Breathing analyzer, artificial respiration device and method for breathing analysis

Abstract

A breathing analyzer (1) for detecting breathing events, comprising an electronic computing and storage unit (2) for receiving a signal (3) of an artificial breathing pressure and / or a breathing flow and / or a breathing volume of a breathing gas and detecting the curve of the signal by means of a curve analyzer (4) during a predetermined analysis period. The computing and storage unit is for determining a signal segment (5) from the signal curve, for classifying the signal segment into one of a plurality of breathing categories by means of a comparator (7) and for storing the frequency of occurrence of the classified breathing category during the analysis period in a frequency counter (8) of the storage unit (2), and for storing at least one of the signal segments, but with a lower frequency of occurrence than detected in the frequency counter, for each classified breathing category in the storage unit. An artificial respiration device for artificial respiration, comprising a breathing analyzer according to the invention. A method for detecting breathing events.

Description

Technical Field

[0001] The invention relates to a breathing analyzer for identifying breathing events of a person who is artificially respiring with breathing gas, for example by means of an artificial respiration device. The invention also relates to an artificial respiration device for artificially respiring a person with breathing gas, and a method for identifying breathing events of a person who is artificially respiring with breathing gas. Background Art

[0002] In artificial respiration devices for artificially respirating a person with breathing gas, it is generally advantageous if there is a possibility to detect breathing disturbances during artificial respiration. Such a detection allows, on the one hand, the effectiveness of the artificial respiration to be controlled by the artificial respiration device, and, on the other hand, allows better identification of underlying diseases and secondary diseases. Furthermore, the control of the artificial respiration device can be adapted (dynamically) to the detected breathing patterns or events and individually to the person to be artificially respirated.

[0003] For example, EP 1 136 094 B1 discloses an artificial respiration device with a device for monitoring artificial respiration parameters. The time curve of the artificial respiration parameter to be monitored is detected by a sensor and provided to an analyzer, which performs pattern recognition and is coupled to a classifier for supporting the evaluation and to a memory, which provides the required comparison patterns for the pattern recognition. Summary of the invention

[0004] Against this background, the object of the present invention is to provide a breath analyzer for identifying breathing events of a person who is artificially breathing with breathing gas, an artificial respiration device for artificially respiring a person with breathing gas, and a method for identifying breathing events of a person who is artificially respiring with breathing gas, which significantly reduces the time required for evaluating the breathing events identified during the entire analysis period. In addition, the breath analyzer, the artificial respiration device and the breathing analysis method should be compact and can be implemented using relatively uncomplicated technical means and thus cost-effectively.

[0005] This object is achieved by a respiration analyzer having the features of the invention, by an artificial respiration device having the features of the invention, and by a method for respiration analysis having the features of the invention. The corresponding subsequent description discloses further particularly advantageous embodiments of the invention.

[0006] It should be pointed out that the features listed individually in the claims can be combined with one another in any technically meaningful manner and reveal further embodiments of the invention. The description, in particular in conjunction with the drawings, additionally characterizes and explains the invention in detail.

[0007] Furthermore, it should be pointed out that the conjunction "and / or" used here between two features and connecting these two features to each other should always be understood as meaning that in a first configuration scheme of the object according to the present invention only the first feature can exist, in a second configuration scheme only the second feature can exist, and in a third configuration scheme both the first feature and the second feature can exist.

[0008] According to the invention, a breathing analyzer for identifying breathing events of a person who is artificially ventilating with breathing gas has an electronic computing and storage unit, which can be realized, for example, in the form of a processor-based electronic control device (e.g., a microprocessor, a microcontroller, a DSP, etc.) and at least one electronic memory (e.g., a RAM, a ROM, an EPROM, a flash memory, a magnetic memory, etc.) cooperating with the control device. According to the invention, the computing and storage unit is designed and arranged to receive a signal, in particular an electronic signal, corresponding to the artificial breathing pressure and / or the breathing flow and / or the breathing volume of the breathing gas supplied to the person and to detect the curve of the signal (hereinafter also referred to as signal curve) by means of a curve analyzer during a predetermined analysis period.

[0009] Furthermore, according to the invention, the computing and storage unit is designed and arranged to determine a signal segment from the ascertained signal curve (in particular immediately after the ascertainment of the signal curve), the signal segment being determined according to a segment duration predefined or identified from the signal curve, wherein the segment duration is significantly shorter than the (entire) analysis duration. The analysis duration can, for example, be in the range of a few minutes to one or several hours or even to one or several days, but the segment duration preferably regularly includes a duration in the range of one or several seconds to one minute, in particular preferably up to a maximum of one minute, two minutes or a maximum of three minutes.

[0010] Furthermore, the computing and storage unit is designed and arranged to classify the determined signal segments into one of a plurality of breathing classes (also referred to here as a classified / identified breathing pattern or also simply referred to as a breathing pattern) by means of a comparator and to store the corresponding occurrence frequency of the classified (i.e. occurring) breathing class during the analysis period or the entire analysis time period in a frequency counter provided for each breathing class in the storage unit and to store, for each classified (i.e. occurring) breathing class, at least one of the signal segments determined during the analysis period, but always less than the occurrence frequency detected in the frequency counter (i.e. the total number of identified signal segments of the breathing class determined during the entire analysis time period) in the storage unit. Classification of the signal segment into a breathing class means that one or more determined features of the signal curve in the signal segment coincide with or are at least similar to features of one of the plurality of breathing classes.

[0011] Even in the case of a relatively long analysis period, for example several hours to several days, a diagnosis for a person who is performing artificial respiration can be generated with very little time expenditure based on the frequency counter from a small number of, but relevant, stored analysis results. In addition, the frequency counter (even in the case of a large number of different breathing categories) requires very few storage resources to be provided. Even if one or several stored signal segments are added for each breathing category, the storage requirement of the breathing analyzer according to the invention as a whole is still only a very small part of the originally received curve of the signal representing the artificial breathing pressure and / or breathing flow and / or breathing volume of the breathing gas supplied to the person who is performing artificial respiration. In addition, according to the invention, one and the same breathing category is always stored less than all the signal segments identified in the analysis period, so that the diagnosis based on the stored signal segments is limited to a number (several) relevant exemplary signal segments, which significantly reduces the time expenditure for diagnosis, especially when there is a breathing state that is critical to the health of the person who is performing artificial respiration. If a (wired or wireless) data transfer from the storage unit to an evaluation unit or an expert user (e.g. a doctor) which is arranged remotely from the breath analyzer is provided or required for diagnosis, the time expenditure for such data transfer is advantageously also significantly reduced. In addition, the storage unit itself for relatively long-term monitoring scenarios and breath analysis can be equipped with a small memory, which not only saves installation space for implementing the breath analyzer, but also saves costs for its production.

[0012] It should be understood that the curve analyzer and / or comparator do not necessarily have to be a specific functional unit, but can be implemented as, for example, a software-based functional unit, which is implemented in a corresponding manner by a computing unit to provide the functionality described herein.

[0013] According to an advantageous embodiment of the invention, the computing and storage unit is designed and arranged to store only a predetermined maximum number of signal segments in the storage unit for each breathing class. In other words, the maximum number of signal segments to be stored for each breathing class can be predefined once at the beginning of the analysis period or before the breathing analyzer is put into use for the first time, for example a maximum of 3, 5 or 10 signal segments. Preferably, the breathing analyzer stores a maximum of 5 signal segments for each breathing class, particularly preferably a maximum of 3. Thus, on the one hand, there are enough examples of relevant signal curves identified by each breathing class for diagnosis, and on the other hand, the storage requirement for such a small maximum number of signal segments to be stored is always very low.

[0014] Depending on the application, it can be provided that the maximum number of signal segments to be stored per breathing class is the first breathing pattern identified and classified during the entire analysis period or, for example, the last (most recent) breathing pattern identified during the analysis period.

[0015] In order to prioritize the breathing categories, for example, with regard to the severity of the patient's health-related complication states associated with the respective breathing categories, an advantageous embodiment of the present invention provides that the number of a plurality of signal segments to be stored can be determined separately for at least two different breathing categories. Thus, for breathing categories that are particularly relevant in terms of health, for example, a higher maximum number of breathing patterns to be stored, identified during the analysis period, and classified can be predefined, whereas for less critical breathing categories a smaller number of signal segments to be stored can be determined. Thus, although a higher number of stored breathing patterns can be required for an accurate diagnosis in the case of a specific breathing category than for other breathing categories, an optimized, efficient memory utilization can be achieved while the memory requirement for storing all signal segments is still low.

[0016] Another advantageous configuration of the present invention proposes that the segment duration of the signal segment can be determined according to at least one predetermined signal curve standard. This signal curve standard is preferably selected from the following standard group: the standard group includes the slope of the detected signal curve and / or the slope curve, the occurrence of at least one maximum and / or minimum value of the signal curve and the like. Compared with the segment duration that is fixedly pre-set once at the beginning of the analysis duration, such a dynamic and automatic determination of the segment duration is significantly more flexible and can automatically adapt to the immediate breathing condition of the person who is to be artificially respired. Therefore, for example, the start and end of at least one complete breath can be identified using the aforementioned standard and the at least one complete breath can be determined as the signal segment to be analyzed. Similarly, only one determined signal curve part of the breath can also be included in the determined signal segment. The slope can be understood as the change of the signal curve, for example, in time. From the signal curve, it can be inferred that the signal first increases and then decreases in the signal curve, and vice versa, when the sign changes (the slope curve passes through the zero point). Therefore, the original signal curve can also be contoured according to the signal curve.

[0017] Particularly preferably, the segment duration can be measured such that a signal segment contains at least one complete breath, preferably 1 to a maximum of 10 breaths, particularly preferably at least 3 to a maximum of 5 breaths.

[0018] A further advantageous embodiment of the invention provides that the stored signal segments are provided with a time stamp defining them, for example a time and / or a date, so that in a subsequent diagnosis the occurrence of a respectively identified breathing pattern can be easily determined and correspondingly taken into account in the evaluation.

[0019] In addition, another advantageous embodiment of the invention provides that the comparator is designed and arranged to classify the signal segment into the corresponding respiration class by means of pattern recognition and / or by means of a comparison of the signal curve in the signal segment with a predetermined number of predefined reference signal curves prestored in a memory unit and / or by means of a detected slope and / or slope curve of the signal curve in the signal segment and / or by means of one or more detected maximum and / or minimum values ​​of the signal curve in the signal segment. The slope or the slope curve can determine a specific profile of the signal curve in the signal segment.

[0020] According to another advantageous embodiment of the invention, the comparator is designed and arranged to store a predetermined number of selected signal segments that are classified into the corresponding breathing class (i.e., identified or occurring) during the analysis period as additional, newly added, adaptively learned reference signal curves in the storage unit and to take them into account when classifying subsequently detected signal segments into the breathing class. Thus, during the analysis period, it is possible to automatically and dynamically adapt the breath analyzer to the individual breathing of the person in order to further improve the quality of the analysis result (classification rate and classification confidence).

[0021] According to another preferred configuration of the present invention, the breathing category at least includes the breathing types of inspiration, exhalation, pause and cough. Preferably, the breathing category also additionally includes the following breathing types, which represent inspiratory flow limitation and / or expiratory flow limitation and / or intrinsic PEEP (positive end-expiratory pressure) and / or breathing asynchronous type and / or double breathing effort (for example, additional inspiratory effort at the beginning of exhalation due to too short inspiratory time, which is also referred to as double trigger here), and the breathing asynchronous type is, for example, inefficient breathing effort (for example, inefficient inspiratory effort, lack of inspiratory trigger). The breathing asynchronous type especially represents the asynchrony between the breathing of the person performing artificial respiration and the artificial respiration pressure of the breathing gas (for example, predetermined by the artificial respiration device). For example, by comparing the signal curve in the signal segment with a predetermined number of predefined and / or adaptively learned reference signal curves prestored in a memory unit and / or by a detected slope and / or slope curve of the signal curve in the signal segment and / or by one or more detected maximum and / or minimum values ​​of the signal curve in the signal segment, inspiratory flow limitation, missing inspiratory trigger, coughing episodes, expiratory flow limitation, intrinsic PEEP and similar events can be identified by the respiratory analyzer.

[0022] According to another advantageous embodiment of the invention, the comparator is designed and arranged to receive additional information about the type of breathing, for example, spontaneous breathing or forced breathing and / or additional information about the rotation speed of an artificial respiration blower supplying breathing gas to the person and / or additional information about leakage losses of the breathing gas supplied to the person, and to take this additional information into account when classifying the signal segment into a breathing class. This additional information can be used, for example, to subject the signal curve in the signal segment to a corresponding preprocessing in advance (for example, normalization of the signal curve) in order to more accurately classify the correct breathing type by the comparator.

[0023] According to the invention, internal and / or optionally external or adaptable sensors can be used to provide information about the respiratory effort or respiratory gas values ​​or respiratory gas parameters. In particular, for the recognition of respiratory effort or respiratory effort, acceleration sensors on the trunk, strain or stress sensors in a belt around the trunk, EMG sensors at the respiratory muscles or diaphragm, pressure sensors in the respiratory tract, impedance sensors in a belt around the trunk, measurement of chest impedance via electrodes, measurement of respiratory murmurs, measurement of SpO2 or transcutaneous CO2 or CO2 in the exhaled air, measurement of pulse or EKG can be used individually or cumulatively. According to the principle of the invention, the signal curves of these sensors can be evaluated and used individually in combination to improve the pre-determination of the pressure and / or flow and / or volume of the artificial respiration device so that fewer or less severe respiratory events occur.

[0024] It should be noted that the evaluation of the signal curve in the signal segment can be performed based on the time curve of the signal (e.g., respiratory flow over time). Alternatively or additionally, an evaluation of the signal about the artificial respiration pressure, such as respiratory flow or respiratory volume (so-called P / V curve) can also be performed.

[0025] As another aspect of the present invention, an artificial respiration device for artificially respiring a person using a respiratory gas is provided, which has a sensor unit for determining the artificial respiration pressure and / or the respiratory flow and / or the respiratory volume of the respiratory gas supplied to the person. In addition, the artificial respiration device according to the present invention has a respiratory analyzer according to any of the above-mentioned configuration schemes, wherein the sensor unit is coupled to the computing and storage unit of the respiratory analyzer in a data-transferring manner and is configured to generate a signal corresponding to the determined artificial respiration pressure and / or the determined respiratory flow and / or the determined respiratory volume of the respiratory gas and supply the signal to the computing and storage unit.

[0026] With regard to the definitions of terms relating to the artificial respiration device and with regard to the effects and advantages that correspond to the characteristics of the artificial respiration device, reference is made in all respects to the above explanations regarding the meaningful definitions, effects and advantages of the breath analyzer according to the invention. The disclosure herein regarding the breath analyzer according to the invention should be able to be used in a meaningful way to define the artificial respiration device according to the invention, as long as this is not explicitly excluded here. Likewise, the disclosure herein regarding the artificial respiration device according to the invention should be able to be used in a meaningful way to define the breath analyzer according to the invention, as long as this is not also explicitly excluded here. In this regard, a repetition of the explanations regarding the breath analyzer according to the invention disclosed herein and the artificial respiration device according to the invention disclosed herein, which have the same features, their effects and advantages, is largely avoided in order to facilitate a more compact description.

[0027] Another advantageous configuration of the present invention proposes that the artificial respiration device has at least one data transmission interface, which is constructed and arranged to transmit the content of the frequency counter of each breathing category stored in the storage unit and / or at least one signal segment of each classified (i.e., occurring during the analysis duration) breathing category stored in the storage unit to a data receiver or a data receiving unit. Here, data transmission can be performed wired or wirelessly. The data transmission interface can be, for example, an electronic data transmission bus, a network interface (e.g., LAN), a data transmission modem, a USB interface, a radio transmission interface (e.g., infrared, Bluetooth, WIFI, GSM / LTE, and the like) or can also be a replacement storage medium, such as a memory card (flash memory), a USB memory stick / hard disk, and the like. The data receiver or the data receiving unit can be, for example, a (remote) data processing device, an expert user (e.g., a doctor), a display device (e.g., a display screen, a monitor), and the like, outside the device with an evaluation unit (e.g., evaluation software). Therefore, the data receiving unit can diagnose the analysis result at a later time point, especially when the artificial respiration device does not have an evaluation unit inside the device. Alternatively or additionally, however, the respirator can have an internal evaluation unit.

[0028] According to another advantageous embodiment of the invention, the artificial respiration device has a further sensor unit which is arranged and constructed to detect and supply to the breath analyzer the rotational speed of the artificial respiration blower and / or the leakage loss of breathing gas and / or the type of breathing, such as spontaneous breathing or forced breathing, when artificially respiring a person.

[0029] According to another advantageous embodiment of the present invention, the artificial respiration device has an evaluation unit, which is constructed and arranged to evaluate the analysis results of the breath analyzer in terms of the health-critical complication status of the person performing artificial respiration and store them in a storage unit and / or display them on a display device and / or transmit them to a data receiving unit outside the device. For this purpose, data transmission can be performed in particular via the above-mentioned data transmission interface. The data receiving unit can be understood in particular as a monitoring unit, a monitor in a hospital or sanatorium and as a remote monitoring server in the case of home artificial respiration. If the frequency and / or severity of the determined complications exceed a determined limit value, the artificial respiration device can also be constructed to output an alarm, for example, output an alarm on a display device and / or send an alarm to a data receiving unit.

[0030] According to a further advantageous embodiment of the invention, it is provided that the breathing class of inefficient breathing effort is identified in that the breathing flow during the exhalation phase is evaluated and the breathing flow during the exhalation phase at least temporarily has a sign inversion.

[0031] According to another advantageous configuration of the present invention, it is proposed to identify the breathing category of inefficient breathing effort in the following manner: analyze whether the exhaled air flow exceeds a certain threshold value (for example, in l / min) within a first time window at the beginning of exhalation, and then reduce it within a second time window until the exhaled air flow does not exceed a certain threshold value or even briefly produces a positive breathing flow, that is, the inspiratory breathing flow or at least the derivative of the respiratory flow shows a decrease in the expiratory breathing flow, wherein it is then checked whether the exhaled air flow increases again within a third time window before the next inhalation is identified, that is, a stronger exhaled air flow is regenerated or at least the derivative of the respiratory flow shows that the expiratory breathing flow has increased again.

[0032] According to a further advantageous embodiment of the invention, provision is made for the breathing class of inefficient breathing effort to be identified by identifying the inefficient breathing effort by pattern recognition which compares the curve of the respiratory flow with a previously stored test pattern representing an inefficient inspiratory effort.

[0033] According to a further advantageous embodiment of the invention, it is provided that the breathing class of inefficient breathing effort is identified by comparing the exhalation time with a plurality of preceding (typical) exhalation times or with an average exhalation time or with a predefined exhalation time to identify the breathing class of a double trigger.

[0034] According to another advantageous embodiment of the invention, it is proposed to identify the breathing category of inefficient breathing effort in the following manner: the breathing category of a double trigger is identified by comparing the measured flow with a theoretical breathing flow, wherein the measured flow continues to increase relative to the theoretical breathing flow after switching to the artificial respiration pressure for exhalation.

[0035] According to a further advantageous embodiment of the invention, it is provided that the breathing class of the intrinsic PEEP is identified on the basis of a stored exemplary curve of the flow or volume representing the intrinsic PEEP, wherein for this purpose the current curve is compared with the stored curve.

[0036] According to a further advantageous embodiment of the invention, it is provided that the breathing class of the intrinsic PEEP is identified by evaluating the flow or volume distribution over time during exhalation, wherein a high exhalation flow is identified at the beginning of exhalation, the exhalation flow then decreases and transitions to a low exhalation flow during the further course of exhalation.

[0037] According to another advantageous embodiment of the invention, it is proposed to identify the breathing class of the intrinsic PEEP by identifying the patient's inspiratory effort and evaluating the expiratory air flow within the time range of the inspiratory effort, wherein the remaining expiratory respiratory flow within the time range of triggering the next inspiratory phase represents the degree of the intrinsic PEEP.

[0038] According to a further advantageous embodiment of the invention, the breathing class of a coughing attack is identified by pattern recognition of the exhalation curve of the flow or volume based on stored curves representing a coughing attack, wherein for this purpose the current curve is compared with the stored curves.

[0039] According to another advantageous embodiment of the present invention, it is proposed to identify the breathing category of a coughing attack by comparing the curve of the current exhaled airflow, the exhaled airflow peak value or the exhaled volume peak value with the patient's typical previous exhaled airflow, the exhaled airflow peak value or the exhaled volume peak value, wherein the comparison is used to identify forced exhalation indicating a coughing attack, wherein a coughing attack is identified if the current breathing reaches 70% or 140% or more than 200% of the typical breathing.

[0040] According to another aspect of the invention, a method for analyzing respiratory events of a person who is artificially breathing with a respiratory gas is provided, wherein a signal corresponding to the artificial breathing pressure and / or the respiratory flow and / or the respiratory volume of the respiratory gas supplied to the person is generated by means of a sensor unit and the time curve of the signal is determined during a predetermined analysis period. A signal segment is then determined from the determined signal curve (in particular immediately after the determination of the signal curve), which is determined according to a segment duration that is shorter than the analysis period and is predefined or identified from the signal curve. The determined signal segment is then classified into one of a plurality of respiratory classes, wherein the respective occurrence frequency of the classified respiratory class during the analysis period is stored in a frequency counter provided for each respiratory class and at least one of the signal segments determined during the analysis period, but with a lower occurrence frequency than that detected in the frequency counter, is stored for each classified respiratory class.

[0041] As another aspect of the present invention, the artificial respiration device can also react intelligently or automatically adaptively to minimize the occurrence of at least some respiratory events. The artificial respiration device is arranged and constructed to identify individual respiratory events and to change the artificial respiration pressure and / or flow and / or volume of the respiratory gas for inspiration and exhalation according to the frequency or importance of the respiratory event or the respiratory event, so that the frequency or importance of the respiratory event or the respiratory event is weakened or reduced. For example, when asynchronous is identified, that is, when the patient's respiratory effort stops or occurs at an undesirable time point or a time point outside the desired window, the trigger sensitivity for identifying the patient's respiratory effort can be adapted. The expiratory pressure can be increased or reduced, especially at the beginning of exhalation, then the expiratory pressure is increased or reduced at the end of exhalation. When switching from the inspiratory pressure to the expiratory pressure (supporting the airway in exhalation, Schienung der Atemwegein der Expiration), the pressure curve can be adapted to minimize the intrinsic PEEP.

[0042] As another aspect of the present invention, when identifying inspiratory flow limitation and / or expiratory flow limitation and / or inherent PEEP and / or inefficient breathing effort and / or double breathing effort (double trigger), the artificial respiration device is able to generate at least one notification through the data transmission interface and / or on the display screen or store the at least one notification in the data and / or change at least one setting of the artificial respiration, the setting being selected from IPAP and / or EPAP and / or inspiratory time and / or expiratory time and / or trigger sensitivity.

[0043] It should be noted again that with regard to the definitions of terms relating to the method and with regard to the effects and advantages that correspond to the features of the method, reference is made in all respects to the above explanations regarding the meaningful definitions, effects and advantages of the device according to the invention. Correspondingly, the disclosures herein regarding the device according to the invention can be used in a meaningful way to define the method according to the invention, and the disclosures herein regarding the method according to the invention can be used in a meaningful way to define the device according to the invention. Thus, a repeated explanation of features that are identical in meaning, their effects and advantages is largely avoided. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] Further features and advantages of the invention are apparent from the following description of an embodiment of the invention, which should not be understood as limiting and is explained in more detail below with reference to the accompanying drawings. In the drawings, schematically shown:

[0045] Figure 1 Functional diagram of an embodiment of a breath analyzer according to the invention,

[0046] Figure 2 With the help of Figure 1 A diagram showing the analysis results of a short period of time obtained by a breath analyzer.

[0047] Figure 3 With the help of Figure 1 Another overview diagram of the analysis results obtained by the breath analyzer over a long period of time,

[0048] Figure 4a , b, c exemplary signal curves of signals corresponding to artificial respiration pressure, respiration flow and respiration volume of the first respiration category,

[0049] Figure 5a , b an exemplary signal curve of the signals corresponding to the artificial respiration pressure, the respiration flow and the respiration volume of the second respiration category,

[0050] Figure 6a , b an exemplary third signal curve of the signals corresponding to the artificial respiration pressure, the respiration flow and the respiration volume of the third respiration category,

[0051] Figure 7a , b an exemplary fourth signal curve of signals corresponding to artificial respiration pressure, respiration flow and respiration volume for a fourth respiration category, and

[0052] Figure 8 A perspective view of an embodiment of an artificial respiration device according to the invention.

[0053] Parts which are equivalent in terms of their function are always provided with the same reference numerals in the various figures, so that these reference numerals are usually also only indicated once. DETAILED DESCRIPTION

[0054] Figure 1 The functional diagram of an embodiment of a breathing analyzer 1 according to the invention for detecting breathing events of a person (not shown) who is artificially ventilating with breathing gas is schematically shown. The breathing analyzer 1 shown has an electronic (in the present case processor-based) computing and storage unit, wherein the storage unit 2 is stored in Figure 1, which is only schematically shown by a dotted frame, is constructed and arranged to receive a signal 3 corresponding to an artificial breathing pressure and / or a breathing flow and / or a breathing volume of the breathing gas supplied to the person and to determine the curve of the signal 3 during a predetermined analysis period by means of a curve analyzer 4. The illustrated computation and storage unit 2 of the breathing analyzer 1 is further constructed and arranged to: determine a signal segment 5 from the signal curve, which signal segment is determined in the present case based on a segment duration 6 identified from the signal curve, which segment duration is much shorter than the analysis period and can, for example, include one or also several complete breaths (for example 1 to 3 breaths); and classify the determined signal segment 5 into one of a plurality of predefined breathing categories by means of a comparator 7; and store the corresponding occurrence frequency of the classified breathing category during the analysis period in a frequency counter 8 provided for each breathing category in the memory unit 2; and store in the memory unit 2 for each classified breathing category at least one of the signal segments 5 determined during the analysis period, but a signal segment with a frequency less than the total occurrence frequency detected in the frequency counter 8, in order to store the signal segment 5 in the memory unit 2, wherein the signal segment 5 is determined in the memory unit 2, and the signal segment 5 is stored in the memory unit 2, wherein the signal segment 5 is stored in the memory unit 2, and ... Figure 1 The signal segments are stored in the storage area 9 of the storage unit 2 for signal segments.

[0055] from Figure 1 It is also known that the breath analyzer 1 has a plurality of predefined reference signal curves 10. Figure 1 , three of which are shown by way of example, and the reference signal curves are stored in the storage unit 2. It should be understood that a plurality of reference signal curves 10.1 to 10.n can be stored for each breathing category. The predefined reference signal curves 10.1 to 10.n can be used by the comparator 7 to compare the currently detected signal segment 5, so as to classify the signal segment 5 into the breathing category represented by the reference signal curve.

[0056] In addition, Figure 1 The respiratory analyzer 1 shown shows that, optionally (dash-dotted frame), additional reference signal curves 11 for adaptive learning are stored in the memory unit 2 during the operation or analysis period for each respiratory class in which the respiratory class is Figure 1, three adaptively learned reference signal curves 11.1, 11.2, and 11.3 for a breathing category are shown by way of example. It should be understood that, depending on the purpose of the breathing analyzer 1, not only the number of predefined reference signal curves 10 for each breathing category but also the number of adaptively learned reference signal curves 11 for each breathing category can be predetermined accordingly. In order to adaptively learn the reference signal curve 11, the breathing analyzer 1 optionally has an adaptive learning unit 12. In order to learn and expand the adaptively learned reference signal curve 11, the adaptive learning unit receives signal segment data from the comparator 7 on the one hand and provides the learned reference signal curve 11 to the comparator 7 on the other hand, for classifying the signal segment 5 into the corresponding breathing category.

[0057] In addition, Figure 1 , it is indicated that the signal 3 is generated by a sensor unit 13 outside the analyzer. The respiratory analyzer 1 can optionally also be supplied with information, for example, about the type of breathing (in particular spontaneous breathing or forced breathing) and / or about the rotational speed of an artificial respiration blower (not shown) of the respiratory gas supplied to the person and / or about leakage losses of the respiratory gas supplied to the person via a further, optional second sensor unit 14.

[0058] Furthermore, at least the analysis data stored in the storage unit 2 and obtained during the analysis period can be transferred to the storage unit 2 via the optional data transfer interface 15. Figure 1 In the data receiver (not shown), the evaluation data are in particular the contents of a corresponding frequency counter 8 for each breathing class and / or the contents of a storage area 9 which stores the signal segments identified for each breathing class during the evaluation period.

[0059] Figure 2 In a relatively short total analysis time, for example, Figure 1An overview of the analysis results obtained by the respiratory analyzer 1 in FIG. 1 , wherein the relatively short total analysis duration is currently 4.5 hours. The overview diagram generally shows the timestamps during the analysis duration (currently the time and date of the performed analysis) as well as the additional leakage information (currently "low"). In addition, the overview shown has three different breathing categories, which are referred to here as breathing types or asynchronous type 1 ("inefficient inspiratory effort"), asynchronous type 2 ("double inspiratory effort / double trigger") and breathing type 3 ("cough attack"), wherein for each breathing category, for example, the first five (stored) signal segments 5 identified during the analysis duration are shown. In addition, each signal segment 5 is also marked with a timestamp (currently the time at which the corresponding signal segment 5 was determined). Similarly, for each of the three breathing categories shown, the absolute frequency (stored in Figure 1 and the relative frequency in percentage relative to the total number of breaths identified during the analysis period.

[0060] Figure 3 It is shown that in a long analysis period, Figure 1 1, the long analysis period is currently seven days, wherein on each day an analysis with an analysis period of different lengths is performed, as shown in FIG. Figure 3 According to the corresponding timestamp (date and duration), they are respectively as follows Figure 2 The breathing categories shown show the analysis results, where the current Figure 3 The absolute frequency of the identified breathing classes during the respective analysis duration is plotted per day in the overview diagram of .

[0061] 4, 5, 6 and 7 show exemplary signal curves of signals corresponding to artificial breathing pressure, breathing flow and breathing volume for the first, second, third and fourth breathing categories, respectively. In the corresponding view (a), the signal curves of artificial breathing pressure, breathing flow, breathing volume and breathing phases (inspiratory phase 16 and expiratory phase 17) are shown for an inconspicuous signal curve.

[0062] Figure 4b The respiratory flow at position 18 is shown for Figure 2 and 3, i.e. a signal curve typical for a missed breath of a person who is being artificially ventilated with the artificial ventilator during the exhalation phase 17 of the artificial ventilator. An "inefficient inspiratory effort" 18 is an inefficient breathing effort of the patient, which corresponds to an attempt to inhale, but is usually unconscious and therefore does not trigger a switch to the phase of inspiration (by the artificial ventilator).

[0063] An "inefficient inspiratory effort" 18 is identified by analyzing the respiratory flow during the exhalation phase 17. Typically, the respiratory flow drops permanently below the zero line during the exhalation phase 17, i.e. the sensor records a (negative) flow directed toward the artificial respiration device. An "inefficient inspiratory effort" has the following curve, in which a negative flow is first presented (within 0.1 to 4 seconds or within 10% to 90% of the inspiratory flow). Then, a sign reversal of the flow is identified, the flow approaches the zero line or temporarily becomes a positive value. Subsequently, the flow reverses again and becomes negative again and drops below the zero line. An "inefficient inspiratory effort" can usually be identified by at least temporarily identifying a sign reversal of the flow and / or the flow at least temporarily exceeding the zero line again in the respiratory flow during the exhalation phase 17.

[0064] In one embodiment, this can be determined in such a way that it is checked whether the exhaled air flow exceeds a certain threshold value (e.g. in l / min) at the beginning of exhalation within a first time window, for example within a fixed time or within a portion of a typical exhalation duration, and then decreases within a second time window until the exhaled air flow does not exceed a certain threshold value (e.g. in l / min) or even briefly produces a positive respiratory flow, i.e. the inspiratory respiratory flow or at least the derivative of the respiratory flow shows a threshold value reduction for the expiratory respiratory flow. It is then checked whether the exhaled air flow increases again within a third time window before the next inhalation is detected, i.e. a stronger exhaled air flow is produced again or at least the derivative of the respiratory flow shows a threshold value increase for the expiratory respiratory flow again.

[0065] In this case, the time window is preferably selected in each case in the range between 0.1 and 3 seconds or in each case in the range between 10% and 90% of a typical exhalation duration.

[0066] Alternatively, an inefficient inspiratory effort can also be identified by means of pattern recognition, which analyses the curve of the respiratory flow and has been previously trained to identify inefficient inspiratory efforts according to a test pattern.

[0067] Alternatively, according to Figure 4cIt is also possible to determine an ideal or passive respiratory flow curve 43, which can be obtained from the measured pressure curve if a pressure curve is applied to the patient's lungs during purely passive artificial respiration, i.e., without the patient's own respiratory effort. Here, the patient's lung parameters, such as the compliance and resistance of different lung regions, are determined in advance or estimated by the artificial respiration device. The passive respiratory flow curve is then compared with the actually measured respiratory flow curve 44.

[0068] If the actually measured respiratory flow curve deviates from the determined passive respiratory flow curve, the patient's respiratory effort can be estimated from the difference. If a temporary inspiratory effort 18 exceeding a previously defined threshold is detected within an exhalation, an inefficient inspiratory effort can also be inferred. If an inspiratory effort 18 that successfully triggers the inspiratory trigger of the artificial respiration device, i.e., an effective inspiratory effort, is detected, the time difference between the inspiratory effort 18 and the increase in the treatment pressure to the inspiratory treatment pressure can be determined as the reaction time of the inspiratory trigger.

[0069] Figure 4c An ideal or passive respiratory flow curve 43 and an actual respiratory flow curve 44 are shown in the absence of respiratory effort and airway obstruction, and when the curve of the treatment pressure is applied to a lung model having at least one compliance and resistance, the ideal or passive respiratory flow curve is obtained. The inspiratory effort 18 is identified in the following manner: although the treatment pressure and thus the theoretical passive respiratory flow have not increased, the measured respiratory flow increases. The time difference between the two flow curves is obtained as the trigger delay time. If the trigger delay time is high, for example, higher than 200ms, the patient's respiratory work increases and there is a risk of inefficient inspiratory effort 18. If the trigger delay time is very low, for example, lower than 50ms, there is a risk of triggering breathing even without the patient's inspiratory effort 18, i.e., the so-called self-triggering of the artificial respiration device. Both situations where the trigger delay time is too high and too low will damage the subjective and objective success of artificial respiration.

[0070] The message of "inefficient inspiratory effort" and preferably about the trigger delay time is transmitted to the doctor through telemedicine and / or displayed or stored in the device. In addition, the device or an external display device generates a treatment suggestion or automatically executes this treatment suggestion. Here, it is preferred to change the trigger sensitivity or shorten the exhalation duration or change the inspiratory treatment pressure to change the respiratory drive or the spontaneous respiratory frequency by changing the respiratory volume (Tidal volume), so that the inspiratory effort 18 either appears earlier (higher inspiratory pressure) or later (lower inspiratory pressure). Alternatively, when the trigger delay time is extended or inefficient inspiratory effort 18 occurs, the treatment pressure of exhalation can also be increased. This is particularly advantageous in the following cases: the patient has an increased lung pressure (intrinsic PEEP) at the end of exhalation due to the air contained in the lungs, and the lung pressure is higher than the treatment pressure of the exhalation of the artificial respirator, so that the pressure difference (iPEEP minus PEEP) of the artificial respirator must be overcome by the patient in order to trigger the trigger of the artificial device. For example, in such a case, the treatment pressure of exhalation will be at least temporarily higher than the inspiratory pressure.

[0071] A more sensitive trigger is achieved, for example, by lowering the flow threshold for triggering the trigger; or increasing the therapeutic pressure of inspiration to prolong the exhalation phase without re-inspiratory breathing drive; or increasing the therapeutic pressure of exhalation to reduce the difference between the iPEEP and PEEP of the artificial respiration device that must be overcome in order to trigger the inhalation trigger. The preferred rule for automatically selecting the more sensitive trigger is:

[0072] - Identified inefficient inspiratory effort

[0073] - Exceeding a defined number of inefficient inspiratory efforts within a time window, e.g. more than 2 inefficient inspiratory efforts within 2 minutes

[0074] - Improved trigger delay

[0075] - Increased average trigger delay time within a time window (e.g. 2 minutes)

[0076] - Exceeding a defined number of breaths with increased trigger delay time within a time window

[0077] - The sum of the number of inefficient breathing efforts and the number of breaths with increased trigger delay exceeds a defined number within a time window

[0078] - The number of inefficient inspiratory efforts per time window is greater than the number of breaths with very short trigger delays within a time window

[0079] - The sum of the number of inefficient breathing efforts and the number of breaths with increased trigger delay times is greater than the number of breaths with very short trigger delay times within a time window

[0080] A more sensitive trigger is predefined in the following ways: for example, by lowering the flow threshold for triggering the trigger; or increasing the therapeutic pressure of inspiration to prolong the expiratory phase without re-inspiratory breathing drive; or increasing the therapeutic pressure of exhalation to reduce the difference between iPEEP and PEEP of the artificial respiration device that must be overcome in order to trigger the inhalation trigger.

[0081] The reduction in trigger sensitivity is predetermined, for example, by increasing the flow threshold for triggering the trigger, or reducing the inspiratory therapy pressure, or reducing the expiratory airway pressure, with corresponding inverse rules applying to the reduction in trigger sensitivity.

[0082] Figure 5b The respiratory flow at position 19 is shown for Figure 2 and 3 1 , 14 , 15 , 16 , 17 , 18 , 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 , 32 , 33 , 34 , 36 , 37 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 46 , 47 , 48 , 49 , 50 , 51 , 52 , 53 , 54 , 56 , 57 , 58 , 59 , 60 , 61 , 62 , 63 , 64 , 65 , 67 ,

[0083] Such a double trigger can be detected, for example, by an exhalation time that is greatly shortened compared to a typical (eg average or predefined) exhalation time, either once or in x breaths within a time window.

[0084] Such a double trigger can also be detected, for example, by a measured flow rate which, after switching to the artificial respiration pressure for exhalation, remains increased relative to the theoretical passive respiration flow rate.

[0085] When such a double trigger is detected, the caregiver is informed via remote monitoring or on a display screen or in the form of stored data.

[0086] When such a double trigger is recognized, alternatively or additionally, the sensitivity of the exhalation trigger is automatically reduced, i.e., the exhalation trigger is triggered later and / or the predetermined inhalation time is automatically prolonged and / or a trigger lock time is automatically used, during which no new inhalation trigger is allowed to be triggered after switching to the artificial respiration pressure for exhalation, and / or the artificial respiration pressure for inhalation is increased so that the lungs fill with air more quickly and the patient's inhalation time becomes shorter.

[0087] Figure 6bAt position 20 of the respiratory flow, a typical signal curve is shown for the breathing class “intrinsic PEEP”, which means that during the exhalation phase 17 the air volume inhaled by a person who is being ventilated using a ventilator is not fully exhaled.

[0088] The intrinsic PEEP 20 is identified, for example, based on a stored exemplary curve of the flow or volume, which represents the intrinsic PEEP 20. For this purpose, the current curve is compared with the stored curve.

[0089] Alternatively, intrinsic PEEP 20 can also be detected by evaluating the time profile of the flow or volume during exhalation. It is typical for intrinsic PEEP 20 that a high exhalation flow is detected at the beginning of exhalation, which then suddenly decreases and transitions to a very low (eg below threshold) exhalation flow during the further course of exhalation.

[0090] Alternatively, intrinsic PEEP 20 is identified by comparison / correlation of the curve of the expiratory flow with stored comparison curves of the expiratory flow during normal exhalation and during exhalation with increased resistance which results in iPEEP.

[0091] Intrinsic PEEP 20 is also detected by comparing the actual exhaled airflow with a theoretical passive exhaled airflow based on the curve of the artificial respiration pressure and the patient's lung parameters that are predefined or estimated during operation. If the exhaled airflow is smaller than the ideal passive exhalation, i.e. the lungs empty more slowly, the additional resistance in the exhalation must be hindering the respiratory flow. That is, there must be an obstruction of the airways, the so-called exhaled airflow limitation, which usually leads to iPEEP.

[0092] The identification of the patient's inspiratory effort 18 and the evaluation of the expiratory airflow during the inspiratory effort 18 or when using the background rate of the ventilator can also be used to identify the intrinsic PEEP 20. The greater the remaining expiratory respiratory flow when the next inspiration is triggered, the greater the intrinsic PEEP or the amount of air contained in the lungs.

[0093] According to the invention, it is also provided that breaths with a high degree of expiratory flow limitation are counted in a defined period of between a few minutes and a day (or taking into account the iPEEP) in order to identify the intrinsic PEEP 20 .

[0094] Furthermore, an average degree of expiratory flow limitation / iPEEP can be formed over a defined period of time between a few minutes and a day in order to identify the intrinsic PEEP 20 .

[0095] When intrinsic PEEP 20 is detected, the caregiver is informed via remote monitoring and / or on a display screen or in stored data.

[0096] Especially in the case of rapid changes, ie, a significant increase in the expiratory flow limitation within hours to days, a warning (notification) with high priority is advantageous, since either the quality of the artificial respiration or the patient's condition has deteriorated.

[0097] Alternatively, the pressure ramp from IPAP to EPAP, i.e. the transition from the artificial respiration pressure of inspiration to the artificial respiration pressure of expiration, is automatically adjusted when intrinsic PEEP 20 is detected. If this ramp is implemented flatter or is implemented flatter in the second part of expiration, or if the pressure in the second part of expiration increases again, the collapse of the lower airway is prevented, the expiratory flow limitation and thus the iPEEP are weakened.

[0098] Alternatively, when intrinsic PEEP 20 is detected, the expiratory pressure level is increased in order to support the lower airways and to reduce the respiratory effort for triggering the inspiratory trigger. The expiratory pressure can be temporarily higher than the inspiratory pressure.

[0099] Figure 7b The respiratory flow at position 21 is shown for Figure 2 and 3 Typical signal curve for the respiratory class “coughing episode” (type 3), ie, an increased, forced and prolonged inhalation and exhalation of a person who is being artificially ventilated using an artificial respirator, is given in FIG.

[0100] For example, a coughing episode is detected by pattern recognition of the exhalation curve using stored example curves 21. For this purpose, the current curve is compared with the stored curves.

[0101] Coughing episodes 21 are identified by comparing / correlating the curve of the expiratory flow, the expiratory flow peak or the expiratory volume peak with the patient's typical (e.g. average) expiratory flow, the expiratory flow peak or the expiratory volume peak, in order to identify forced exhalations that indicate a coughing episode. For example, a coughing episode is considered to be present if the current exhalation reaches more than 120%, 150% or 200% of a typical breath.

[0102] Alternatively, the coughing episode 21 can be detected by additionally detecting (measured in terms of volume or flow or flow peak) whether an increased exhalation is also preceded by an increased inspiration, which confirms the consideration of a coughing episode.

[0103] In addition, intrinsic PEEP 20 is detected by comparing the actual exhaled airflow with a theoretical passive exhaled airflow based on the curve of the artificial respiration pressure and the patient's lung parameters that are predefined or estimated during operation. If the exhaled airflow is greater than the ideal passive exhalation, i.e. the lungs empty faster, there must be forced exhalation. For example, the threshold value for forced exhalation can be determined by the volume difference between the actual exhalation and the ideal passive exhalation or by the difference of the two flow curves. If the difference in volume or flow is above a certain value, a coughing attack is detected.

[0104] Coughing episodes are also identified by comparing the derivative of respiratory flow according to the mentioned criteria21.

[0105] When a coughing attack 21 is detected, the caregiver is informed by remote monitoring and / or on a display or with stored data. In particular, a high-priority warning (information) is advantageous when rapid changes occur, i.e. when the expiratory flow limitation increases significantly over a period of hours to days, because either the quality of the artificial respiration or the patient's condition has deteriorated.

[0106] Especially in the case of rapid changes, ie a marked increase in cough frequency over a period of hours to days, a high-priority warning is advantageous, since either the quality of the artificial respiration or the patient's condition has deteriorated.

[0107] Figure 8 A three-dimensional view of an embodiment of an artificial respiration device 30 according to the invention for artificially respiring a person using breathing gas from a breathing gas source (not shown), which can be arranged in the interior space of a device housing 31. An operating element 32 and a display device 33 are arranged in the area of ​​the device housing 31, wherein the display device 33 can also be configured as a combined operating and display device with a touch-sensitive input unit, for example. A connecting hose 35 is attached via a coupling 34. An additional pressure measuring hose 36 can extend along the connecting hose 35, which can be connected to the device housing 31 via a pressure input cannula 37. In order to realize data transmission, the artificial respiration device 30 has at least one or more data transmission interfaces 15, 38.

[0108] exist Figure 8 In the artificial respiration device 30 shown in Figure 1 The respiratory analyzer 1 according to the invention shown in FIG. 1 is implemented as part of an artificial respiration device 30. Figure 8 It is further known that the breath analyzer 1 and Figure 1 The sensor unit 13 shown in FIG. 1 , which is currently associated with the artificial respiration device 30 , is coupled to the data transmission interface 15 or 38 .

[0109] Furthermore, a humidifier known per se (not shown) can be adapted. An exhalation element 39 is arranged in the region of the end of the connecting hose 35 facing away from the device housing 31. An exhalation valve (not explicitly shown) can also be used.

[0110] from Figure 8 In the figure, it is further seen that the patient interface is configured as an artificial respiration mask 40, which is currently implemented as a nasal mask. It can be fixed in the head area of ​​the person who will perform artificial respiration by a hood 41. The patient interface 40 has a coupling element 42 in its end area facing the connecting hose 35.

[0111] For example, data, such as the dead space volume, can be input and / or output via the interface 38. The data transmission interfaces 15, 38 can be implemented wired, can be implemented as an infrared interface, a Bluetooth interface or a USB. Preferably, a memory card slot (not shown) is also provided. The interfaces 15, 38 can also be implemented as a LAN interface or other data transmission interface for connection to a network, the Internet, etc. In the area of ​​the device housing 31, an oxygen connection valve (not shown) can be adapted to the artificial respiration device 30. It is conceivable to transmit breathing gas that also contains oxygen in order to improve the care of the patient.

[0112] Via the interface 38, which is designed, for example, as a memory card slot or USB, non-therapeutic data can also be loaded into the respirator 30 according to the invention or from this respirator to a data receiver outside the device. If the respirator 30 recognizes an external storage medium, the user must confirm the query via the operating area 32 on the display device 33 and then optionally store or execute the data in the area of ​​the respirator 30.

[0113] Telemedicine data can be input and / or output via the interface 38. For this purpose, for example, mobile radio data or near field radio data or WLAN data or Bluetooth data or network data can be received / sent via the data transmission interface 38.

[0114] The artificial respiration device 30 is designed so that it can be connected to a person or patient who needs to be artificially respired via a hose 35 and a patient interface 40 so that artificial respiration is provided by the device 30. The breathing gas source can be configured, for example, as an electric motor having a fan impeller driven by the electric motor (neither of which is shown). The sensor unit 13 is configured to determine the artificial respiration pressure and / or the respiration flow and / or the respiration volume of the respiration gas. The artificial respiration device 30 has an electronic control unit (not shown) which is configured to determine the breathing gas pressure for each breath based on a predetermined value of the patient and / or based on a measurement signal of the sensor unit 13 and to adjust the breathing gas source so that the desired breathing gas pressure or artificial respiration pressure is generated.

[0115] A particularly preferred use of the respiratory analyzer or the respiratory analysis method disclosed here focuses on monitoring the exhalation phase of a person who is to be artificially ventilated, for example during sleep. The respiratory flow curve during exhalation can be used to identify a number of complications during artificial respiration. Early response (for example, by medication or optimization of artificial respiration settings) can significantly improve the result of artificial respiration.

[0116] Thus, if the artificial respiration device 30 has a corresponding evaluation unit, for example, the expiratory respiratory flow curve of the person who is being artificially ventilated can already be evaluated in this artificial respiration device, wherein preferably at least two of the four complication states are identified: the severity of the expiratory flow limitation, the severity of the intrinsic PEEP, the frequency of missed breaths (missing inspiratory triggers or inefficient inspiratory efforts), the frequency of coughing attacks. For example, by pattern recognition, comparison of the expiratory curve with a stored reference curve, evaluation of the respiratory flow (at a specific time point in the expiratory curve) and / or the slope of the respiratory flow or the exhaled volume, the corresponding respiratory class can be identified from the determined signal segment. The storage and processing of the identified complications can be carried out in the artificial respiration device itself and / or transmitted to the data receiver via the data transmission interface 38. Preferably, the identified complication states are transmitted to a monitoring unit (for example, a monitoring unit in a hospital or nursing home), to a monitor, and in the case of home artificial respiration to a remote monitoring server. In addition, if the frequency or severity of the complication exceeds a specific limit value, a visually displayed / acoustically generated prompt or alarm can be triggered at the monitoring unit.

[0117] In order to detect the aforementioned respiratory events, the corresponding respiratory flow during the exhalation phase 17 of the artificial respiration device 30 can be evaluated, optionally in combination with the artificial respiration pressure or the artificial respiration blower of the respiratory gas source, which is detected by the sensor unit 14 ( Figure 1 ) detected rotational speed and / or detected leakage loss.

[0118] According to the invention, for each of the patterns or each of the previously mentioned breathing events that can be recognized during exhalation, there are different recognition possibilities. These recognition possibilities can be used individually or in combination and can also be regarded as just an example. It is also not necessary to search for all of the patterns described, but it is necessary to search for at least one pattern.

[0119] The evaluation, for example by an evaluation unit arranged in the artificial respiration device 30 or alternatively / additionally by an evaluation unit external to the device (neither shown), can include one or a combination of the following method steps:

[0120] a) Evaluation of the expiratory flow amplitude, especially the maximum and minimum values ​​and their occurrence in time:

[0121] Maximum value during the first part of the exhalation phase indicates: too short an inspiratory time, additional inspiratory effort

[0122] The maximum value during the further course of the exhalation phase represents: missed breaths

[0123] A minimum value lower than in the previous breath indicates: a coughing episode

[0124] Negative respiratory flow at the beginning of inspiration indicates intrinsic PEEP

[0125] b) Evaluation of the expiratory volume and, if applicable, the inspiratory volume which can be measured during the expiratory phase:

[0126] Positive volume in the segment of the exhalation phase indicates: missed breaths

[0127] Increased exhaled volume compared to previous exhalations indicates: coughing episode

[0128] · Exhaled volume less than inspiratory volume indicates: inherent PEEP

[0129] c) Evaluation of the derivative (= slope) of respiratory flow:

[0130] A rise followed by a fall during the first part of the exhalation phase indicates: too short an inspiratory time, additional inspiratory effort

[0131] A rise followed by a fall during the further course of the exhalation phase indicates: missed breaths

[0132] A steeper drop than in the previous breath indicates a coughing fit

[0133] d) Evaluation of the volume distribution of expired air during the exhalation phase:

[0134] Volume components primarily at the beginning and end of a breath represent: missed breaths

[0135] The volume component that is strongly concentrated at the beginning of breathing represents: intrinsic PEEP

[0136] e) Evaluation of the profile of the respiratory flow and / or volume curve:

[0137] Correlation or other similarity measure with stored or learned reference profiles

[0138] Pattern recognition, such as using neural networks or support vector machines or principal component analysis or fuzzy logic

[0139] Instead of the evaluation of the respiratory flow as a function of time described here, the respiratory flow and / or the respiratory volume can alternatively also be evaluated as a function of the artificial respiration pressure (P / V diagram).

[0140] The respiratory analyzer, the artificial respirator and the method for respiratory analysis according to the invention disclosed herein are not limited to the embodiments disclosed in this case, but also include other embodiments having the same effect, which result from other technically meaningful combinations of the features of the respiratory analyzer, the artificial respirator and the respiratory analysis method described here. In particular, the features and feature combinations mentioned above in the general description and the description of the figures and / or shown individually in the figures can be used not only in the respective combinations explicitly indicated here, but also in other combinations or in isolation without departing from the scope of the invention.

[0141] In a preferred embodiment, the breath analyzer according to the invention is used in a respirator in order to automatically identify different breathing events of a person being respired by the respirator with breathing gas during sleep, for example as described here, and to make them available for subsequent processing.

Claims

1. A breathing analyzer for identifying breathing events of a person who is artificially breathing with breathing gas, comprising an electronic computing and storage unit (2) which is designed and arranged to receive a signal (3) corresponding to the artificial breathing pressure and / or the breathing flow and / or the breathing volume of the breathing gas supplied to the person and to detect the curve of the signal (3) during a predetermined analysis period by means of a curve analyzer (4), characterized in that The computing and storage unit (2) is designed and arranged to: determine a signal segment (5) from a signal curve, the signal segment being determined according to a segment duration (6) that is shorter than an analysis duration and that is predefined or identified from the signal curve, wherein the segment duration (6) of the signal segment (5) can be determined according to at least one predetermined signal curve criterion, namely according to a slope and / or a slope curve of the signal curve; classify the determined signal segment (5) into one of a plurality of breathing categories by means of a comparator (7); and store the respective occurrence frequency of the classified breathing category during the analysis duration in a frequency counter (8) provided for each breathing category in the storage unit (2); and store, for each classified breathing category, in the storage unit (2, 9) at least one of the signal segments (5) determined during the analysis duration, but having a smaller occurrence frequency than that detected in the frequency counter (8), wherein the breathing categories at least include the breathing types of inhalation, exhalation, pause and cough.

2. The breath analyzer according to claim 1, characterized in that The calculation and storage unit (2) is designed and arranged to store only a predetermined maximum number of signal segments (5) in the storage unit (2, 9) for each breathing class.

3. The breath analyzer according to claim 2, characterized in that The number of signal segments (5) to be stored can be determined individually for at least two different respiration classes.

4. The breath analyzer according to any one of claims 1 to 3, characterized in that: The segment duration (6) of the signal segment (5) can be determined according to at least one predetermined signal curve criterion, which is selected from a group having the occurrence of maxima and / or minima of the signal curve.

5. The breath analyzer according to any one of claims 1 to 3, characterized in that: The segment duration (6) is measured such that the signal segment (5) contains at least one complete breath.

6. The breath analyzer according to claim 5, characterized in that The segment duration (6) is measured such that the signal segment (5) contains 1 to 3 breaths or at least 3 to a maximum of 10 breaths.

7. The breath analyzer according to any one of claims 1 to 3, characterized in that: The stored signal segment (5) is provided with a time stamp which identifies the signal segment.

8. The breath analyzer according to any one of claims 1 to 3, characterized in that: The comparator (7) is designed and arranged to classify the signal segment (5) into the breathing class by means of pattern recognition and / or by means of a comparison of the signal curve in the signal segment (5) with a predetermined number of predefined reference signal curves (10) prestored in the memory unit (2) and / or by means of a detected slope and / or slope curve of the signal curve in the signal segment (5) and / or by means of one or more detected maximum and / or minimum values ​​of the signal curve in the signal segment (5).

9. The breath analyzer according to any one of claims 1 to 3, characterized in that: The comparator (7) is designed and arranged to store a predetermined number of selected signal segments (5) that are assigned to the breathing class during the analysis period as newly added, adaptively learned reference signal curves (11) in the memory unit (2) and to take the selected signal segments (5) into account when assigning future detected signal segments (5) to the breathing class.

10. The breath analyzer according to any one of claims 1 to 3, characterized in that: The breathing categories also additionally include breathing types and / or breathing asynchrony types which represent inspiratory flow limitation and / or expiratory flow limitation and / or intrinsic PEEP ( 20 ).

11. The breath analyzer according to claim 10, characterized in that The types of respiratory asynchrony include inefficient respiratory effort (18) and / or double respiratory effort, ie double triggers (19).

12. The breath analyzer according to claim 11, characterized in that The breathing class of inefficient breathing effort (18) is identified in that the breathing flow during the exhalation phase (17) is analyzed and the breathing flow during the exhalation phase (17) has at least temporarily a sign reversal.

13. The breath analyzer according to claim 11, characterized in that The breathing category of inefficient breathing effort (18) is identified in the following manner: an analysis is made as to whether the expiratory airflow exceeds a certain threshold value within a first time window at the beginning of exhalation, and the expiratory airflow is then reduced within a second time window until the expiratory airflow falls below a certain threshold value or even briefly generates a positive respiratory flow, i.e., an inspiratory respiratory flow, or at least the derivative of the respiratory flow shows a decrease in the expiratory respiratory flow, wherein it is then checked whether the expiratory airflow increases again within a third time window before the next inhalation is identified, i.e., a stronger expiratory airflow is generated again or at least the derivative of the respiratory flow shows a renewed increase in the expiratory respiratory flow.

14. The breath analyzer according to claim 13, characterized in that The determined threshold value is in l / min.

15. The breath analyzer according to claim 11, characterized in that The breathing category of inefficient breathing effort (18) is identified by pattern recognition comparing the curve of the respiratory flow with a previously stored test pattern representing an inefficient inspiratory effort.

16. The breath analyzer according to claim 11, characterized in that The breathing class of the double trigger (19) is detected by comparing the exhalation time with a plurality of preceding typical exhalation times or with an average exhalation time or with a predefined exhalation time.

17. The breath analyzer according to claim 11, characterized in that The breathing class of a double trigger (19) is detected by comparing a measured flow with a theoretical breathing flow, wherein after switching to an artificial respiration pressure for exhalation, the measured flow continues to increase relative to the theoretical breathing flow.

18. The breath analyzer according to claim 10, characterized in that The respiratory class of the intrinsic PEEP (20) is identified based on a stored exemplary curve of flow or volume that is representative for the intrinsic PEEP (20), wherein for this purpose the current curve is compared with the stored curve.

19. The breath analyzer according to claim 10, characterized in that The breathing class of the intrinsic PEEP (20) is identified by evaluating the flow or volume distribution over time during exhalation, wherein a high exhalation flow is identified at the beginning of exhalation, which then decreases and transitions to a low exhalation flow during the further course of exhalation.

20. The breath analyzer according to claim 11, characterized in that The breathing category of the intrinsic PEEP (20) is identified by identifying the patient's inefficient breathing effort (18) and evaluating the expiratory airflow within a time range of the inefficient breathing effort (18), wherein the remaining expiratory breathing flow within a time range for triggering the next inspiratory phase represents the degree of the intrinsic PEEP.

21. The breath analyzer according to any one of claims 1 to 3, characterized in that: The respiratory class of a coughing episode (21) is identified by pattern recognition of an exhalation curve of the flow or volume based on stored curves that represent a coughing episode, wherein for this purpose the current curve is compared with the stored curves.

22. The breath analyzer according to claim 21, characterized in that The breathing class of a coughing attack (21) is identified by comparing the curve of the current exhaled airflow, the exhaled airflow peak or the exhaled volume peak with the patient's typical previous exhaled airflow, the exhaled airflow peak or the exhaled volume peak in order to identify forced exhalations that indicate a coughing attack, wherein a coughing attack is identified if the current exhalation reaches more than 70%, 140% or 200% of a typical breath.

23. The breath analyzer according to any one of claims 1 to 3, characterized in that The comparator (7) is designed and arranged to receive additional information about the breathing type and / or additional information about the rotational speed of an artificial respiration blower supplying the breathing gas to the person and / or additional information about leakage losses of the breathing gas supplied to the person and to take this additional information into account when assigning the signal segment (5) to the breathing class.

24. The breath analyzer according to claim 23, characterized in that The breathing type includes spontaneous breathing or forced breathing.

25. An artificial respiration device for artificially respiring a person with the aid of a breathing gas, comprising a sensor unit (13) for determining an artificial respiration pressure and / or a breathing flow and / or a breathing volume of the breathing gas supplied to the person, characterized in that The artificial respiration device has a respiratory analyzer (1) according to any of the above claims, wherein the sensor unit (13) is coupled to a computing and storage unit (2) of the respiratory analyzer (1) in a data transmission manner and is configured to generate a signal (3) corresponding to a determined artificial respiration pressure and / or a determined respiratory flow and / or a determined respiratory volume of the respiratory gas and to supply the signal to the computing and storage unit (2).

26. The artificial respiration device according to claim 25, characterized in that It has at least one data transmission interface (15, 38) which is designed and provided for transmitting the contents of a frequency counter (8) stored in the memory unit (2) for each breathing class and / or at least one signal segment (5) stored in the memory unit (2, 9) for each assigned breathing class to a data receiving unit.

27. The artificial respiration device according to claim 25 or 26, characterized in that A further sensor unit (14) is provided and designed to detect and to supply the respiratory analyzer (1) with information on the rotational speed of the respiratory blower and / or the leakage loss of the respiratory gas and / or the type of breathing during artificial respiration of a person.

28. The artificial respiration device according to claim 27, characterized in that The breathing type includes spontaneous breathing or forced breathing.

29. The artificial respiration device according to claim 25 or 26, characterized in that It has an evaluation unit which is designed and arranged to evaluate analysis results of the breath analyzer (1) with respect to health-critical complication states of a person undergoing artificial respiration and to store the analysis results in the storage unit (2) and / or to display them on a display device (33) and / or to transmit them to a data receiving unit outside the device.

30. The artificial respiration apparatus according to claim 26, characterized in that When an inspiratory flow limitation and / or an expiratory flow limitation and / or an intrinsic PEEP (20) and / or an inefficient breathing effort (18) and / or a double breathing effort (19), i.e. a double trigger, is detected, at least one notification is made via the data transmission interface and / or on a display device or at least one notification is stored in the data.

31. The artificial respiration device according to claim 25 or 26, characterized in that When an inspiratory flow limitation and / or an expiratory flow limitation and / or an intrinsic PEEP (20) and / or an inefficient breathing effort (18) and / or a double breathing effort (19), i.e. a double trigger, is detected, at least one setting of the artificial respiration is changed, the setting being selected from IPAP and / or EPAP and / or an inspiratory time and / or an expiratory time and / or a trigger sensitivity.

32. A method for identifying respiratory events in a person who is being artificially ventilated using a respiratory gas, wherein: A signal (3) corresponding to an artificial breathing pressure and / or a breathing flow and / or a breathing volume of breathing gas supplied to a person is generated by means of a sensor unit (13) and a time curve of the signal (3) is determined during a predetermined analysis period, characterized in that a signal segment (5) is determined from the time curve of the signal, the signal segment being determined according to a segment duration (6) which is shorter than the analysis period and which is predefined or identified from the time curve of the signal, and the determined signal segment (5) is classified into one of a plurality of breathing categories, wherein the respective occurrence frequency of the classified breathing category during the analysis period is stored in a frequency counter (8) provided for each breathing category, and at least one of the signal segments (5) determined during the analysis period, but having a smaller occurrence frequency than that detected in the frequency counter (8), is stored for each classified breathing category.

Images