Method and signal processing unit for determining respiratory activity of a patient

Abstract

The invention relates to a method and to a signal processing unit (5) for determining a spontaneous breathing activity of a patient (P) artificially ventilated by a ventilation device (1). The signal processing unit carries out a first ventilation process in which a ventilation device parameter is set to a first setting value. From measured values which have been measured in the case of the first setting value, the signal processing unit generates a first signal value group. The signal processing unit derives a first breathing activity value and for this uses a predefined lung mechanics model (20) and the first signal value group. The signal processing unit calculates a measure of reliability that the first breathing activity value corresponds to a corresponding actual breathing activity measure. Depending on the measure of reliability, the signal processing unit checks whether a predefined triggering criterion is fulfilled. If the predefined triggering criterion is fulfilled, the signal processing unit triggers a change process (run-through) in which the ventilation device parameter is set to a second setting value. The signal processing unit carries out a further ventilation process in which the ventilation device parameter is set to the second setting value.

Description

Technical Field

[0001] The present invention relates to a method and a signal processing unit for automatically approximating the magnitude of a patient’s own respiratory activity, particularly during artificial ventilation of a patient. Background Technology

[0002] Ventilation devices support a patient's own breathing activities (spontaneous breathing), or, if the patient is sedated, temporarily and completely replace their own breathing activities. Anesthesia devices are a special case of ventilation devices. At least one actuator in a ventilation device is operated to induce artificial ventilation of the patient, typically through a series of ventilation strokes. For the ventilation achieved by the device to be automatically synchronized with, for example, the patient's own breathing activities (spontaneous breathing) and to achieve, proportional amplification, the patient's own breathing activities need to be understood as well as possible. This can be irregular and / or change over time. The transmission path from the respiratory muscles on the patient's side to the sensors measuring signals from the respiratory system is typically affected by additional, time-varying signals. These same signals, which also typically affect the patient's body, are usually also present in the body. Despite these unavoidable effects, ventilation devices must operate with high safety and good synchronization with the patient's own breathing activities.

[0003] A method for automatically controlling ventilation equipment is described in DE 102007062214 C5. The signal u is determined using electrodes. emg The signal u emg This indicates the patient's own respiratory activity. The respiratory muscle pressure P generated by the patient's respiratory muscles. mus It is calculated, more precisely, either from measurements of airway pressure and volumetric flow, as well as lung mechanics parameters, or as negative airway pressure during interruptions of artificial ventilation, or by means of a measured pressure P in the esophagus. es The probe is used to calculate the respiratory activity signal u. emg Converted into a pressure signal P emg This makes it related to the respiratory muscle pressure P. mus The deviation is minimized. The control unit of the ventilation equipment sets the patient's breathing effort pressure (Atemanstrengungs-Druck) p. pat Calculated as P mus and P emg The weighted average. Based on the supplied airway pressure P. aw Previous actual values ​​and based on the respiratory effort pressure p for the patient pat The control device calculates the desired airway pressure to be supplied by the ventilation equipment, based on the previous value.

[0004] WO 2018 / 143844 A1 describes a ventilation device 1 and a device for providing artificial ventilation to a patient 3. Artificial ventilation is provided to the patient 3 in at least two different modes (different levels). In each ventilation mode, a sample is measured containing a pressure P in the airway. aw Multiple measurements of the volumetric flow rate during artificial ventilation, changes in lung volume over time, and electrical respiratory signals are used. Using these two samples, at least one physiological parameter (e.g., neuromechanical efficiency) is calculated.

[0005] US 9114220 B2 describes a method for determining a patient's spontaneous breathing effort based on measured ventilation pressure and measured volumetric flow. For this purpose, a pre-given relationship is used relating to at least one so-called provisional value. This provisional value is updated at least once. Based on the determined spontaneous breathing effort, a cycle is triggered when the patient is given artificial ventilation.

[0006] EP 3424407 A1 describes an apparatus and method for physiological monitoring of a person, particularly monitoring a person's health. A physiological sensor 17 supplies a biomedical signal S17. A feature extractor 11 acquires the biomedical signal 17 and supplies a feature signal S17A, which is displayed, for example. A quality estimator 10 estimates the quality of the biomedical signal S17 and, for example, replaces the deviation value with a statistical average. Summary of the Invention

[0007] The objective of this invention is to provide a method and a signal processing unit that, during at least temporary artificial ventilation of a patient, approximates a measure of the patient's own respiratory activity, wherein determining the measure of respiratory activity is safer than known methods and signal processing units.

[0008] The task is solved by the method and signal processing unit proposed in this invention. Advantageous construction schemes are also described in this invention. Advantageous construction schemes described with respect to the method according to the invention can be used in a corresponding manner with the signal processing unit according to the invention, and are advantageous construction schemes for the signal processing unit, and vice versa.

[0009] The computer-implemented method according to the invention and the signal processing unit according to the invention for data processing can approximately determine the measure of a patient's own respiratory activity (spontaneous breathing)—more precisely: automatically determine the measure associated with the patient's own respiratory activity.

[0010] The patient is at least temporarily ventilated by a ventilation device. Anesthesia equipment is a specific example of a ventilation device. The ventilation device operates according to a first variable ventilation device parameter. This first variable ventilation device parameter affects the control of the flow of gas to and / or from the patient and / or affects the pressure of that gas. It is possible that the ventilation device additionally operates according to at least one other variable ventilation device parameter. A signal processing unit may be an integral part of the ventilation device or may be spatially separate from it.

[0011] A lung mechanics model is pre-given in a computer-usable form for the method according to the invention. The signal processing unit according to the invention has at least temporary read access to a data memory in which the lung mechanics model is stored. The lung mechanics model describes at least one, and optionally multiple, relationships between the following:

[0012] - A measure of the patient's own breathing activity (spontaneous breathing), i.e., a respiratory activity measure, which is correlated with the patient's breathing activity, and

[0013] - At least one measurable signal, preferably at least one of the following measurable signals, which is superimposed on the body’s respiratory activity and artificial ventilation.

[0014] The method according to the present invention includes the following steps, and the signal processing unit according to the present invention is constructed to perform the following steps:

[0015] The signal processing unit executes at least one ventilation process. During this ventilation process, or each ventilation process, the parameters of the first ventilation device are set to a set value (Einstell-Wert). This set value may vary depending on the ventilation process.

[0016] The ventilation process, or at least one ventilation process (preferably each ventilation process), performed under the determined set values ​​of the parameters of the first ventilation device includes the following steps:

[0017] - The signal processing unit receives at least one value for at least one measurable signal appearing in the lung mechanics model. Preferably, the signal processing unit receives one value for each measurable signal in the lung mechanics model, and more preferably receives multiple values ​​for each measurable signal or each measurable signal sequentially. During the period when the first ventilation device parameters are set to the determined setpoint, the signal value or each value is measured.

[0018] - The signal processing unit generates at least one set of signal values, each set including the signal value of each measurable signal from the lung mechanics model, and relating to a sampling time. Preferably, the signal processing unit generates multiple sets of signal values ​​for different sampling times. To generate the sets of signal values, the signal processing unit uses measurements taken under a defined setpoint.

[0019] The signal processing unit derives at least one respiratory activity value for a measure of respiratory activity associated with the patient's own respiratory activity. To derive this respiratory activity value, the signal processing unit uses a lung mechanics model and at least one set of signal values. Using measurements taken at set values, the signal processing unit generates the used set of signal values ​​or each used set of signal values.

[0020] - The signal processing unit controls the ventilation device. The control is performed with the aim of supporting or replacing the patient's breathing activities. During control, first ventilation device parameters are set to predetermined values.

[0021] The signal processing unit performs at least one first ventilation process. The first ventilation process includes the steps of the ventilation process just listed. During the first ventilation process, the parameters of the first ventilation device are set to a first set value. In particular, at the first set value, at least one measurement value, preferably multiple measurement values, are measured, and at least one group of signal values, preferably multiple group of signal values, are generated.

[0022] During the first ventilation process, that is, under the first set value, the signal processing unit derives the first respiratory activity value, that is, the first value of the respiratory activity measurement.

[0023] Furthermore, the signal processing unit calculates a reliability metric that measures the reliability of a derived first respiratory activity value that corresponds to a metric for the patient's actual respiratory activity during the first ventilation process. The first respiratory activity value has been derived during the first ventilation process.

[0024] The signal processing unit checks whether a pre-defined triggering criterion is met. This triggering criterion and the result of the check depend on a reliability metric. This reliability metric describes the reliability of the derived first respiratory activity value as the reliability that the value matches the actual respiratory activity value.

[0025] If the signal processing unit has detected that the trigger criteria are met, then the following steps are performed:

[0026] - The signal processing unit triggers a change process. During this change process, the parameters of the first ventilation device are set to a second set value. This second set value deviates from the first set value, that is, from the value that existed when the first ventilation process was executed.

[0027] - The signal processing unit executes at least one other ventilation process. In the other ventilation process, the steps described above for the ventilation process are executed again. In the other ventilation process, the parameters of the first ventilation device are set to a second set value, and not to the first set value as in the first ventilation process.

[0028] Furthermore, the present invention relates to an apparatus comprising a signal processing unit according to the invention, a ventilation device, and a data storage device. A computer-usable lung mechanics model is stored in the data storage device. The ventilation device is capable of providing artificial ventilation to a patient, at least temporarily, and operates according to first ventilation device parameters. The signal processing unit according to the invention has read access to the data storage device. The signal processing unit enables the ventilation device to calculate an actuation value (Stellwert) for controlling the ventilation device, more precisely, to calculate the actuation value for controlling the ventilation device based on a determined measure of the patient's own respiratory activity. The signal processing unit is capable of automatically controlling the ventilation device based on the actuation value and / or outputting the actuation value in a human-perceptible form.

[0029] The signal processing unit according to the invention receives measurements of signals that are also measurable and generally time-varying, and generates a set of signal values ​​through signal preprocessing. These measurable signals are associated with physical variables, in this invention with the patient's cardiac activity and / or spontaneous breathing and / or with the patient's artificial ventilation, and are generated by at least one signal source in the patient's body or by a ventilation device. Hereinafter, "signal" is to be understood as the change in a directly or indirectly measurable and time-varying variable in the time domain or also in the frequency domain, said variable being associated with physical variables, preferably with anthropological variables. Respiratory signals are associated with the patient's respiratory activity, and cardiac signals are associated with the patient's cardiac activity.

[0030] According to the present invention, the signal processing unit controls the ventilation device, and the controlled ventilation device performs at least one ventilation process, wherein during the ventilation process, the parameters of the first ventilation device remain set at the same set value.

[0031] Ideally, the ventilation device is operated during ventilation so that it operates in perfect synchronization with the patient's own respiratory activity, which has been determined according to the invention. Therefore, regulation is performed, wherein the patient's own respiratory activity supplies the reference variable or supplies the reference variable. In practice, however, achieving perfect synchronization is rarely possible.

[0032] According to the invention, during the ventilation process, the parameters of the first ventilation device are kept set to the same set value. Preferably, each ventilation process includes at least one ventilation stroke, and more preferably, it includes multiple ventilation strokes.

[0033] The patient's own respiratory activity is described by respiratory activity measurements, preferably by pneumatic measurements. For example, the measurement is the pressure P generated by the respiratory muscles. mus Specifically, the pressure P in the esophagus es Or it could be the gastric pressure P in the patient's stomach. ga Because of this invention, it is not necessary to continuously and directly measure this respiratory activity. Such direct measurement may often be impossible, or only possible in special circumstances, especially in cases of occlusion (where artificial ventilation is set for a short period of time).

[0034] According to the present invention, a lung mechanics model is pre-defined. This lung mechanics model includes at least one relationship between respiratory activity measurements and at least one measurable signal, preferably multiple measurable signals. Preferably, the relationship or at least one relationship of the lung mechanics model is a model equation. According to the present invention, the lung mechanics model is stored in a data memory, and a signal processing unit has at least temporary read access to the data memory. The signal processing unit receives at least one measurable signal for the lung mechanics model, preferably the measurable signal for the lung mechanics model, or a measurement value for each measurable signal, repeatedly generates signal value sets from these measurements, each signal value set having a signal value for each measurable signal, and derives a first respiratory activity value and optionally other respiratory activity values. Due to the features of the present invention, it is not necessary to directly measure respiratory activity values. This is absolutely impossible in many cases or situations, or only possible in cases of significant time delay, or may place a heavy burden on the patient, or be excessively time-consuming in clinical practice.

[0035] According to the invention, the signal processing unit calculates a measure of reliability that a first respiratory activity value corresponds to the patient's actual respiratory activity value, which has been derived from at least one set of signal values ​​using a pre-given lung mechanics model. Therefore, the signal processing unit not only supplies the estimated respiratory activity value but also additionally supplies a statement regarding the reliability of this signal value, i.e., the signal quality index (SQI). In particular, the invention enables the derived respiratory activity value to be used with a constant setpoint when reliability is sufficiently high, while not using the derived respiratory activity value when reliability is too low, or setting the ventilation device to a different setpoint even though the derived respiratory activity value is used. In some cases, this effect facilitates meeting the statutory requirements for medical devices.

[0036] According to the present invention, the signal processing unit automatically determines at least once after deriving respiratory activity values ​​whether it has triggered a change process. Preferably, for example, the signal processing unit repeats this determination after each ventilation process, after each derived respiratory activity value, after each change process, and / or after each breath of the patient.

[0037] According to the present invention, the signal processing unit triggers a change process under conditions of low reliability metric—more generally: the signal processing unit triggers a change process when a predefined trigger criterion is met and thus occurs. This predefined trigger criterion relates at least to the last calculated reliability metric, and optionally additionally to a previously calculated reliability metric.

[0038] During the modification process, the parameters of the first ventilation device acquire different values ​​than before. Therefore, the ventilation device operates differently than before. This modification process for the first ventilation device parameters can be referred to as a manoever during ventilation device operation. In many cases, this manoever results in respiratory activity values ​​being derived with higher reliability based on measurements taken after the modification process than before the manoever. Higher reliability is often achieved when at least one set of signal values ​​generated not only before the modification process but also at least one set of signal values ​​generated after the modification process are used for derivation.

[0039] In some cases, respiratory activity measurements may be determined directly after the process has been modified, or more precisely, preferably without using a lung mechanics model applied in the derivation. This avoids the error that a pre-defined lung mechanics model is merely a simplification of reality.

[0040] On the one hand, the present invention causes a change in the parameters of the first ventilation device, and thereby generally, when triggering criteria are met, and especially when the reliability measure is below a first reliability limit, changes the manner and method of artificial ventilation of the patient. It is possible that the signal processing unit uses multiple measurements to derive respiratory activity values, measured under different setpoints. If the measurements used are taken under different setpoints and the respiratory activity values ​​are derived from these measurements obtained under different setpoints, then in many cases, the respiratory activity values ​​derived in this way are more reliable than those taken when the same setpoint is maintained for a longer period and only measurements are measured and used under that single setpoint. This higher reliability results from a stronger change in the effect that the ventilation device has on the patient's own breathing when the setpoint changes. Therefore, the present invention improves the reliability of the ventilation device in many cases.

[0041] On the other hand, the present invention enables the maintenance of currently used settings, such as standard settings for the parameters of the first ventilation device, for as long as possible, especially when the respiratory activity values ​​derived from or under these settings are sufficiently reliable. This avoids placing a greater burden on the patient than necessary by frequently changing the parameters of the first ventilation device, i.e., by frequent repetitions. Thus, the ventilation device is also less frequently and intensely loaded.

[0042] This invention demonstrates an understandable and recordable approach to why the signal processing unit triggers or also stops the change of a first ventilation device parameter. This invention can also be applied to ventilation devices with multiple changeable ventilation device parameters. The signal processing unit then preferably determines which ventilation device parameter is involved in the change process, i.e., which ventilation device parameter acquires an additional set value during the change process.

[0043] If multiple ventilation device parameters can be changed, this invention provides a way of indicating which ventilation device parameter is changed and why it is not changed. To improve the reliability of the derivation and thereby improve the consistency of the measure between the derived and actual respiratory activity values, the necessity of modifying the first ventilation device parameter, or optionally other ventilation device parameters, based solely on "intuition" or, for example, a pre-given, general empirical formula to be applied to each patient, is avoided. The characteristic of ventilation device parameters being changed, and thus understood and systematically changed, is particularly advantageous when changing the process and / or the second setpoint, or each second setpoint, imposes a burden on the patient and / or can only be maintained for a short period. This facilitates the recording of the ventilation device's operation.

[0044] According to the present invention, the signal processing unit uses a pre-given lung mechanics model. This lung mechanics model may consist of model equations, or may include multiple model equations. Preferably, in each of the model equations or at least one model equation of the lung mechanics model, a measure of respiratory activity to be determined appears. Preferably, in each of the model equations or at least one model equation, at least one measurable signal also appears.

[0045] According to the present invention, the signal processing unit calculates a metric for reliability that the derived first respiratory activity value corresponds to the patient's actual respiratory activity. In one implementation, the signal processing unit calculates an estimated signal value as the first respiratory activity value and calculates a metric for the estimation uncertainty, the derivation of the first respiratory activity value being related to the estimation uncertainty. When the metric for the estimation uncertainty is above an uncertainty limit, the signal processing unit triggers a change process. A reliability metric below the reliability limit is equivalent to a metric for the estimation uncertainty being above the uncertainty limit.

[0046] According to the present invention, the signal processing unit automatically determines whether to trigger a change process based on a calculated reliability metric. When a pre-defined triggering criterion is met, more specifically, at least when the reliability metric is below a first reliability limit, the signal processing unit triggers the change process or initiates a change process. In one implementation, the signal processing unit derives respiratory activity values ​​multiple times sequentially under a first set value and calculates a reliability metric for each derived value. In another implementation, the signal processing unit also triggers a change process when multiple successive reliability metrics become worse and approach the first reliability limit from above, more specifically, preferably before the reliability metric drops below the first reliability limit.

[0047] According to the present invention, the signal processing unit triggers the ventilation process at least once, preferably repeatedly, during which the parameters of the first ventilation device are set to a set value that deviates from the previous value.

[0048] According to the invention, the signal processing unit derives a first respiratory activity value during the first ventilation process. Preferably, the signal processing unit also derives the respiratory activity value after the change process, i.e., during operation at a second set value, more precisely, derives the respiratory activity value from at least one set of signal values ​​generated under the second set value. In one implementation, the respiratory activity value is derived only from a set of signal values ​​generated under the current set value (more precisely: generated from measurements taken under the current set value). To derive the respiratory activity value, the signal processing unit uses at least one set of signal values, preferably multiple sets of signal values ​​generated under the current set value.

[0049] In the alternative configuration, the signal processing unit derives at least one respiratory activity value based on a set of signal values ​​generated under the current settings, and additionally based on a set of signal values ​​generated under previously used settings, preferably based on a set of signal values ​​generated under the settings to which the first ventilation device parameter was set before the final change process. Due to this alternative configuration, more signal value sets are available for derivation compared to when only the set of signal values ​​generated under the current settings might be used. In many cases, this improves the reliability of the derivation, especially when statistical methods are applied, and avoids the need for additional change processes.

[0050] According to the present invention, the signal processing unit derives a first respiratory activity value and calculates a reliability measure used to derive the first respiratory activity value. At least in cases of low reliability measure, additional ventilation procedures are performed, more specifically, in cases of deviation from a second set value. These additional ventilation procedures supply additional measurements, from which the signal processing unit generates additional signal value sets. The signal processing unit then determines a second respiratory activity value.

[0051] In one configuration, the signal processing unit uses a set of signal values ​​generated under a second setpoint, optionally using a set of signal values ​​under a previous setpoint, and applies a pre-defined lung mechanics model to derive the second respiratory activity value in the same manner as the first respiratory activity value. Preferably, the signal processing unit calculates a reliability metric for deriving the second respiratory activity value.

[0052] In another implementation, the signal processing unit determines the second respiratory activity value in a different manner, such as by direct measurement that was impossible before the change but became possible after the change, or more precisely, preferably by determining the second respiratory activity value without using a lung mechanics model. Alternatively, the signal processing unit uses another lung mechanics model, particularly one that describes reality better after the change and / or better than previously used lung mechanics models. It may be possible, but not necessarily, for the signal processing unit to also calculate a reliability metric for determining the second respiratory activity value.

[0053] According to the invention, the ventilation device operates according to first ventilation device parameters. In one configuration, the first ventilation device parameters affect the amount of gas delivered to the patient. If the ventilation device operates in a volume-controlled manner, the ventilation device parameters affect, for example, the amount of required air volume flow to the patient or the amount of lung filling level. If the ventilation device operates in a pressure-controlled manner, the ventilation device parameters affect, for example, the required respiratory air pressure to be generated by the ventilation device.

[0054] Typically, the required volumetric flow or required pressure is related to the patient's own respiratory activity. According to the invention, when triggering criteria are met, particularly when the reliability measure of the calculated first respiratory activity value is below a first reliability limit, the signal processing unit triggers a change process. In the configuration of the first ventilation device parameters just described, this change process preferably consists of or includes the following steps: the ventilation device temporarily reduces or restricts gas delivery to the patient, or may temporarily increase gas delivery to the patient. Preferably, this is followed by other change processes in which the ventilation device again increases gas delivery to the patient, or cancels or reduces the restriction again, particularly to the old set value. A particular example of this configuration is that the ventilation device is fully set to provide artificial ventilation (occlusion) to the patient for a predetermined time interval, preferably less than 5 seconds, particularly preferably less than 1 second.

[0055] In one configuration, the triggered change process causes the airway pressure and / or established volumetric flow generated by the ventilation device to remain either below or above a predetermined limit, either only within a predetermined time limit, or only during patient inhalation, or only during patient exhalation. A specific example of this configuration is that the ventilation device terminates artificial ventilation (occlusion) after the change process. Preferably, after a predetermined time interval, typically less than 5 seconds, a repeat change process is performed, during which the ventilation device restarts artificial ventilation.

[0056] In one configuration, the signal processing unit controls the ventilation device with the following regulatory objective: the airway pressure actually established by the ventilation device, or the patient's lung filling level actually induced by the ventilation device, equals a pre-given desired airway pressure or a pre-given desired filling level, wherein this pressure or filling level may be variable over time. A triggered change process alters the desired airway pressure or desired filling level. In one configuration, after the change process, a pre-given time-varying process of the desired airway pressure or desired filling level is used, which is not necessarily related to the patient's own respiratory activity. In particular, control, rather than regulation, is thus performed. In an alternative configuration, the change process results in the derivation of the desired time-varying process used as a reference variable in a manner different from that prior to the change process of the patient's own respiratory activity.

[0057] In the modified configuration, the alteration process includes the following steps: manipulating the ventilation equipment such that, after the alteration process, the flow rate, i.e., the volume of air delivered per unit time, remains below a predetermined limit. Preferably, a repeat alteration process is then performed, after which the flow rate can again be above the limit.

[0058] In one configuration, a signal processing unit can control a ventilation device to selectively perform either pressure-regulated ventilation or volume-regulated ventilation of the patient. In pressure-regulated ventilation, a predetermined time-varying process of the desired pressure to be generated by the ventilation device is given, and the signal processing unit controls the ventilation device such that the actual pressure follows the predetermined change process of the desired pressure. In volume-regulated ventilation, a predetermined time-varying process of the patient's lung filling level (volume) is given, and the signal processing unit controls the ventilation device such that the flow rate (volume flow) of gas between the ventilation device and the patient causes the actual filling level to follow the predetermined desired change process. In one configuration, the alteration process, or the triggered alteration process, includes the step of changing the type of regulation, i.e., either the ventilation device operates in a pressure-regulated manner before the alteration process and subsequently operates in a volume-regulated manner, or vice versa.

[0059] In one configuration, the ventilation device operates in a proportionally regulated manner, at least before the change process; that is, the measurement of the variable for artificial ventilation is proportional to the corresponding variable for the patient's own respiratory activity, which is preferably determined according to the invention. Therefore, the stronger the patient's breathing, the stronger the support provided by the ventilation device through artificial ventilation. In one configuration, the change process includes the step of: after the change process, the ventilation device is no longer proportionally regulated. In another configuration, the ventilation device also operates in a proportionally regulated manner after the change process, but after the change process, the scaling factor (support factor) is different from the previous scaling factor, especially a smaller scaling factor. In this configuration, the scaling factor therefore acts as the setpoint or serves as a set value.

[0060] In one configuration, the ventilation device performs a series of ventilation cycles, the execution of which is related to, or relating to, calculated respiratory activity values. Setpoints define parameters of the ventilation cycles, such as amplitude or frequency, or a time delay between the patient's own respiratory activity and the ventilation cycle. Changing the process results in different setpoints, and thus different amplitudes, frequencies, or time delays.

[0061] The respiratory activity values ​​derived according to the invention, or each respiratory value derived according to the invention, can be used for various purposes. In one implementation, the signal processing unit uses the calculated respiratory activity values, or at least one calculated respiratory activity value, to control the ventilation device. For example, the signal processing unit performs control to achieve the adjustment objective that the artificial ventilation induced by the ventilation device is completely synchronized with the patient's own respiratory activity. The signal processing unit uses the derived respiratory activity values, or at least one derived respiratory activity value, to control the ventilation device according to this adjustment objective.

[0062] The signal processing unit controls the ventilation device based on respiratory activity values, including at least one of the following steps: the signal processing unit

[0063] - Triggers the ventilation cycle of the ventilation equipment.

[0064] - Set the frequency and / or amplitude of successive ventilation strokes of the ventilation device to a pre-defined value, or trigger this setting, or

[0065] - Establish a pre-defined time-varying process of airway pressure to be established.

[0066] According to the invention, a change process is executed when a pre-defined trigger criterion is detected. Preferably, the signal processing unit controls the ventilation device based on the derived first respiratory activity value only when the trigger criterion is not met, for example, when the reliability measure used to derive the respiratory activity value and optionally, at least one previously calculated reliability measure, exceed a first reliability limit. For example, the signal processing unit derives the respiratory activity value based on a set of signal values ​​generated by sensors measured closer to the signal source, particularly measuring electrodes on the patient's skin and / or optical sensors spaced apart from the patient or pneumatic sensors in the patient's esophagus.

[0067] If the reliability measure is below a first reliability limit (generally: the triggering criterion is met at a first set value), the signal processing unit in one configuration does not use the derived first respiratory activity value for manipulation. Specifically, in one configuration, the signal processing unit manipulates the ventilation device based on signals for flow rate and / or pressure, where the flow rate or pressure occurs in the circulation of gas between the ventilation device and the patient. Signals for flow rate and / or pressure can typically be directly measured with the aid of measurement preprocessing, especially without using a pre-defined lung mechanics model. However, these signals are more strongly superimposed with interfering signals compared to the measurable signals or each measurable signal present in the lung mechanics model, and / or the sensors used only measure the corresponding signals with a time delay. Especially in many cases, the sensors for flow rate or pressure are arranged within or on the ventilation device, while the volumetric flow rate or pressure is measured at the patient's mouth, in the patient's airway, or in the patient's esophagus, and interference effects may occur along the path between the ventilation device and the patient. Furthermore, there is a time delay between the formation of a signal in or on the patient’s body and the measurement location in the ventilation device, and this time delay can usually only be approximated and is also usually variable over time.

[0068] For all these reasons, artificial ventilation, which is solely regulated based on signals for flow rate and / or pressure, is less synchronized with the patient's own respiratory activity than when respiratory activity values ​​might be measured using sensors (e.g., a set of measuring electrodes) located close to the body. Therefore, it is advantageous to regulate artificial ventilation based on respiratory activity values ​​derived from measurements taken from sensors located close to the body. However, this must be sufficiently reliable.

[0069] For example, the measurable signal or measurable energy is measured using measuring electrodes positioned on the patient's skin. The measurable signal or measurable energy is electromyography (EMG) or myometry (MMG). Respiratory activity measurements are aerodynamic variables, such as the aerodynamic pressure P generated by the respiratory muscles. mus Furthermore, the aerodynamic variables are associated with electromyography or muscle mapping based on a pre-defined lung mechanics model, and optionally with other measurable signals, such as volumetric flow and / or volume.

[0070] In one configuration, the derived or determined respiratory activity value, or at least one derived or determined respiratory activity value, is preferably output along with the calculated reliability metric, and especially in a human-perceptible form, such as visually on the output unit. In another configuration, a duct (Schlauch) is placed on the output unit around the time-varying process of the respiratory activity metric, wherein the wider the duct, the lower the reliability.

[0071] Preferably, this output is executed continuously. Alternatively, the signal processing unit checks whether the derived respiratory activity value, or the change in the derived respiratory activity value over time, meets predefined criteria, such as the value being outside a predefined range, or the change occurring faster than a predefined limit. If the predefined criteria are met, the signal processing unit triggers an alarm.

[0072] In other configurations, the derived or determined respiratory activity values ​​are transmitted to other devices, such as anesthesia equipment or other medical devices, or to a central data processing system. These other medical devices use the transmitted respiratory activity values ​​for their own operation. The central data processing system preferably evaluates data transmitted from different medical devices, such as data about the same patient.

[0073] According to the invention, a signal processing unit derives a first respiratory activity value, and for the purpose of said deriving, at least one group of signal values, preferably multiple groups of signal values, are generated under a first set value. "Generated under a set value" means that the measurement value used to generate has been measured under that set value. The signal processing unit derives a second respiratory activity value using at least one group of signal values, preferably multiple groups of signal values, generated under the second set value. In one implementation, the signal processing unit uses the measurement value measured under the second set value and a lung mechanics model to derive the second respiratory activity value. Preferably, the signal processing unit calculates a reliability measure for the second respiratory activity value, said reliability measure being a measure of reliability that the derived second respiratory activity value is consistent with actual respiratory activity.

[0074] In one configuration, a signal processing unit adjusts the ventilation device based on a plurality of derived and / or determined respiratory activity values, which are derived or determined by applying the method according to the invention. The preferred objective of such adjustment is that the ventilation device operates synchronously with the patient's own respiratory activity, i.e., the gas flow to and / or from the patient induced by the ventilation device is synchronized with the patient's own respiratory activity. In this adjustment, for example, the degree of lung filling (i.e., volume) is a reference variable that can be varied over time (volume-controlled adjustment of the ventilation device). Volume flow (i.e., the flow of gas into or out of the lungs) is a control variable. Alternatively, a predetermined required pressure in the airway is a reference variable (pressure-controlled adjustment), which can also be varied over time. The actual pressure in the airway is measured. The pressure generated by the ventilation device is a control variable.

[0075] According to the invention, the signal processing unit derives a first respiratory activity value and uses at least one set of signal values ​​that has been measured at a first set value. Preferably, the first ventilation device parameter remains set at the first set value as long as no predetermined triggering criterion for triggering the change process is detected, particularly as long as the calculated reliability metric is above a first reliability limit. The signal processing unit preferably also performs a ventilation process at the first set value, and in this case generates at least one other set of signal values ​​that has been measured later at the first set value. Using the other set of signal values, or at least one other set of signal values ​​and optionally the first set of signal values, the signal processing unit derives other respiratory activity values. This configuration avoids the step of performing the change process when it is not required.

[0076] According to the present invention, when using at least one set of signal values ​​that have been measured under a first set value, the signal processing unit derives a first respiratory activity value. Optionally, when using at least one other set of signal values ​​that have been measured under other set values, the signal processing unit derives other respiratory activity values. In one configuration, the signal processing unit generates multiple sets of signal values, wherein all measurements of the multiple sets of signal values ​​were measured under the same set value. The signal processing unit calculates a reliability measure based on the multiple sets of signal values ​​used for derivation. Preferably, the signal processing unit applies a statistical method to derive the reliability measure. This configuration reduces the impact of measurement errors and biases that occur only at a single sampling time.

[0077] In an improved version of this construction scheme, in the step of deriving the first respiratory activity value, the signal processing unit applies a regression method, more specifically, applies the regression method to the lung mechanics model and to multiple groups of signal values ​​that have so far been obtained with the current settings of the first ventilation device parameters. Preferably, the signal processing unit applies the regression method to all groups of signal values ​​obtained so far with the current settings. Preferably, the signal processing unit also applies the regression method when deriving at least one other respiratory activity value. Preferably, the regression method includes the step of calculating and minimizing the sum of squared errors.

[0078] According to the present invention, when a triggering criterion is met under a first set value, the signal processing unit triggers a change process in which a first ventilation device parameter is set to a second set value deviating from the first set value. In one configuration, the second set value is related to a calculated reliability metric. Preferably, the further the reliability metric is from a first reliability limit, the more strongly the second set value deviates from the first set value. Alternatively, depending on which of two pre-defined ranges the reliability metric falls below the first reliability limit, the first ventilation device parameter is set to one of two possible second set values. It is also possible, naturally, that during the change process, the first ventilation device parameter is set to one of at least three different possible set values.

[0079] In one configuration, in addition to a first reliability limit, a smaller second reliability limit is pre-defined. If the reliability measure used to derive the first respiratory activity value is between the two reliability limits, the derived first respiratory activity value is used to regulate the ventilation device. However, the ventilation device deviates from normal operation, for example, by reducing or limiting the support factor, volumetric flow, or pressure. If the reliability measure is below the second reliability limit, the first respiratory activity value is not used; instead, the signal processing unit, for example, causes the ventilation device to temporarily set to artificial ventilation (occlusion), or the ventilation device uses signals for volumetric flow and / or pressure instead of the first respiratory activity value, or controls the ventilation device instead of regulating it.

[0080] Alternatively, other reliability limits below the first reliability limit may be predetermined. If the reliability measure falls between the first and other reliability limits, the first ventilation device parameter is set to a second set value. If the reliability measure falls even below the other reliability limits, additional ventilation device parameters are set to a deviated second set value.

[0081] In one construction scheme, a pre-defined lung mechanics model has at least one model parameter, which is typically variable over time and not known in advance. The current value of this model parameter is not known beforehand. For example, the parameter value varies depending on the patient and / or changes during artificial ventilation. To derive a first respiratory activity value, the signal processing unit derives the parameter value at least once for the model parameter of the lung mechanics model, or at least one model parameter, preferably for each model parameter of the lung mechanics model. To derive the model parameter value, the signal processing unit uses at least one group of signal values, preferably multiple groups of signal values, which are generated with a first set value. Using the model parameter value, or at least one model parameter value and at least one signal value, the signal processing unit derives the respiratory activity value.

[0082] The derivation of model parameter values ​​typically involves uncertainty. The signal processing unit calculates a measure of reliability for the model parameters, or for each model parameter individually, based on the reliability with which the values ​​for the model parameters are derived. This reliability measure is used to calculate a reliability measure for deriving the first respiratory activity value; it is used, for example, as a reliability measure for the derivation.

[0083] In an improved version of this construction scheme, to calculate the reliability measure used to derive the model parameter values, a probability distribution is pre-given for each model parameter, or for at least one model parameter. In the step of calculating the reliability measure used to derive the model parameter values ​​with respect to their pre-given probability distributions, the following steps are performed:

[0084] - The signal processing unit generates multiple signal value groups.

[0085] - For the model parameters or model parameters with a pre-defined probability distribution, the signal processing unit calculates confidence intervals and / or standard deviations and / or empirical scattering or variance. Alternatively, the signal processing unit performs statistical tests.

[0086] - For the calculation, the signal processing unit uses a pre-defined probability distribution of the model parameters. Furthermore, the signal processing unit uses the set of signal values ​​used to derive respiratory activity values.

[0087] - The reliability metric sought involves the derivation of respiratory activity values ​​and is calculated based on the calculated confidence interval or the calculated standard deviation / scattering / variance.

[0088] In one configuration, the lung mechanics model has first model parameters and at least one second model parameter. A signal processing unit calculates a first reliability metric and a second reliability metric. Each reliability metric is a measure of reliability that the derived value for either the first or second model parameter is sufficiently consistent with reality. If the first reliability metric is below a first reliability limit, the signal processing unit triggers a first modification process. If the second reliability metric is below the first reliability limit, the signal processing unit triggers a second modification process. These two modification processes may be consistent or different. For example, the two modification processes may involve different ventilation device parameters. Alternatively, the first modification process may result in a second setpoint for the first ventilation device parameter that differs from the second modification process. This configuration enables targeted acquisition of measurements to derive values ​​with high safety for defined model parameters.

[0089] In one configuration, a signal processing unit derives at least one respiratory activity value from multiple sets of signal values. At least one first set of signal values ​​is generated with a first set of values, and at least one second set of signal values ​​is generated with a second set of values. For each set of signal values ​​used, the signal processing unit calculates a weighting factor and additionally uses the weighting factor of the signal value set to derive the respiratory activity value. This configuration leads to high reliability in many cases.

[0090] In a preferred configuration, before the change of process, i.e., under a first setpoint, the ventilation equipment operates in a normal mode; and after the change of process, i.e., under a second setpoint, the ventilation equipment operates in a special mode, which is typically maintained for only a short period. The signal value set generated under the second setpoint, or each signal value set generated under the first setpoint, receives a higher weighting factor compared to the signal value set generated under the first setpoint, or each signal value set generated under the second setpoint. For example, the fewer signal value sets measured under the setpoint, the larger the weighting factor of the signal value sets generated under that setpoint. Due to this configuration, the signal value sets generated under the second setpoint (i.e., in the special operating mode) have a relatively large impact on the derivation, even if the special operating mode is used only for a relatively short period. Therefore, this configuration makes it easy to set special operating modes for short-term and specific measurement and derivation. This, in particular, allows for a higher evaluation of such signal value sets generated during targeted short-term exercises.

[0091] In another configuration, under the second setpoint, it is possible to measure a respiratory activity measure rather than deriving it. For example, the respiratory activity measure is a pneumatic measure, and under the second setpoint, the ventilation device does not support the patient's respiratory activity ("occlusion"), so that external pressure is caused solely by the patient's own respiratory activity. In this other configuration, the step is unnecessary and preferably not performed; a second respiratory activity value or a reliability measure for the second respiratory activity value is derived from the set of signal values ​​using a lung mechanics model. The signal processing unit compares the determined second respiratory activity value with the derived first respiratory activity value to calculate the reliability measure used to derive the first respiratory activity value. In one configuration, when such a comparison yields a low reliability measure, the signal processing unit automatically applies another pre-given lung mechanics model or changes the model parameter values.

[0092] According to the present invention, a signal processing unit derives a first respiratory activity value from at least one set of signal values ​​measured at a first set value. If a triggering criterion is met, the signal processing unit triggers a change process in which a first ventilation device parameter is set to a second set value. In one implementation, when the first ventilation device parameter is set to the second set value, it is possible to measure a respiratory activity measure. For example, a respiratory activity measure is the aerodynamic pressure P generated by the patient's respiratory muscles. mus Furthermore, the measurable signal is the pneumatic pressure P in the ventilation cycle between the patient and the ventilation device. aw Furthermore, under the second set value, the ventilation device does not perform manual ventilation. In this case, for example, P is applied. mus =P aw In one construction scheme, consider P mus With P aw The correction factor and / or delay factor between them.

[0093] In one configuration, a signal processing unit determines a second respiratory activity value by processing at least one measurement value, preferably a measurement value from a pneumatic sensor, measured under a second set value. A lung mechanics model is preferably not used for this determination. In another configuration, the signal processing unit compares a first respiratory activity value derived under a first set value with a second respiratory activity value determined under a second set value. The signal processing unit calculates a reliability metric for the first respiratory activity value and uses the result of this comparison.

[0094] According to the present invention, a pre-defined lung mechanics model is stored in a data memory and describes at least one relationship between a respiratory activity measure and at least one measurable signal. The respiratory activity measure is preferably an aerodynamic measure P of the pressure generated by the patient's respiratory muscles. mus In the lung mechanics model, at least one of the following signals is preferably used:

[0095] - Airway pressure (P aw ),

[0096] - Pressure in the esophagus (P es ),

[0097] - Respiratory flow (Flow, ),

[0098] - Lung volume (Vol)

[0099] - The amount of carbon dioxide (CO2) in exhaled air, and / or

[0100] - The oxygen content in the blood.

[0101] In one construction scheme, the following two linear model equations in the model parameters are pre-defined as the lung mechanics model:

[0102] ,and

[0103] .

[0104] in this case,

[0105] - P mus (t) is the sought and time-varying measure of respiratory activity, namely the aerodynamic pressure generated by the patient's respiratory muscles.

[0106] - P aw (t) is the airway pressure measured during the patient's circulation, which is used as a measurable signal and is derived from the superposition of the patient's own respiratory activity and ventilation via a ventilation device.

[0107] - R is a factor describing the respiratory resistance of the patient's airway against volumetric flow Vol'.

[0108] - E is a factor targeting the elasticity of the patient's lungs, and

[0109] - P0 is a variable that is treated as a constant, such as a measure of the effect of incomplete exhalation (iPEEP) on a patient.

[0110] The signal Sig(t) is also related to the aerodynamic pressure P generated by the patient's respiratory muscles.mus The respiratory signal is associated with, and is measured, for example, by means of measuring electrodes (EMG sensors) or myographic sensors (MMG sensors) on the skin, and is therefore either electrical or mechanical.

[0111] The measured electrorespiratory signals are correlated with electrical impulses that cause contraction of the respiratory muscles, which in turn contribute to the patient's own respiratory activity. Factor k eff It is a scaling factor between pneumatic pressure and the electrical signal from the measuring electrodes, and describes the so-called electromechanical efficiency, i.e., how well the electrical impulse is converted into muscle activity. In this example, the factors R, E, and k... eff The addend P0 consists of four model parameters whose values ​​may change during patient ventilation. Parameters R, E, and P0 are lung mechanics parameters. These two model equations of the lung mechanics model provide two pathways for deriving the respiratory activity measure P. mus .

[0112] In one construction scheme, model equations are used. In this case, the first respiratory activity value is derived, and preferably using the model equation and / or model equation In this case, and calculate the reliability metric used for the derivation—or vice versa. Attached Figure Description

[0113] The invention will now be described with reference to embodiments. In this case:

[0114] Figure 1 The illustration schematically shows which sensors measure which different signals, which are used to deduce the patient's own respiratory activity;

[0115] Figure 2 It shows which signals are derived from the measurements of which sensors;

[0116] Figure 3 An exemplary weighting function is shown for weighting multiple groups of signal values;

[0117] Figure 4 An exemplary weighting of a group of signal values ​​based on the frequency of the signal values ​​is shown;

[0118] Figure 5 The first part of the flowchart is shown: deriving respiratory activity values ​​and determining whether they meet pre-defined triggering criteria;

[0119] Figure 6 The second part of the flowchart is shown: routine operation when respiratory activity values ​​are sufficiently reliable;

[0120] Figure 7The third part of the flowchart is shown: conducting a lighter exercise;

[0121] Figure 8 The fourth part of the flowchart is shown: Executing a major exercise;

[0122] Figure 9 The fifth part of the flowchart is shown: deriving model parameter values ​​based on the set of signal values ​​generated in the exercise;

[0123] Figure 10 The sixth part of the flowchart is shown: deriving the respiratory activity values ​​in the exercise and calculating the reliability of the derivation of the respiratory activity values;

[0124] Figure 11 Part 7 of the flowchart is shown: Determining how to continue artificial ventilation after the exercise. Detailed Implementation

[0125] In this embodiment, patient P is at least temporarily ventilated by ventilation device 1. The ventilation is synchronized with patient P's own respiratory activity. Ventilation device 1 is adjusted according to patient P's own respiratory activity.

[0126] In one configuration, the ventilation device 1 operates in a pressure-regulated manner. In this case, the reference variable in the regulation is the required time-varying process of the pneumatic pressure of respiration (preferably in the patient P's airway). The control variable is the pneumatic pressure achieved by artificial ventilation. This desired pressure change is synchronized with the time-varying pressure achieved by the patient P's own respiratory activity, and thus the desired change is related to the patient's own respiratory activity. In another configuration, the reference variable in the regulation is the required time-varying process of volume, i.e., the required time-varying process of the lung filling level of the patient P. The control variable is the airflow entering and exiting the lungs, which is achieved by artificial ventilation. Also in this configuration, the desired volume change is synchronized with the patient P's own respiratory activity.

[0127] For synchronization, what is needed in the regulation of these two methods is to determine the preferred aerodynamic quantity, such as pressure P, for the patient P's own respiratory activity. mus The patient P's own respiratory activity is related to the pressure generated by the patient P's respiratory muscles. During artificial ventilation, the time-varying and preferably pneumatic quantity P cannot be directly measured. mus Instead, the time-varying and preferably aerodynamic measure P is determined in the following manner. mus At each sampling time t i

[0128] - Measure multiple variable events that occur during the ventilation cycle.

[0129] - Generate a set of signal values ​​from a single measurement value for each measurable signal, and

[0130] - From at least one set of generated signal values, preferably from multiple sets of signal values, repeatedly derive the respiratory activity measure P for the preferred pneumatic system. mus The value, that is, the estimated respiratory activity value P. mus,est (t i ).

[0131] When adjusting the ventilation device 1 proportionally, under ideal conditions, at each sampling time t i The pressure P generated by ventilation device 1 art (t i ) and the estimated respiratory activity value P mus,est (t i Proportional, that is

[0132] ,

[0133] Where P mus,est (t i ) is the estimated respiratory activity value, and x is a pre-given scaling factor. This scaling factor x is also referred to as the support provided by ventilation device 1. Under ideal synchronization, .

[0134] The signal processing unit that performs data processing executes the higher-level regulation just described, such as pressure-regulated or volume-regulated regulation, and for this purpose uses the estimated value P for the respiratory activity measure. mus,est (t i ), wherein the value P mus,est (t i The signal processing unit derives the values ​​using a set of signal values. In the higher-level regulation, the signal processing unit calculates the values ​​for the pressure and / or volumetric flow currently to be generated by the ventilation device. Furthermore, the signal processing unit performs lower-level regulation to derive the control intervention (Stelleingriffe) for the control element of ventilation device 1 from the required values ​​for the pressure to be generated, wherein the control element induces artificial ventilation for patient P.

[0135] Figure 1 This illustration shows which sensors measure patient P's spontaneous breathing and artificial ventilation. It shows:

[0136] - Patient P,

[0137] - The esophagus (Sp) and diaphragm (Zw) of patient P,

[0138] - Ventilation device 1, which provides artificial ventilation to patient P at least temporarily, and includes a signal processing unit 5 for data processing.

[0139] - Data memory 9, signal processing unit 5 has at least temporary read access to data memory 9, and computer-usable lung mechanics model 20 is stored in data memory 9.

[0140] - Four sets of sensors 2.1.1 through 2.2.2, each having at least one measuring electrode, wherein measuring electrode sets 2.1.1 and 2.1.2 are arranged parallel to the sternum, and measuring electrode sets 2.2.1 and 2.2.2 are arranged on the costal arch.

[0141] - Pneumatic sensor 3, which measures the airway pressure P in front of the patient P's mouth. aw And the volumetric flow of breathing air entering and leaving the lungs of patient P. ,

[0142] - Optional optical sensor 4, which includes an image recording device and an image evaluation unit, and is aligned with the chest region of patient P, and

[0143] - An optional pneumatic sensor 6, which is in the form of a probe or balloon, is placed in the esophagus Sp and close to the diaphragm Zw of the patient P. This optional pneumatic sensor 6 measures the pressure P in the esophagus Sp. es .

[0144] Measuring electrodes 2.1.1 through 2.2.2, and an electrode for electrical ground (not shown), enable non-invasive electromyography (EMG) measurements. Alternatively, the sensor may be positioned on the patient P's body as close as possible to a signal source capable of myometry (MMG) measurements.

[0145] Figure 2 This clarifies which signals are derived from the measurements of which sensors. The possible sources of these signals and measurement errors are described below.

[0146] The four groups of measuring electrodes 2.1.1 through 2.2.2 and the electrode directed to ground supply the measurement values. These measurements are preprocessed, and the preprocessing provides at least one electrical signal associated with an electrical impulse generated in the body of patient P. Several of these electrical impulses cause contraction of the respiratory muscles of patient P, thereby inducing the movement of breathing air into and out of the lungs. The electrically excited respiratory muscles generate a pressure that is related to the desired aerodynamic quantity P for the patient's own respiratory activity. musCorrelation. Other electrical pulses in these pulses cause the patient P's heart to beat.

[0147] The measurements from the four measuring electrode groups 2.1.1 through 2.2.2 are therefore preprocessed, and a summed electrical signal, derived from the superposition of the respiratory and cardiac signals, is supplied after preprocessing. The respiratory signal is sought. The influence of the cardiac signal on the summed electrical signal is compensated for as much as possible computationally, for example, by applying methods described in DE 10 2015 015 296 A1, DE 10 2007 062214 B3, or in M. Ungureanu and WMWolf, “Basic Aspects Concerning the Event-Synchronous Interference Canceller” (IEEE Transactions on Biomedical Engineering, Vol. 53, No. 11 (2006), pp. 2240-2247). This computational compensation supplies a time-varying electrical respiratory signal, Sig. This electrorespiratory signal Sig is acquired near the signal source, i.e., the respiratory muscle, and is associated with electrical pulses that cause the respiratory muscles of patient P to move, and this electrorespiratory signal Sig is therefore associated with the aerodynamic quantity P. mus Related.

[0148] After computational preprocessing and compensation, the electrorespiratory signal Sig may still be superimposed with interference signals, such as those induced by electrochemical effects on the contact surfaces between the patient P's skin and the measuring electrodes 2.1.1 to 2.2.2. Furthermore, the patient P may change his / her body posture during measurement, and the effects of cardiac signals may not have been fully or correctly compensated for computationally.

[0149] The pneumatic sensor 3 measures the following values: these values ​​are the result of the superposition of the patient P's own respiratory activity and artificial ventilation. Only when artificial ventilation is interrupted are these values ​​solely caused by the patient's own respiratory activity. The airway pressure P is derived from these values. aw and volume flow That is, the flow of breathing air entering and leaving the lungs of patient P per unit time.

[0150] Patient P's respiratory activity is influenced by lung mechanics parameters. The values ​​of lung mechanics parameters and volumetric flow cannot be approximated simultaneously using only a single pneumatic sensor. Furthermore, especially to meet hygiene requirements in hospitals, the pneumatic sensor 3 is not positioned directly in front of or even within patient P's mouth, but rather spaced apart from patient P within or on the ventilation device 1. Consequently, a transmission channel exists between patient P's airway and the pneumatic sensor 3, specifically including a conduit between patient P and the ventilation device 1 and a mouthpiece in patient P's mouth. Therefore, a time delay inevitably exists between the formation of pressure in patient P's body and the moment when the pneumatic sensor 3 measures the pressure. For these two reasons—lack of observability and time delay—artificial ventilation generally cannot be ideally synchronized with patient P's respiratory activity solely based on the measurements of the pneumatic sensor 3.

[0151] Optical sensor 4 can determine the geometry of patient P's body through image processing, and the determined body geometry is related to the current lung filling level Vol, but also to other parameters. Therefore, optical sensor 4 can typically measure the lung filling level alone, only approximately and with considerable uncertainty.

[0152] Optional pneumatic sensor 6 measures the pressure P in the esophagus Sp of patient P. es However, in many cases, it is undesirable to deliver the pneumatic sensor 6 into the patient's esophagus (Sp), especially since placement and removal of the pneumatic sensor 6 takes relatively long and may, in some cases, place an additional burden on the patient (P). Furthermore, the pneumatic sensor 6 in the esophagus (Sp) only measures the pneumatic quantity (P) of respiratory activity under time delay. mus Moreover, it is achieved through the superposition of interference signals.

[0153] For the reasons mentioned above, it is worthwhile, on the one hand, to measure P's own respiratory activity based on aerodynamic parameters. mus To perform artificial ventilation for the patient, the estimated value P is derived using measurements from sensors located near the signal source. mus,est (t i In this invention, the estimated value P is derived by means of the measured values ​​of measuring electrodes 2.1.1 to 2.2.2. mus,est (t i On the other hand, current respiratory activity P musThe method must be derived with sufficiently high reliability so that artificial ventilation is reliably synchronized with the patient P's own breathing. Therefore, in this embodiment, artificial ventilation is adjusted based on the measurements of measuring electrodes 2.1.1 through 2.2.2, as well as the measurements of pneumatic sensor 3 and optionally other sensors 4 and / or pneumatic sensor 6.

[0154] In one construction scheme, at each sampling time t i Generate volumetric flow that varies with time. signal value And through numerical integration, the signal value Vol(t) for the current volume Vol, that is, the current filling level of the lung, is derived. i It is also possible that the signal value Vol(t) for the current volume can be derived additionally or alternatively from the measurements of optical sensor 4. i Note: Sampling time t i Is it a signal value or a measurement P? mus The value refers to the time in which it relates. This value itself can be calculated later.

[0155] According to the present invention, a lung mechanics model 20 is pre-defined and stored in a data storage 9 in a computer-usable form. The lung mechanics model 20 includes at least one relationship, particularly a model equation. The relationship or at least one relationship of the lung mechanics model 20 describes the variable P associated with the patient P's own respiratory activity. mus The relationship between multiple measurable signals, especially at least several of the following:

[0156] - Airway pressure (pressure in the airway) P aw ), obtained from the measurement value of sensor 3,

[0157] - Esophageal pressure (pressure in the esophagus) P es ), obtained from the measurement value of pneumatic sensor 6,

[0158] - Respiratory flow (flow) Similarly, the values ​​are obtained from the measurements of sensor 3.

[0159] - Lung volume (Vol), from airway flow Derived from or obtained from the measurement values ​​of sensor 4, and / or

[0160] - The amount of carbon dioxide (CO2) in the exhaled air.

[0161] In one construction scheme, the following two linear model equations are given in advance as lung mechanics model 20:

[0162] ,and

[0163] .

[0164] in this case,

[0165] - P mus (t) is the sought and time-varying measure of respiratory activity, which is related to the aerodynamic pressure generated by the respiratory muscles of patient P at time t.

[0166] - P aw (t) is the airway pressure measured during the patient's circulation, preferably as the pressure difference relative to ambient pressure, where the airway pressure P aw It is used as a measurable signal and is derived from the superposition of the patient P's own respiratory activity and the ventilation performed through the ventilation device 1 during artificial ventilation, while in other cases it is derived solely from the patient's own respiratory activity.

[0167] - R is the lung mechanics factor, which describes the airway resistance volumetric flow in patient P. breathing resistance,

[0168] - E is a pulmonary mechanical factor that targets the elasticity of the lungs in patient P.

[0169] - P0 is the lung mechanics constant, which is, for example, a pneumatic measure of the effect of incomplete exhalation (iPEEP) on patient P.

[0170] - Sig(t) is the electrorespiratory signal (EMG signal) described above or it is also a myocardial signal (MMG signal), which is determined by evaluating the measurements of the measuring electrodes 2.1.1 through 2.2.2 or the MMG sensor, and

[0171] - k eff It is under pneumatic pressure P mus The scaling factor between the electrical respiratory signal Sig or the mechanical respiratory signal measured by electrodes 2.1.1 through 2.2.2, where the factor k eff This describes so-called electromechanical efficiency, that is, how well electrical impulses are converted into muscle activity in the body of patient P.

[0172] Substitute (3) into (2) to supply the following model equations:

[0173] .

[0174] Model (4) is only approximately applicable. Model (4) has four model parameters, namely the lung mechanics factors R, E, and k. eff And the addend P0. The values ​​of these model parameters are usually not known beforehand, and vary from patient to patient, and even change over time for the same patient P. Therefore, the values ​​of the model parameters are approximately derived from multiple sets of signal values, which is further described below. In many cases, the addend P0 can be assumed to be constant over time. In a preferred construction scheme, the model equation is then differentiated once with respect to time beforehand, thereby eliminating the assumed constant addend P0. The differentiation provides the following model equation:

[0175] .

[0176] Only three model parameter values ​​still need to be estimated. Signal P aw , The values ​​of Vol and Sig can then be calculated using numerical integration.

[0177] In other construction schemes, model equations with other addends and other lung mechanics parameters are given in advance, such as the following model equations:

[0178] .

[0179] in this case,

[0180] - Q describes the resistance to airflow, which is generated by turbulence in the catheter between the ventilation device 1 and the patient P.

[0181] - S describes the stretchability of the lungs and / or pleural cavity according to changes in volume (Vol), and

[0182] - I describes the resistance to the acceleration of the breathing air, where the resistance I is negligible when the acceleration is sufficiently small.

[0183] In other construction schemes, the same model equations (3) and (4) with potentially different model parameter values ​​are used, one for inspiration (index ins) and one for exhalation (index exp), resulting in the use of the following two model equations:

[0184]

[0185] and

[0186]

[0187] These two model equations each have a set of model parameters for inhalation and exhalation, respectively.

[0188] It is also possible that the value of R is calculated only for the model parameters R and E respectively. ins and E ins Sum R exp and E exp The value R ins and E ins It is effective for inhalation, the value R exp and E exp It is effective for exhalation. For the remaining model parameters, calculate a unique value that is effective for both inhalation and exhalation.

[0189] To derive the estimated values ​​of the model parameters for model equation (7), only the set of signal values ​​generated from measurements taken during inhalation are used. Correspondingly, the values ​​of the model parameters for model equation (8) are estimated using only the set of signal values ​​generated during exhalation.

[0190] In another construction scheme, the following linear relationships are given in advance as model equations:

[0191] ,and

[0192] .

[0193] Pes(t) is the esophageal pressure, which is measured, for example, by a pneumatic sensor 6 in the esophagus Sp. Factor E CW Describe the elasticity of the chest wall due to patient P. Substitute (3) into (9) to supply the following model equation:

[0194] .

[0195] The model equations (2) up to (10) mentioned above are only applicable under ideal conditions. The lung mechanics model 20, defined by at least one model equation, only approximates reality, and the signal is superimposed with interference signals and affected by measurement errors. Therefore, the values ​​of the model parameters can only be approximately derived, and thus the derivation of the model parameter values ​​and therefore the derivation of the values ​​for respiratory activities necessarily involve estimation uncertainties.

[0196] The following descriptions involve model equations

[0197] .

[0198] The methods described thereafter can also be applied in a corresponding manner to other model equations belonging to lung mechanics model 20.

[0199] At each sampling time t i A set of signal values ​​is generated from the measured values, i.e., a signal value set.

[0200] .

[0201] Using lung mechanics model 20 and a set of signal values, the estimated values ​​are derived for (in this case, four) model parameters. .

[0202] In a preferred embodiment, in order to derive the respiratory activity value P mus,est (t i The regression method is applied to the pre-given model equation (4). Preferably, the sum of squared errors is minimized.

[0203] In one construction scheme, the model parameters in model equation (4) It is considered to be constant in time, and all the signal values ​​generated so far are used to derive the values ​​for the model parameters.

[0204] In another construction scheme, the values ​​of these model parameters may change over time. In one possible implementation, N sampling times are pre-defined. The estimated values ​​{R} for the model parameters are derived only using N sets of signal values ​​that are the latest in time. est (t i E est (t i ), k eff,est (t i ), P0 est (t i That is to say, until sampling time t i The last N sampling times (inclusive) form the evaluation time window. The number N is chosen to be large enough to perform sufficiently reliable regression analysis, while also being small enough that the model parameters {R, E, k} are... eff , P0} can be considered to be constant in time during the evaluation time window.

[0205] In one possible construction scheme, the last N signal value groups are weighted equally over time, i.e., using, for example, a weighting factor. To weight them. In another construction scheme, the older the signal value group, the higher the weighting factor of that signal value group. The smaller it is.

[0206] In other construction schemes, the specific time point within a single respiration is determined for the set of signal values. A weighting function is pre-defined, describing the weighting factors as a function of the measurement time point within a single respiration. The duration of a single respiration is preferably normalized. Figure 3 An exemplary weighting function is shown, where time t is denoted on the x-axis, and time-related weighting factors are... Record it on the y-axis. In this case...

[0207] - The interval from 0 to T indicates the normalized or typical time interval for a single breath.

[0208] - T_I indicates the start of inhalation.

[0209] - T_E indicates the start of the exspiration, and

[0210] - x1, x2 and x3 indicate three pre-given weighting factors, where, for example, x3=2, x2=1 and x1=0.5 are used.

[0211] In the third construction scheme, the signal value group is weighted according to the corresponding set value of the first ventilation device parameter and / or according to the frequency of the signal value, preferably as follows: the fewer signal value groups determined under the given set value and / or the rarer the signal value appears in the signal value group used for the current estimation, the higher the weighting factor for the signal value group under the current estimation.

[0212] Example: In the last N sampling times t1, ..., t N During this period, N1 signal value groups were determined under the standard setting value, N2 signal value groups were determined under the second setting value that deviates from the standard setting value, and N3 signal value groups were determined under the third setting value that deviates from both the standard setting value and the second setting value. Therefore, applying N = N1 + N2 + N3, the signal value groups determined under the standard setting value are weighted by a factor. The weighting factor is obtained from the signal value group determined under the second set value. Furthermore, the signal value group determined under the third set value obtains the weighting factor. .

[0213] Figure 4 An example of this weighting based on the frequency of setpoints and signal values ​​is shown. At time interval T_O, occlusion has been performed (no artificial ventilation, and the patient's own breathing has ceased), and the group of signal values ​​generated during the occlusion is weighted particularly highly.

[0214] Figure 4 The weighted AND signal P shown aw , The frequency of the signal values ​​(Vol, Sig) is related to the frequency of the signal values. Signal value groups with rare occurrences receive higher weights compared to signal value groups with frequent occurrences. These frequency-related weights of the signal values ​​are combined into the total weight of the signal value group. Figure 4 The time-varying process of this total weight is shown in the figure. .

[0215] These implementations can be combined. For example, each weighting factor All are calculated as products

[0216] ,

[0217] Among them, the first factor The second factor is related to the age of the signal value group. It relates to the relative time during a single breath, and the third factor Related to the number of signal value groups and / or the number of signal values ​​determined under this setting, see [link / reference]. Figure 4 Preferably, the weighting factors of the N signal value groups are normalized so that their sum is, for example, equal to 1.

[0218] In the deviated construction scheme, a regression method is applied after the first N sampling times, where at sampling time t i Previously, four model parameter values ​​R(t) were derived. i-1 ), E(t) i-1 ), k eff (t) i-1 ) and P0(t i-1 More precisely, based on t i-1 As the N final sampling times, these four model parameter values ​​were derived, along with the values ​​at sampling time t. i Then, using the previously obtained four model parameter values ​​{R} est (t i-1 ), E est (t i-1 ), k eff,est (t i-1 ), P0 est (t i-1 )} and the current signal value group {P aw (t i ), , Vol(t i ), Sig(t i In the case of )}, four updated model parameter values ​​{R} are derived. est (ti ), E est (t i ), k eff,est (t i ), P0 est (t i )}. The index est indicates that this is the estimated value. This recursive method saves computation time and can be combined with the use of a weighting factor.

[0219] In a construction scheme, at each sampling time t i , the estimated value for the pneumatic measure P mus is derived as follows, see model equation (3):

[0220] .

[0221] The respiratory activity value P mus (t i ) is derived with uncertainty, more precisely with uncertainty in two factors k eff,est (t i ) and Sig(t i ). To increase the reliability of the derivation, in a construction scheme of the present invention, a so-called exercise is carried out in case of low reliability. In the case of this exercise, in this embodiment, the first operating parameter BG of the ventilation device 1 is temporarily set from the standard set value EW_Std to at least one deviating set value and then set back to the standard set value EW_Std. For example, for a single breath of a patient P, the exercise is performed. In the case of the standard set value EW_Std, the ventilation device 1 is adjusted such that artificial ventilation is synchronized with the patient P's own respiratory activity as well as possible, for example such that it applies:

[0222] .

[0223] In the case of the deviating set value, in addition, according to the derived respiratory activity value P mus,est (t i ), but deviating from the normal operation, the ventilation device 1 is adjusted as follows, for example with at least one of the following deviations from the normal operation:

[0224] - The proportional adjustment according to (12) is performed with a smaller support degree x1 < x - or also with a larger support degree x2 > x.

[0225] - The volume flow of the breathing air flowing from the ventilation device 1 to the patient P is restricted to a maximum value.

[0226] - The ventilation device 1 uses pneumatic pressure P to provide artificial ventilation to patient P. art It is limited to the maximum value.

[0227] - Ventilation device 1 fills the lungs of patient P only up to a pre-defined volume limit. Patient P can achieve further lung volume increase solely through their own respiratory activities.

[0228] - The amplitude and / or frequency of the ventilation stroke performed by ventilation device 1 are reduced and / or limited.

[0229] - Ventilation device 1 is switched from pressure-regulated ventilation to volume-regulated ventilation, wherein the pressure-regulated ventilation is performed at a standard set value and the volume-regulated ventilation is performed at a deviation from the set value.

[0230] - Ventilation device 1 is switched from volume-regulated ventilation to pressure-regulated ventilation, wherein the volume-regulated ventilation is performed at the standard set value EW_Std, and the pressure-regulated ventilation is performed at a deviation from the set value.

[0231] The exercise may also include: ventilation equipment 1 being completely unadjusted, but controlled or deactivated, or, although adjusted, not based on the estimated respiratory activity value P. mus,est (t i Instead of being regulated, it is regulated as follows:

[0232] - Based on respiratory pressure P aw (t i ) and / or based on the volumetric flow measured by pneumatic sensor 3 And / or the esophageal pressure P measured by pneumatic sensor 6 es (t i Adjust the ventilation device 1. As described above, it is disadvantageous to continuously adjust the ventilation device 1 in this manner. However, in some cases it is meaningful to conduct an exercise in which the ventilation device 1 is adjusted in this manner for a short period of time, and then routinely adjusted as described above thereafter.

[0233] - The ventilation device 1 is controlled and is not adjusted according to the patient P's own respiratory activity. In control, the ventilation device 1 uses, for example, a pressure P to be generated during artificial ventilation. aw Or volume flow The predetermined expected change process.

[0234] - Ventilation device 1 is fully set to provide artificial ventilation (occlusion) for patient P, and patient P's own respiratory activity is interrupted, for example, by closing a valve on ventilation device 1, thereby preventing patient P from breathing. This occlusion is performed for a maximum of 5 seconds, preferably a maximum of 1 second, and there is no danger to patient P for such a short duration.

[0235] Preferably, this occlusion is performed at predetermined relative times during one breath of patient P, for example, at the end of inhalation (end-inspiratorische occlusion) or at the end of exhalation (end-exspiratorische occlusion). Even in the case of occlusion, model equations are applied in a construction scheme.

[0236] .

[0237] During the blockage, volumetric flow The negligible smallness makes it applicable In the case of end-expiratory occlusion, the remaining volume is included in the addend P0, making Vol(t) = 0 applicable. Therefore, in this case, the following applies:

[0238] .

[0239] Therefore, P can be measured well during the occupancy period. mus However, due to the present invention, occlusion is only required when it is necessary.

[0240] According to the present invention, the reliability metric ZM(t) is calculated. i The reliability metric ZM(t) i ) refers to respiratory activity values ​​(in this invention, i.e., P) mus,est (t i The derivation of )) is a reliable evaluation. For such calculations, for example, in time, it is a sequence of the last M+1 estimated model parameter values.

[0241] Used.

[0242] In one construction scheme, at each sampling time t i From the final model parameter values, the covariance matrix is ​​calculated, or more precisely, according to the calculation rules.

[0243]

[0244] Calculate the covariance matrix.

[0245] High cross-correlation between two different model parameters (e.g., for Cov(E,R) between E and R at sampling time t) i The large value means that, based on the existing set of signal values, the effects of the two model parameters E and R can only be distinguished by their differences.

[0246] In one construction scheme, the aerodynamic quantity P is calculated based on model equation (3) and the estimated respiratory signal Sig. mus At sampling time t i The value of P mus,est (t i That is, according to

[0247]

[0248] To calculate the value P mus,est (t i As for sampling time t i A measure of the uncertainty of the estimate is preferably calculated using empirical variance (empirical scattering).

[0249] .

[0250] Alternatively, other measures for estimation uncertainty can be used.

[0251] In the deviation scheme, a measure of the estimation uncertainty is calculated either after the end of a breath or after a pre-given time interval. For example, if there are M sampling times t... i+1 ... t i+M Within the time interval of one breath, the arithmetic mean, median, or other mean is calculated with respect to M empirical variances. It is calculated and used as a measure of the uncertainty of the estimate.

[0252] As described above, in a construction scheme, empirical variance

[0253]

[0254] Used as a measure of estimation uncertainty, in other construction schemes regarding empirical variance. Calculate the arithmetic mean or other average. In other construction schemes, bias and measurement error are combined into a time-varying error. If model equation (4) could accurately describe reality and there could be no measurement error, then err(t) could be 0 at any given time. In reality, this does not apply, and err(t) changes over time. Signal processing unit 5 calculates the reliability metric ZM(t). iAnd for this purpose, preferably, the model equation (16) given above for the time-variable error err(t) is used for N signal value sets for the last N sampling times in time. Preferably, the signal processing unit 5 applies statistical methods to calculate the reliability metric ZM(t). i In this embodiment, after ventilation has begun and as long as the signal processing unit 5 has not detected a predefined trigger criterion E1 being met, the ventilation device 1 is operated using the standard setting value EW_Std. In the case of the standard setting value EW_Std, for example, the pressure P generated by the ventilation device 1 during artificial ventilation... art (t i )equal The scaling factor (level of support) x remains constant. For example, with the standard setting EW_Std, ventilation device 1 always operates under pressure regulation.

[0255] Once trigger criterion E1 is met, signal processing unit 5 triggers a change process. The pre-defined trigger criterion E1 that triggers the change process is related to at least one calculated reliability metric, and is met, for example, when at least one of the following events is detected:

[0256] - Used to derive respiratory activity values, i.e., for the aerodynamic quantity P mus The value of P mus,est (t i The reliability metric ZM(t) is the last calculated time dimension of the time. i Within pre-defined reliability limits. This is agreed upon when deriving the respiratory activity value P. mus,est (t i The measure of the uncertainty of the estimate is above a pre-given uncertainty limit.

[0257] - The reliability measure ZM(t) calculated for the last M time intervals. i ZM(t) i-1 ...and get smaller and smaller, and approach the reliability limit from above.

[0258] - At least one of the last calculated reliability metrics ZM(t) i The reliability measure ZM(t) is significantly smaller than at least one, preferably multiple, previously calculated reliability measures. i-n ), ..., ZM(t) i-1 ).

[0259] According to the present invention, when the signal processing unit 5 has detected that the trigger criterion E1 has been met, especially when the last calculated reliability metric ZM(t) is obtained... iWhen the estimated uncertainty exceeds a pre-defined reliability limit, or when the estimated uncertainty exceeds a pre-defined uncertainty limit, signal processing unit 5 triggers a rehearsal, i.e., a change in process. The rehearsal includes the following steps: ventilation equipment 1 temporarily operates at a setting deviating from the standard setting value EW_Std. An example of the rehearsal has been described above.

[0260] The exercise was conducted with the following objective: to derive, with greater reliability, the P-value for respiratory activity during and / or after the exercise. mus The estimated value P mus,est (t i Regarding the aerodynamic measurement P mus The value of P mus,est (t i The following signal value set is derived when it deviates from a set value; and preferably additionally, the following signal value set is used to derive the aerodynamic quantity P. mus The value of P mus,est (t i The signal value set was generated before the exercise, i.e., under the standard setting value EW_Std.

[0261] The signal processing unit 5 terminates the exercise once it detects that a predefined termination criterion E3 has been met. For example, termination criterion E3 is met when at least one of the following events occurs:

[0262] - Since the start of the exercise, for example, since the start of the blockade, a predetermined time limit has passed, and it is not permissible to continue the exercise for an extended period of time.

[0263] - The last P calculated reliability metrics in time are above the pre-defined reliability limits, meaning that the cause for the exercise no longer exists.

[0264] The exercise did not result in an increase in reliability metrics. Therefore, it is preferable to perform another exercise instead of the one currently being performed.

[0265] The triggering and execution of the exercise are illustrated below.

[0266] In this example, a first estimation uncertainty limit, for example, 1 mbar, and a larger second estimation uncertainty limit, for example, 2 mbar, are pre-defined. Ventilation device 1 operates at the standard setting EW_Std as long as the estimation uncertainty measure is below the first estimation uncertainty limit. If the estimation uncertainty measure is between the two estimation uncertainty limits, a milder exercise is performed, in which ventilation device 1 also operates based on the estimated respiratory activity value P. mus,est (ti ) is adjusted. For example, lighter exercises include at least one of the following steps:

[0267] - During the exercise, the support level x is suddenly or also smoothly reduced to a smaller value x1 < x, that is, the ventilation device 1 operates according to to run.

[0268] - For a single breath, let the support pressure P art be below the maximum value, or the support pressure P art is otherwise reduced.

[0269] - The support pressure or the volume flow is restricted.

[0270] If the estimated uncertainty measure is even above the larger estimated uncertainty bound, a major exercise is performed, in which the estimated respiratory activity value P mus,est (t i ) is not used, but instead it is based on P aw (t i ) and / or perform, for example, occlusion or control or regulation. In one configuration, which major exercise is performed depends on the estimated uncertainty measure, for example on how strongly the estimated uncertainty measure is above the larger estimated uncertainty bound.

[0271] For example, artificial ventilation is fully set within a short period, and the patient P's own breathing is blocked. During the occlusion, the airway pressure P measured by the sensor 3 aw is only related to the patient P's own respiratory activity, for example, P mus = P aw is applicable. After the occlusion ends, for the current value of the pneumatic measure P mus it is again derived as just described above in the case of using the signal P aw , and Vol and the model equation (4), where for the derivation but additionally the following signal values are used: the signal values that have been measured during the occlusion.

[0272] In one configuration, the following exercise is related to the covariance matrix Cov(t i ) shown in formula (14) or to another measure of the correlation between different model parameters: in the case of an estimated uncertainty measure above the larger estimated uncertainty bound, the exercise is performed. For example, if the cross - correlation Cov(R, k est and k eff,est ) between two estimates R eff )(ti If the ventilation equipment is large, then the breathing airflow caused by the ventilation equipment 1 will be reduced during the exercise. In the model equations

[0273]

[0274] In this case, the reduction should be applied to the addend. It takes effect, but only for the addend k. eff *Sig(t) has a significantly smaller effect. If in the two estimates E est (t i ) and k eff,est (t i The cross-correlation between Cov(E,k) and E) eff )(t i Or in two estimates of R est (t i ) and E est (t i The cross-correlation between Cov(R,E)(t) i If the volume of the ventilation device is large, then the exercise aims to maintain a constant volume of Vol (i.e., lung filling level) over a predetermined period. In the model equation above, this exercise acts on the addend E*Vol(t), but not on the addend k. eff *Sig(t) has a significantly smaller effect.

[0275] Figures 5 to 11 A flowchart is shown, which illustrates the method according to the invention and an exemplary construction scheme of the signal processing unit according to the invention.

[0276] Figure 5 The first part of the flowchart illustrates how the estimated respiratory activity value P is derived. mus,est (t i And how to determine whether the triggering criterion E1 is met. Figure 6 The second part of the flowchart illustrates the normal operation of ventilation equipment 1, that is, its operation under the standard setting value EW_Std. Figure 7 The third part of the flowchart illustrates how to conduct easier exercises. Figure 8 It demonstrates how to conduct a major exercise. Figure 9 It shows how to generate sets of signal values ​​during an exercise and how to derive model parameter values ​​using these sets of signal values. Figure 10 It shows how respiratory activity values ​​are derived during an exercise. Figure 11 It shows how to check in multiple steps whether and how to continue artificial ventilation for patient P.

[0277] The flowchart is described below.

[0278] At the start of artificial ventilation, the first ventilation device parameter BG is set to a pre-defined standard setting value EW_Std. As long as this setting is maintained, ventilation device 1 operates in normal operation. After the exercise, ventilation device 1 is also adjusted for normal operation. In this normal operation, ventilation device 1 preferably operates according to the aerodynamic quantity P. mus The standard support factor x is adjusted. At each sampling time ti, the estimated value Pmus,est(ti) or P is derived as described above. mus,est m (t i This value is used as a respiratory activity value. The index m above indicates that the corresponding value has been calculated or derived during the exercise, which is further described below.

[0279] In step S1, signal processing unit 5 receives measurement values ​​from sensors 2.1.1 through 2.2.2 and 3, and optionally receives measurement values ​​from optical sensor 4 and / or pneumatic sensor 6. Signal processing unit 5 preprocesses these measurement values. This preprocessing is performed for each sampling time t. i Separately supply signal value groups .

[0280] In step S2, the signal processing unit 5 processes data from the last N+1 sampling times t. i-N Until t i The estimated model parameter values ​​are derived from the set of signal values ​​{R}. est (t i ), E est (t i ), k eff,est (t i ), P0 est (t i To this end, the signal processing unit 5 uses the lung mechanics model 20, for example, a pre-given model equation.

[0281]

[0282] and

[0283] .

[0284] In step S3, for example, according to the model equation

[0285] ,

[0286] Signal processing unit 5 derives the estimated value P of the respiratory activity of patient P. mus,est (t i), and for this purpose, at least one estimated model parameter value is used.

[0287] In step S4, the signal processing unit 5 calculates the respiratory activity value P used to derive it. mus,est (t i The reliability metric ZM(t) i For example, signal processing unit 5 calculates a measure for the estimation uncertainty. The calculated reliability measure ZM(t) i Or the measurement of uncertainty may also be related to the previous sampling time t. i-1 t i-2 ...related to the calculated values.

[0288] The signal processing unit 5 automatically makes a decision. Does it meet the pre-defined trigger criterion E1? When used to derive the respiratory activity value P mus,est (t i When the reliability of ) is low, especially when the recently calculated reliability metric ZM(t) is low. i If the reliability threshold is lowered below a pre-defined limit or becomes significantly smaller, trigger criterion E1 is met. Furthermore, when trigger criterion E1 is met, signal processing unit 5 determines whether to execute a lighter exercise (branch "leg") or a more significant exercise (branch "grav").

[0289] If the trigger criterion E1 is not met (branch "No"), then the reliability metric ZM(t) i It is large enough. Maintain normal operation. Figure 6 The steps performed during normal operation are shown. In step S5, the signal processing unit 5 calculates the respiration activity value P based on the derived respiration activity value P. mus,est (t i ) Execute the adjustments made by the superior. For example, according to the rules:

[0290] ,

[0291] Signal processing unit 5 calculates the expected pressure P to be generated when ventilation device 1 provides artificial ventilation to patient P. art (t i ).

[0292] In step S6, the signal processing unit 5 performs the next-level adjustment and adjusts according to the expected pressure value P. art (t i ) Calculate the required or each required control intervention SE(t) i The control intervention SE(t) is implemented with the following objectives. i Ventilation equipment 1 actually reaches this pressure P. art (ti )。For the next sampling instant t i+1 = t i + Δ, the steps described so far are executed again.

[0293] Figure 7 Shows the steps performed in the lighter exercise (decision branch "leg").

[0294] In step S7, the signal processing unit 5 specifies a setpoint EW_leg(t i ) for the first ventilation device parameter BG that deviates from the standard setpoint EW_Std. i ) This deviating setpoint EW_leg(t i ) may be related to the calculated reliability measure ZM(t

[0295] In step S8, the signal processing unit 5 performs the lighter exercise. In this case, the first ventilation device parameter BG is set to the deviating setpoint EW_leg(t i ), and the ventilation device 1 operates accordingly.

[0296] In the lighter exercise, the respiratory activity value P i inferred at the sampling instant t mus,est (t i ) is also used to regulate the ventilation device 1. However, the ventilation device 1 (contrary to normal operation) operates according to the deviating setpoint EW_leg(t i ). For example, the support level is reduced to x1 < x, or the pressure P art or the volume flow is restricted.

[0297] In step S9, based on the derived respiratory activity value P mus,est (t i ), and optionally additionally based on the deviating setpoint EW_leg(t i ), the signal processing unit 5 performs the higher-level regulation. The signal processing unit 5 calculates the pressure setpoint P art m (t i ) again. The index m indicates that this occurs during the exercise.

[0298] In step S6, the signal processing unit 5 calculates the required control intervention SE m (t i ) in the lighter exercise, more precisely based on the pressure setpoint P art m (t i)Calculate the required control intervention SE m (t i Further details are provided below for the next sampling time t. i+1 The continuation.

[0299] Figure 8 The steps performed during a major exercise are shown (in...) Figure 5 In the judgment The branch "grav"). In the case of major exercises (as opposed to lighter exercises), the derived respiratory activity value P, which carries significant uncertainty, is not used. mus,est (t i ).

[0300] In step S10, the signal processing unit 5 calculates the set value EW_grav(t) for the deviation from the critical exercise. i The setting is EW_grav(t) i The value EW_leg(t) is calculated in step S7 for the lighter exercise. i ( ) A more pronounced deviation from the standard setting value EW_Std, or otherwise causing a significant deviation in the operation of ventilation device 1.

[0301] In step S11, signal processing unit 5 triggers the following steps: ventilation device 1 performs a major exercise, wherein the first ventilation device parameter BG is set to the set value EW_grav(t) i ).

[0302] In step S12, the signal processing unit 5 determines the airway pressure P measured during the exercise. aw m (t i ) and / or based on volumetric flow To execute the superior's regulation, that is, not to use the respiratory activity value P derived in step S3. mus,est (t i This adjustment may also affect the ventilation device 1 or trigger a blockage. The adjustment may also deviate from the setpoint EW_grav(t). i (This is related to step S12, which involves resupplying the expected pressure P.) art m (t i ).

[0303] In step S6, the signal processing unit 5 calculates the required control intervention SE. m (t i See also Figure 6 .

[0304] In both lighter and heavier exercises, the signal processing unit 5 generates at least one set of signal values ​​based on the measurements taken during the exercise, and then derives model parameter values ​​and respiratory activity values.

[0305] Figure 9 The steps are shown for execution in both lighter and heavier exercises. In step S13, signal processing unit 5 generates a set of signal values. In step S14, the signal processing unit 5 calculates the estimated set of model parameter values. And for this, the signal value group of step S13 is used and optionally an older signal value group is used.

[0306] If blocking is not executed during the exercise (decision) If the branch is "No"), then the following steps are performed: Using the lung mechanics model 20 and at least one model parameter value, the signal processing unit 5 derives the respiratory activity value P. mus,est m (t i )( Figure 10 Step S3). The signal processing unit 5 then calculates the value P for the derived respiratory activity. mus,est m (t i ZM, a measure of reliability m (t i )( Figure 10 Step S4).

[0307] If blockade is executed during the exercise (determination) If the branch "is", then artificial ventilation for patient P is set for a short period of time, and patient P's own respiratory activity is inhibited, and respiratory activity P can be directly measured. mus In step S16, the signal processing unit 5 receives the measurement value from the sensor 3 and generates a signal value. Therefore, when the occlusion does not occur at the end of a breath and the volume Vol cannot be ignored, signal processing unit 5 uses the estimated value E for factor E derived prior to the occlusion. est (t i And the estimate of the addend P0 est (t i Signal processing unit 5 processes these signal values. and optionally from model parameter values ​​E est (t i ) and P0 est (t i The respiratory activity value P was derived from this. mus m (ti Instead of using the respiratory signal Sig, the signal processing unit 5 calculates the reliability metric ZM in step S17. m (t i This is used to derive the estimated respiratory activity value P. mus,est m (t i Furthermore, the measured respiratory activity value P was used for this purpose. mus m (t i ).

[0308] Figure 11 The three decisions are shown in sequence. , and In determining The decision is made to determine whether to continue or terminate treatment for patient P. In the middle, signal processing unit 5 determines whether to end the current exercise and whether to return to normal operation. The reason for ending the exercise is that the reliability metric ZM calculated during the exercise... m (t i The value is large enough. Other reasons include, for example, the pre-defined time interval for the blockade has elapsed. If the exercise is to be terminated ( If the branch "is", then the signal processing unit 5 will reset the ventilation device parameter BG to the standard setting value EW_Std again in step S18. Otherwise ( The branch "No"), the signal processing unit 5 determines The decision was made to continue with a lighter-scale exercise. The branch "leg" will continue with major exercises. The branch "grav"). Preferably, for the next ventilation step in routine operation ( Figure 6 In step S15), the signal processing unit 5 uses the derived respiratory activity value P. mus,est m (t i Or the measured respiratory activity value P mus,est (t i The signal processing unit has derived the respiratory activity value P during the exercise. mus,est m (t i Or the respiratory activity value P mus,est (t i ).

[0309] List of reference numerals

[0310] 1 The ventilation equipment, used to manually ventilate patient P, includes a signal processing unit 5 and a pneumatic sensor 3. 2.1.1、2.1.2、2.2.1、2.2.2 Measuring electrodes on the skin of patient P, together with a grounding electrode, supply measurement values, from which a respiratory signal Sig is generated. 3 <![CDATA[Pneumatic sensor in front of the mouth of patient P, measuring respiratory tract pressure P aw and volumetric flow > 4 An optional optical sensor with an image recording device and an image processing unit measures the geometry of the patient P's body and derives the current lung filling level Vol from said geometry. 5 The signal processing unit, performing the steps of the method according to the present invention, has read access to the data memory 9. 6 <![CDATA[Optional pneumatic sensor in the esophagus Sp, measuring the pneumatic pressure P in the esophagus Sp es > 9 The data memory stores the lung mechanics model 20 with the model equations used, and has read access to the signal processing unit 5. 20 A pre-defined lung mechanics model, including at least one model equation, is stored in data storage 9 in a computer-usable format. <![CDATA[For the weighting factor of the signal value group that has been determined at sampling time t i > BG <![CDATA[The first ventilation device parameter is set to the standard set value EW_Std during normal operation and to a deviated set value EW_leg(t i ) or EW_grav(t i )]]> Δ <![CDATA[At the interval between two sampling instants t i and t i+1 > E Model parameters in the form of lung mechanical factors: elasticity of the lungs of patient P <![CDATA[E est (t i )]]> <![CDATA[The estimated value of the model parameter E at sampling time t i is derived in the case of the standard set value EW_Std]]> <![CDATA[E est m (t i )]]> <![CDATA[The estimated value of the model parameter E at sampling time t i derived during the exercise]]> E1 A pre-defined trigger criterion, upon detection of the trigger criterion, triggers an exercise. Determination: Does it meet trigger criterion E1? Determination: Should patient P continue treatment? E3 Determination: Does it meet the termination criteria used to end the current exercise? E4 Determine whether to conduct a lighter exercise or a more significant exercise. <![CDATA[EW_grav(t i )]]> <![CDATA[Set for the deviation of the first ventilation equipment parameter BG from the set value during a major exercise at sampling time t i Set]]> <![CDATA[EW_leg(t i )]]> <![CDATA[Setting of the deviation of the first ventilation device parameter BG from the set value during a lighter exercise, at the sampling time t i Set]]> EW_Std The standard setting values ​​for parameter BG of the first ventilation device are used in the routine operation of ventilation device 1. <![CDATA[k eff ]]> <![CDATA[Model parameters in the form of factors of neuromuscular efficiency, i.e., how well the respiratory muscles of patient P convert electrical pulses into respiratory activity that results in pneumatic pressure P mus > <!-- 27 --> <![CDATA[k eff,est (t i )]]> <![CDATA[Model parameter k eff At sampling time t i The estimated value of, derived under the standard set value EW_Std]]> <![CDATA[k eff,est m (t i )]]> <![CDATA[Model parameter k eff at sampling instant t i estimated value of, derived during the exercise]]> Determination: Should blocking be performed? P <![CDATA[A patient with an esophagus Sp and diaphragm Zw is artificially ventilated, at least temporarily, by a ventilation device 1 due to a pressure P generated by the patient's own respiratory activity.]]> mus > <![CDATA[P art ]]> Support pressure, the pressure generated through artificial ventilation. <![CDATA[P art (t i )]]> <![CDATA[The current value of P during normal operation, calculated according to P]] art at mus,est (t i )]]> <![CDATA[P art m (t i )]]> <![CDATA[The current value of P during the exercise, calculated according to P art art or P mus,est mus,est (t i i ) or P mus mus m (t i i )]]> <![CDATA[P aw ]]> <![CDATA[Respiratory pressure, generated by the superposition of the patient P's own breathing activity and artificial ventilation P through the ventilation device 1, is measured by the sensor 3]]> art > <![CDATA[P aw (t i )]]> <![CDATA[The signal value of the respiratory tract pressure P aw is generated at the sampling moment t i during normal operation]]> <![CDATA[P aw m (t i )]]> <![CDATA[The signal value of the respiratory tract pressure P aw at the sampling moment t during the exercise i generated]]> <![CDATA[P es ]]> The pressure in the esophagus Sp of patient P was measured using a pneumatic sensor 6. <![CDATA[P mus ]]> aerodynamic measurements of patient P's own respiratory activity <![CDATA[P mus (t i )]]> <![CDATA[At sampling time t i actual respiration activity value]]> <![CDATA[P mus m (t i )]]> <![CDATA[Respiratory activity values derived by measurement during the exercise at sampling time t i > <![CDATA[P mus,est (t i )]]> <![CDATA[For the derived estimated value of the pneumatic measure P mus derived in the case of the standard setpoint EW_Std, which serves as the respiratory activity value]]> <![CDATA[P mus,est m (t i )]]> <![CDATA[For the derived estimated value of the pneumatic measure P mus derived during the exercise, which serves as a respiratory activity value]]> P0 Model parameters in the form of lung mechanics addends: residual pressure after incomplete exhalation by patient P. <![CDATA[P0 est (t i )]]> <![CDATA[The estimated value of the model parameter P0 at the sampling time t i is derived in the case of the standard set value EW_Std]]> <![CDATA[P0 est m (t i )]]> <![CDATA[The estimated value of the model parameter P0 at the sampling time t i derived during the exercise]]> R Model parameters in the form of lung mechanical factors: airway resistance volumetric flow in patient P breathing resistance <![CDATA[R est (t i )]]> <![CDATA[The estimated value of the model parameter R at sampling time t i derived in normal operation]]> <![CDATA[R est m (t i )]]> <![CDATA[The estimated value of the model parameter R at sampling time t i derived during the exercise]]> S1 <![CDATA[Steps: Receive and preprocess measurement values to generate a signal value group {P aw (t i ), , Vol(t i ), Sig(t i )}]]> S2 Steps: Calculate the set of estimated model parameter values S3 <![CDATA[Step: Derive the estimated respiratory activity value P mus,est (t i )]]> S4 <![CDATA[Step: Calculate the reliability measure ZM(t mus,est (t i ) for deriving P i )]]> S5 <![CDATA[Steps: Perform the adjustment of the superior in normal operation, and calculate the expected pressure value P mus,est (t i ) according to P art (t i )]]> S6 <![CDATA[Step: Perform the regulation of the subordinate level, and calculate the control intervention SE(t) or SE(t) according to the pressure expectation value P(t) or P(t).]]> art (t i ) or P art m (t i ) Calculate the control intervention SE(t i ) or SE m (t i )]]> S7 <![CDATA[Step: In a lighter exercise, a set value EW_leg(t i )]]> S8 <![CDATA[Steps: Perform a lighter exercise, set the first ventilation equipment parameter BG to a deviated set value EW_leg(t i )]]> S9 <![CDATA[Steps: Execute the regulation of the superior in a relatively light exercise, and calculate the expected pressure value P mus,est (t i ) and the set value EW_leg(t i ) to calculate the expected pressure value P art m (t i )]]> S10 <![CDATA[Steps: In major exercises, set the set value EW_grav(t i )]]> S11 <![CDATA[Steps: Conduct a major drill and set the first ventilation equipment parameter BG to a deviated set value EW_grav(t i )]]> S12 <![CDATA[Steps: Execute the adjustment of the superior in a major exercise, and calculate the expected pressure value P according to the measured signal value Calculate the expected pressure value P art m (t i )]]> S13 <![CDATA[Steps: Generate a signal value group {P aw m (t i ), , Vol m (t i ), Sig m (t i )} <!-- 28 -->]]> S14 <![CDATA[Steps: Derive the model parameter values {R est m (t i ), E est m (t i ), k eff,est m (t i ), P0 est m (t i )} in the exercise. For this, use the signal value group and N + 1 previous signal value groups ,..., > S15 <![CDATA[Steps: Execute the adjustment of the superior during normal operation, according to P mus m (t i ) or P mus,est (t i ) calculate the expected pressure value P art (t i )]]> S16 <![CDATA[Steps: During occlusion, derive the respiratory activity value P from the signal value mus m (t i )(direct measurement)]]> S17 <![CDATA[Steps: Calculate the reliability measure ZM mus,est m (t i ) for deriving P during the exercise m (t i ), and for this purpose use P mus m (t i )]]> S18 Steps: End the exercise and set the first ventilation device parameter BG to the standard setting value EW_Std. <![CDATA[SE(t i )]]> <![CDATA[Control intervention during normal operation in the case of EW_Std, calculated according to P art (t i ) in the lower-level regulation]]> <![CDATA[SE m (t i )]]> <![CDATA[Control intervention during the exercise, calculated according to P in the regulation of the subordinate]]> art m (t i ) calculation]]> Sig The electrical respiratory signal (EMG signal) for the respiratory activity of patient P is generated from the measurements taken by measuring electrodes 2.1.1 to 2.2.2. <![CDATA[Sig(t i )]]> <![CDATA[The signal value of signal Sig at sampling time t i is generated during normal operation]]> <![CDATA[Sig m (t i )]]> <![CDATA[The signal value of signal Sig at sampling time t i is generated during the exercise]]> Sp Patient P's esophagus <![CDATA[t i ]]> Sampling time T_E The moment when patient P begins to exhale (exhale) T_I The moment when patient P begins to inhale (inspire). T_O Execution of blocking time period Vol The volume of patient P's lungs (current filling level) is a volumetric flow. The integral over time is measured by optical sensor 4 in one configuration. The airflow entering or leaving patient P's lungs per unit time is the derivative of volume Vol with respect to time, for example, as measured by sensor 3. x <![CDATA[The degree of support, which is a proportionality factor for artificial ventilation in the case of proportional regulation during normal operation, that is, the ventilation device 1 operates according to P art (t i ) = x * P mus,est (t i )]]> x1 A smaller level of support was used in lighter exercises. Zw Patient P's diaphragm

Claims

1. A computer-implemented method for approximating the determination of a measure associated with a patient's (P) own respiratory activity. The method described therein automatically uses - Ventilation equipment (1), and - Signal processing unit (5) for data processing Executed under the following circumstances The ventilation device (1) mentioned above - Provide artificial ventilation to the patient (P) at least temporarily, and - Operates according to the variable parameters of the first ventilation device. The parameters of the first ventilation device affect the control of the gas flow to and / or from the patient (P) and / or the pressure of the gas. Wherein a pre-given lung mechanics model (20) is provided, the lung mechanics model (20) describes at least one in - Respiratory activity measurement and - At least one measurable signal The relationship between them The method includes the following steps: During the period when the parameters of the first ventilation device are set to the set value, the signal processing unit (5) performs at least one ventilation process. Wherein, under the set value, the ventilation process or at least one ventilation process includes the following steps: the signal processing unit (5) - For at least one measurable signal appearing in the lung mechanics model (20), at least one value is received, said at least one value being measured during the period when the first ventilation device parameters are set to said set value. - When using the measured values ​​obtained under the set values, at least one group of signal values ​​is generated, wherein the at least one group of signal values ​​has a signal value for each measurable signal of the lung mechanics model (20). - Derive at least one respiratory activity value for the respiratory activity measure. - For the purpose of derivation, the lung mechanics model (20) is used, and a set of signal values ​​generated under the given conditions (or at least one) is used. - The ventilation device (1) is operated with the following objective: the ventilation device (1) supports the patient's (P) spontaneous breathing activities, wherein the first ventilation device parameters are set to the set values. The signal processing unit (5) mentioned above - Perform at least one first ventilation procedure, in which the parameters of the first ventilation device are set to a first preset value. - During the first ventilation process, the first respiratory activity value was derived, and - Calculate a measure of reliability that is consistent with the first respiratory activity value and the corresponding actual respiratory activity measure of the patient (P). The method includes the following additional steps: The signal processing unit (5) checks whether the pre-given triggering criteria are met. The triggering criterion is related to a calculated reliability metric used to derive the first respiratory activity value, and Wherein, if the calculated reliability measure used to derive the first respiratory activity value is below a pre-given first reliability limit, then at least the triggering criterion is satisfied, and In response to the detection of the triggering criteria being met, the signal processing unit (5) - Triggering a change process, during which the parameters of the first ventilation device are set to a second set value that deviates from the first set value, and - Perform at least one other ventilation procedure in which the parameters of the first ventilation device are set to the second set value instead of the first set value.

2. The method according to claim 1, Its features are, During the other ventilation processes performed under the second set value, the signal processing unit (5) derives or determines the second respiratory activity value. The signal processing unit (5) mentioned above - To derive or determine the second respiratory activity value, at least one set of second signal values ​​is used, which is generated using measurements taken at the second set value. - Additionally, the lung mechanics model (20) is used for derivation.

3. The method according to claim 2, Its features are, In the other ventilation process performed under the second set value, in order to derive the second respiratory activity value, in addition to the signal value group generated when using the measurement value measured under the second set value, the signal processing unit (5) uses the signal value group generated when using the measurement value measured before the change process, including the use of the signal value group generated when the measurement value of the signal value group was generated under the first set value.

4. The method according to any one of claims 1-3, Its features are, As parameters of the first ventilation device, measures of gas delivery to the patient (P) are used, including the use of a pre-defined desired volumetric flow rate or a pre-defined desired respiratory air pressure, and The signal processing unit (5) - During or in the process of the change, the following steps are triggered: reducing or increasing the gas delivery to the patient (P), and - This then triggers other alteration processes to increase or decrease the gas delivery to the patient (P) again.

5. The method according to claim 1, Its features are, The signal processing unit (5) operates the ventilation device (1) during the first ventilation process by the following steps: the signal processing unit (5) operates the ventilation device (1) as long as the reliability measure calculated for deriving the respiratory activity value is above the first reliability limit. - During the first ventilation process, the parameters of the first ventilation device can be set to the first set value, and - The ventilation device (1) is operated with the goal of supporting ventilation for the patient (P) based on at least one respiratory activity value derived under the first set value.

6. The method according to claim 1, Its features are, The signal processing unit (5) controls the ventilation device (1) by means of the following steps: at least when the calculated reliability measure is below the first reliability limit or below a smaller second reliability limit, then after the change process, the signal processing unit (5) controls the ventilation device (1) with the goal of supporting ventilation to the patient (P) based on signals of the flow rate and / or the pressure in the gas circulation between the ventilation device (1) and the patient (P).

7. The method according to claim 5 or 6, Its features are, The signal processing unit (5) performs regulation of the ventilation device (1) with the following regulation objectives: The gas flow to and / or from the patient (P) caused by the ventilation device (1) is synchronized with the patient's (P) own respiratory activity. Or it may lead to proportional control. The signal processing unit (5) repeatedly performs the ventilation process during the adjustment to achieve the adjustment target, and The signal processing unit (5) performs the following steps during at least one ventilation process: - Calculate the corresponding reliability metric. - When the calculated reliability metric falls below a pre-defined reliability limit, a process for changing the parameters of the first ventilation device is triggered, and - Then, the modified settings are used to perform other ventilation procedures.

8. The method according to any one of claims 1-3, Its features are, When the calculated reliability metric is above the reliability limit, then the signal processing unit (5) executes at least one additional ventilation process, in which... - The parameters of the first ventilation device remain set at the first set value. - The signal processing unit (5) generates other signal value groups, and - In the case of using the other signal value set and the signal value set previously generated under the first set value, the signal processing unit (5) derives other respiratory activity values ​​and calculates a measure of the reliability of the derivation of the other respiratory activity values.

9. The method according to any one of claims 1-3 above, Its features are, The first ventilation process or the at least one other ventilation process includes the following steps: The signal processing unit (5) - For each measurable signal appearing in the lung mechanics model (20), multiple measurement values ​​are received, wherein the multiple measurement values ​​are measured under the same set value. - When using the received measurement values, multiple signal value groups are generated, each group having a signal value for each measurable signal. - In the case of using multiple sets of signal values ​​generated so far under the stated set value, derive the one or at least one respiratory activity value, and - When applying statistical methods, the reliability measure for the respiratory activity value is calculated based on the set of signal values ​​used for derivation.

10. The method according to any one of claims 1-3, Its features are, When the triggering criterion is met, the signal processing unit (5) triggers the change process, such that the second set value is related to the calculated reliability measure. The more the reliability measure is below the first reliability limit, the more strongly the second set value deviates from the first set value. The triggering criterion is related to the reliability measure used to derive the first respiratory activity value.

11. The method according to any one of claims 1-3, Its features are, The pre-given lung mechanics model (20) has variable first model parameters. In the step of deriving respiratory activity values ​​under set values, the signal processing unit (5) - Using the set or at least one set of signal values ​​generated under the given set values, derive the values ​​of the first model parameters for the pre-given lung mechanics model (20), and - Using the values ​​of the first model parameters and the lung mechanics model (20), the respiratory activity values ​​are derived, and In the step of calculating the reliability measure for respiratory activity values, the signal processing unit (5) calculates the reliability measure for deriving the values ​​of the first model parameters.

12. The method according to claim 11, Its features are, The lung mechanics model (20) has a first model parameter and a second model parameter. The signal processing unit (5) mentioned above - Calculate a reliability measure used to derive the values ​​of the first model parameters as a first reliability measure, and calculate a reliability measure used to derive the values ​​of the second model parameters as a second reliability measure. - When the first reliability metric falls below the first reliability limit, then a first change process is triggered, and - When the second reliability metric falls below the first reliability limit, then the second change process is triggered. The first modification process involves the parameters of the first ventilation device, and the second modification process involves the parameters of another ventilation device, and / or The first change process results in a set value that is different from the second change process.

13. The method according to any one of claims 1-3, Its features are, During at least one ventilation process, when deriving the said or at least one respiratory activity value, the signal processing unit (5) uses the lung mechanics model (20). - Applied to at least one first signal value group, and - Applied to at least one group of second signal values, Wherein, under the first set value of the first ventilation device parameters, the measured value of either the first signal value group or the measured value of each first signal value group is measured. In cases where the measured value of the second set value or another set value deviates from the first set value, the measured value of the second signal value group is generated. The signal processing unit (5) calculates a weighting factor for each set of signal values ​​used for derivation, and The signal processing unit (5) uses the weighting factor when deriving the respiratory activity value.

14. The method according to claim 13, Its features are, The signal processing unit (5) calculates the weighting factor, such that - The fewer signal value sets generated at a given set value, the larger the weighting factor for the signal value sets generated at that set value, and / or - The sum of the weighting factors for those signal value groups measured under set values ​​is equal to a pre-given share value, and / or - Each weighting factor is related to which stage of the sequence of changes in at least one breath of the patient (P) produced the set of signal values, or to whether the set of signal values ​​was measured when the patient (P) inhaled or exhaled.

15. The method according to claim 13, Its features are, The signal processing unit (5) during at least one ventilation process - Individual values ​​of respiratory activity are calculated for at least two different setpoints used so far, and in this case, at least one set of signal values ​​is generated using measurements taken with respect to the setpoints used during the ventilation process. - Combine the individual respiratory activity values ​​into a single respiratory activity value using the weighting factor.

16. The method according to any one of claims 1-3, Its features are, The respiratory activity measurement can be measured under the second set value or under the second set value. The signal processing unit (5) mentioned above - Determine the second respiratory activity value based on the second set value, and - In this case, by performing at least one measurement under the second set value, a signal value for the respiratory activity measure is generated, and In the step of calculating a measure of the reliability of the derived first respiratory activity value, the signal processing unit (5) compares the derived first respiratory activity value with the determined second respiratory activity value.

17. A signal processing unit (5) for approximately determining a measure associated with a patient's (P) spontaneous respiratory activity, wherein the signal processing unit (5) is at least temporarily connected to or is capable of being connected to a ventilation device (1). The ventilation device (1) described herein is constructed as follows: - Provide artificial ventilation to the patient (P) at least temporarily, and - Operates according to variable first ventilation device parameters, wherein the first ventilation device parameters affect the control of the gas flow to and / or from the patient (P) and / or the pressure of the gas. The signal processing unit (5) has at least temporary access to the data memory (9), in which a lung mechanics model (20) is stored, the lung mechanics model (20) describing at least one lung... - Respiratory activity measurement and - At least one measurable signal The relationship between them, and The signal processing unit (5) is configured to perform at least one ventilation process during the period when the parameters of the first ventilation device are set to a determined set value. The signal processing unit (5) is configured to, during the ventilation process or in at least one ventilation process, - For at least one measurable signal appearing in the lung mechanics model (20), at least one value is received, said at least one value being measured during the period when the first ventilation device parameters are set to the determined set value. - When using the measured values ​​obtained under the set values, at least one group of signal values ​​is generated, wherein the at least one group of signal values ​​has one signal value for each measurable signal of the lung mechanics model (20). - Derive at least one respiratory activity value for the respiratory activity measure, using the lung mechanics model (20) and using the set of signal values ​​generated at the set value, and - The ventilation device (1) is operated with the following objective: the ventilation device (1) supports the respiratory activities of the patient (P), wherein the first ventilation device parameters are set to the determined set values. The signal processing unit (5) is configured as follows: - Perform at least one first ventilation procedure, in which the parameters of the first ventilation device are set to a first preset value. - The first respiratory activity value was derived during the first ventilation process, and - Calculate a measure of reliability that the derived first respiratory activity value corresponds to the corresponding actual respiratory activity measure of the patient (P), and The signal processing unit (5) is configured as follows: Check whether a pre-defined triggering criterion is met, the pre-defined triggering criterion being related to a calculated reliability measure used to derive the first respiratory activity value, and The triggering criterion occurs at least when the calculated reliability measure used to derive the first respiratory activity value is below a pre-defined first reliability limit. The signal processing unit (5) is configured to, in response to the detection of a condition that meets the triggering criteria, - Triggering a change process, during which the parameters of the first ventilation device are set to a second set value that deviates from the first set value, and - Perform at least one other ventilation procedure in which the parameters of the first ventilation device are set to the second set value instead of the first set value.

18. A computer program product having a computer program that can be implemented on a signal processing unit (5), and when implemented on the signal processing unit (5) and connected to a ventilation device (1), the computer program causes the signal processing unit (5) to perform the method according to any one of claims 1 to 16, wherein the ventilation device (1) - Provide artificial ventilation to the patient (P) at least temporarily, and - Operates according to variable first ventilation device parameters, which affect the control of the gas flow to and / or from the patient (P) and / or the pressure of the gas.

Images