WO2022133942A1 - Medical ventilation device and ventilation monitoring method - Google Patents

Abstract

A medical ventilation device and a ventilation monitoring method. The ventilation monitoring method comprises: acquiring at least one type of parameter data capable of representing the occurrence of a patient-ventilator asynchrony event of a patient, the parameter data comprising at least one type of parameter data among device ventilation parameter data and patient physiological parameter data; enabling algorithm modules associated with the at least one type of parameter data to acquire recognition results of the patient-ventilator asynchrony event during the ventilation process of the patient, there being a plurality of algorithm modules, and the different algorithm modules performing calculation on the basis of different data combinations formed by the at least one type of parameter data, so as to acquire the recognition results; determining the patient-ventilator asynchrony event occurring to the patient according to the recognition results respectively acquired by the different algorithm modules; and outputting the patient-ventilator asynchrony event occurring to the patient. Applying the ventilation monitoring method to a medical ventilation device allows for accurate determination of whether a patient-ventilator asynchrony event has occurred.

Description

Medical ventilation equipment and ventilation monitoring method technical field

The invention relates to the field of medical equipment, in particular to medical ventilation equipment and a ventilation monitoring method

Background technique

patient-ventilator confrontation Asynchrony), also known as man-machine confrontation, refers to the phenomenon of asynchrony between the supply of ventilation equipment (such as a ventilator) and the patient's spontaneous breathing during the process of mechanical ventilation. The occurrence of man-machine confrontation will affect the patient's condition and cause patient discomfort, such as increased work of breathing and prolonged on-machine time. If the human-machine confrontation is serious, it will lead to accidents such as aggravated lung injury and increased fatality rate of patients. Therefore, it is very important to identify the phenomenon of man-machine confrontation in the process of patients using a ventilator for respiratory support, and to remind the medical staff of this phenomenon.

At present, the method for identifying man-machine confrontation is to monitor the patient's ventilation waveform in real time, and automatically identify the man-machine confrontation phenomenon that occurs in the patient through the waveform characteristics. Among them, the monitored ventilation waveforms are usually three waveforms of airway pressure, flow rate, and volume. The monitoring object of this method is fixed, which has shortcomings such as insufficient recognition accuracy and limited types of man-machine confrontation that can be identified, and the recognition result of man-machine confrontation is easily affected by the waveform signal strength and monitoring quality, which is not conducive to man-machine confrontation. accurate identification.

technical solutions

According to a first aspect, the present invention provides a medical ventilation device, comprising:

Air source interface, used to connect air source;

A patient interface for connecting to the patient's respiratory system;

A breathing circuit for connecting the air source interface and the patient interface to deliver the gas provided by the air source to the patient;

Respiratory assistance device, used to provide respiratory support power to control the delivery of gas provided by the gas source to the patient;

a processor, configured to acquire at least one parameter data capable of characterizing a human-machine confrontation event in a patient, the parameter data including at least one type of device ventilation parameter data and patient physiological parameter data;

The processor is further configured to, when receiving the parameter data, enable an algorithm module associated with the parameter data to obtain the identification result of the patient-machine confrontation event during the ventilation process; wherein the The algorithm module includes multiple, and different algorithm modules perform calculations based on different data combinations formed by at least one parameter data to obtain the identification result of the patient's human-machine confrontation event during the ventilation process; the identification result includes: identifying the man-machine confrontation event or the man-machine confrontation event is not recognized;

And according to the recognition results obtained by different algorithm modules, determine and output the man-machine confrontation event that occurs in the patient.

According to a second aspect, the present invention provides a ventilation monitoring method, comprising the steps of:

acquiring at least one parameter data that can characterize the occurrence of a human-machine confrontation event in a patient, the parameter data including at least one type of device ventilation parameter data and patient physiological parameter data;

enabling an algorithm module associated with the at least one parameter data to obtain an identification result of a human-machine confrontation event in the patient during ventilation, wherein the algorithm module includes a plurality of different algorithm modules based on at least one A different data combination formed by parameter data is calculated to obtain a recognition result, and the recognition result includes: identifying the man-machine confrontation event or not recognizing the man-machine confrontation event;

Determine the human-machine confrontation event that occurs in the patient according to the recognition results obtained by the different algorithm modules;

The man-machine confrontation events that occurred in the patient are output.

According to a third aspect, the present invention provides a computer-readable storage medium, which is characterized by comprising a program, and the program can be executed by a processor to implement the method described in any one of the foregoing aspects.

beneficial effect

In the above embodiment, the corresponding algorithm module is activated according to at least one parameter data, and the identification results of each algorithm module are combined to determine the human-machine confrontation event that occurs in the patient during the ventilation process. Compared with the existing method:

(1) The output human-machine confrontation events are based on at least one parameter data, so more human-machine confrontation types can be identified.

(2) Different algorithm modules can use different algorithms for human-machine confrontation recognition, and then integrate the recognition results of multiple algorithm modules to confirm human-machine confrontation events, so that the recognition accuracy and robustness of human-machine confrontation events are also higher. it is good.

Description of drawings

1 is a schematic diagram of a medical ventilation device according to an embodiment;

Fig. 2 is the schematic diagram of the algorithm module of an embodiment and its data combination;

3 is a waveform diagram of corresponding parameter data when an invalid trigger event occurs according to an embodiment;

FIG. 4 is a waveform characteristic diagram of the airway pressure waveform and the airway flow velocity waveform at the moment when the invalid trigger event occurs in FIG. 3;

5 is a waveform diagram of corresponding parameter data when a double trigger event occurs according to an embodiment;

6 is a waveform diagram of corresponding parameter data when an invalid trigger event occurs in another embodiment;

7 is a waveform diagram of corresponding parameter data when an invalid trigger event occurs according to another embodiment;

8 is a schematic diagram of a judgment module corresponding to various types of man-machine confrontation events according to an embodiment;

9 is a schematic diagram of marking a human-machine confrontation event on a waveform according to an embodiment;

10 is a display interface for displaying prompt information and a statistical rate of man-machine confrontation according to an embodiment;

11 is a flowchart of a ventilation monitoring method according to an embodiment;

10. Air source interface;

20. Respiratory aids;

30, breathing circuit; 30a, inspiratory passage; 30b, expiratory passage;

31. Carbon dioxide receiver; 32. One-way valve;

40. Patient interface;

50. Processor;

60. Parameter measuring device;

70. Display.

Embodiments of the present invention

The present invention will be further described in detail below through specific embodiments in conjunction with the accompanying drawings. Wherein similar elements in different embodiments have used associated similar element numbers. In the following embodiments, many details are described so that the present application can be better understood. However, those skilled in the art will readily recognize that some of the features may be omitted under different circumstances, or may be replaced by other elements, materials, and methods. In some cases, some operations related to the present application are not shown or described in the specification, in order to avoid the core part of the present application from being overwhelmed by excessive description, and for those skilled in the art, these are described in detail. The relevant operations are not necessary, and they can fully understand the relevant operations according to the descriptions in the specification and general technical knowledge in the field.

Additionally, the features, acts, or characteristics described in the specification may be combined in any suitable manner to form various embodiments. At the same time, the steps or actions in the method description can also be exchanged or adjusted in order in a manner obvious to those skilled in the art. Therefore, the various sequences in the specification and drawings are only for the purpose of clearly describing a certain embodiment and are not meant to be a necessary order unless otherwise stated, a certain order must be followed.

The serial numbers themselves, such as "first", "second", etc., for the components herein are only used to distinguish the described objects, and do not have any order or technical meaning. The "connection" and "connection" mentioned in this application, unless otherwise specified, include both direct and indirect connections (connections).

At present, the existing technology of man-machine confrontation identification mainly focuses on identifying the man-machine confrontation characteristics on airway pressure or flow rate, but it is impossible to judge false triggering and reverse triggering only from the pressure or flow rate. When there is interference, it will affect the accuracy of its judgment on events such as invalid triggering and handover advance. If the trans-diaphragmatic pressure and the electrical signal of the diaphragm are mainly used to judge the man-machine confrontation, the results are very easily affected by the quality of signal monitoring. For example, the monitoring of transdiaphragmatic pressure requires the use of esophageal pressure and intragastric pressure measurement accessories, and the measurement accessories need to be placed in a suitable measurement position in the patient's body. The accuracy and stability of the pressure monitoring results are very dependent on the measurement accessories themselves and operating procedures. Whether normative impact. In the process of monitoring the signal, the intensity of the signal noise, the patient's involuntary swallowing, or the change of the body position can easily lead to misjudgment of the result. Similarly, the monitoring of diaphragm electromyography also requires corresponding measurement accessories and placement procedures. If the patient’s ability to breathe spontaneously is weak, the intensity of the electromechanical signal of the diaphragm will be low, and the signal cycle characteristics will not be obvious, which will lead to poor signal recognition for human-machine confrontation. . In addition, only relying on trans-diaphragmatic pressure or diaphragm electromyography cannot identify human-machine confrontation events such as low patient flow rate and slow pressure rise time.

The parameter data referred to in the present invention includes analog data and digital data, etc., which may be structured data or unstructured data, and the signal referred to hereinafter is also included in the parameter data referred to in the present invention.

The feature extraction referred to in the present invention includes the process of calculating the signal or data. The calculation includes the process of operating the signal to obtain a new intermediate signal. Possible methods include basic mathematical operations, such as signal addition. , subtraction, and various forms of mathematical calculation; new parameters are extracted from the signal to assist the judgment of human-machine confrontation identification, such as calculating the mean, variance, standard deviation, central moments of each order, standard moments of each order, etc. It belongs to the meaning category of feature extraction; various preprocessing or post-processing of the signal, such as filtering, normalization, etc., also belongs to the meaning category of feature extraction.

The carbon dioxide module referred to in the present invention refers to a module for monitoring carbon dioxide concentration, and its monitoring principle is mostly through infrared absorption spectroscopy technology, using the relationship between the carbon dioxide concentration in the gas and the absorption rate to calculate the carbon dioxide concentration in the patient's exhaled breath.

Please refer to the embodiment shown in FIG. 1 , which provides a medical ventilation device (such as a ventilator, an anesthesia machine, etc.), the medical ventilation device includes an air source interface 10 , a breathing assistance device 20 , a breathing circuit 30 , a patient Interface 40 and processor 50.

The gas source interface 10 is used for connecting with a gas source (not shown in the figure), and the gas source is used for supplying gas. Oxygen, air, etc. can be used as the gas. In some embodiments, the gas source may use a compressed gas cylinder or a central gas supply source, and supply gas to the medical ventilation equipment through the gas source interface 10, and the gas supply types include oxygen O2 and air. The gas source interface 10 may include conventional components such as a pressure gauge, a pressure regulator, a flow meter, a pressure reducing valve, and a proportional control protection device, which are respectively used to control the flow of various gases (eg, oxygen and air). The gas input from the gas source interface 10 enters the breathing circuit 30 and forms a mixed gas with the original gas in the breathing circuit 30 .

The respiratory assistance device 20 is used to provide power for the patient's involuntary breathing and maintain airway patency, that is, to drive the gas input from the air source interface 10 and the mixed gas in the breathing circuit 30 to the patient's breathing system, and to drain the patient's exhaled gas. into the breathing circuit 30, thereby improving ventilation and oxygenation, preventing hypoxia and accumulation of carbon dioxide in the patient's body. In a specific embodiment, the breathing assistance device 20 generally includes a machine-controlled ventilation module, and the airflow conduit of the machine-controlled ventilation module communicates with the breathing circuit 30 . When the patient does not resume spontaneous breathing during the operation, the machine-controlled ventilation module is used to provide the patient with the power to breathe.

The breathing circuit 30 includes an inspiratory passage 30a, an expiratory passage 30b and a carbon dioxide absorber 31. The inspiratory passage 30a and the expiratory passage 30b communicate to form a closed circuit, and the carbon dioxide absorber 31 is arranged on the pipeline of the expiratory passage 30b. The mixed gas of fresh air introduced by the air source interface 10 is input through the inlet of the inspiratory passage 30a, and provided to the patient through the patient interface 40 disposed at the outlet of the inspiratory passage 30a. The patient interface 40 may be a mask, nasal cannula, or endotracheal cannula. In a preferred embodiment, the inhalation passage 30a is provided with a one-way valve 32, and the one-way valve 32 is opened in the inhalation phase and closed in the expiratory phase. The expiratory passage 30b is also provided with a one-way valve 32, and the one-way valve 32 is closed during the inspiratory phase and opened during the expiratory phase. The inlet of the exhalation passage 30b is communicated with the patient interface 40. When the patient exhales, the exhaled gas enters the carbon dioxide absorber 31 through the exhalation passage 30b, and the carbon dioxide in the exhaled gas is filtered by the substances in the carbon dioxide absorber 31. The gas from which the carbon dioxide has been filtered is recirculated into the intake passage 30a. In some embodiments, the breathing circuit 30 is further provided with a flow sensor and/or a pressure sensor, which are respectively used to detect the gas flow and/or the pressure in the pipeline.

The processor 50 is used to execute instructions or programs to control various control valves in the breathing assistance device 20, the air source interface 10 and/or the breathing circuit 30, or to process the received data to generate the required calculations or judgments As a result, visual data or graphs are generated, and the visual data or graphs are output to the display 70 for display.

In this embodiment, the processor 50 receives at least one parameter data measured by an external device that can characterize a patient-machine confrontation event, and the parameter data includes at least one type of device ventilation parameter data and patient physiological parameter data. The equipment ventilation parameter data includes the control parameter data set by the medical ventilation equipment itself, and also includes the ventilation-related parameter data monitored by the medical ventilation equipment itself, and these parameter data include numerical data, waveform data, and the like. Similarly, the physiological parameter data also includes numerical data, waveform data, and the like. For example, the device ventilation parameter data includes at least one of airway pressure, airway flow rate, and gas volume of the medical ventilation device, and the physiological parameter data includes the patient's esophageal pressure, intragastric pressure, transdiaphragmatic pressure, and carbon dioxide concentration during ventilation. At least one of , diaphragm muscle electricity, etc., the above-mentioned external equipment can be various types of sensors placed in the patient's body or various types of plug-ins or modules arranged on medical ventilation equipment, such as an esophageal pressure sensor placed in the patient's esophagus, used for measuring A carbon dioxide module for carbon dioxide concentration, etc. In other embodiments, the medical ventilation device itself may further include a parameter measurement device 60 for acquiring the above-mentioned parameter data. It should be noted that, in the present invention, device ventilation parameter data and patient physiological parameters are two types of parameter data, while airway pressure and the like are one kind of parameter data. If the processor 50 receives the airway pressure measured by the external device and airway flow rate, then what is obtained is two parameter data in one type.

Upon receiving the above-mentioned parameter data, the processor 50 activates the algorithm module associated with the received parameter data to obtain the identification result of the patient-machine confrontation event during the ventilation process. The algorithm module includes a plurality of algorithm modules, and different algorithm modules perform calculations based on different data combinations formed by at least one parameter data, so as to obtain the identification result of the human-machine confrontation event during the ventilation process of the patient. An association can be defined between an algorithm module and the parametric data on which it is calculated. The above data combination refers to a combination in a broad sense, and when there is only one parameter data, it can be called a data combination, that is, the algorithm module can perform calculations based on a data combination with only one parameter data. In some embodiments, the relationship between the algorithm module and its corresponding data combination is shown in FIG. 2 , wherein, the data combination corresponding to the algorithm module A includes only one parameter data of airway pressure, and the data combination corresponding to the algorithm module D includes: Two parameters, airway pressure and gas volume, were obtained. In this embodiment, the processor 50 will enable the corresponding algorithm module only when the acquired at least one kind of parameter data meets the requirements of the algorithm module for the data combination of the parameter data. For example, taking the example shown in FIG. 2 , if the processor 50 obtains the parameter data of the airway pressure, then the algorithm module A is activated, but the algorithm module B is not activated based on the airway pressure. In other embodiments, the requirements of the data combination for the parameter data also include the requirements for the validity of the parameter data. Taking FIG. 2 as an example, if the processor 50 obtains the airway pressure, the processor 50 will perform a valid operation on the airway pressure. According to the judgment, if the airway pressure is valid data, the processor 50 activates the algorithm module A. The validity requirements of the above parameter data may be requirements such as data range and periodicity. For example, if the acquired airway pressure is not within a preset range, it is determined that the airway pressure is not valid.

FIG. 2 shows a method of data combination formed by parameter data. In other embodiments, the data combination of each algorithm module may include airway flow rate and/or airway pressure, or any other number of parameter data. The combination makes the features of man-machine confrontation more obvious and easier to obtain.

The above-mentioned algorithm modules may be corresponding computer programs stored in the memory, the number of the algorithm modules may also be increased or decreased, and the algorithm modules may also be manually turned on or off. The algorithm module calculates and obtains the recognition result based on the parameter data, which may include recognizing the human-machine confrontation event or not recognizing the human-machine confrontation event, that is to say, the algorithm module A in Fig. Whether the machine-machine confrontation event occurs, the algorithm module C can judge whether the human-machine confrontation event occurs according to the airway pressure and the airway flow rate and based on a certain calculation method. Although both algorithm module A and algorithm module C use the parameter data of airway pressure, the calculation methods of algorithm module A and algorithm module C can be independent, that is, the identification results of algorithm module A and algorithm module C can be independent.

In some embodiments, the identification result further includes the type of the identified man-machine confrontation event, that is, the algorithm module A can not only identify whether the man-machine confrontation event occurs according to the airway pressure, but also identify the occurrence of the man-machine confrontation event according to the airway pressure. What type of man-machine confrontation is. The types of man-machine confrontation mentioned above include one or more of invalid trigger events, double trigger events, false trigger events, reverse trigger events, trigger delay events, switching advance events, switching delay events, and low flow rate events. Hereinafter, the algorithm module C and the algorithm module D are taken as examples to describe how the algorithm module obtains the above identification result.

As shown in FIG. 3 , the corresponding airway pressure waveform and airway flow rate waveform generated by the processor 50 according to the airway pressure and the airway flow rate are shown. If a human-machine confrontation event occurs in a patient, corresponding "traces" will be left on the airway pressure waveform and airway flow velocity waveform, which is manifested as abnormal waveform characteristics, and different types of human-machine confrontation event waveform characteristics. different. Taking an invalid trigger event as an example, Figure 4 shows the waveform characteristics of the airway pressure waveform and the airway flow velocity waveform when the invalid trigger event occurs. The upper part in Figure 4 is the airway pressure waveform, and the lower part in Figure 4 is the Airway velocity waveform. If a sudden rise of the airway flow velocity waveform and a sudden drop of the airway pressure waveform occur simultaneously at a certain moment, and some waveform characteristics in the waveform change meet the preset threshold conditions, it means that an invalid trigger event occurs in the patient. The above-mentioned waveform characteristics may include the change amplitude, change rate (first-order derivative), second-order derivative, and change duration of the waveform. The algorithm module C performs feature extraction on the airway pressure waveform and the airway flow velocity waveform, that is, calculates the parameter data. If the extracted feature satisfies the preset threshold condition, the algorithm module C will obtain the identification result of the invalid trigger event of the patient. In some embodiments, it can be set that if the waveform characteristics of the airway flow velocity waveform and the waveform characteristics of the airway pressure satisfy corresponding threshold conditions at the same time, the algorithm module C obtains the identification result of the invalid trigger event of the patient. In other embodiments, it may also be within a certain time range when one of the waveform characteristics of the airway flow velocity waveform and the waveform characteristics of the airway pressure satisfies the corresponding threshold condition, and the other waveform characteristic also satisfies the corresponding threshold condition, The algorithm module C obtains the identification result of the invalid triggering event of the patient.

FIG. 5 shows the corresponding airway pressure waveform and gas volume waveform generated by the processor 50 according to the airway pressure and the gas volume. Taking the double-trigger event as an example, the algorithm module D can calculate the duration of the patient's breathing cycle according to the obtained airway pressure. A double-trigger event is considered to exist during this breath cycle. In this embodiment, the preset threshold may be a result obtained by jointly limiting multiple related threshold conditions. For example, the average inspiratory time and average expiratory time of multiple (for example, 12) breathing cycles before this breathing cycle can be calculated. If the expiratory time of this breathing cycle is less than half of the average inspiratory time, the average expiratory time The minimum value between half of the time and a fixed time threshold (for example, 500ms), the recognition result of the algorithm module D is the occurrence of a double-triggered event in this breathing cycle. In addition, the algorithm module D can also determine whether there is a double trigger event in combination with the gas volume. Because there will be a very short expiratory phase in the breathing cycle with double trigger events, resulting in a small exhaled tidal volume of the patient, the algorithm module D can calculate the exhaled tidal volume and/or the inhaled tidal volume according to the obtained gas volume. When it is recognized that the expiratory tidal volume is less than a threshold (for example, 1/2 of the inspiratory tidal volume in the current breathing cycle), or the result of subtracting the expiratory tidal volume from the inspiratory tidal volume is greater than IBW*k (IBW is ideal body weight, k is the coefficient threshold, for example k=1ml/kg), then the recognition result of algorithm module D is the occurrence of a double-triggered event.

The above example illustrates how the algorithm module obtains the recognition result of the human-machine confrontation event. Among them, the algorithm module C can not only identify invalid trigger events, but also identify other types of man-machine confrontation events according to airway pressure and airway flow rate. For a certain type of human-machine confrontation event, there may be multiple algorithm modules that can determine whether this type of human-machine confrontation event occurs. For example, in addition to algorithm module C, algorithm module E, algorithm module G, algorithm module H, The algorithm module I and the algorithm module K can also identify whether an invalid trigger event occurs, wherein, a plurality of algorithm modules can be calculated according to different parameter data using different calculation methods. Based on this, the above and the following relate to The threshold condition, threshold time, threshold, etc., can have different setting values or preset values according to different algorithm modules and/or different types of human-machine confrontation events identified, so as to avoid repeating the same judgment logic for different algorithm modules ( calculation and determination method). The invention identifies man-machine confrontation events from different angles, so that the identification result has better robustness and higher accuracy.

For example, in the embodiment shown in FIG. 6, the airway pressure caused by the invalid trigger event of the patient has a small change range. If only relying on the airway pressure to determine whether the invalid trigger event occurs, it may occur that the characteristics of the airway pressure are insufficient. Obviously lead to inaccurate judgment. Figure 6 also combines esophageal pressure to jointly determine whether an invalid trigger event occurs. When the patient breathes spontaneously, the patient's diaphragm, intercostal muscles and other respiratory muscles actively contract, causing the pressure in the pleural cavity to drop. Clinically, the monitoring of this esophageal pressure is approximately equivalent to monitoring the patient's pleural pressure, so the identification of the esophagus The change in pressure can distinguish the patient's spontaneous breathing state. In Figure 6, it can be seen that there is a change in the esophageal pressure caused by the patient's spontaneous inhalation, and there is also a slight pressure amplitude change in the airway pressure. These two points are also invalid triggers for the patient. the moment of the event. In this embodiment, the characteristics of the patient's spontaneous inhalation stage can be obtained through the identified characteristics of the esophageal pressure and the downward pressure swing. Combining this characteristic with the identified characteristics of the airway pressure, an invalid trigger event can be determined together, that is, If the patient's spontaneous breathing effort is detected through a change in esophageal pressure within a threshold time after the airway pressure is identified as having the characteristics of the invalid trigger event, then it can be determined that the patient has an invalid trigger event. Two algorithm modules can be set to obtain the recognition result of the human-machine confrontation event based on the airway pressure and the esophageal pressure respectively, or in the embodiment shown in FIG. 2, an algorithm module E is set, and the algorithm module E can Combined with esophageal pressure to judge whether human-machine confrontation events (such as invalid trigger events) occur, thus improving the accuracy of judgment.

For another example, the embodiment shown in FIG. 7 also combines the parameter data of carbon dioxide concentration to jointly determine whether an invalid trigger event occurs. In some embodiments, the carbon dioxide concentration of the patient can be monitored through an external carbon dioxide module. When the patient is in the exhalation phase, the carbon dioxide waveform curve can be seen rising, and when the patient is in the inhalation phase, the carbon dioxide waveform curve can be seen. decline. Therefore, when it is detected that the carbon dioxide waveform has obvious waveform decline characteristics (for example, the decline amplitude and the decline duration meet a certain threshold), it means that the patient has inspiratory effort. event, the time corresponding to the arrow in Figure 7 is the time when the invalid trigger event occurs. It can be seen that judging invalid triggering by carbon dioxide waveform characteristics can be combined with the original method of judging invalid triggering by airway pressure, airway flow rate, gas volume curve, etc., thereby improving the accuracy of judgment. By extracting the features of the carbon dioxide waveform through the algorithm module, that is to say, calculating the change of carbon dioxide concentration, it is possible to detect that the carbon dioxide waveform has obvious waveform decline characteristics, so as to obtain the recognition result of the human-machine confrontation event. For example, as shown in Figure 2 In the embodiment of the invention, the algorithm module H is enabled to jointly determine whether the invalid trigger event occurs with other algorithm modules.

The processor 50 is further configured to determine the human-machine confrontation events that occur in the patient according to the identification results obtained by the different algorithm modules. In some embodiments, the human-machine confrontation events that occur in the patient are distinguished according to different types, that is, it is determined that the patient has experienced human-machine confrontation events. What types of man-machine confrontation events. Figure 8 shows a schematic diagram of different algorithm modules comprehensively judging whether a certain type of human-machine confrontation event occurs. Each type of human-machine confrontation event in the figure has a corresponding judgment module. For each algorithm module of the same type of human-machine confrontation event, the output result of the judgment module is that the corresponding human-machine confrontation event occurs or does not occur, that is to say, the judgment module can identify the same type of human-machine confrontation event. As a result, judging whether the corresponding type of man-machine confrontation event occurred, and which types of man-machine confrontation events occurred in the patient can be determined according to the output results of each judgment module, which is in turn determined based on the recognition results of each internal algorithm module. For example, in the invalid trigger judgment module, the recognition result of some algorithm modules may be that an invalid trigger event has occurred, and the recognition result of another part of the algorithm module may be that no invalid trigger event has occurred. If the final output result of the invalid trigger judgment module is that an invalid trigger event has occurred If the trigger event is invalid, it means that the result of the comprehensive judgment of the algorithm module enabled by the processor 50 is that the patient has an invalid trigger event. The judgment modules corresponding to other types of human-machine confrontation events will also obtain corresponding output results, so as to determine various types of human-machine confrontation events that occur in patients.

In some embodiments, the output result of the judgment module can be determined in the following manner:

Obtain the credibility of the first algorithm module, wherein the first algorithm module is the algorithm module whose recognition result is that the human-machine confrontation event is recognized, for example, in the invalid trigger judgment module, the algorithm module C, the algorithm module E and the algorithm module G are If the identification result is that an invalid trigger event is identified, the three algorithm modules are defined as the first algorithm module. In other embodiments, the first algorithm module is an algorithm module that identifies the type of human-machine confrontation event corresponding to the judgment module. Then, according to the credibility of each first algorithm module, it is judged whether the same type of human-machine confrontation event identified by each first algorithm module occurs. The reliability determines whether the invalid trigger event occurs, and the reliability may be related to at least one of the signal-to-noise ratio, the degree of feature clarity, and the degree of regularity of the parameter data associated with the algorithm module. For example, when airway pressure and airway flow rate are used to identify invalid trigger events, the degree of clarity of the feature can be divided according to the magnitude of the identified features such as the pressure drop range and the change range of the first derivative of the flow rate. The larger the variation range of the first derivative of the flow velocity is, the higher the degree of feature definiteness and the higher the reliability. When using physiological signals of patients such as esophageal pressure or diaphragm electromyography for human-machine confrontation identification, the quality of signal monitoring and the signal-to-noise ratio will affect the reliability of the corresponding algorithm module. For example, when it is detected that the signal fluctuation range is weak, or there is no regular periodic motion, the reliability is low. On the contrary, if the signal fluctuation is greater than a certain threshold, and it is detected that there is a regular fluctuation period in the recent period of time, the reliability is high. .

According to the credibility of the first algorithm module, the processor 50 may generate a credibility score corresponding to the credibility of the first algorithm module, and then combine the corresponding weight coefficients of the first algorithm module to determine the credibility of the first algorithm module. For different types of human-machine confrontation events, different algorithm modules are preset with corresponding weight coefficients. For example, the weight coefficient can be at least based on the difference between the algorithm module and the human-machine confrontation event. The degree of correlation of the type is determined. For example, when identifying the trigger delay event, considering that the diaphragmatic electromyographic signal is generally earlier than the esophageal pressure signal, a higher weight coefficient is given to the algorithm module that uses the diaphragmatic electromyography for identification. The benefits of doing so Yes, it is possible to modify the algorithm results to a certain extent based on clinical consensus or the essential characteristics of parameters to make the algorithm results more credible. Calculate the sum of the credibility scores of the first algorithm modules according to the revised credibility score, and then determine whether the sum of the credibility scores is greater than the preset threshold. The same type of man-machine confrontation event occurs, otherwise, the same type of man-machine confrontation event identified by each first algorithm module does not occur. The sum of the credibility scores of the above-mentioned first algorithm modules refers to the sum of the credibility scores of the first algorithm modules in the same judgment module. If the credibility of each first algorithm module in the judgment module is If the sum of the scores is greater than the preset threshold, the output result of the judgment module is that a human-machine confrontation event of the corresponding type occurs.

The above method comprehensively considers the credibility of the identification results of each first algorithm module. When the total credibility of each first algorithm module in the judgment module satisfies a certain condition, the output result of the judgment module is the corresponding type of man-machine confrontation. event happens.

In other embodiments, according to the ratio and/or quantity relationship between the first algorithm module and the second algorithm module, it is determined whether the same type of human-machine confrontation event identified by each algorithm module occurs, wherein the first algorithm module is The recognition result is an algorithm module for recognizing the human-machine confrontation event, and the second algorithm module is an algorithm module for which the recognition result is that the human-machine confrontation event is not recognized. For example, in the invalid trigger judgment module, the identification results of the algorithm module C, the algorithm module E and the algorithm module G are that the invalid trigger event is identified, then these three algorithm modules are defined as the first algorithm module, the algorithm module H, the algorithm module I and the algorithm module K do not identify an invalid trigger event, then these three algorithm modules are the second algorithm module. In this example, there are three first algorithm modules. If there are more than two first algorithm modules in a judgment module, the output result of the judgment module is that a man-machine confrontation event of the corresponding type occurs, and the judgment is invalid. The output of the module is that an invalid trigger event has occurred. The advantage of this embodiment is that misidentification can be avoided as much as possible, because too many misidentifications will cause user information fatigue and weaken the original prompting effect.

After determining the human-machine confrontation event occurred in the patient, the processor 50 further outputs the human-machine confrontation event occurred in the patient to the display interface of the display 70 or other display devices.

In some embodiments, the processor 50 marks the identified man-machine confrontation event near the corresponding feature of the corresponding parameter data waveform in the display interface, for example, as shown in FIG. The form indicates the occurrence of the human-machine confrontation event (Paw is the airway pressure, Pes is the esophageal pressure), the triangle mark is used to indicate the position of the corresponding feature of the human-machine confrontation event on the waveform, and the string below the triangle mark indicates the human-machine The type name of the confrontation event, (IE is short for invalid trigger event, DT is short for double trigger event, RT is short for reverse trigger event). The marking method can also be in the form of any symbol, color, or character string to distinguish different types of human-machine confrontation events. The advantage of this is that the types of human-machine confrontation events can be matched with the corresponding features on the waveform. Experienced doctors can directly judge the accuracy of the recognition results from the marking results. The markings on the waveform are also related to the occurrence rate of human-machine confrontation events. The changes of the monitoring value are consistent, and the doctor can clearly understand the meaning of the monitoring value of the incidence of human-machine confrontation. For general medical staff, they can also learn by comparing the waveform annotation. If the waveform annotation is not performed, there is only the monitoring value of the occurrence rate of human-machine confrontation events, and the doctor cannot know whether the recognition algorithm is accurate. Another advantage of annotating results directly on the waveform interface is that the user can still observe the latest ventilation waveform without the need to freeze the waveform or switch to other interfaces. In addition, the user does not need to perform additional operations to see the recognition results and waveforms Features, easy to use.

In some embodiments, as shown in FIG. 10 , according to the identification result of the human-machine confrontation event, the occurrence rate of various types of human-machine confrontation events is counted. Such statistics can be statistics of the occurrence rate in a recent period of time, or statistics of the occurrence rate of human-machine confrontation events in a certain number of recent respiratory cycles. When the occurrence rate of a certain type of human-machine confrontation event is higher than a certain threshold, a specific area on the main interface prompts the user that there are too many human-machine confrontation events of this type, and gives operation suggestions. Taking invalid triggers as an example, Figure 10 shows a prompt method when the occurrence rate of invalid trigger events is detected to be higher than 10%. This method mainly embodies two pieces of information. The first one is to remind the user of the type name of too many man-machine confrontation events, and the second one is to give operation suggestions. For example, the prompt information in Figure 10 is based on the trigger sensitivity in the current ventilation parameters. Threshold, prompts the user to lower the threshold setting.

Compared with the prior art, the advantage of doing this is that the user is not prompted when the man-machine confrontation occurs, but only when the occurrence rate of the man-machine confrontation exceeds a certain level, because prompting the user frequently will cause user information or visual effects. fatigue. In addition, another advantage compared with the prior art is that the prior art only judges human-machine confrontation events based on abnormal monitoring parameters, and gives operation prompts (US9027552), while the definition and identification of human-machine confrontation in clinical practice are not. It is based on the waveform characteristics, and the algorithm module in the present invention also performs feature extraction on the waveform of the parameter data to judge the human-machine confrontation, so as to provide an operation prompt for improving this human-computer confrontation, which is more consistent with the operation of clinical doctors. For approximation, prompt results and guidance information are also more meaningful.

The present invention also provides a ventilation monitoring method, as shown in Figure 11, comprising the steps:

Step 1000: Acquire at least one parameter data that can characterize the human-machine confrontation event of the patient.

The parameter data includes at least one type of device ventilation parameter data and patient physiological parameter data. For example, the ventilation parameter data includes at least one of airway pressure, airway flow rate and gas volume of the medical ventilation device, and the physiological parameter data includes the patient's esophageal pressure, intragastric pressure, transdiaphragmatic pressure, carbon dioxide concentration, At least one of diaphragm electromyography and the like. It should be noted that equipment ventilation parameter data and patient physiological parameters are two types of parameter data, while airway pressure is a kind of parameter data. If the airway pressure and airway flow rate are obtained, then the obtained data is one type two parameter data.

The above parameter data can be measured by external equipment. The external equipment can be various sensors placed in the patient's body or various plug-ins or modules set on the medical ventilation equipment, such as an esophageal pressure sensor placed in the patient's esophagus for measuring. A carbon dioxide module for carbon dioxide concentration, etc. In other embodiments, the medical ventilation device itself may further include a parameter measurement device 60 for acquiring the above-mentioned parameter data.

Step 2000: Activating an algorithm module associated with at least one parameter data to obtain the identification result of the human-machine confrontation event of the patient during the ventilation process. Wherein, the algorithm module includes a plurality of algorithm modules, and different algorithm modules perform calculation based on different data combinations formed by at least one parameter data, so as to obtain the identification result of the human-machine confrontation event during the ventilation process of the patient.

In this step, an association can be defined between the algorithm module and the parameter data on which the calculation is based. The above data combination refers to a combination in a broad sense, and when there is only one parameter data, it can be called a data combination, that is, the algorithm module can perform calculations based on a data combination with only one parameter data. In some embodiments, the relationship between the algorithm module and its corresponding data combination is shown in FIG. 2 , wherein, the data combination corresponding to the algorithm module A includes only one parameter data of airway pressure, and the data combination corresponding to the algorithm module D includes: Two parameters, airway pressure and gas volume, were obtained. In this embodiment, the corresponding algorithm module is enabled only when the acquired at least one parameter data meets the requirements of the algorithm module for the data combination of the parameter data. For example, taking the example shown in FIG. 2 , if the parameter data of the airway pressure is obtained, then the algorithm module A is activated, but the algorithm module B is not activated based on the airway pressure. In other embodiments, the requirements of the data combination for the parameter data also include the requirements for the validity of the parameter data. Also taking FIG. 2 as an example, if the airway pressure is obtained, the validity of the airway pressure can be judged. If the track pressure is valid data, then enable algorithm module A. The validity requirements of the above parameter data may be requirements such as data range and periodicity. For example, if the acquired airway pressure is not within a preset range, it is determined that the airway pressure is not valid.

FIG. 2 shows a method of data combination formed by parameter data. In other embodiments, the data combination of each algorithm module may include airway flow rate and/or airway pressure, or any other number of parameter data. The combination makes the features of man-machine confrontation more obvious and easier to obtain.

The above-mentioned algorithm modules can be corresponding computer programs stored in the memory, and the number of algorithm modules can also be increased or decreased. The algorithm module calculates and obtains the recognition result based on the parameter data, which may include recognizing the human-machine confrontation event or not recognizing the human-machine confrontation event, that is to say, the algorithm module A in Fig. Whether the machine-machine confrontation event occurs, the algorithm module C can judge whether the human-machine confrontation event occurs according to the airway pressure and the airway flow rate and based on a certain calculation method. Although both algorithm module A and algorithm module C use the parameter data of airway pressure, the calculation methods of algorithm module A and algorithm module C can be independent, that is, the identification results of algorithm module A and algorithm module C can be independent

In some embodiments, the identification result further includes the type of the identified man-machine confrontation event, that is, the algorithm module A can not only identify whether the man-machine confrontation event occurs according to the airway pressure, but also identify the occurrence of the man-machine confrontation event according to the airway pressure. What type of man-machine confrontation is. The types of man-machine confrontation mentioned above include one or more of invalid trigger events, double trigger events, false trigger events, reverse trigger events, trigger delay events, switching advance events, switching delay events, and low flow rate events. In the following, the algorithm module C and the algorithm module D are taken as examples to describe how the algorithm module obtains the above identification results.

Figure 3 shows the corresponding airway pressure waveform and airway flow rate waveform generated according to the airway pressure and airway flow rate. If a human-machine confrontation event occurs in a patient, corresponding "traces" will be left on the airway pressure waveform and airway flow velocity waveform, which is manifested as abnormal waveform characteristics, and different types of human-machine confrontation event waveform characteristics. different. Taking an invalid trigger event as an example, Figure 4 shows the waveform characteristics of the airway pressure waveform and the airway flow velocity waveform when the invalid trigger event occurs. The upper part in Figure 4 is the airway pressure waveform, and the lower part in Figure 4 is the Airway velocity waveform. If a sudden rise of the airway flow velocity waveform and a sudden drop of the airway pressure waveform occur simultaneously at a certain moment, and some waveform characteristics in the waveform change meet the preset threshold conditions, it means that an invalid trigger event occurs in the patient. The above-mentioned waveform characteristics may include the change amplitude, change rate (first-order derivative), second-order derivative, and change duration of the waveform. The algorithm module C performs feature extraction on the airway pressure waveform and the airway flow velocity waveform, that is, calculates the parameter data. If the extracted feature satisfies the preset threshold condition, the algorithm module C will obtain the identification result of the invalid trigger event of the patient. In some embodiments, it can be set that if the waveform characteristics of the airway flow velocity waveform and the waveform characteristics of the airway pressure satisfy corresponding threshold conditions at the same time, the algorithm module C obtains the identification result of the invalid trigger event of the patient. In other embodiments, it may also be within a certain time range when one of the waveform characteristics of the airway flow velocity waveform and the waveform characteristics of the airway pressure satisfies the corresponding threshold condition, and the other waveform characteristic also satisfies the corresponding threshold condition, The algorithm module C obtains the identification result of the invalid triggering event of the patient.

Figure 5 shows the corresponding airway pressure waveform and gas volume waveform generated according to the airway pressure and gas volume. Taking the double-trigger event as an example, the algorithm module D can calculate the duration of the patient's breathing cycle according to the obtained airway pressure. A double-trigger event is considered to exist during this breath cycle. In this embodiment, the preset threshold may be a result obtained by jointly limiting multiple related threshold conditions. For example, the average inspiratory time and average expiratory time of multiple (for example, 12) breathing cycles before this breathing cycle can be calculated. If the expiratory time of this breathing cycle is less than half of the average inspiratory time, the average expiratory time The minimum value between half of the time and a fixed time threshold (for example, 500ms), the recognition result of the algorithm module D is the occurrence of a double-triggered event in this breathing cycle. In addition, the algorithm module D can also determine whether there is a double trigger event in combination with the gas volume. Because there will be a very short expiratory phase in the breathing cycle with double trigger events, resulting in a small exhaled tidal volume of the patient, the algorithm module D can calculate the exhaled tidal volume and/or the inhaled tidal volume according to the obtained gas volume. When it is recognized that the expiratory tidal volume is less than a threshold (for example, 1/2 of the inspiratory tidal volume in the current breathing cycle), or the result of subtracting the expiratory tidal volume from the inspiratory tidal volume is greater than IBW*k (IBW is ideal body weight, k is the coefficient threshold, for example k=1ml/kg), then the recognition result of algorithm module D is the occurrence of a double-triggered event.

The above example illustrates how the algorithm module obtains the recognition result of the human-machine confrontation event. Among them, the algorithm module C can not only identify invalid trigger events, but also identify other types of man-machine confrontation events according to airway pressure and airway flow rate. For a certain type of human-machine confrontation event, there may be multiple algorithm modules that can determine whether this type of human-machine confrontation event occurs. For example, in addition to algorithm module C, algorithm module E, algorithm module G, algorithm module H, The algorithm module I and the algorithm module K can also identify whether an invalid trigger event occurs, wherein, a plurality of algorithm modules can be calculated according to different parameter data using different calculation methods. Based on this, the above and the following relate to The threshold condition, threshold time, threshold, etc., can have different setting values or preset values according to different algorithm modules and/or different types of human-machine confrontation events identified, so as to avoid repeating the same judgment logic for different algorithm modules ( calculation and determination method). The invention identifies man-machine confrontation events from different angles, so that the identification result has better robustness and higher accuracy.

For example, in the embodiment shown in FIG. 6, the airway pressure caused by the invalid trigger event of the patient has a small change range. If only relying on the airway pressure to determine whether the invalid trigger event occurs, it may occur that the characteristics of the airway pressure are insufficient. Obviously lead to inaccurate judgment. Figure 6 also combines esophageal pressure to jointly determine whether an invalid trigger event occurs. When the patient breathes spontaneously, the patient's diaphragm, intercostal muscles and other respiratory muscles actively contract, causing the pressure in the pleural cavity to drop. Clinically, the monitoring of this esophageal pressure is approximately equivalent to monitoring the patient's pleural pressure, so the identification of the esophagus The change in pressure can distinguish the patient's spontaneous breathing state. In Figure 6, it can be seen that there is a change in the esophageal pressure caused by the patient's spontaneous inhalation, and there is also a slight pressure amplitude change in the airway pressure. These two points are also invalid triggers for the patient. the moment of the event. In this embodiment, the characteristics of the patient's spontaneous inhalation stage can be obtained through the identified characteristics of the esophageal pressure and the downward pressure swing. Combining this characteristic with the identified characteristics of the airway pressure, an invalid trigger event can be determined together, that is, If the patient's spontaneous breathing effort is detected through a change in esophageal pressure within a threshold time after the airway pressure is identified as having the characteristics of the invalid trigger event, then it can be determined that the patient has an invalid trigger event. Two algorithm modules can be set to obtain the recognition result of the human-machine confrontation event based on the airway pressure and the esophageal pressure respectively, or in the embodiment shown in FIG. 2, an algorithm module E is set, and the algorithm module E can Combined with esophageal pressure to judge whether human-machine confrontation events (such as invalid trigger events) occur, thus improving the accuracy of judgment.

For another example, the embodiment shown in FIG. 7 also combines the parameter data of carbon dioxide concentration to jointly determine whether an invalid trigger event occurs. In some embodiments, the carbon dioxide concentration of the patient can be monitored through an external carbon dioxide module. When the patient is in the exhalation phase, the carbon dioxide waveform curve can be seen rising, and when the patient is in the inhalation phase, the carbon dioxide waveform curve can be seen. decline. Therefore, when it is detected that the carbon dioxide waveform has obvious waveform decline characteristics (for example, the decline amplitude and the decline duration meet a certain threshold), it means that the patient has inspiratory effort. event, the time corresponding to the arrow in Figure 7 is the time when the invalid trigger event occurs. It can be seen that judging invalid triggering by carbon dioxide waveform characteristics can be combined with the original method of judging invalid triggering by airway pressure, airway flow rate, gas volume curve, etc., thereby improving the accuracy of judgment. By extracting the features of the carbon dioxide waveform through the algorithm module, that is to say, calculating the change of carbon dioxide concentration, it is possible to detect that the carbon dioxide waveform has obvious waveform decline characteristics, so as to obtain the recognition result of the human-machine confrontation event. For example, as shown in Figure 2 In the embodiment of the invention, the algorithm module H is enabled to jointly determine whether the invalid trigger event occurs with other algorithm modules.

Step 3000: Determine the human-machine confrontation event that occurs in the patient according to the recognition results obtained by different algorithm modules.

In some embodiments, step 3000 may include:

Step 3100: Determine whether the same type of human-machine confrontation event identified by each algorithm module occurs according to the identification result of each algorithm module capable of identifying the same type of human-machine confrontation event.

In some embodiments, as shown in FIG. 8 , each type of man-machine confrontation event in the figure has a corresponding judgment module, and each judgment module includes each algorithm module capable of identifying the same type of man-machine confrontation event, The output result of the judgment module is that the corresponding human-machine confrontation event occurs or does not occur, that is, the judgment module judges the corresponding type of human-machine confrontation event according to the recognition results of each algorithm module that can identify the same type of human-machine confrontation event. does it happen. Based on this, step 3100 specifically includes:

Step 3110: Obtain the credibility of the first algorithm module. The first algorithm module is an algorithm module whose recognition result is that a human-machine confrontation event is recognized. For example, in the invalid trigger judgment module, the recognition results of the algorithm module C, the algorithm module E and the algorithm module G are that the invalid trigger event is recognized, then The three algorithm modules are defined as the first algorithm module. In other embodiments, the first algorithm module is an algorithm module that identifies the type of human-machine confrontation event corresponding to the judgment module.

Step 3120: According to the credibility of each first algorithm module, determine whether the same type of human-machine confrontation event identified by each first algorithm module occurs.

In this embodiment, whether the invalid triggering event occurs is judged according to the reliability of the algorithm module C, the algorithm module E and the algorithm module G. The reliability can be related to the signal-to-noise ratio of the parameter data associated with the algorithm module, the degree of feature clarity and at least one correlation of regularity, etc. For example, when airway pressure and airway flow rate are used to identify invalid trigger events, the degree of clarity of the feature can be divided according to the magnitude of the identified features such as the pressure drop range and the change range of the first derivative of the flow rate. The larger the variation range of the first derivative of the flow velocity is, the higher the degree of feature definiteness and the higher the reliability. When using physiological signals of patients such as esophageal pressure or diaphragm electromyography for human-machine confrontation identification, the quality of signal monitoring and the signal-to-noise ratio will affect the reliability of the corresponding algorithm module. For example, when it is detected that the signal fluctuation range is weak, or there is no regular periodic motion, the reliability is low. On the contrary, if the signal fluctuation is greater than a certain threshold, and it is detected that there is a regular fluctuation period in the recent period of time, the reliability is high. .

In some embodiments, step 3120 may include:

According to the credibility of the first algorithm module, a credibility score corresponding to the credibility of the first algorithm module is generated.

Combined with the weight coefficient corresponding to the first algorithm module, the credibility score of the first algorithm module is modified to obtain a modified credibility score.

Wherein, for different types of human-machine confrontation events, different algorithm modules are preset with corresponding weight coefficients. For example, the weight coefficient can be determined based on at least the degree of association between the algorithm module and the type of human-machine confrontation event, for example, when identifying a trigger delay event , considering that in general, the diaphragm electromyographic signal will be earlier than the esophageal pressure signal, so a higher weight coefficient is given to the algorithm module that uses the diaphragm electromyography for identification. The advantage of this is that it can be based on clinical consensus or parameter essential characteristics. Modify the algorithm results to a certain extent to make the algorithm results more credible.

According to the revised reliability score, the sum of the reliability scores of each first algorithm module is calculated.

Determine whether the sum of the reliability scores is greater than the preset threshold. If it is greater than the preset threshold, the same type of man-machine confrontation event identified by each first algorithm module occurs, otherwise, the same type of man-machine identified by each first algorithm module occurs. The confrontation did not take place.

The sum of the credibility scores of the above-mentioned first algorithm modules refers to the sum of the credibility scores of the first algorithm modules in the same judgment module. If the credibility of each first algorithm module in the judgment module is If the sum of the scores is greater than the preset threshold, the output result of the judgment module is that a human-machine confrontation event of the corresponding type occurs.

The above method comprehensively considers the credibility of the identification results of each first algorithm module. When the total credibility of each first algorithm module in the judgment module satisfies a certain condition, the output result of the judgment module is the corresponding type of man-machine confrontation. event happens.

In other embodiments, step 3100 specifically includes: according to the ratio and/or quantity relationship between the first algorithm module and the second algorithm module, judging whether the same type of human-machine confrontation event identified by each algorithm module occurs, wherein, The first algorithm module is an algorithm module whose recognition result is that a man-machine confrontation event is recognized, and the second algorithm module is an algorithm module whose recognition result is that no man-machine confrontation event is recognized. For example, in the invalid trigger judgment module, the identification results of the algorithm module C, the algorithm module E and the algorithm module G are that the invalid trigger event is identified, then these three algorithm modules are defined as the first algorithm module, the algorithm module H, the algorithm module I and the algorithm module K do not identify an invalid trigger event, then these three algorithm modules are the second algorithm module. In this example, there are three first algorithm modules. If there are more than two first algorithm modules in a judgment module, the output result of the judgment module is that a man-machine confrontation event of the corresponding type occurs, and the judgment is invalid. The output of the module is that an invalid trigger event has occurred. The advantage of this embodiment is that misidentification can be avoided as much as possible, because too many misidentifications will cause user information fatigue and weaken the original prompting effect.

Step 3200: Determine various types of human-machine confrontation events that occur in the patient according to whether different types of human-machine confrontation events occur.

What types of human-machine confrontation events have occurred in the patient can be determined according to the output results of each judgment module, and the output results are determined based on the recognition results of each internal algorithm module.

Step 4000: Output the human-machine confrontation events that occur in the patient.

In some embodiments, the identified man-machine confrontation event is marked near the corresponding feature of the corresponding parameter data waveform, for example, as shown in FIG. , the triangle mark is used to indicate the position of the corresponding feature of the human-machine confrontation event on the waveform, and the string below the triangle mark indicates the type name of the human-machine confrontation event, (IE is the abbreviation of invalid trigger event, DT is the double trigger event Short for , RT is short for reverse trigger event). The marking method can also be in the form of any symbol, color, or character string to distinguish different types of human-machine confrontation events. The advantage of this is that the types of human-machine confrontation events can be matched with the corresponding features on the waveform. Experienced doctors can directly judge the accuracy of the recognition results from the marking results. The markings on the waveform are also related to the occurrence rate of human-machine confrontation events. The changes of the monitoring value are consistent, and the doctor can clearly understand the meaning of the monitoring value of the incidence of human-machine confrontation. For general medical staff, they can also learn by comparing the waveform annotation. If the waveform annotation is not performed, there is only the monitoring value of the occurrence rate of human-machine confrontation events, and the doctor cannot know whether the recognition algorithm is accurate. Another advantage of annotating results directly on the waveform interface is that the user can still observe the latest ventilation waveform without the need to freeze the waveform or switch to other interfaces. In addition, the user does not need to perform additional operations to see the recognition results and waveforms Features, easy to use.

In some embodiments, as shown in FIG. 10 , according to the identification result of the human-machine confrontation event, the occurrence rate of various types of human-machine confrontation events is counted. Such statistics can be statistics of the occurrence rate in a recent period of time, or statistics of the occurrence rate of human-machine confrontation events in a certain number of recent respiratory cycles. When the occurrence rate of a certain type of human-machine confrontation event is higher than a certain threshold, a specific area on the main interface prompts the user that there are too many human-machine confrontation events of this type, and gives operation suggestions. Taking invalid triggers as an example, Figure 10 shows a prompt method when the occurrence rate of invalid trigger events is detected to be higher than 10%. This method mainly embodies two pieces of information. The first one is to remind the user of the type name of too many man-machine confrontation events, and the second one is to give operation suggestions. For example, the prompt information in Figure 10 is based on the trigger sensitivity in the current ventilation parameters. Threshold, prompts the user to lower the threshold setting.

The invention adopts multiple algorithm modules according to at least one parameter data, and uses different algorithms to comprehensively analyze and judge the human-machine confrontation event, so that the recognition result of the human-machine confrontation event is more accurate and robust.

The above specific examples are used to illustrate the present invention, which are only used to help understand the present invention, and are not intended to limit the present invention. For those skilled in the art, according to the idea of the present invention, the above-mentioned specific embodiments can be changed.

Claims (26)

1. A medical ventilation device, characterized in that it includes:

   Air source interface, used to connect air source;

   A patient interface for connecting to the patient's respiratory system;

   A breathing circuit for connecting the air source interface and the patient interface to deliver the gas provided by the air source to the patient;

   Respiratory assistance device, used to provide respiratory support power to control the delivery of gas provided by the gas source to the patient;

   a processor, configured to acquire at least one parameter data capable of characterizing a human-machine confrontation event in a patient, the parameter data including at least one type of device ventilation parameter data and patient physiological parameter data;

   The processor is further configured to, when receiving the parameter data, enable an algorithm module associated with the parameter data to obtain the identification result of the patient-machine confrontation event during the ventilation process; wherein the The algorithm module includes multiple, and different algorithm modules perform calculations based on different data combinations formed by at least one parameter data to obtain the identification result of the patient's human-machine confrontation event during the ventilation process; the identification result includes: identifying the man-machine confrontation event or the man-machine confrontation event is not recognized;

   And according to the recognition results obtained by different algorithm modules, determine and output the man-machine confrontation event that occurs in the patient.
2. The medical ventilation device of claim 1, wherein enabling an algorithmic module associated with the parameter data comprises:

   When it is determined that the acquired at least one kind of parameter data satisfies the requirement of the algorithm module for the data combination of the parameter data, the algorithm module is activated.
3. 3. The medical ventilation device of claim 2, wherein the data combination requirements for parameter data include validity requirements for the parameter data.
4. The medical ventilation device according to claim 1, wherein the identification result of the algorithm module further includes the type of the identified human-machine confrontation event.
5. The medical ventilation device according to claim 4, wherein the types of the human-machine confrontation events include invalid trigger events, double trigger events, false trigger events, reverse trigger events, trigger delay events, switching advance events, switching events One or more of a delay event and an underflow event.
6. The medical ventilation device according to claim 4, wherein determining the human-machine confrontation event that occurs in the patient according to the identification results obtained by different algorithm modules, comprising:

   According to the recognition results of each algorithm module capable of recognizing the same type of human-machine confrontation event, determine whether the same type of human-machine confrontation event identified by each algorithm module occurs;

   According to whether different types of man-machine confrontation events occur, various types of man-machine confrontation events occurring in the patient are determined.
7. The medical ventilation device according to claim 6, wherein, according to the identification result of each algorithm module capable of identifying the same type of human-machine confrontation event, it is determined whether the same type of human-machine confrontation event identified by each algorithm module occurs, including:

   Obtaining the credibility of the first algorithm module, wherein the first algorithm module is an algorithm module whose identification result is that the human-machine confrontation event is identified;

   According to the credibility of each of the first algorithm modules, it is determined whether the same type of human-machine confrontation event identified by each of the first algorithm modules occurs.
8. The medical ventilation device according to claim 7, wherein the reliability is related to at least one of a signal-to-noise ratio, a degree of feature definiteness, and a degree of regularity of the parameter data associated with the algorithm module.
9. The medical ventilation device according to claim 7, wherein, according to the credibility of each first algorithm module, judging whether the same type of human-machine confrontation event identified by each first algorithm module occurs, comprising:

   generating a credibility score corresponding to the credibility of the first algorithm module;

   Combining the corresponding weight coefficients of the first algorithm module, the credibility score of the first algorithm module is revised to obtain a revised credibility score, wherein, for different types of human-machine confrontation events, different algorithms are used. The module is preset with the corresponding weight coefficient;

   Calculate the sum of the credibility scores of each first algorithm module according to the revised credibility score;

   Judging whether the sum of the reliability scores is greater than the preset threshold, if it is greater than the preset threshold, the same type of human-machine confrontation event identified by each first algorithm module occurs, otherwise, the same type identified by each first algorithm module. The man-machine confrontation incident did not take place.
10. The medical ventilation device according to claim 9, wherein the weight coefficient is determined based on at least the degree of association between the algorithm module and the type of human-machine confrontation event.
11. The medical ventilation device according to claim 6, wherein, according to the identification result of each algorithm module capable of identifying the same type of human-machine confrontation event, it is determined whether the same type of human-machine confrontation event identified by each algorithm module occurs, including:

    According to the ratio and/or quantity relationship between the first algorithm module and the second algorithm module, it is judged whether the same type of human-machine confrontation event identified by each algorithm module occurs, wherein the first algorithm module is the identification result of identifying The algorithm module of the human-machine confrontation event, and the second algorithm module is an algorithm module whose recognition result is that the human-machine confrontation event is not recognized.
12. The medical ventilation device according to claim 1, wherein acquiring at least one parameter data capable of characterizing a human-machine confrontation event in a patient comprises:

    receive said parameter data measured by an external device, or

    The medical ventilation equipment further includes a parameter measurement device for acquiring at least one parameter data that can characterize a patient-machine confrontation event, including the parameter measurement device acquiring the parameter data through measurement.
13. The medical ventilation device of claim 1, wherein the ventilation parameter data includes at least one of airway pressure, airway flow rate, and gas volume;

    The physiological parameter data includes at least one of esophageal pressure, intragastric pressure, transdiaphragmatic pressure, carbon dioxide concentration, and diaphragmatic muscle electricity.
14. A method for monitoring ventilation, comprising the steps of:

    acquiring at least one parameter data capable of characterizing a human-machine confrontation event in a patient, the parameter data including at least one type of device ventilation parameter data and patient physiological parameter data;

    enabling an algorithm module associated with the at least one parameter data to obtain an identification result of a human-machine confrontation event of the patient during ventilation, wherein the algorithm module includes a plurality of different algorithm modules based on at least one A different data combination formed by parameter data is calculated to obtain a recognition result, and the recognition result includes: identifying the man-machine confrontation event or not recognizing the man-machine confrontation event;

    Determine the human-machine confrontation event that occurs in the patient according to the recognition results obtained by the different algorithm modules;

    The man-machine confrontation events occurring in the patient are output.
15. 15. The ventilation monitoring method of claim 14, wherein enabling an algorithm module associated with the at least one parameter data comprises:

    When it is determined that at least one kind of parameter data satisfies the parameter data requirements of the data combination of the algorithm module, the algorithm module is activated.
16. 16. The ventilation monitoring method of claim 15, wherein the data combination requirements for parameter data include validity requirements for the parameter data.
17. The ventilation monitoring method according to claim 14, wherein the identification result of the algorithm module further includes the type of the identified man-machine confrontation event.
18. The ventilation monitoring method according to claim 17, wherein the types of the man-machine confrontation events include invalid trigger events, double trigger events, false trigger events, reverse trigger events, trigger delay events, switching advance events, switching events One or more of a delay event and an underflow event.
19. The ventilation monitoring method according to claim 17, wherein determining the human-machine confrontation event that occurs in the patient according to the identification results obtained by the different algorithm modules, comprising:

    According to the recognition results of each algorithm module capable of recognizing the same type of human-machine confrontation event, determine whether the same type of human-machine confrontation event identified by each algorithm module occurs;

    According to whether different types of man-machine confrontation events occur, various types of man-machine confrontation events occurring in the patient are determined.
20. The ventilation monitoring method as claimed in claim 19, wherein, according to the identification result of each algorithm module capable of identifying the same type of man-machine confrontation event, judging whether the same type of man-machine confrontation event identified by each algorithm module occurs, including:

    Obtaining the credibility of the first algorithm module, wherein the first algorithm module is an algorithm module whose identification result is that the human-machine confrontation event is identified;

    According to the credibility of each of the first algorithm modules, it is determined whether the same type of human-machine confrontation event identified by each of the first algorithm modules occurs.
21. 21. The ventilation monitoring method according to claim 20, wherein the reliability is related to at least one of a signal-to-noise ratio, a degree of feature definiteness, and a degree of regularity of the parameter data associated with the algorithm module.
22. The ventilation monitoring method according to claim 20, wherein, according to the credibility of each first algorithm module, judging whether the same type of human-machine confrontation event identified by each first algorithm module occurs, comprising:

    generating a credibility score corresponding to the credibility of the first algorithm module;

    Combining the corresponding weight coefficients of the first algorithm module, the credibility score of the first algorithm module is revised to obtain a revised credibility score, wherein, for different types of human-machine confrontation events, different algorithms are used. The module is preset with the corresponding weight coefficient;

    Calculate the sum of the credibility scores of each of the first algorithm modules according to the revised credibility score;

    Judging whether the sum of the reliability scores is greater than the preset threshold, if it is greater than the preset threshold, the same type of human-machine confrontation event identified by each first algorithm module occurs, otherwise, the same type identified by each first algorithm module. The man-machine confrontation incident did not take place.
23. The ventilation monitoring method according to claim 22, wherein the weight coefficient is determined at least based on the degree of correlation between the algorithm module and the type of human-machine confrontation event.
24. The ventilation monitoring method as claimed in claim 19, wherein, according to the identification result of each algorithm module capable of identifying the same type of man-machine confrontation event, judging whether the same type of man-machine confrontation event identified by each algorithm module occurs, including:

    According to the ratio and/or quantity relationship between the first algorithm module and the second algorithm module, it is judged whether the same type of human-machine confrontation event identified by each algorithm module occurs, wherein the first algorithm module is the identification result of identifying The algorithm module of the human-machine confrontation event, and the second algorithm module is an algorithm module whose recognition result is that the human-machine confrontation event is not recognized.
25. The ventilation monitoring method of claim 14, wherein the ventilation parameter data includes at least one of airway pressure, airway flow rate and gas volume;

    The physiological parameter data includes at least one of esophageal pressure, intragastric pressure, transdiaphragmatic pressure, carbon dioxide concentration, and diaphragmatic muscle electricity.
26. A computer-readable storage medium, characterized by comprising a program that can be executed by a processor to implement the method according to any one of claims 14-25.