CN115414026A - Automatic breath segmentation method and system based on flow velocity waveform - Google Patents

Abstract

The invention discloses a method, system, device and computer-readable storage medium for automatic breathing segmentation based on a flow velocity waveform. The method includes: acquiring the respiratory waveform data of a subject to be tested; performing preprocessing on the respiratory waveform data to obtain The preprocessed respiratory waveform data; cutting the preprocessed respiratory waveform data into respiratory data with multiple respiratory cycles to obtain preprocessed respiratory data with multiple respiratory cycles; The above segmentation is to find all points where the flow velocity crosses zero and the derivative is greater than zero; perform feature extraction on the preprocessed respiratory data with multiple respiratory cycles, and obtain the respiratory data with multiple respiratory cycles after feature extraction as Feature data: input the feature data of the subject into the classification model to obtain the classification result of the feature data.

Description

A method and system for automatic breathing segmentation based on flow velocity waveform

technical field

The present invention relates to the field of respiratory monitoring, and more particularly, to a method and system for automatic respiratory segmentation based on flow velocity waveforms.

Background technique

Mechanical ventilation refers to the use of a ventilator to replace or assist the work of the respiratory muscles when the respiratory organs cannot maintain normal gas exchange, that is, when respiratory failure occurs. Mechanical ventilation strives for treatment time and creates conditions for clinical respiratory failure caused by various reasons, as well as other diseases that require respiratory function support.

Invasive mechanical ventilation is an important treatment for critically ill patients with respiratory failure. However, inappropriate invasive mechanical ventilation may cause ventilator-related lung injuries such as barotrauma and volutrauma. Proper monitoring during mechanical ventilation is helpful. Helps avoid adverse reactions to mechanical ventilation. During the implementation of invasive mechanical ventilation monitoring, the collected flow velocity and pressure waveforms are usually continuous waveforms. However, most monitoring parameters, such as airway peak pressure, airway plateau pressure, and positive end-expiratory pressure, are based on single-mouth breathing. Therefore, it is very important to accurately identify and segment each breath in the continuous ventilator waveform data, and the segmentation results directly affect the acquisition of monitoring parameters during invasive mechanical ventilation.

In addition, the patient-ventilator asynchrony (PVA) phenomenon caused by the mismatch between the magnitude and phase of the patient's needs and the help provided by the ventilator during mechanical ventilation will also cause harm to the patient. It helps the identification and classification of PVA, thereby improving the accuracy of the PVA automatic identification algorithm and making it possible to give auxiliary treatment decisions for different PVA types.

During invasive mechanical ventilation treatment, the ventilator mainly relies on the opening and closing of the inspiratory valve, the exhalation valve and the data monitored by the flow sensor to determine inhalation and exhalation. A complete inspiratory process plus a complete breathing process It will be judged as a complete breath, the beginning of the inspiratory phase is defined as the beginning of the breath, and the end of the expiratory phase (that is, the beginning of the inspiratory phase of the next breath) is defined as the end of the breath. At present, there is no technology for automatic breathing segmentation based on the flow velocity waveform; for the analysis of offline data, the methods of respiratory segmentation mainly include manual segmentation and respiratory segmentation based on the periodic changes and characteristics of flow velocity or airway pressure waveform . The above method mainly has the following defects: 1. The algorithm and monitoring system that comes with the ventilator can automatically split the breath, but the algorithms of different ventilator manufacturers are different and not public, resulting in the result of automatic splitting of breath between different brands of ventilators Different, not comparable. In addition, the results of automatic segmentation are difficult to export, which brings difficulties for subsequent offline analysis and data processing; 2. The method of manual segmentation has the highest accuracy, but it is time-consuming and laborious. Massive respiratory monitoring generated at the bedside It is difficult to segment the data one by one; 3. Respiration segmentation based on the rules of the flow rate or airway pressure periodic changes and characteristics can achieve a certain degree of automation, but for the presence of waveform disturbances, especially when there is a PVA phenomenon, accurate Insufficient rate.

Contents of the invention

The present invention aims to solve at least one of the technical problems existing in the prior art. For this reason, the present invention provides an automatic breathing segmentation method based on flow velocity waveform, which can realize automatic breathing segmentation based on real-time or offline flow velocity waveform, and effectively ensure the accuracy of breathing segmentation, which is helpful for the identification and classification of PVA , and provide the possibility for adjuvant treatment decisions for different types of PVA.

The present application discloses a breath automatic segmentation method based on flow velocity waveform, including:

Obtain the respiratory waveform data of the subject;

Preprocessing the respiratory waveform data to obtain preprocessed respiratory waveform data; cutting the preprocessed respiratory waveform data into respiratory data with multiple respiratory cycles to obtain preprocessed respiratory waveform data And there are respiratory data of multiple respiratory cycles; the segmentation is to find all flow velocity zero-crossing points and points whose derivatives are greater than zero;

Perform feature extraction on the preprocessed respiratory data with multiple respiratory cycles, and obtain the feature-extracted respiratory data with multiple respiratory cycles as feature data;

The feature data of the subject is input into the classification model to obtain the classification result of the feature data.

The respiratory waveform data includes: respiratory flow rate data based on breathing time;

The pre-processed respiratory waveform data includes: from the respiratory flow velocity waveform of the respiratory flow velocity data based on the breathing point time, find the point where the flow velocity crosses zero and the derivative is greater than zero, and record the time index of the point respectively , the obtained preprocessed respiratory waveform data;

Optionally, the preprocessing is to traverse the respiratory waveform data sequentially, at least once.

The preprocessed respiratory data with multiple respiratory cycles includes: segmenting the adjacent points respectively, and segmenting to form a waveform corresponding to each breath, and obtain the respiratory data of the multiple respiratory cycles;

Optionally, the nth and n+1th adjacent points are respectively taken, and the interval between the nth and n+1th points is defined as a single breathing cycle; wherein, 1≤n<N-1, N is The total number of time indices for all said points.

The characteristic data include: the breathing time interval of a single breathing cycle, the flow rate change of a single breathing cycle, the ratio of the exhalation duration of a single breathing cycle to the total breathing time of a single breathing cycle, and the rising slope of the expiratory waveform of a single breathing cycle;

Optionally, the breathing time interval of the single breathing cycle is: the time interval a between the time indexes respectively corresponding to the adjacent points;

Optionally, the value range of a is 0s<a<1s;

Optionally, the flow velocity change of the single respiratory cycle is: the flow velocity change b between the time indexes respectively corresponding to the adjacent points;

Optionally, the value range of b is 0L/min<b<20L/min;

Optionally, the ratio of the exhalation duration of the single breathing cycle to the total breathing time of the single breathing cycle is: the exhalation duration between the time indexes corresponding to the adjacent points respectively accounts for the ratio of the exhalation duration between the adjacent points The proportion c% of the total breathing time between the time indices corresponding to the points respectively;

Optionally, the range of c% is 20%-80%;

The rising slope of the expiratory waveform of the single respiratory cycle is: the rising slope d of the expiratory waveform between the time indexes corresponding to the adjacent points; optionally, within the first 100ms of inspiration, the d's The value range is 10-50L/min/s.

The method or step of inputting the characteristic data of the subject into the classification model and obtaining the classification result of the characteristic data of the subject comprises:

Input the respiratory time interval of the single respiratory cycle of the subject into the classification model, and judge whether the respiratory time interval of the single respiratory cycle of the subject falls within the range of a; If the respiratory time interval of the single respiratory cycle of the subject falls within the range of a, the output is yes, and enters the stage of inputting the flow rate change of the single respiratory cycle of the subject to the classification model; otherwise, the output is no , the operation is terminated;

Input the flow velocity change of the single respiratory cycle of the subject into the classification model, and judge whether the flow velocity change of the single respiratory cycle of the subject falls within the range of b; In the case that the flow velocity change of the single breathing cycle falls within the range of b, the output is, and enter the ratio of the exhalation duration of the single breathing cycle of the test subject to the total breathing time of the single breathing cycle into the classification the stage of the model; otherwise the output is NO and the run terminates;

Input the ratio of the exhalation duration of the single breathing cycle of the subject to the test to the total breathing time of the single breathing cycle into the classification model, and judge the difference between the exhalation duration of the single breathing cycle of the test subject and the ratio of the total breathing time of the single breathing cycle. Whether the proportion of the total breathing time of the cycle falls within the range of c; when the ratio of the exhalation duration of the single breathing cycle of the subject to the test to the total breathing time of the single breathing cycle falls within the range of c, the output Yes, and enter the stage of adding the rising slope of the expiratory waveform of the single respiratory cycle of the subject to the classification model; otherwise output No, the operation is terminated;

Input the rising slope of the expiratory waveform of the single respiratory cycle of the subject into the classification model, and judge whether the rising slope of the expiratory waveform of the single respiratory cycle of the subject falls within the range of d; When the upward slope of the expiratory waveform of the single respiratory cycle of the test subject falls within the range of d, the output is yes, and the classification result of the test subject's respiratory waveform data is output; otherwise, the output is no, and the operation is terminated.

The respiratory waveform data is respiratory waveform data with a continuous waveform signal.

A device for automatically segmenting breath based on a flow velocity waveform, the device comprising: a memory and a processor;

The memory is used to store program instructions;

The processor is used for invoking program instructions, and when the program instructions are executed, is used for executing the above-mentioned automatic breathing segmentation method based on the flow velocity waveform.

An analysis system for automatic breath segmentation based on flow velocity waveform, including:

an acquisition unit, configured to acquire the respiratory waveform data of the subject;

The first processing unit is configured to preprocess the respiratory waveform data to obtain the preprocessed respiratory waveform data; segment the preprocessed respiratory waveform data into breaths with multiple respiratory cycles Data, obtain the preprocessed respiratory data with multiple respiratory cycles; the segmentation is to find all flow velocity zero-crossing points and points whose derivatives are greater than zero;

The second processing unit is configured to perform feature extraction on the preprocessed respiratory data with multiple respiratory cycles, and obtain the feature-extracted respiratory data with multiple respiratory cycles as feature data;

The classification unit is used to input the characteristic data of the subject into the classification model to obtain the classification result of the characteristic data.

A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned automatic breathing segmentation method based on a flow velocity waveform is realized.

Any of the following applications:

The application of the above-mentioned equipment in the detection of human-machine imbalance;

The application of the above-mentioned equipment in respiratory parameter extraction; optional, after each breath is segmented based on the real-time or offline flow velocity waveform, many characteristic parameters can be extracted from the airway pressure, esophageal pressure, and volume changes of each breath , for the monitoring of respiratory therapy.

The application has the following beneficial effects:

1. This application innovatively discloses an automatic breathing segmentation method based on real-time or offline flow velocity waveforms, which can effectively ensure the accuracy of single-mouth breathing segmentation, standardize the breathing segmentation results, and be comparable; it is helpful for the identification of PVA and classification, so as to improve the accuracy of PVA automatic identification algorithm, and provide the possibility to give auxiliary treatment decisions for different PVA types;

2. The automatic breathing segmentation method proposed by this application overcomes the time-consuming and labor-intensive defects of manual segmentation, and the lack of accuracy when performing breathing segmentation based on the rules and characteristics of the periodic change of flow velocity or airway pressure;

3. This application can automatically segment the respiratory waveform into single-mouth breathing, which is convenient for doctors to quickly identify whether the patient's breathing is normal, greatly shortens the doctor's diagnosis time, and strives for valuable treatment time for ICU patients; in addition, referring to this segmentation rule, you can use a computer The program realizes the automatic segmentation of respiratory waveforms, which facilitates the subsequent identification and monitoring of abnormal respiratory waveforms.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings that need to be used in the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained based on these drawings without creative work.

FIG. 1 is a schematic flow chart of a method for automatically segmenting breath based on a flow velocity waveform provided by an embodiment of the present invention;

Fig. 2 is a schematic diagram of a breath automatic segmentation device based on a flow velocity waveform provided by an embodiment of the present invention;

Fig. 3 is a schematic flow chart of the breath automatic segmentation system based on the flow velocity waveform provided by the embodiment of the present invention;

Fig. 4 is a classification flow chart of the automatic breathing segmentation method based on the flow velocity waveform provided by the embodiment of the present invention;

Fig. 5 is a schematic diagram of the point where the flow velocity crosses zero and the derivative is greater than zero in the breath automatic segmentation method based on the flow velocity waveform provided by the embodiment of the present invention;

Fig. 6 is a schematic diagram of the time interval a of the breath automatic segmentation method based on the flow velocity waveform provided by the embodiment of the present invention;

Fig. 7 is a schematic diagram of the flow velocity change b of the breath automatic segmentation method based on the flow velocity waveform provided by the embodiment of the present invention;

Fig. 8 is a schematic diagram of the ascending slope d of the expiratory waveform in the method of automatic breathing segmentation based on the flow velocity waveform provided by the embodiment of the present invention.

detailed description

In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention.

In some processes described in the specification and claims of the present invention and the above-mentioned drawings, a plurality of operations appearing in a specific order are contained, but it should be clearly understood that these operations may not be performed in the order in which they appear herein Execution or parallel execution, the serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. Additionally, these processes can include more or fewer operations, and these operations can be performed sequentially or in parallel. It should be noted that the descriptions of "first" and "second" in this article are used to distinguish different messages, devices, modules, etc. are different types.

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts belong to the protection scope of the present invention.

Fig. 1 is a schematic flowchart of a method for automatically segmenting breath based on a flow velocity waveform provided by an embodiment of the present invention. Specifically, the method includes the following steps:

101: Obtain the respiratory waveform data of the subject;

In one embodiment, the respiratory waveform data includes: respiratory flow rate data based on the breathing time; the respiratory flow data based on the breathing time is Time (s) on the abscissa and Flow (L/min) on the ordinate The flow velocity waveform data; the breathing time is Time (s), and Time is set according to the sampling frequency hz of the ventilator and the serial number number of the data, and the calculation formula is: Time=(number-1)/hz; number is the sampling point serial number.

In one embodiment, the respiratory waveform data also includes: airway pressure waveform data based on the breathing time; Airway pressure waveform data;

In one embodiment, the respiratory waveform data further includes: esophageal pressure waveform data based on breathing time; the esophageal pressure waveform data is the esophageal pressure waveform whose abscissa is Time, and whose ordinate is esophageal pressure data Pes (cmH2O) data;

In one embodiment, the respiratory waveform data further includes: volumetric waveform data based on respiration time; the volumetric waveform data has Time as the abscissa and Volume (ml) as the ordinate; The formula for calculating volume data is:

102: Perform preprocessing on the respiratory waveform data to obtain preprocessed respiratory waveform data; segment the preprocessed respiratory waveform data into respiratory data with multiple respiratory cycles to obtain preprocessed respiratory waveform data The respiratory data that has been passed and has multiple respiratory cycles; the segmentation is to find all the points where the flow velocity crosses zero (one point is less than or equal to 0, and the other point is greater than or equal to 0) and the derivative is greater than zero. This point is circled in Figure 5 shown;

In one embodiment, the pre-processed respiratory waveform data includes: from the respiratory flow velocity waveform of the respiratory flow velocity data based on the breathing point time, find the points where the flow velocity crosses zero and the derivative is greater than zero, and record them respectively The time index of the point is obtained from the preprocessed respiratory waveform data; the preprocessing is to traverse the respiratory waveform data in order, at least once.

In one embodiment, the preprocessed respiratory data with multiple respiratory cycles includes: respectively segmenting the adjacent points to form a waveform corresponding to each breath, and the obtained multiple Respiration data for the respiration cycle;

Each mouthful of breath is a single breath, the definition of a single breath: a complete inhalation plus a complete expiratory phase; the inspiratory flow rate (Flow) changes from negative to positive as the beginning of a breath; the end of the breath is the next The previous sample point at which a breath was started.

Optionally, the nth and n+1th adjacent points are respectively taken, and the interval between the nth and n+1th points is defined as a single breathing cycle; wherein, 1≤n<N-1, N is The total number of time indices for all said points.

103: Perform feature extraction on the preprocessed respiratory data with multiple respiratory cycles, and obtain the feature-extracted respiratory data with multiple respiratory cycles as feature data;

In one embodiment, the feature data includes: the breathing time interval of a single breathing cycle, the flow rate change of a single breathing cycle, the ratio of the exhalation duration of a single breathing cycle to the total breathing time of a single breathing cycle, and the breathing time of a single breathing cycle. Gas waveform rising slope;

Optionally, the breathing time interval of the single breathing cycle is: the time interval a between the time indexes corresponding to the adjacent points, that is, the time interval between two adjacent circles, as shown in Figure 6 indicated by the arrow;

Optionally, the value range of the a is 0s<a<1s; the a is preferably 0s<a<0.6s; the respiratory rate of a normal person is about 8-20 times/min, and the patient’s breathing can be Up to 30-60 times/min, so the value range of breathing time a is 0s<a<1s, and the extreme case of breathing frequency>100 times is extremely rare, so the preferred value range of a is 0s<a<0.6s.

Optionally, the flow velocity change of the single respiratory cycle is: the flow velocity change b between the time indexes corresponding to the adjacent points; as shown by the arrow in FIG. 7 ;

Optionally, the value range of b is 0L/min<b<20L/min; the b is preferably 0L/min<b<10L/min; the inspiratory flow rate of normal people is 40-60L/min, children It is 5-10L/min, so the value range of the flow rate change b between one breath is 0L/min<b<20L/min, considering that some patients have insufficient inspiratory force, the preferred value range of b is 0L/min< b<10L/min.

Optionally, the ratio of the exhalation duration (the time when the flow rate is negative) of the single breathing cycle to the total breathing time (the time between two adjacent circles) of the single breathing cycle is: the adjacent points The exhalation duration between the corresponding time indexes accounts for the proportion c% of the total breathing time between the time indexes corresponding to the adjacent points;

Optionally, the value range of c% is 20%-80%; the c is preferably 30%-50%; according to the definition of double triggering, the expiratory time between two inspiratory cycles is less than the average inspiratory cycle. Half of the gas time, the value range of c is 20%-80%, wherein the preferred value range of c is 30%-50%. This setting can well avoid the influence of double or multiple triggers on the results; when there is clinical shortness of breath and abnormal breathing, take 2 breaths in a short period of time, and the accuracy of calculating a single breath by this segmentation method is better; the middle Make a judgment, and this reference can effectively assist how to deal with abnormal situations.

Optionally, the rising slope of the expiratory waveform of the single respiratory cycle is: the rising slope d of the expiratory waveform between the time indexes corresponding to the adjacent points; as shown in the box in Figure 8

Optionally, within the first 100ms of inhalation, the value range of d is 10-50L/min/s; the d is preferably 5-25L/min/s; the flow rate trigger setting value is about 1-5L/min/s min, within the first 100ms of inhalation, the value range of d is about 10-50L/min/s, and considering the partial trigger delay, the preferred range of d is 5-25L/min/s.

104: Input the feature data of the subject into a classification model to obtain a classification result of the feature data.

In one embodiment, the method or step of inputting the characteristic data of the subject into the classification model and obtaining the classification result of the characteristic data of the subject comprises:

Input the respiratory time interval of the single respiratory cycle of the subject into the classification model, and judge whether the respiratory time interval of the single respiratory cycle of the subject falls within the range of a; If the respiratory time interval of the single respiratory cycle of the subject falls within the range of a, the output is yes, and enters the stage of inputting the flow rate change of the single respiratory cycle of the subject to the classification model; otherwise, the output is no , the operation is terminated; a in this step is recorded as 1 in accompanying drawing 4;

Input the flow velocity change of the single respiratory cycle of the subject into the classification model, and judge whether the flow velocity change of the single respiratory cycle of the subject falls within the range of b; In the case that the flow velocity change of the single breathing cycle falls within the range of b, the output is, and enter the ratio of the exhalation duration of the single breathing cycle of the test subject to the total breathing time of the single breathing cycle into the classification The stage of the model; otherwise the output is No, and the operation is terminated; b in this step is recorded as 2 in accompanying drawing 4;

Input the ratio of the exhalation duration of the single breathing cycle of the subject to the test to the total breathing time of the single breathing cycle into the classification model, and judge the difference between the exhalation duration of the single breathing cycle of the test subject and the ratio of the total breathing time of the single breathing cycle. Whether the proportion of the total breathing time of the cycle falls within the range of c; when the ratio of the exhalation duration of the single breathing cycle of the subject to the test to the total breathing time of the single breathing cycle falls within the range of c, the output Yes, and enter the stage of increasing the slope of the expiratory waveform of the single respiratory cycle of the subject to the classification model; otherwise the output is no, and the operation is terminated; c in this step is recorded as 3 in accompanying drawing 4;

Input the rising slope of the expiratory waveform of the single respiratory cycle of the subject into the classification model, and judge whether the rising slope of the expiratory waveform of the single respiratory cycle of the subject falls within the range of d; When the upward slope of the expiratory waveform of the single respiratory cycle of the test subject falls within the range of d, the output is yes, and the classification result of the test subject's respiratory waveform data is output; otherwise, the output is no, and the operation is terminated. D in this step is recorded as 4 in accompanying drawing 4;

In one embodiment, the respiratory waveform data is respiratory waveform data with a continuous waveform signal.

The method for determining the classification model includes:

Obtain the respiratory waveform data of normal people;

Preprocessing the respiratory waveform data to obtain preprocessed respiratory waveform data; cutting the preprocessed respiratory waveform data into respiratory data with multiple respiratory cycles to obtain preprocessed respiratory waveform data and have breathing data of multiple breathing cycles;

Perform feature selection or feature extraction on the preprocessed respiratory data with multiple respiratory cycles, and obtain the respiratory data with multiple respiratory cycles after feature selection or feature extraction as feature data;

Feature extraction is performed on the feature data by using a machine learning method to obtain feature data after feature extraction, and a classification model is constructed using the feature data after feature extraction to obtain a constructed classification model.

Fig. 2 is a schematic diagram of an automatic respiratory segmentation device based on a flow velocity waveform provided by an embodiment of the present invention, the device includes: a memory and a processor;

The memory is used to store program instructions;

The processor is used for invoking program instructions, and when the program instructions are executed, is used for executing the above-mentioned automatic breathing segmentation method based on the flow velocity waveform.

Fig. 3 is a schematic flow chart of the breath automatic segmentation system based on the flow velocity waveform provided by the embodiment of the present invention, including:

An acquisition unit 301, configured to acquire respiratory waveform data of the subject;

The first processing unit 302 is configured to preprocess the respiratory waveform data to obtain the preprocessed respiratory waveform data; segment the preprocessed respiratory waveform data into Respiratory data, obtaining preprocessed and respiration data with multiple respiratory cycles; the segmentation is to find all flow velocity zero-crossing points and points whose derivatives are greater than zero;

The second processing unit 303 is configured to perform feature extraction on the preprocessed respiratory data with multiple respiratory cycles, and obtain the feature-extracted respiratory data with multiple respiratory cycles as feature data;

The classification unit is used to input the characteristic data of the subject into the classification model to obtain the classification result of the characteristic data.

A computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the above-mentioned automatic breathing segmentation method based on a flow velocity waveform is realized.

The validation results of this validation example show that assigning intrinsic weights to indications can moderately improve the performance of the method relative to the default setting.

Those skilled in the art can clearly understand that for the convenience and brevity of the description, the specific working process of the above-described system, device and unit can refer to the corresponding process in the foregoing method embodiment, which will not be repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or May be integrated into another system, or some features may be ignored, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above-mentioned embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the storage medium can include: Read Only Memory (ROM, Read Only Memory), Random Access Memory (RAM, Random Access Memory), magnetic disk or optical disk, etc.

Those of ordinary skill in the art can understand that all or part of the steps in the method of the above-mentioned embodiments can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium, and the above-mentioned storage The medium can be read-only memory, magnetic or optical disk, etc.

The computer equipment provided by the present invention has been introduced in detail above. For those of ordinary skill in the art, according to the idea of the embodiment of the present invention, there will be changes in the specific implementation and application range. In summary, , the contents of this specification should not be construed as limiting the present invention.

Claims (10)

1. A respiration automatic segmentation method based on a flow velocity waveform comprises the following steps:

acquiring respiratory waveform data of a person to be measured;

preprocessing the respiratory waveform data to obtain preprocessed respiratory waveform data; cutting the preprocessed respiratory waveform data into respiratory data with a plurality of respiratory cycles to obtain preprocessed respiratory data with a plurality of respiratory cycles; the cutting is to find all the flow velocity zero crossing points and the point of which the derivative is greater than zero;

performing feature extraction on the preprocessed respiratory data with a plurality of respiratory cycles to obtain the respiratory data with a plurality of respiratory cycles after feature extraction as feature data;

and inputting the characteristic data of the person to be measured into a classification model to obtain a classification result of the characteristic data.

2. The method according to claim 1, wherein the respiratory waveform data comprises: respiratory flow rate data based on breath strike time;

the preprocessed respiratory waveform data comprises: searching a point with a flow velocity zero crossing point and a derivative larger than zero from the respiratory flow velocity waveform of the respiratory flow velocity data based on the respiratory dotting time, and respectively recording the time indexes of the point to obtain the preprocessed respiratory waveform data;

optionally, the preprocessing is to sequentially traverse the breathing waveform data at least once.

3. The method of claim 2, wherein the pre-processed respiration data having a plurality of respiration cycles comprises: respectively segmenting adjacent points to form a waveform corresponding to each breath to obtain breath data of a plurality of breath cycles;

optionally, the adjacent points of the nth and n +1 are respectively taken, and a single breathing cycle is defined between the points of the nth and n + 1; wherein N is more than or equal to 1 and less than N-1,N is the total number of the time indexes of all the points.

4. The method according to claim 1, wherein the characteristic data comprises: the breath time interval of a single breath cycle, the flow rate change of the single breath cycle, the ratio of the expiration time of the single breath cycle to the total breath time of the single breath cycle, and the rising slope of the expiration waveform of the single breath cycle;

optionally, the breathing time interval of the single breathing cycle is: the time interval a between the time indexes respectively corresponding to the adjacent points; the value range of a is preferably made of 10 s-a-1s;

optionally, the flow rate variation of the single breathing cycle is: the flow rate change b between the time indexes respectively corresponding to the adjacent points; the value range of b is preferably 0L/min < b <20L/min;

optionally, the ratio of the expiration time of the single respiratory cycle to the total respiratory time of the single respiratory cycle is: the expiration time between the time indexes respectively corresponding to the adjacent points accounts for the proportion c% of the total respiration time between the time indexes respectively corresponding to the adjacent points; the value range of c% is preferably 20% -80%.

5. The method according to claim 4, wherein the rising slope of the expiratory waveform of the single respiratory cycle is: the rising slope d of the expiratory waveform between the time indexes respectively corresponding to the adjacent points;

optionally, within the initial 100ms of inspiration, the value range of d is 10-50L/min/s.

6. The method for automatic segmentation of respiration based on flow velocity waveform according to claim 5, wherein the method or step of inputting the feature data of the person to be measured into a classification model to obtain the classification result of the feature data of the person to be measured comprises:

inputting the breathing time interval of the single breathing cycle of the person to be detected into a classification model, and judging whether the breathing time interval of the single breathing cycle of the person to be detected falls into a range; in the case where the breathing time interval of said single breathing cycle of said subject falls within a range, outputting yes and entering a phase of inputting the variation of the flow rate of said single breathing cycle of said subject into a classification model; otherwise, the output is no, and the operation is terminated;

inputting the flow rate change of the single breathing cycle of the person to be tested into a classification model, and judging whether the flow rate change of the single breathing cycle of the person to be tested falls into a range b or not; if the change of the flow rate of the single respiratory cycle of the testee falls into the range b, outputting yes, and inputting the ratio of the expiration time of the single respiratory cycle of the testee to the total respiration time of the single respiratory cycle into a classification model; otherwise, the output is no, and the operation is terminated;

inputting the ratio of the expiration time of the single respiratory cycle of the person to be detected to the total respiratory time of the single respiratory cycle into a classification model, and judging whether the ratio of the expiration time of the single respiratory cycle of the person to be detected to the total respiratory time of the single respiratory cycle falls within a range c; under the condition that the ratio of the expiration time of the single respiration period of the testee to the total respiration time of the single respiration period falls into a range c, outputting yes, and entering a stage of ascending a slope of an expiration waveform of the single respiration period of the testee to a classification model; otherwise, the output is not, and the operation is terminated;

inputting the rising slope of the expiratory waveform of the single respiratory cycle of the person to be detected into a classification model, and judging whether the rising slope of the expiratory waveform of the single respiratory cycle of the person to be detected falls into a range d; if the rising slope of the expiratory waveform of the single respiratory cycle of the person to be detected falls within the range d, outputting yes and outputting a classification result of the respiratory waveform data of the person to be detected; otherwise, outputting no, and ending the operation.

7. The method according to claim 1, wherein the respiratory waveform data is respiratory waveform data having a continuous waveform signal.

8. An apparatus for automatic breath segmentation based on a flow rate waveform, the apparatus comprising: a memory and a processor;

the memory is to store program instructions;

the processor is configured to invoke program instructions that, when executed, perform the method for breath autosegmentation based on flow rate waveforms of any of claims 1 to 7.

9. An automatic breath segmentation system based on a flow rate waveform, comprising:

the acquisition unit is used for acquiring respiratory waveform data of a person to be measured;

the first processing unit is used for preprocessing the respiratory waveform data to obtain preprocessed respiratory waveform data; cutting the preprocessed respiratory waveform data into respiratory data with a plurality of respiratory cycles to obtain preprocessed respiratory data with a plurality of respiratory cycles; the cutting is to find all the flow velocity zero crossing points and the point of which the derivative is greater than zero;

the second processing unit is used for performing feature extraction on the preprocessed respiratory data with a plurality of respiratory cycles to obtain the respiratory data with a plurality of respiratory cycles after feature extraction as feature data; and the classification unit is used for inputting the characteristic data of the person to be tested into a classification model to obtain a classification result of the characteristic data.

10. A computer-readable storage medium, on which a computer program is stored, which, when being executed by a processor, implements the method for flow-waveform-based automatic breath segmentation according to any one of claims 1 to 7.

Images