signal, and obtaining the breathing waveform according to the time interval of the same characteristic event. Zh

It can be seen from FIG. 3 that the “J” peak of each heartbeat can be easily detected based on the BCG signal. Affected by the cardiopulmonary coupling effect, the "J" peak also exhibits a "high and low" contour along with the exhalation and inspiration process. As shown in Figure 3, after detecting each "J" peak based on the BCG signal, the time interval between each "J" peak and the adjacent pre-order "J" peak is calculated, that is, the "JJ" time interval, that is, the same characteristic event. time interval. Each time interval has been identified in Figure 3. The "J-J" time interval at the first "J" peak is the time interval with the adjacent pre-order "J" peak. The time at which each "J" peak is located is taken as the abscissa, and the "J-J" time interval is taken as the ordinate, and the time interval change of the same characteristic event that changes with time is plotted. The respiratory waveform shown in Figure 3 was extracted based on cubic spline fitting. Compared with the breathing profile of the original vibration signal, the frequency of the extracted breathing waveform is basically the same. The basic parameter breathing frequency can be calculated, and according to the influence of the cardiopulmonary coupling, it can be judged that the waveform goes to the lower part as the inhalation process, and the waveform goes to the higher part For the exhalation process, you can do more calculations and analysis based on the signal during the exhalation / inhalation phase. Here, the “J” peak is used as the characteristic peak for heartbeat division. The “I” valley and the “K” valley on the left and right sides of the “J” peak, the characteristic peak and valley groups (MC, AVO, etc.) corresponding to the systole, The peaks and valleys (AVC, MO, etc.) corresponding to the diastolic phase and other distinctive peaks and valleys are used for classification.

As shown in FIG. 4, a time-domain waveform of a BCG signal is generated based on the original vibration signal, and a waveform obtained by subjecting the BCG signal to second-order differential transformation is illustrated. The waveform obtained by the second-order differential transformation also has obvious characteristic peaks, which are long and narrow and are significantly higher than other peaks in a single heartbeat, such as the "J ^" peak.

FIG. 5 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the BCG signal subjected to the second-order differential transformation, and obtaining the breathing waveform according to the time interval of the same characteristic event. As shown in FIG. 5, after detecting each “J ^” peak based on the waveform of the second-order differential transformation of the BCG signal, the time interval between each “J ^” peak and the adjacent pre-order “J ^” peak is calculated, that is, "J ^-J ^" time interval, that is, the time interval of the same characteristic event. Each time interval has been identified in FIG. 5. The "J ^ -J ^" time interval at the first "J ^" peak is the time interval with the adjacent pre-order "J ^" peak. The time at which each "J ^" peak is located is taken as the abscissa, and the "J ^-J ^" time interval is taken as the ordinate, and the time interval change of the same characteristic event that changes with time is plotted. The breathing waveform shown in Figure 5 was extracted based on cubic spline fitting. Compared with the breathing profile of the original vibration signal, the frequency of the extracted breathing waveform is basically the same. The basic parameter breathing frequency can be calculated, and according to the influence of the cardiopulmonary coupling, it can be judged that the waveform goes to the lower part as the inhalation process, and the waveform goes to the higher part. For the exhalation process, you can do more calculations and analysis based on the signal during the exhalation / inhalation phase. Here, the “J ^” peak is used as a characteristic peak to perform heartbeat division, and the signal characteristics may also be used to perform division based on other peaks or valleys with obvious characteristics.

FIG. 6 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the ECG signal, and obtaining the breathing waveform according to the time interval of the same characteristic event. Zh

It can be seen from Fig. 6 that the most obvious characteristic peak is a narrow and tall "R" peak. The “R” peak of each heartbeat can be easily detected based on the ECG signal. Affected by the cardiopulmonary coupling effect, the "R" peak also presents a "high and low" profile along with the exhalation and inspiration process. As shown in Figure 6, after detecting each "R" peak based on the ECG signal, the time interval between each "R" peak and the adjacent preamble "R" peak is calculated, that is, the "R-R" time interval, that is, the same characteristics The time interval of the event. Each time interval has been identified in Figure 6. The "R-R" time interval at the first "R" peak is the time interval with the adjacent pre-order "R" peak. The time at which each "R" peak is located is taken as the abscissa, and the "R-R" time interval is taken as the ordinate, and the time interval change of the same characteristic event that changes with time is plotted. The breathing waveform shown in Figure 6 was extracted based on cubic spline fitting. Compared with the breathing profile of the original vibration signal, the frequency of the extracted breathing waveform is basically the same. The basic parameter breathing frequency can be calculated, and according to the influence of the cardiopulmonary coupling, it can be judged that the waveform goes to the lower part as the inhalation process, and the waveform goes to the higher part. For the exhalation process, you can do more calculations and analysis based on the signal during the exhalation / inhalation phase. Here, the “R” peak is used as a characteristic peak to perform heartbeat division, and it may also be classified based on other peaks or valleys with obvious characteristics according to signal characteristics. The ECG signal can also be transformed by integral transformation, differential transformation (such as second-order differential transformation), and other transformation methods that do not affect the distribution characteristics of the time interval of the same characteristic event on its time domain signal, and then based on the characteristics of the transformed ECG signal Peaks and valleys are divided into heartbeats. The ECG signal shown in FIG. 6 is a lead II signal collected through five leads, and other lead signals can be analogized based on this embodiment.

FIG. 7 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the PPG signal, and obtaining the breathing waveform according to the time interval of the same characteristic event. Zh

It can be seen from FIG. 7 that the most obvious characteristic peak is the narrow and tall “P” peak. It is easy to detect the “P” peak of each heartbeat based on the PPG signal. Affected by the cardiopulmonary coupling effect, the "P" peak also presents a "high and low" contour along with the exhalation and inspiration process. As shown in Fig. 7, after detecting each "P" peak based on the PPG signal, the time interval between each "P" peak and the adjacent previous "P" peak is calculated, that is, the "P-P" time interval, that is, the same characteristics The time interval of the event. Each time interval has been identified in Figure 7. The "P-P" time interval at the first "P" peak is the time interval from the adjacent "P" peak. The time at which each "P" peak is located is taken as the abscissa, and the "P-P" time interval is taken as the ordinate, and the time interval change of the same characteristic event that changes with time is plotted. The breathing waveform shown in Figure 7 was extracted based on cubic spline fitting. Compared with the breathing profile of the original vibration signal, the frequency of the extracted breathing waveform is basically the same. The basic parameter breathing frequency can be calculated, and according to the influence of the cardiopulmonary coupling, it can be judged that the waveform goes to the lower part as the inhalation process, and the waveform goes to the higher part. For the exhalation process, you can do more calculations and analysis based on the signal during the exhalation / inhalation phase. Here, the “P” peak is used as a characteristic peak to perform heartbeat division, and it can also be divided based on other peaks or valleys with obvious characteristics according to the signal characteristics. The PPG signal can also be transformed by integral transformation, differential transformation (such as second-order differential transformation), and other transformation methods that do not affect the distribution characteristics of the time interval of the same characteristic event on its time domain signal, and then based on the characteristics of the transformed PPG signal Peaks and valleys are divided. The PPG signal shown in FIG. 7 is a PPG signal collected by a finger, and signals of other parts can be analogized based on this embodiment.

FIG. 8 is a schematic diagram of obtaining the time interval of the same characteristic event according to the characteristic peak of the waveform of the SCG signal, and obtaining the breathing waveform according to the time interval of the same characteristic event. Zh

It can be seen from Fig. 8 that the most obvious characteristic peak is the narrow and tall "S" peak. It is easy to detect the "S" peak of each heartbeat based on the SCG signal. Affected by the cardiopulmonary coupling effect, the "S" peak also presents a "high and low" contour along with the exhalation and inspiration process. As shown in FIG. 8, after detecting each “S” peak based on the SCG signal, the time interval between each “S” peak and the adjacent “S” peak is calculated, that is, the “S-S” time interval, that is, the same characteristics The time interval of the event. Each time interval has been identified in Figure 8. The "S-S" time interval at the first "S" peak is the time interval from the adjacent "S" peak. The time at which each “S” peak is located is taken as the abscissa, and the “S-S” time interval is taken as the ordinate, and the time interval change of the same characteristic event that changes with time is plotted. The respiratory waveform shown in Figure 8 was extracted based on cubic spline fitting. Compared with the breathing profile of the original vibration signal, the frequency of the extracted breathing waveform is basically the same. The basic parameter breathing frequency can be calculated, and according to the influence of the cardiopulmonary coupling, it can be judged that the waveform goes to the lower part as the inhalation process, and the waveform goes to the higher part. For the exhalation process, you can do more calculations and analysis based on the signal during the exhalation / inhalation phase. Here, the “S” peak is used as a characteristic peak to perform heartbeat division, and it may also be classified based on other peaks or valleys with obvious characteristics according to signal characteristics. The SCG signal can also be transformed by integral transformation, differential transformation (such as second-order differential transformation), and other transformation methods that do not affect the distribution characteristics of the time interval of the same characteristic event on its time domain signal, and then based on the characteristics of the transformed SCG signal Peaks and valleys are divided. The SCG signal shown in FIG. 8 is an SCG signal collected through the apex of the chest when lying down, and the signals of other parts can be analogized based on this embodiment.

Embodiment two:

Referring to FIG. 9, a breathing signal extraction device provided in Embodiment 2 of the present invention includes:

An acquisition module 21 for acquiring a waveform of a heart beat monitoring signal;

The respiratory waveform acquisition module 22 is configured to acquire the time interval of the same characteristic event according to the waveform of the heart beat monitoring signal, and acquire the respiratory waveform according to the time interval of the same characteristic event.

The breathing signal extraction device provided in the second embodiment of the present invention and the breathing signal extraction method provided in the first embodiment of the present invention belong to the same concept, and the specific implementation process thereof is detailed in the entire description, and will not be repeated here.

Embodiment three:

Embodiment 3 of the present invention provides a computer-readable storage medium. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the breathing signal extraction method provided by Embodiment 1 of the present invention is implemented. A step of.

Embodiment 4:

As shown in FIG. 10, Embodiment 4 of the present invention provides a breathing signal extraction processing device 100. The breathing signal extraction processing device 100 includes: one or more processors 101, a memory 102, and one or more A computer program in which the processor 101 and the memory 102 are connected by a bus, the one or more computer programs are stored in the memory 102 and configured to be executed by the one or more processors 101 When the processor 101 executes the computer program, the steps of the method for extracting the breathing signal provided by the first embodiment of the present invention are implemented.

Embodiment 5:

Referring to FIG. 11, a breathing signal extraction system provided in Embodiment 5 of the present invention includes:

A generation module 11 configured to generate a waveform of a heart beat monitoring signal; and

The breathing signal extraction and processing device 100, which is connected to the generating module, is provided by the fourth embodiment of the present invention.

In the fifth embodiment of the present invention, when the heartbeat monitoring signal is a BCG signal, a PCG signal or an SCG signal, the generating module is a vibration sensor; when the heartbeat monitoring signal is an ECG signal, the generating module is an electrocardiograph; when the heartbeat monitoring is When the signal is a PPG signal, the generating module is a PPG signal collector.

A person of ordinary skill in the art may understand that all or part of the steps in the various methods of the foregoing embodiments may be implemented by a program instructing related hardware. The program may be stored in a computer-readable storage medium. The storage medium may include: Read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disks or optical disks, etc.

The above description is only the preferred embodiments of the present invention, and is not intended to limit the present invention. Any modification, equivalent replacement, and improvement made within the spirit and principle of the present invention shall be included in the protection of the present invention. Within range.

Claims

A method for extracting a breathing signal, wherein the method includes:

Obtain the waveform of the heart beat monitoring signal;

Obtain the time interval of the same characteristic event according to the waveform of the heart beat monitoring signal, and obtain the breathing waveform according to the time interval of the same characteristic event.
The method according to claim 1, wherein the heartbeat monitoring signal is an ECG signal or a PPG signal.
The method according to claim 1, wherein the heartbeat monitoring signal is a BCG signal, a PCG signal, or an SCG signal.
The method according to claim 3, wherein the heartbeat monitoring signal is obtained through a vibration sensor, and the vibration sensor is an acceleration sensor, a speed sensor, a displacement sensor, a pressure sensor, a strain sensor, or an acceleration or a speed. One or more of the sensors that convert the equivalent of a physical quantity based on pressure, pressure, or displacement.
The method according to claim 4, wherein the vibration sensor is placed on a contact surface behind a human body lying supine, a contact surface behind a human body lying supine within a predetermined tilt angle range, or a reclining human body that can lean on an object Contact surface behind.
The method according to claim 4, wherein the acquiring the waveform of the heart beat monitoring signal is specifically: filtering and scaling the original vibration signal obtained by the vibration sensor to generate a heart beat monitoring signal waveform.
The method according to claim 6, characterized in that when filtering the original vibration signal, an IIR filter, a FIR filter, a wavelet filter, a zero-phase bidirectional filter, a polynomial simulation is used according to the requirements of the filtered signal characteristics. Combine one or more of the smoothing filters to filter and denoise the original vibration signal.
The method according to claim 6, further comprising: when filtering the original vibration signal:

Determine whether the original vibration signal carries a power frequency interference signal, and if so, filter the power frequency noise through a power frequency notch.
The method according to claim 1, wherein the obtaining the time interval of the same characteristic event according to the waveform of the heart beat monitoring signal, and obtaining the breathing waveform according to the change of the time interval of the same characteristic event with time are:

Obtain the time interval of the same characteristic event according to the same characteristic peak / valley of the waveform of the heart beat monitoring signal or the waveform of the same characteristic peak / valley after the waveform of the heart beat monitoring signal is transformed, and according to the time interval of the same characteristic event, The change in time acquires the breathing waveform.
The method according to claim 9, wherein the waveform of the waveform of the heart beat monitoring signal is transformed: the waveform of the heart beat monitoring signal does not affect the time of each of the same characteristic events on its time domain signal Waveform of interval distribution feature transformation.
The method according to claim 10, wherein the transformation mode of the distribution characteristics that does not affect the time interval of each of the same characteristic events on the time domain signal is an integral transformation or a differential transformation.
The method according to claim 1, wherein the obtaining a breathing waveform according to a change in time interval of the same characteristic event is based on a change in time interval of the same characteristic event using time, using linear interpolation, cubic spline Respiratory waveform extracted by fitting or polynomial fitting.
The method according to claim 9, characterized in that the same characteristic peak / valley of the waveform based on the waveform of the heart beat monitoring signal, or the same characteristic peak / valley of the waveform transformed according to the waveform of the heart beat monitoring signal obtains the same The time interval of the characteristic event is specifically:

Based on the waveform of the heart beat monitoring signal or the waveform of the heart beat monitoring signal after transformation, each identical characteristic peak / valley is detected, and the time of each identical characteristic peak / valley and the same characteristic peak / valley of adjacent preambles is calculated Interval, and the time interval is used as the time interval of the same characteristic event corresponding to the same characteristic peak / valley.
A breathing signal extraction device, characterized in that the extraction device includes:

An acquisition module for acquiring a waveform of a heart beat monitoring signal; and

The breathing waveform acquisition module is configured to obtain the time interval of the same characteristic event according to the waveform of the heart beat monitoring signal, and obtain the breathing waveform according to the time interval of the same characteristic event.
A computer-readable storage medium storing a computer program, wherein when the computer program is executed by a processor, extraction of the breathing signal according to any one of claims 1 to 13 is performed Method steps.
A breathing signal extraction processing device includes:

One or more processors;

Memory; and

One or more computer programs, wherein the one or more computer programs are stored in the memory and configured to be executed by the one or more processors, characterized in that the processors execute the The computer program implements the steps of the breathing signal extraction method according to any one of claims 1 to 13.
A respiratory signal extraction system, characterized in that the extraction system includes:

A generation module configured to generate a waveform of a heart beat monitoring signal; and

The breathing signal extraction and processing device according to claim 16 connected to the generating module.