LU102031B1 - Method and device for automatically determining the target frequency of a ventilator - Google Patents

Abstract

The invention relates to a method, a signal processing unit and a ventilator which automatically define a target value for the frequency at which the ventilator performs ventilation strokes and thereby artificially ventilates a patient. An alveolar or proximal minute volume is specified. A lung time constant for the patient's lungs is determined. The volume of dead space in a fluid connection between the lung and the processing device is determined. A mandatory target rate (fset.mand) for the patient's mandatory ventilation is calculated. An ideal rate (fspon) at which the patient can achieve minute ventilation through spontaneous breathing is calculated. The target ventilation rate is calculated as the weighted average of the mandatory target rate (fset.mand) and the ideal rate (fspon). The averaging depends on the determined intensity of the patient's spontaneous breathing.

Description

The invention relates to a method and a device for automatically setting a target value for the frequency at which a ventilator performs ventilation strokes and thereby artificially ventilates a patient.

Such a method and such a device are described in EP 3332827 A1.

The frequency at which a patient should be artificially ventilated is calculated automatically.

The calculation depends on a target minute volume and a given functional dead space, as well as a given lung model, a stored lung time constant, e.g.

R*C, and a predetermined time duration ratio, e.g.

Inspiration | to expiration E.

The frequency is calculated so that a parameter dependent on the frequency is minimized.

In one embodiment, this parameter is the sum of W'c and Wr, i.e. the power W'c required to stretch the lungs and the power Wr required to overcome the pneumatic resistance in the airway.

Optimization is carried out in order to define the target frequency, in the exemplary embodiment an iterative optimization.

US 20090007915 A1 describes a device and a method for automatically calculating values for parameters of artificial respiration.

A desired rate RR® of ventilation is established and a rate RRspon of spontaneous breathing is determined.

The two frequencies are compared with each other.

Depending on the deviation, a target value for ventilation is set.

In particular, the target value can be the total alveolar ventilation or the absolute or relative minute volume.

If the rate of spontaneous breathing deviates significantly from the target rate, the target value for ventilation is increased. 20201001LU 02.09.2020

-2/45- The object of the invention is to provide a method and a device which automatically determine the target ventilation frequency of a ventilator, this determination being valid for a relatively wide range of applications and whereby it should not be necessary to use the ventilator switch between different modes. The object is achieved by a method having the features of claim 1, by a signal processing unit having the features of claim 12 and by a ventilator having the features of claim 13. Advantageous configurations are specified in the dependent claims. Advantageous configurations of the method according to the invention are also configurations of the signal processing unit according to the invention and the ventilator according to the invention and vice versa.

A fluid connection is established or can be established at least temporarily between the lungs of a patient who is to be artificially ventilated and the ventilator according to the invention. The ventilator is at least temporarily connected or connectable to at least one patient sensor. The or each patient sensor is able to measure at least one parameter of the patient to be artificially ventilated or to be ventilated.

The signal processing unit according to the invention and the ventilator according to the invention include a data memory or at least temporarily have read access to a data memory. A measure of a desired volume flow into the patient's lungs is stored in this data memory. Such a desired volume flow is specified for the method according to the invention.

A volumetric flow is the volume of fluid flowing into and/or out of a space in a given unit of time. The desired volume flow is in particular a required alveolar or proximal minute volume.

The method according to the invention comprises the following steps, and the signal processing unit according to the invention and the processing device according to the invention are designed and set up to carry out the following steps: 20201001LU 02.09.2020

-3/45- — A lung time constant for the patient's lungs is determined. Measured values of at least one parameter of the patient are used for this, in particular values from the or at least one patient sensor.

— The volume of a dead space is determined. This dead space is in the fluid communication between an area of the lungs suitable for gas exchange with the patient's blood and the ventilator. The dead space is preferably located between a measuring point at the patient's mouth or in an area of the fluid connection which is between the mouth and the ventilator and the area of the lungs suitable for gas exchange.

— A mandatory target rate for mandatory ventilation of the patient by the ventilator is calculated. This mandatory setpoint frequency is calculated as a function of the specified desired volume flow, the determined lung time constant and the determined dead space volume.

— An ideal spontaneous breathing rate for the patient's spontaneous breathing is calculated. With this ideal spontaneous breathing frequency, the patient is able to achieve the desired volume flow, and this solely through spontaneous breathing. This ideal spontaneous breathing frequency is calculated as a function of the determined lung time constant and the determined dead space volume, preferably using a model for a patient's respiratory activity.

— A measure of the current intensity of the patient's spontaneous breathing is determined.

— The desired target ventilation rate is calculated as the weighted average of the mandatory target rate and the ideal spontaneous ventilation rate. Which weighting, in particular which weighting factor, is used in this weighted average depends on the measure determined for the intensity of the patient's spontaneous breathing.

The ventilator is able to carry out ventilation strokes depending on the calculated setpoint ventilation frequency, in particular to carry out ventilation strokes with the calculated setpoint ventilation frequency.

20201001LU 02.09.2020

"4/45 - According to the invention, a target ventilation rate for the LU102031 ventilator is automatically calculated.

Therefore, the invention avoids the need to manually set a target respiratory rate.

It is possible for the processing device to be operated with this automatically calculated setpoint ventilation frequency, in particular to carry out ventilation strokes in accordance with this setpoint ventilation frequency.

It is preferably possible for the automatically calculated target ventilation rate to be displayed to a user of the ventilator and for user input to be recorded and evaluated, in particular whether the user confirms the displayed target ventilation rate or overwrites it with a different value.

According to the invention, a measure for a desired volume flow into and out of the patient's lungs is specified, ie the volume of gas flowing into and out of the lungs per unit of time.

The target volume flow is preferably a target alveolar or proximal minute volume.

Preferably, this target volume flow is the flow of volumes into and out of that area of the lungs that is available for gas exchange with the patient's blood.

The gases are in particular O2 and CO2 and optionally at least one anesthetic.

According to the invention, a mandatory setpoint frequency for the mandatory ventilation of the patient and an ideal spontaneous breathing frequency for the patient's spontaneous breathing are calculated.

Mandatory ventilation is the artificial ventilation of a completely sedated patient, i.e. a patient who is currently not breathing spontaneously at all.

With the ideal spontaneous breathing frequency for spontaneous breathing, the patient is able to generate the desired volume flow without artificial respiration, with the ideal spontaneous breathing frequency preferably optimizing a predetermined target function.

The ideal spontaneous breathing frequency is preferably that frequency with which the patient is able to achieve the desired volume flow with the least work or the lowest effort, particularly preferably the frequency that leads to a 20201001LU 02.09.2020

-5/45- average mechanical power during inspiration (energy efficient LU102031 frequency). According to the invention, the target ventilation rate is calculated as a weighted summary, in particular as a weighted average, of the mandatory target rate and the ideal spontaneous breathing rate for spontaneous breathing.

The setpoint respiration frequency therefore depends on a weighting factor, this weighting factor in turn depending on a measure determined for the current spontaneous respiration of the patient.

The larger this measure is for spontaneous breathing, i.e. the more intense the spontaneous breathing is at the moment, the larger the weighting factor and thus the contribution of the ideal spontaneous breathing frequency to the calculated setpoint ventilation frequency.

The feature according to the invention, that a weighted summary is calculated, takes into account the fact that a patient is often not completely sedated throughout the artificial respiration, but breathes spontaneously at least at times.

Artificial respiration supports this spontaneous breathing.

The patient then breathes due to a superposition of spontaneous breathing and mechanical ventilation.

As a rule, a spontaneous breath triggers a ventilation stroke of the ventilator.

If there is weak or no spontaneous breathing, the processing device carries out additional ventilation strokes, i.e. ventilation strokes that are not triggered by spontaneous breathing.

How many such additional ventilation strokes are carried out depends on the setpoint ventilation frequency calculated according to the invention and the spontaneous breathing.

The risk is reduced that too few or too many ventilation strokes are carried out, i.e. the patient receives too little or too much air.

Artificial ventilation can be mandatory ventilation (patient does not breathe spontaneously at all) or assisted ventilation (automatic ventilation supports his spontaneous breathing).

Thanks to the invention, it is not necessary to switch the ventilator between at least two different modes, namely at least one mode for mandatory ventilation and at least one mode for supportive ventilation

| -6/45- Ventilation.

Such switching could result in an abrupt change in FI artificial respiration.

Instead, thanks to the invention, the artificial respiration can be adapted to the current intensity of the spontaneous respiration, even if the desired volume flow remains the same and the intensity of the spontaneous respiration changes.

The setpoint ventilation frequency calculated according to the invention can also be used for the two special cases that the patient is completely sedated and that the ventilation strokes are triggered exclusively by the patient's spontaneous breathing.

In the first case, no ventilation stroke is triggered by the spontaneous respiration, in the second case every ventilation stroke.

Thanks to the invention, these two special cases do not necessarily have to be detected in order to set the target respiration rate.

It is sufficient to detect the measure of spontaneous respiration.

The device according to the invention and the method according to the invention can be used equally well for the two special cases and for many situations between these two special cases, ie relatively weak spontaneous respiration.

According to the invention, the mandatory setpoint frequency is calculated as a function of the desired volume flow and of the determined dead space volume and the determined lung time constant.

The ideal spontaneous breathing frequency depends at least on the lung time constant and the dead space volume.

For this reason, the calculated mandatory target rate depends on the pneumatic properties of the lungs and on a desired volume flow into and out of the lungs.

Therefore, the mandatory target rate is adapted to the lungs of a patient to be artificially ventilated.

Because the patient's lung time constant is determined and used according to the invention, the following two dangers in artificial ventilation are reduced: Too little breathing air reaches the alveolar lung space due to a ventilation frequency that is too high.

This undesired effect occurs in particular in a lung with a high lung resistance (resistance) R and therefore a large lung time constant 71 .

In particular, the 20201001LU 09/02/2020

-7/45- undesired situation that during expiration too little LU102081 used breathing air flows out of the lungs. The lungs keep getting inflated. This occurs primarily in congested airways, especially in a smoker's lungs. — The lungs are overstretched due to a ventilation rate that is too low, especially because the tidal volume achieved is too large. This undesired effect occurs in particular in a lung with a low elasticity (compliance) C and therefore a small lung time constant 1 . The invention makes it possible, but saves the need to determine the elasticity or the compliance or the pneumatic resistance of the lungs separately for the calculation of the mandatory setpoint frequency. However, such a determination is often not possible in practice, involves relatively large errors and/or puts a considerable strain on the patient. Rather, thanks to the invention, it is sufficient to determine a lung time constant.

The ideal spontaneous breathing frequency calculated according to the invention for the patient's spontaneous breathing also depends on the patient, namely on the determined dead space volume, which occurs to a considerable extent in the patient's body, and on the desired volume flow.

In a preferred embodiment, the mandatory setpoint frequency is calculated as a function of an ideal mandatory setpoint frequency and an upper limit for the mandatory setpoint frequency. The mandatory setpoint frequency is particularly preferably equal to the ideal mandatory setpoint frequency or the upper limit, depending on which of these two values is lower.

The determined upper limit depends on the determined lung time constant and preferably additionally on the determined dead space volume and/or on the desired volume flow. The upper limit can be the minimum of several individual upper limits, with a first individual limit depending on the lung time constant and a further individual limit depending on the dead space volume and/or the volume flow.

Because the mandatory target frequency is not greater than this upper limit, the risk is further reduced that the lungs will be damaged by too high a frequency

-8/45- or is subjected to excessive mechanical stress due to an excessive tidal volume, or 1102081 does not receive enough air to breathe.

In particular, it is avoided that the artificial respiration only moves breathing air in both directions through the dead space due to a frequency that is too high, without a sufficient amount of breathing air reaching the area of the lungs suitable for gas exchange and without a sufficient amount of used air being removed from the lungs is carried away.

However, if the upper limit were always used as the mandatory target frequency, there would be a risk that the lungs would be subjected to a greater than necessary load due to the artificial ventilation, i.e. greater than necessary to achieve the desired volume flow.

Therefore, the mandatory setpoint frequency can also be lower than the upper limit, namely in particular equal to a calculated ideal mandatory setpoint frequency.

The ideal mandatory setpoint frequency is calculated as a function of a predetermined required inhalation fraction.

The proportion of inhalation provides a measure of the proportion of time in a respiratory cycle that an inhalation process has on average.

It is possible to specify the ratio between the average duration of an inspiration process and the average duration of an expiration process.

It is also possible to specify the average duration of an inhalation process or that of an exhalation process.

The duration of the inhalation process

The exhalation process and the ventilation frequency determine an inhalation proportion.

In a preferred embodiment, in order to calculate the ideal mandatory setpoint frequency, a lung model is specified in a form that can be evaluated by a computer and stored in the data memory.

This lung model approximately describes the pneumatic behavior of a person's lungs and thus also of the lungs of the artificially ventilated patient.

This lung model preferably contains at least one model parameter, in particular the lung time constant and/or the dead space volume.

Using this lung model, it is possible to predict what resistive power and what elastic power will be at a given actual 20201001LU 09/02/2020

-9/45- Ventilation frequency affect the lungs.

The ideal mandatory target rate is 10102081 calculated using the default lung model and the default inspiration fraction.

Using the given lung model and the required inspiration fraction, a resistive power and an elastic power are calculated.

The resistive power is the work per unit of time that has to be expended during an inhalation process in order to overcome the pneumatic resistance of the lungs.

Elastic power is the work per unit time required to stretch the lungs.

In order to calculate the ideal mandatory target frequency, it is predicted which frequency leads to which resistive power and to which elastic power according to the lung model and with the specified inhalation fraction. | In a particularly preferred embodiment, this ideal mandatory target frequency is defined such that the resistive power produced differs from the elastic power produced by at most a predetermined power factor, with the power factor preferably being the quotient of the two powers produced.

The power factor can be chosen so small that the resistive power is only slightly larger than the elastic power.

In many cases, this configuration leads to a relatively low load on the lungs, in particular because each tidal volume is relatively small.

Nevertheless, the configuration brings about a sufficiently large volume flow into and out of the lung area suitable for gas exchange, in particular the alveolar lung space.

The advantageous embodiment with the power factor saves the need to minimize a function that describes the total work or power that acts on the lungs during artificial respiration.

In particular, the need to minimize such a function at runtime is avoided.

Minimizing a function is time consuming.

If an iterative process is used during minimization and if the process is aborted at runtime when an abort criterion is met, the minimization can lead to an unfavorable result.

Without a 20201001LU 09/02/2020

- 10/45 - suitable abort criterion, the minimization can take too long.

This advantageous embodiment avoids the disadvantages of such a minimization. In a preferred implementation, an initialization phase and a subsequent usage phase are carried out.

A constant is calculated in an initialization phase.

To calculate the constant, the lung model and preferably the optionally specified power factor are used.

In the subsequent usage phase, the ideal mandatory target frequency is calculated.

For this purpose, the proportion of inhalation, the determined lung time constant and the constant that was calculated in the initialization phase are used.

This refinement often leads to less computing effort in the use phase.

It is possible to recalculate the constant in the usage phase, for example due to a changed specification.

It is possible to use the constant obtained in the initialization phase in several usage phases, also for several patients.

In a development of this implementation, a first signal processing unit calculates the constant in the initialization phase.

A second signal processing unit calculates the ideal mandatory setpoint frequency and from this the setpoint ventilation frequency during the usage phase.

The first - signal processing unit is not necessarily a part of the ventilator.

According to an embodiment that has just been described, an ideal mandatory setpoint frequency is calculated as a function of a predefined required inhalation fraction.

The ideal mandatory setpoint frequency is preferably the greater, the greater the predetermined required inhalation proportion.

According to the invention, a mandatory setpoint frequency for the mandatory ventilation of the patient and an ideal spontaneous breathing frequency for the spontaneous breathing of the patient are calculated.

The mandatory setpoint frequency is preferably calculated in such a way that it is greater than or equal to the ideal spontaneous breathing frequency.

Then the specified target ventilation frequency is at least as large as the ideal spontaneous ventilation 20201001LU 02.09.2020

- 11/45 - frequency. This avoids a ventilation frequency that is too low. LU 102057 In particular, it is ensured that at least every spontaneous breath taken by the patient triggers a ventilation stroke. According to the invention, a target respiration rate is calculated. A parameter that is often set on a ventilator is a required tidal volume. This required tidal volume is a value for that volume which the ventilator should feed into the fluid connection during a ventilation stroke. The required tidal volume is calculated depending on the desired volume flow, the determined dead space volume and the target ventilation frequency calculated according to the invention. The ventilator is controlled with the aim of regulating that the achieved tidal volume is equal to the calculated target tidal volume for at least one ventilation stroke, preferably for each ventilation stroke. An actuator of the ventilator is preferably controlled accordingly.

During a ventilation stroke, the ventilator injects fluid into the fluid connection. The pressure at a measuring point in the fluid connection increases until it reaches a maximum value. In one embodiment, the pressure that is actually achieved is regulated, with a time profile of the required pressure being specified and with the control aim of the regulation being that the actual time profile of the pressure is equal to the specified time profile.

If the predetermined time profile of the pressure suddenly reaches the maximum value, there is a risk that the patient's lungs will be damaged. A required ramp time is therefore preferably calculated for at least one ventilation stroke, preferably for each ventilation stroke. In the required time profile of the pressure, this ramp time defines the period of time that should elapse between — the beginning of the ventilation stroke and — the point in time at which the maximum value of the pressure is reached. The regulation of the actual pressure profile over time is carried out in such a way that the ramp time actually achieved is equal to the calculated setpoint ramp time. 20201001LU 02.09.2020

-12/45- The target ramp time is preferably calculated as a function of the determined HU102031 lung time constant of the patient.

It is also possible to use an ideal weight (ideal body weight) of the patient in addition to or instead of the lung time constant.

This ideal weight can be derived from easily measurable patient parameters.

The invention is described below using an exemplary embodiment.

Here shows

FIG. 1 shows an example of an artificially ventilated patient, a ventilator and a number of sensors;

FIG. 2 shows the resistive and the elastic power as a function of the ventilation frequency;

FIG. 3 shows an exemplary time course of pressure, volume flow, resistive work and elastic work with a relatively low lung resistance;

FIG. 4 shows an exemplary time profile of pressure, volume flow, resistive work and elastic work with a relatively high lung resistance and relatively high ventilation frequency;

FIG. 5 shows an exemplary time profile of pressure, volume flow, resistive work and elastic work with a relatively low lung resistance, the ramp time at the start of a ventilation stroke being taken into account;

FIG. 6 shows how the measure for the intensity of the patient's spontaneous breathing is determined as a function of the respiratory work performed;

20201001LU 02.09.2020

- 13/45 -

FIG. 7 shows how the measure for the intensity of spontaneous respiration 10106081 is determined as a function of the pressure-time product that occurs;

FIG. 8 shows the total power and the power for spontaneous respiration as a function of the ventilation frequency;

In the exemplary embodiment, the invention is used to artificially ventilate a patient.

Fluid communication is established between the patient's lungs and a ventilator.

Breathing air is supplied to the patient through this fluid connection.

Optional is this | Breathing air mixed with at least one anesthetic so that the patient is partially or completely sedated.

Optionally, the fluid connection is associated with a breathing circuit that recycles exhaled air, particularly when an anesthetic is loaded.

In the case of a breathing circuit, carbon dioxide (CO2) is filtered out of the exhaled air.

FIG. 1 shows an example of a patient P who is artificially ventilated.

The lungs Lu, the esophagus Sp, the stomach Ma and the diaphragm Zw of patient P are shown schematically.

A ventilator 1 with a display and control unit 12 and a data-processing signal processing unit 10 artificially ventilates the patient P.

The ventilation hoses between the ventilator 1 and the patient P are not shown.

A flexible connection piece 4 is arranged in the mouth of the patient P during ventilation.

In one embodiment, a flexible measuring catheter 6 is placed in the esophagus of the patient P, the measuring catheter 6 beginning in the connecting piece 4 .

Various sensors measure various vital parameters of the patient P and/or parameters of the fluid connection between the ventilator 1 and the patient P's lungs Lu.

All of these sensors need not necessarily be present in order to practice the invention.

The following sensors are shown in Figure 1: 20201001LU 02.09.2020

- 14/45 - — A pneumatic sensor 2 comprises a transducer 2.1 LU102031 comprising an orifice located near the mouth of the patient P and sampling air from the fluid connection. The sampled air is transmitted via a hose to a pressure sensor 2.2, which measures a measure of the airway pressure Paw (pressure in airway) in the fluid connection and optionally a measure of the volume flow Vol'. In one embodiment, the measured value recorder 2.1 is arranged in or on a Y-piece near the connection piece 4, i.e. near the mouth of the patient P.

- Optionally, a sensor 15 in or on the ventilator 1 measures a measure of the volume Vol' per unit time of the flow of respiratory air from the ventilator 1 to the patient P (inspiratory minute volume) and/or back from the patient P to the ventilator 1 (expiratory minute volume).

- A probe 3 in the esophagus Sp of the patient P, preferably comprising a measuring balloon, measures a measure of the time-varying pneumatic pressure Pes (pressure in esophagus) in the esophagus Sp Connector 4 or is part of the measuring catheter 6.

- Another measuring balloon of the probe 3 or an optional gastric probe 7 in the form of a measuring balloon placed in the stomach Ma measures a measure of the gastric pressure Pga in the stomach Ma.

— Several measuring electrodes are attached to patient P's chest. FIG. 1 shows an example of a set 5.1.1, 5.1.2 of measuring electrodes close to the heart and a set 5.2.1, 5.2.2 of measuring electrodes close to the diaphragm. An electrocardiogram (ECG) and/or an electromyogram (EMG) of the patient P is generated with the aid of these optional measuring electrodes 5.1.1, . . . 522 .

The signal processing unit 10 is able to automatically determine when air is flowing into the breathing system of the patient P and when air is flowing out of the breathing system again, ie it is able to detect every inspiration phase and every expiration phase. For this purpose, the signal processing unit 10 uses measured values from the sensors 2 and 15 and optionally from the measuring electrodes 5.1.1 to 5.2.2.

20201001LU 02.09.2020

-15/45- In one application, patient P is completely sedated and does not breathe spontaneously, so 1192087 does not breathe at all or at most a few spontaneous breaths. In this application, the ventilator 1 performs mandatory ventilation. In another embodiment, the patient P himself breathes spontaneously, at least at times, and the ventilator 1 supports his ventilation. Temporal transitions between these two ventilation modes are also possible.

In both modes of ventilation, the ventilator 1 performs a sequence of ventilation strokes. With each ventilation stroke, the ventilator 1 feeds breathing air into the fluid connection. Artificial ventilation is characterized by several parameters, e.g.

- a required time profile of the pressure that the ventilator 1 generates during the ventilation strokes, - a required time profile of the volume of breathing air that is fed into the fluid connection during the ventilation strokes, - a frequency of the ventilation strokes or also - the amplitude, i.e. the maximum pressure difference during ventilation strokes.

A person takes in air during spontaneous breathing or during artificial respiration performed at a frequency fist. The volume taken in with one breath is called the (actual) tidal volume or tidal volume Volriaist. The alveolar lung space is available for the exchange of oxygen and carbon dioxide between the inhaled air and the blood. The volume of gas that actually flows through the alveolar lung space is referred to below as the achieved alveolar lung volume. The process of gas flowing through the alveolar lung space is often referred to as "alveolar ventilation". A dead space with the entire volume Vo, whereby gas flows through this dead space in both directions, but not 20201001LU 02.09.2020

- 16/45 - is available for the exchange of oxygen and carbon dioxide.

The alveolar lung volume achieved per unit of time, in [l/min] in the exemplary embodiment, is referred to as the actual alveolar minute volume V'actual HU102031.

Then (1) Vaist= fist * (Volria,ist - Vb). The actual alveolar minute volume V'aist correlates with the volume flow Vol', which is measured at a measuring point, for example using the measuring sensor 2.1. A dead space occurs in the fluid connection between the ventilator 1 and the lungs of the patient P, in which no gas exchange takes place.

Only that device-side dead space that is located between the measuring point 2.1 and the patient P and through which gas flows in both directions is to be taken into account.

In many cases, this device-side dead space has a negligibly small volume Vo,Ger, so that only the volume Vo pat of the patient-side dead space is used as the total volume Vo alveolar lung space flushed and therefore for the exchange of oxygen and carbon dioxide with the blood of the patient P is available.

This minute volume is given, for example, in liters per minute and is denoted by V'areq.

This desired alveolar minute volume V'Areq is specified, for example by a user using the display and control unit 12 on the ventilator 1, or automatically or derived by a user, for example depending on a required time profile of the CO2 content in the exhaled air.

Such a course over time is also referred to as a capnogram or capnography curve and can be measured using a CO2 sensor.

Instead of a target alveolar minute volume V'a req, a target proximal minute volume can also be specified.

The actual proximal minute volume V'p is the volume flow at the patient P's mouth, specifically the volume flow at a tube connector. 20201001LU 02.09.2020

- 17 / 45 - The volume flow is divided into the alveolar air flow and the flow through the LU102081 dead space. The volume Vo,Ger of the device-side dead space is negligibly small at a measuring point on the patient P. The actual proximal minute volume V'Eiss is related to the actual alveolar minute volume V'Aactual as follows: (2) VPactual = V'Aactual + VD factual = V'A,actual + (VD,Pat + VD,Ger) * factual where Vb the total volume of the dead space, which in the exemplary embodiment is equal to the sum of the volume Vp pat of the patient-side dead space and the volume Vo,cer of the device-side dead space, and fist is the actual ventilation frequency of the ventilator 1. The device-side volume Vo,Ger can often be neglected , otherwise it is known due to the construction.

For example, the target alveolar minute volume V'areq or target proximal minute volume V'p req SO is specified so that target values for the concentration or the partial pressure of specific gases in the blood of the patient P are achieved as well as possible. The minute volume V'a req, V'Preq can be specified by a user or by a higher-level automation system or is permanently stored in a data memory of the ventilator 1 .

According to the invention, the signal processing unit 10 automatically calculates a setpoint ventilation frequency fset of the ventilator 1, specifically as a function of a desired alveolar minute volume V'a req. The signal processing unit 10 preferably receives measured values from the sensors 2, 3, 5.1.1, ...,

5.2.2, 7 and carries out the necessary calculation steps depending on the measured values received.

In the exemplary embodiment, the calculated setpoint ventilation frequency fset is used in a subordinate regulation for the pressure. The pressure actually achieved is the controlled variable and is measured, preferably as the airway pressure Paw achieved. A desired time profile Pset of the pressure is specified as a reference variable. The calculated set ventilation frequency fset defines the frequency of the pressure increases and thus the ventilation strokes in this desired time profile of Pset. 20201001LU 02.09.2020

- 18/45 - A control unit, not shown, causes an actuator of the ventilator 10106091 1 to be activated, with this actuator generating the ventilation strokes. As a rule, the ventilation rate fact that is actually achieved is equal to the automatically calculated target ventilation rate fset. Nevertheless, different designations are used in the following for clarification. In the case of mandatory ventilation, a setpoint frequency is specified, while in the case of supportive ventilation, the frequency can depend on the patient P's spontaneous breathing. The following first describes how the target ventilation frequency for mandatory ventilation is calculated. This is referred to as fsetmand in the following. The exchange of oxygen and carbon dioxide with the patient P's lungs Lu during respiration and ventilation takes place only in the alveolar lung space. The fluid connection between the connection of the ventilator 1 on the patient side and the lungs Lu also includes the upper and middle airways of the patient P, in particular his trachea, as well as an optional tube, hoses and/or other fluid-guiding units that connect the patient P to the ventilator 1 connect, as well as optional measuring chambers. Breathing air flows through these during each ventilation stroke, but they do not contribute to the exchange of oxygen or carbon dioxide and are therefore referred to as dead space. This dead space consists of a patient-side dead space and a device-side dead space. The total dead space volume Vo is the sum of the volume VDPat of the dead space on the patient side and the volume Vo,cer of the dead space on the device side.

This dead space volume Vp results in a boundary condition for the calculation of the mandatory setpoint ventilation frequency fset,mand. The ventilation frequency fact must be so low that enough breathing air reaches the alveolar lung volume Va.

During a ventilation stroke, the ventilator 1 feeds breathing air into the fluid connection. With one breath, the patient P takes in the breathing air. The volume of inhaled or inhaled air is called Voltiaist's tidal volume. Ideally, the lungs Lu are 20201001LU 02.09.2020

- 19/45 - Ventilation stroke increased by this tidal volume Voltiais.

In a Ce embodiment, it is required that the tidal volume Voltist be at least twice the dead space volume Vo.

The boundary condition (3) fsetmana <= V'A,req / Vo results from this and from the relationship (1).

In one embodiment of the step of determining the dead space volume Vo pat on the patient side, it is automatically detected when the patient P exhales.

On exhalation, air from the upper and middle airways of the patient P flows out of the body and also air from the lungs Lu flows into the upper and middle airways and then out.

A volume flow sensor, for example sensor 2 or sensor 15, measures the respective volume flow Vol' at a point in the | at several sampling times fluid connection.

This measured volume flow Vol' correlates with the volume flow flowing in and out of the patient P's body.

A CO2 sensor is used to measure the proportion of CO2 in the fluid connection.

Only the alveolar lung space, but not the upper and middle airways, can inject CO2 into the air.

As soon as this proportion is above a predetermined limit, the fluid connection contains air from the lungs Lu.

In the period between the start of an expiratory process and the time at which a relevant proportion of CO2 is contained in the exhaled air, air flows out of the upper middle airway from the patient P.

The volume flow values measured in the period up to the detection of CO2 provide, after up integration, a measure of the volume Vp Pat of the dead space in the upper and middle airways of patient P.

The dead space volume Vp,pat on the patient side can be determined using the Bohr formula, for example.

The volume Vb cer of the device-side dead space in the fluid connection outside the body of the patient P is known with sufficient accuracy from the design of the ventilator 1 and the hoses.

If the volume flow is measured near the patient P's mouth, the volume Vp cer of the device-side dead space can be neglected. 20201001LU 02.09.2020 mm ———

-20/45- In another embodiment, no CO2 sensor is required. Rather, an ideal weight Gewia of the patient P is determined, and the volume Vp pat of the patient-side dead space in the upper and middle airways of the patient P is estimated according to the formula (4) Vo,pat=Y*Geid. Another boundary condition results from the requirement that the actual ventilation frequency factual must be small enough to stretch a less elastic lung Lu, i.e. a lung with a lung time constant of 1, sufficiently. In other words: If the ventilation frequency fact is too high, the ventilation pressure is not sufficient to expand the lungs Lu sufficiently. The lungs Lu can then not absorb enough oxygen and release enough CO2. This boundary condition depends on the patient P's lung Lu.

In order to meet this boundary condition, a so-called lung time constant T is measured on the patient P. Simplistically, the lungs can be viewed as a linear pneumatic system where pressure applied to the lungs overcomes pneumatic resistance and causes the lungs to stretch. Under this simplifying assumption, the following applies to inhalation: (5) T i Vol, , (t) = Vol u,max * [1 -e RC ] Here, R is the lung resistance (resistance) and C is the compliance. Volu(t) is the volume of the lungs at time t, VolLu,max is the maximum lung volume. To put it simply, the lung resistance R (6) is R=AP/ Vol, where AP is the difference between the maximum and minimum pressure over the course of a ventilation stroke and Vol' is the change in volume of the lungs Lu due to this differential pressure. The compliance C is the reciprocal of the elasticity (elastance), where 20201001LU 02.09.2020 applies

-21/45- (7) C=Vol/AP. LU102031 The lung time constant T is the product of lung resistance R and compliance C, i.e. (8) T=R*C.

The lung time constant 1 is a parameter that describes the response of the lungs as a passive pneumatic system to a jump stimulus. In the case of spontaneous breathing, the stimulus to jump results from a sudden increase or reduction in the pressure Pmus generated by the diaphragm Zw, in the case of artificial ventilation from a sudden increase or reduction in the airway pressure Paw. In response to the jump excitation, the volume increases or decreases exponentially, see model equation (5) and definition (8). If the lung time constant T has elapsed since the jump excitation, then two thirds, more precisely 63%, of the total inspiratory tidal volume has passed into the lungs Lu. The lung time constant 1 is specified in [msec] or in [sec], for example. In order to determine the lung time constant 1 of the patient P, at least one volume flow sensor 2, 15 already mentioned and at least one of the pressure sensors 2, 3, 7 are used, which measure the volume flow Vol' or the airway pressure Paw in the fluid connection and optionally measure the esophageal pressure Pes. The volume flow of the lungs Lu over time can be derived from the volume flow Vol. A typical value for the lung time constant 1 of a COPD patient (patient with smoker's lung) is 450 msec. In other patients, the lung time constant T is usually significantly shorter.

The boundary condition is preferably derived from the lung time constant 1 as follows: A factor x is specified and does not depend on the patient P's lungs Lu. The factor x is preferably between 4 and 12 and is, for example, 5.5 or 8. The resulting boundary condition is (9) fset,mand <= 1/(x*7).

The boundary conditions (3) and (9) result in an upper limit fset,mand,max for the mandatory target ventilation frequency fset,mand, namely 20201001LU 09/02/2020

-22/45- (10) fset mand max = MIN { V'A,req / Vo, 1 / (x* T) }. LU102031 It is possible to use a "safety discount", i.e. (1 1) fset,mand,max = min { (1-AM)*V'A req / Vb, 1-A2/ (x* T) }. In this case, 0<M, A2<1, preferably M, A2<=0.2, applies. The following must apply: fsetmand <= fset,mand,max- | The definition of fset mand = fsetmand,max would in many cases lead to an unnecessarily high ventilation frequency fact, namely to a ventilation frequency which puts a greater strain on the lungs Lu than necessary. Therefore, an ideal mandatory setpoint ventilation frequency fset mana a is calculated, which can be smaller than the upper limit fsetmandmax. The mandatory target frequency fsetmand is defined according to the rule (1 2) fsetmand=Min { fset mand, id, fset mand max }-. How the ideal mandatory target ventilation rate fset,mand,id is calculated is described below.

A ventilation stroke causes an inhalation process. In the case of mandatory ventilation, the inspiration process Ti has the same duration as the ventilation stroke that caused it. An expected or desired ratio for the duration Ti of inhalation (inspiration) compared to the duration Te of exhalation (expiration) during a single ventilation process is specified, ie the I:E ratio. Typical values are between 3:5 and 4:5. It is possible that this ratio depends on the patient P's lung time constant T. This results in a predetermined factor D1, which is the required temporal proportion of inspiration in an entire respiration cycle (breathing process), ie (13) D1=Ti/(Ti+Te) This factor D1 acts as the required inhalation proportion. It is also possible to specify the quotient Ti / Te directly as the inhalation portion, i.e. the I:E ratio. It is also possible to specify a required average duration of an inhalation process or an exhalation process. From this 20201001LU 02.09.2020

-23/45- Duration of the inhalation process or

The exhalation process and the Sou-LV102031 ventilation frequency result in a required inhalation proportion D1. The mandatory ventilation of the patient P performs a mechanical work that is divided into a resistive (viscous) work Wr and an elastic work We.

The resistive work Wr overcomes the lung resistance R.

The elastic work Wc stretches the lungs Lu and counteracts the elasticity and thus the compliance C of the lungs Lu.

Both the resistive work Wr and the elastic work Wc depend on the ventilation frequency fact.

As already explained, the lung Lu is modeled in a simplified way as a linear pneumatic system.

This system corresponds to an electrical RC element, i.e. a series connection of an electrical resistance R and an electrical capacitor C.

In simplified terms, it is assumed that the new pressure is present immediately after a ventilation stroke.

In other words: A ventilation stroke is treated as a jump excitation on the pneumatic system.

These simplifying assumptions lead to the following lung mechanical equations for the resistive power W'r and for the elastic power Wc: 1 Vai 09 fist is oD mei f.

Va: 2 T we: — [ee 1-e Sf A z_ Te ( T ) ( RC ) — ( a + Vo) coth(=— ) 1-e "RC. 15 ' (19 . fist Vast 2 2c N f , The two lung mechanical equations (14) and (15) together form two components of a lung model.

They describe the resistive power WR or the elastic power W'c, which the ventilator 1 implements in the case of mandatory ventilation during an inspiration process on the lungs Lu.

They apply with sufficient accuracy to any value that may be considered for the parameters C, R and fist.

According to this lung model (14) and (15), the resistive power Wr is always greater than the elastic power W'c, no matter how large the lung resistance R and the 20201001LU 02.09.2020

- 24 / 45 - lung compliance C and regardless of which ventilation frequency fist is used in 0102031 mandatory ventilation. FIG. 2 shows, by way of example, the resistive power Wr, the elastic power W'c and the total power, ie the sum W'R+Wc, as a function of the respiration frequency factual. The ventilation frequency fact in [1/min] is plotted on the x-axis, the power in [Nm/sec] on the y-axis. In addition, FIG. 2 shows the upper limit for ventilation frequency f actual , which results from boundary condition (9) with the value x=8. FIG. 3 and FIG. 4 show, by way of example, the time profile of the pressure P (above) and the time profile of the work W performed (below) during a ventilation stroke. The time t in seconds is plotted on the x-axis, the pressure P in [mbar] and the volume flow Vol' in [/sec] (top) or the work W in [Nm] (bottom) on the y-axis . The time courses were determined using the two lung mechanical equations (14) and (15). In this example C = 30 ml/mbar. The achieved alveolar minute volume Va,act is V‘a,act = 7.9 l/min. In FIG. 3, R=5 mbar/(I*sec), in FIG. 4, R=mbar/(I*sec). In the example in FIG. 3, the ventilation frequency is factual=15/min, in the example in FIG. 4 factual=30/min.

In the time curves, 20 - Pset designates the pressure over time of the pressure to be generated by the ventilator 1, with the time curve Pset being a reference variable for a subordinate control of the pressure actually achieved, which is preferably measured as the airway pressure Paw achieved, - Pung the resulting pressure Pressure in the lungs Lu, - Vol' the volume flow in the fluid connection to and from the patient P, where a positive value denotes a volume flow from the ventilator 1 to the patient P and a negative value denotes a volume flow from the patient P to the ventilator 1, - WR the resistive work done so far, — We the elastic work done so far, — WRins the resistive work done at the end of an inspiration event, and 20201001LU 02.09.2020

-25/45- — We,ins the elastic work done at the end of an inspiration process.

HU102031 In the example of Figure 3, WRins = 0.76 Nm and Wc ins = 0.76 Nm, in the example of Figure 4, Wr,ins = 0.50 Nm and Wejns = 0.28 Nm.

In the two examples shown in FIG. 3 and FIG. 4, the setpoint pressure Pset immediately increases to the maximum value during a ventilation stroke.

FIG. 3 and FIG. 4 therefore show the response of the pneumatic lung system to a jump excitation by the ventilator 1. In reality, the actual pressure Paw achieved by the subordinate regulation cannot immediately rise to the full value.

In addition, the risk must be ruled out or at least reduced that the lungs Lu of the patient P will be damaged by a pressure increase that is too rapid.

The desired time profile Pset of the pressure is therefore specified in such a way that it increases at the beginning of a ventilation stroke until it reaches the maximum value.

This period of time between the start of the ventilation stroke and reaching the

The maximum value for Paw is called the actual ramp time tr actual.

The desired progression over time Pset of the pressure includes a predefined setpoint ramp time tr set and then a maximum setpoint pressure.

By specifying a target ramp time tr set, the resistive work Wr is reduced compared to a sudden increase in the target pressure Pset, while the elastic work Wc remains unchanged.

One reason for this: The maximum volume flows are reduced.

In a preferred embodiment, the setpoint ramp time tr set is defined as a function of the determined lung time constant 7 of the patient P, for example according to the calculation rule

(16) trset=Y'T In this case, the factor y is specified and stored and is, for example, y=0.5. FIG. 5 shows the curves over time which are also shown in FIG. 3 and FIG. 4, a regulated pressure rise being carried out in the example of FIG.

In the example in Figure 5, the lung compliance is C = 30 ml/mbar, the lung resistance is R = 5 mbar/(l*sec), the ventilation frequency factual = 15/min 20201001LU 02.09.2020

-26/45 - and the ramp time is dreary = 0.25 sec.

A minute volume of Vast = 7.9 L/min LU102031 is achieved.

Compared to ventilation without a ramp time (step response, i.e. tRact = 0), the resistive work Wr is lower when the ramp time tRact is taken into account.

The elastic work Wc remains unchanged.

The resistive work done at the end of an inspiration event is WRins = 0.47 Nm, the elastic work done at the end of an inspiration event is Wc,ins = 0.76 Nm.

The resistive work done at the end of an exhalation is Wr exp = 0.76 Nm, the elastic work done at the end of an exhalation is Wc exp = -0.76 Nm.

Both the resistive power W'r and the elastic power W'c depend on the actual ventilation frequency fact.

The resistive power W'r is always greater than the elastic power Wc.

The ideal mandatory setpoint frequency fsetmand,id Is determined depending on the effect of the ventilation frequency fact on the resistive power W'r and on the elastic power W'c.

In the case of mandatory ventilation, the ventilation strokes each cause patient P to inhale.

It follows that: (17) T; *T — - fist 1 =——— =D1 T,+T, | For the following derivation it is assumed that the actual frequency | 20 factual of the mandatory ventilation is equal to the desired ideal mandatory setpoint ventilation frequency fset,mand,id, so that the following applies: (18) factual = fsetmand,id.

Relationship (17) shows how, under assumption (18), the sought-after ideal ventilation frequency fset,mand,ia, the duration Ti of a ventilation stroke, and the temporal inhalation portion D1 are related.

A certain ventilation frequency fist and a certain duration Ti of the ventilation strokes lead to the pulmonary mechanical model equations 20201001LU September 2nd, 2020

-27/45- (14) and (15) to a specific resistive power W'R and to a specific 0102031 elastic power Wc. In one embodiment, the sought-after ideal mandatory setpoint ventilation frequency fset,mand,id is calculated in such a way that a function that depends on the two powers W'r and W'c is minimized. This function is, for example, the sum of the two powers, i.e. Wr + Wc, or the quotient W'r / Wc from the two powers. In a preferred embodiment of the exemplary embodiment, however, the ideal mandatory setpoint ventilation frequency fset,mand,ia is calculated in such a way that the elastic power Wc is approximately equal to the resistive power W'R. In many cases, this specification leads to a comparatively low mechanical load on the lungs Lu of the patient P. To put it more precisely: a factor a is specified, with a preferably being between 0 and 0.2 and particularly preferably equal to 0.1. The power factor is then 1+a. The boundary condition when determining the ideal mandatory target ventilation rate fset,mand,id is (19) x

WRW Cc From the equations (15) and (19) follows + (20) We T, — = Coth{-— We 2RC If one solves the equations (20) and (17) for Ti under this assumption, the following results Calculation rule for the ideal mandatory 20 — Soli ventilation frequency fset,mand,id: (21) D1 fset,mand,id = * 2*RC*arccoth (1+a) 20201001LU 02.09.2020

-28/45- LU 102057 applies here (22) 1 2+a arccoth (1+a) = == In mm 2 a In addition, the boundary conditions (3) and (9) shown above must be observed, ie the upper limit fset ,mand,max.

The mandatory setpoint ventilation frequency fsetmand is calculated according to the calculation rule (17) for the ideal | 5 mandatory target ventilation frequency fset,mand,id, unless an upper limit fset manda max leads to a lower value, cf. specification (12). This results in the following calculation rule for the mandatory target ventilation frequency fset mand: 23 ' (23) D1 Viareq 1 f = min { TTT A ml } set,mand 2*RC*arccoth (1+a) Vp x*t The lung time constant 7 becomes measured on patient P.

For this purpose, the signal processing unit 10 preferably evaluates measured values from the sensors 3 and and thereby determines the airway pressure Paw and the volume flow Vol'. From these signals, the signal processing unit 10 evaluates an estimate for the product R*C and uses this as the lung time constant 1 of the patient P.

To put it simply, an ideal step increase in pressure can be assumed when determining the lung time constant T, and the lung time constant can then be derived solely from the measured volume flow Vol'.

The procedure of determining the mandatory setpoint ventilation frequency fset,mand according to calculation rule (23) leads to a simple and robust specification of a setpoint value fset,mand for the ventilation frequency factual under mandatory ventilation.

In many cases, a lower target ventilation frequency fset,mand is calculated than with other methods, so that the procedure according to the invention leads to a lower load on the artificially ventilated lung Lu, especially in a patient P with a relatively small lung time constant T

- 29 / 45 - An advantage of the embodiment just described is that it is not necessary to measure the lung resistance R and the lung compliance C separately. In particular when the patient P is breathing spontaneously, this is associated with a relatively large degree of uncertainty. It is sufficient to measure the lung time constant 1, i.e. the product R*C. A sufficiently reliable value is generally obtained after a few spontaneous breaths by the patient P or ventilation strokes by the ventilator 1 . In contrast to other methods that automatically set a ventilation rate, the embodiment just described does not require optimization to be performed at runtime. In particular, it is not necessary to determine the required ventilation frequency at runtime in such a way that a function that is dependent on the ventilation frequency and belongs to a lung-mechanical model is minimized. In other methods, this function describes, for example, the work done during a ventilation stroke or the power that is applied during the ventilation strokes. If an optimization is required at runtime, a relatively high computing capacity and/or a relatively long computing time is required. In practice, an optimization is often carried out using an iterative process, which is terminated when a termination criterion is met. In some cases, the value found can be relatively far from an optimum.

The procedure just described also does not necessarily require a so-called maneuver to be carried out during the artificial ventilation, in which a parameter of the ventilator 1, for example a required time profile of the pressure or the volume flow, is specifically set to a different value for a short period of time to measure a vital parameter of the patient P. In particular, no occlusion is required, in which the artificial respiration is stopped for a short period of time and optionally also the spontaneous breathing of the patient P is stopped for this short period of time in order to measure the time-varying pressure caused by the patient P's spontaneous breathing will. 20201001LU 02.09.2020

| | - 30/45- In many cases, the procedure just described leads to the 17106081 mandatory ventilation being carried out with a ventilation frequency fact that is not higher than required to achieve the target alveolar minute volume V'a req.

For this reason, this procedure reduces the mechanical load on the lungs Lu of the patient P in many cases and therefore reduces the risk of the patient P's lungs Lu being mechanically damaged.

The target alveolar minute volume V'a req is specified for artificial respiration.

Procedures for doing this are given, for example, in J.

Fernandez, D

Miguelena, H.

Mulett, J.

Godoy, and F.

Martinon-Torres, "Adaptive support ventilation: State of the art review," Indian J.

grit

Gare Med., vol. 17, no. 1, p. 16, 2013.

The total dead space volume Vo is also determined at the beginning of the artificial respiration, preferably using one of the two methods described above.

The patient-side dead space volume Vp pat is preferably measured repeatedly during the entire artificial respiration, and a change in the patient-side dead space volume Vo,pat is thus detected and taken into account.

The factors a and x are preferably predefined once and are stored in the ventilator 1 .

In this embodiment, a constant const is determined and stored at the beginning of the artificial respiration of the patient P or beforehand, namely in the exemplary embodiment according to calculation rule (24) D1 1 const = min {——J = } 2*arccoth (1+a ) X This leads to the following calculation rule: (25) Viareq const f=min { Sa } set,mand Vp T In the embodiment described so far, a value for the factor a is specified.

It is also possible to specify n values 01, ..., an for the factor a.

The calculation just described is for each given factor a1, 20201001LU 02.09.2020

-31/45-A. ; LU102031 …, On carried out. This returns n values fset,mand(01), … fset,mand(An). A value that is calculated from these n values by suitable averaging or other aggregation, for example the smallest value, is used as the mandatory target ventilation frequency fset,mand.

The procedure described above shows a way of automatically deriving a setpoint ventilation frequency fset,mand for mandatory artificial ventilation of the patient P. For this purpose, a target alveolar minute volume Va req or proximal minute volume is specified. This setpoint ventilation frequency fsetmand is then used for the subordinate regulation described above when the patient P is completely sedated and therefore not breathing spontaneously. As already mentioned, a target alveolar minute volume V'A req is specified for the procedure. If the patient P is completely sedated, this minute volume V'a req is generated exclusively by ventilation (mandatory ventilation). This mandatory ventilation is carried out with the setpoint ventilation frequency fsetmand, which is defined as just described. However, it is also possible for the patient P to breathe spontaneously and apply a proportion of the desired alveolar minute volume Va req himself. If the _ alveolar minute volume, which the patient P brings up through his own spontaneous breathing, is denoted by V'a spon, then it applies. (26) SML = V'A,spon / V'A,req This factor SML is also determined and used. In one embodiment, a measure of the negative pressure Pmus, which is generated by the activity of the diaphragm Zw and the intercostal muscles of the patient P, is measured. Then both the airway pressure Paw, which is applied to the lungs Lu from the outside, and the negative pressure Pmus acting on the lungs Lu from the inside are known. The alveolar minute volume V'a spon that is exclusively achieved through the patient P's spontaneous respiration is derived from the two signals Paw and Pmus. Because the Paw signal is generated by superimposing the spontaneous respiration and the 20201001LU 02.09.2020

-32/45- artificial respiration causes the signal Prus solely from the HU102031 spontaneous respiration.

In one implementation, to measure a MaR for the negative pressure Pmus, the esophageal pressure Pes is measured.

This other embodiment assumes that the probe 3 is inserted into the esophagus Sp of the patient P and measures a measure of the esophagus pressure Pes.

In another embodiment, signals from the electrodes 5.1.1 to 5.1.2 are used to measure the electrical activity of the muscles of the patient P's respiratory system.

This electrical activity causes the patient P to breathe spontaneously and is not dependent on artificial respiration.

In another embodiment, the fact is exploited that when the spontaneous breathing of the patient P is supported by the processing device 1, spontaneous breaths of the patient P are detected and each detected sufficiently large spontaneous breath triggers a ventilation stroke of the processing device 1.

The stronger his spontaneous breathing, the more ventilation strokes of the ventilator 1 the patient P triggers.

This other configuration does not necessarily require determination of the alveolar minute volume V'a spon.

FIG. 6 and FIG. 7 illustrate two embodiments of how the measure for the intensity of the patient P's spontaneous respiration is determined.

At the beginning of the time period under consideration, the patient P is not breathing spontaneously, and the processing device 1 also does not carry out a ventilation stroke, so that the two pressures Paw and Pes approximately match at the beginning of the time period.

Subsequently, these two pressures differ from each other due to spontaneous respiration and/or artificial respiration.

In the embodiment according to FIG. 6, the work (work of breathing, WOB) expended during one breath for inhalation is used.

The pressure P in [mbar] is plotted on the x-axis, and the volume Vol in [I] flowing into and out of the lungs is plotted on the y-axis. The esophageal pressure Pes, which is measured using the probe 3, is used as a measure of the pressure generated by the diaphragm Zw of the patient P. 20201001LU 02.09.2020

-33/45- The curves shown in Figure 6 are obtained by superimposing a large number (0108081) of breaths, in the exemplary embodiment in which the respective median is formed, alternatively the mean value. The parameters shown are each taken as a normally distributed random variable. It is represented as Measurement uncertainty is the standard deviation, i.e. a tolerance band around the median.

In Figure 6: — PVes denotes the pressure-volume curve developed by the diaphragm Zw, i.e. the volume obtained as a function of the pressure Pmus, — PVaw denotes the pressure-volume curve developed by the ventilator 1, measured with the sensor 2 am Patient P's mouth and optionally the sensor 15 on the ventilator 1, - TL a dividing line passing through the intersection of PVes and PVaw, - WOBpPat the work expended by patient P in one inhalation by his spontaneous breathing, where the work is equal to the area between the pressure-volume curve PVes and the dividing line TL, — WOBvent is the work expended by the ventilator 1 in one inhalation by artificial respiration, where the work is equal to the area between the pressure-volume curve PVaw and the dividing line TL, and - std is the standard deviation around the median.

In one embodiment, the esophagus pressure Pes is measured as a measure of the pressure Pmus that the diaphragm Zw exerts, specifically with the aid of the probe 3 in the esophagus Sp of the patient P. In another embodiment, the time curves of the pressure Paw and of the volume flow Vol' at the mouth of the patient P is measured, preferably using the sensor 2 and optionally the sensor 15, and a measure for the pressure Pmus is derived from these time profiles using a lung model. Averaging over several inhalations is preferred. The total work WOB expended in one inhalation is the sum of WOBPat and WOBvent. The proportion of the work performed by patient P of the total work WOB is calculated and used as a measure of the proportion of spontaneous breathing of patient P, i.e. 20201001LU 02.09.2020

/ -34/45- LU102031 (27) SML = WOBPat / (WOBpat + WOBVvent). In the example in FIG. 6, the factor SML is approximately 41% in the example on the left, and only 13% in the example on the right. In the embodiment just described, an average work is preferably calculated. With this approach, the respiratory activity of the patient P is recorded quantitatively relatively well. In the embodiment according to FIG. 7, the average pressure P is used during an inhalation. PTP means "pressure-time product". The time t in [sec] is plotted on the x-axis, and the pressure in [mbar] for the two pressures Paw and Pes is plotted on the y-axis. The dividing line TL runs through the intersection of the curves of Paw and Pes. The area PTPpat between the time profile of the pressure Pes and the dividing line TL is a measure of the mean pressure Pmus applied by the patient P and correlates with the mechanical effort of spontaneous respiration. The area PTPvent between the time profile of the pressure Paw and the dividing line TL is a measure of the mean pressure Paw applied by the ventilator 1 and correlates with the mechanical performance of the artificial respiration. The following quotient is used as a measure for the portion of the patient P's spontaneous respiration: (28) SML=PTPpat/(PTPpat+PTPvent). In FIG. 7, the proportion of spontaneous respiration is significantly greater in the example shown above than in the example shown below. At the beginning of the artificial respiration, the ventilator 1 is operated in such a way that a desired course over time of the airway pressure Paw or of the volume flow Vol' is achieved during the artificial respiration. A target time profile is specified, and a subordinate control is performed with the aim that the actual time profile is equal to the target time profile. During artificial ventilation, the ventilator 1 performs ventilation strokes one after the other. A part of these ventilation strokes is triggered by the spontaneous breathing of the patient P, the rest automatically by the ventilator 1 in order to achieve the desired time profile of the pressure or volume flow. 20201001LU 02.09.2020

BE

| -35/45- | The factor S ; n / A . LU102031 he factor SML is calculated e.g. according to one of the calculation rules (26), (27), (28) or according to (29) SML = Nspon / Nges.

Here, nspon denotes the number of ventilation strokes that are triggered within a period of time by the patient's spontaneous breathing P, and nges denotes the total number of ventilation strokes carried out in this period.

In addition, an ideal frequency fspon of the patient P's spontaneous breathing is calculated.

Here, it is preferably assumed for the sake of simplicity that the spontaneous respiration of the patient P has a sinusoidal course over time.

If the patient P can take in the desired alveolar minute volume Va req solely through his own spontaneous respiration, then according to this simplification he breathes with the ideal frequency A,req Vo fspon = 2*n2*y This ideal frequency fspon requires the lowest power W' r + Wc to achieve the target alveolar minute volume V'a req through spontaneous breathing.

In other words, any other rate of spontaneous breathing requires more power.

In this calculation, the entire dead space volume Vo is taken into account, ie in addition to the patient-side dead space volume Vp pat also the device-side dead space volume Vb Ger.

A similar calculation rule is already in J.

Mead, "Control of respiratory frequency," J.

appl.

Physiol., vol. 15, no. 3, pp. 325-336, 1960.

Other calculation rules are also possible.

For example, an ideal frequency of spontaneous respiration is determined, which is 20201001LU 02.09.2020

-36/45- “ ; ; ; LL . LU102031 target alveolar minute volume V'a req Leads to minimum amplitude of time-varying generated pressure.

The setpoint ventilation frequency fset of the ventilator 1 is defined as follows: (31) fset=SML*fspon+(1-SML)*fset,mand Here fspon is the ideal frequency of spontaneous breathing, which is calculated according to the calculation rule (30), for example fsetmand the target ventilation frequency for mandatory ventilation and SML defined according to the invention is the factor that is defined, for example, according to the calculation rule (29) or using the esophagus pressure Pes according to the calculation rule (26) or according to (27) or (28). will. The design according to specification (31) avoids an abrupt transition between supportive and mandatory ventilation when the patient's spontaneous breathing changes. Rather, a gradual transition is achieved, with the mandatory ventilation becoming weaker the stronger the spontaneous breathing of the patient P is. If there is a deviation, it is ensured that the set ventilation frequency fset is at least as great as the ideal frequency fspon for spontaneous breathing. The deviating calculation rule is (32) Viareq const Viareq const f = min { — — } if — > fon and — ; set,mand Vp T Vp T fet, mand = fspon @NSONStEN This deviating calculation rule ensures that fset mand >= fspon and therefore also fset >= fspon applies. Even with this deviating calculation rule, the setpoint respiration frequency fset is determined according to the calculation rule (31). A parameter that can be set on the ventilator 1 is the target tidal volume Volrid set, ie the volume that is to be applied to the patient P's mouth during a breathing cycle. This is essentially correct 20201001LU 02.09.2020

-37/45- corresponds to the volume that is to be ejected from 1102081 ventilator 1 during a ventilation stroke. A subordinate control receives the target tidal volume Voltia,set as a specification and controls the ventilator 1 with the control aim that the tidal volume Voltidist actually achieved by the ventilation strokes is equal to the calculated target tidal volume Voltidset. The actual achieved tidal volume TidvoList is ejected into the fluid connection during an inhalation event and is shared between the achieved alveolar lung volume Vaact and the dead space in the fluid connection tubes and in the upper and middle airways of the patient P, with the total dead space having the volume Vo . The actual tidal volume Tidvoiact is intended to generate the required alveolar minute volume Va req and depends on the actual ventilation rate fact and the dead space volume Vp.

The inspiration time Ti and thus the duration of a ventilation stroke depend on the actual ventilation frequency fact and the ratio D1 as follows according to relationship (17): (33) D1 T= — fact It is possible to change the target tidal volume Volria,set by changing the connection (1) as follows: ¢ (34) Via req ey VOlyig set = f D set In one embodiment, the target tidal volume Volrid set is calculated according to the following calculation rule, which leads to a larger target tidal volume Voltia,set and depends on the duration Ti of an inspiration process and on the lung time constant T: 20201001LU 09/02/2020

LE

- 38/45 - 35 ‘ ‘ LU102031 (35) Viareq Vv A,req Fret feet VOoliig cet = A — = T, T, 1 -e RC 1 -e T Figure 8 illustrates the definition of the target ventilation frequency fset.

In this example, the following parameters were presented or measured: - The ideal weight Gewi of patient P is 75 ka. — The factor x is 5.5. — The required alveolar minute volume Va req is 5.1 l/min. — The volume of the dead space is 2.2 ml/kg. — The compliance C of the lungs is 50 ml/mbar. — The resistance R of the lungs is 5 mbar*sec/l. — Thus, the lung time constant is 1 = R*C = 250 msec.

The actual ventilation frequency f in [1/min] is plotted on the x-axis, and the resulting power in [W] on the y-axis. The ideal mandatory target ventilation frequency fset,mand,ia is derived according to calculation rule (21), which leads to a value fset,mand,ia = 21 / min.

This value is above the ideal frequency fspon for spontaneous breathing (fspon = 17 / min) and below the upper limit fset,mand,max for the mandatory target ventilation frequency, because V''areq / Vo = 28 / min and 1 / (x*T) = 43 / min.

Thus, according to calculation rule (32), fset mand=fsetmand,id=21/min is determined.

FIG. 8 also shows that ventilation frequency fminw which leads to a minimum total power W′R+W′c.

This ventilation frequency fminw can be calculated by minimizing.

As already explained, the method according to the invention avoids such a minimization.

As can be seen, the ventilation rate fminw for the minimum performance is greater than the ideal mandatory setpoint ventilation rate fset,mand id. 20201001LU 02.09.2020 Ds

-39/45- An exemplary procedure for automatically calculating values for various parameters LU102031 during artificial respiration is presented below.

This process comprises the following steps: - Factors a and x for mandatory ventilation and factor y for target ramp time tr set and inspiration proportion D1 are specified, for example by being stored in ventilator 1 - A target alveolar minute volume V' a req or proximal minute volume is determined and specified for the procedure.

— The total dead space volume Vo = Vb pat + Vp cer is determined. The volume Vp pat of the patient-side dead space is approximately measured using a CO2 sensor, for example using the Bohr formula. Or it is estimated using relation (4). The volume Vb cer of the device-side dead space is known with sufficient accuracy due to the construction of the ventilator 1 and the connection to the patient P or is negligibly small.

— Artificial respiration is started.

- The lung time constant 1 of the patient P is measured, for which the already mentioned volume flow sensor 2, 15 and at least one of the pressure sensors 2, 3, 7 are preferably used.

- A target ramp time tr,set for the pressure increase during a ventilation stroke is calculated, for example according to the calculation rule (16).

- The factor SML, ie the proportion of the desired alveolar minute volume V'a req that the patient P brings up through his spontaneous breathing, is determined, cf. relationship (26). For example, the SML factor is determined according to the relationship (29) or using the measuring probe 3, which measures the esophageal pressure Pes.

- An ideal frequency fspon of the spontaneous breathing of the patient P is calculated, for example according to the calculation rule (30).

- A setpoint ventilation frequency fset,mand for the mandatory ventilation is calculated, preferably according to calculation rule (23) or (25).

- A setpoint ventilation frequency fset, which takes into account the spontaneous breathing of the patient P, is calculated, preferably according to the calculation rule (31).

20201001LU 02.09.2020

- 40/45 - The duration Ti of an inhalation process and thus the duration Ti of a 10102081 ventilation stroke are calculated, preferably according to the calculation rule (33).

- A target tidal volume VolTia,set is defined, preferably according to the calculation rule (35).

The target ventilation frequency fset,mand calculated according to the invention, the specified target tidal volume VolTia,set, the derived target ramp time tr set and the specified inhalation proportion D1 result in a time profile of the target pressure Pset that the ventilator 1 should achieve.

- A subordinate control is carried out with the control aim that the actual course of the pressure achieved, for example measured as airway pressure Paw, is equal to the course of the pressure Pset.

In one embodiment, the calculated setpoint respiration frequency fset and the specified setpoint tidal volume Volridset are displayed on the display and control unit 12 . A user can confirm these values or overwrite a value with a manually entered value.

20201001LU 02.09.2020

- 41 / 45 - List of reference numbers LU102081 1 ventilator, artificially ventilates the patient P, includes the display and control unit 12 and the signal processing unit 10 2 pneumatic sensor in front of the patient P's mouth, measures the airway pressure Paw and optionally the volume flow Vol', acts as a the airway pressure sensor, includes the components 2.1 and 2.2

2.1 Sensor 2 transducer, samples air from the fluid connection between the patient P's lungs Lu and the ventilator 1 actual sensor 2 pressure sensor 3 Probe in the patient P's esophagus Sp, measures the esophagus pressure Pes AND optionally the gastric pressure Pga 4 Connector in the mouth of the patient P, connected to the measuring catheter 6 in the esophagus Sp

5.1.1, 5.1.2 | set of measuring electrodes close to the heart on the patient's skin

P

5.2.1, 5.2.2 | near-diaphragmatic set of measuring electrodes on the skin of the patient P measuring catheter in the esophagus Sp of the patient P, connected to the measuring probe 3 and the connector 4 7 gastric probe in the stomach Ma of the patent P, measures the gastric pressure Pga data-processing signal processing unit, receives signals from the Sensors 2, 3, 5.1.1 to 5.2.2, 15, calculates a target ventilation frequency fset 20201001LU 09/02/2020

- 42 / 45 - Display and control unit of ventilator 1 Sensor on ventilator 1, measures the volume flow Vol' factor by which the resistive power W'r should be greater than the elastic power Wc compliance of the patient's lungs, equal to Vol / AP D1 D1 = Ti / (Ti + Te), inhalation proportion, time proportion of the inhalation process in a ventilation process fist actual frequency of the patient's spontaneous breathing or artificial ventilation P fminw ventilation frequency, which leads to a minimum total power W'r+W'c fset calculated setpoint ventilation frequency for artificial ventilation of the patient P contains a component fsetmand for mandatory ventilation and a component dependent on fspon for supportive ventilation fset,mand setpoint ventilation frequency for mandatory ventilation by a target alveolar minute volume V'a req TO achieve is automatically calculated fset,mand id ideal mandatory set ventilation rate depends on the factor a ab upper limit for the mandatory target ventilation rate fspon ideal rate of spontaneous breathing of the patient P to achieve the target alveolar minute volume V'a req factor for determining the target ramp time tr set ideal weight of the patient P Ma | Stomach of patient P 20201001LU 09/02/2020

-43/45-

Nspon Number of ventilation strokes within a period | 4192091 be triggered by the patient P's spontaneous breathing

Nges Total number of ventilation strokes performed during this period Patient being artificially ventilated by ventilator 1 has the lungs Lu, the stomach Ma, the esophagus Sp and the diaphragm Zw

Pset Desired time profile of the pressure that is achieved during an inhalation process by a ventilation stroke of the ventilator 1

PTPpat Measure of the mean pressure generated by patient P's spontaneous breathing during an inhalation

PTPvent Measure of the average pressure generated by the artificial respiration of the ventilator 1 during an inhalation

PVaw Pressure-volume curve that the diaphragm Zw and the ventilator 1 apply together

Pressure-volume curve applied by the diaphragm Zw Pneumatic resistance of the patient's lungs, equal to AP / Vol'

SML factor that indicates the proportion of the alveolar minute volume V'a spon, which is generated by the patient P's spontaneous respiration, to the total target alveolar minute volume V/A req

Esophagus of patient P, probe 3 records lung time constant, as product R*C determines duration of exhalation (expiration)

Ti duration of inhalation (inspiration), at the same time duration of a

Ventilation hubs 20201001LU 09/02/2020

- 44 / 45 - TL Separating line that runs through the intersection of PVes and PVaw or 0102031 through the intersection of Pes and Paw tR is the ramp time achieved at the beginning of a ventilation stroke, which elapses until the full pressure Pset is reached Target ramp time VA , is actual alveolar minute volume in [l/min], air volume actually flowing into the alveolar lung space of the patient P V'A,req target alveolar minute volume in [l/min] during artificial ventilation, is specified VA spon alveolar minute volume in [l /min], which the patient P achieves through his own spontaneous breathing V'p is the actual proximal minute volume in [l/min], the volume of air actually flowing through the mouth of the patient P V'P req target proximal minute volume in [l/min] in artificial respiration, Vo dead space volume in the fluid connection between the lungs Lu of the patient P and the ventilator 1, volume of the dead space that occurs with each ventilation h ub is flowed through with air but is not used for exchange with the blood of the patient P, sum of device-side and patient-side dead space volume Vo,Ger device-side dead space volume, volume of the dead space in the area of the fluid connection outside the patient P VD , Pat patient-side Dead space volume, volume of dead space in the patient's upper and middle airway P Vol' Volume flow in the fluid connection, measured by sensor 3 and/or sensor 15 20201001LU 02.09.2020

- 45 / 45 - Volrid,is the actual tidal volume that LU102031 flows into the patient P's breathing system during an inhalation Volrid,set Target tidal volume that is to be applied to the patient's mouth during a breathing cycle, essentially equal to the volume that is to be ejected from the ventilator 1 during a ventilation stroke elastic work elastic power WOBPat by patient P with an inhalation of work expended by his spontaneous breathing WOBvent of ventilator 1 with an inhalation of work expended by the artificial ventilation X Factor for an upper limit of the mandatory target Ventilation rate fset mand We exp elastic work done at the end of an expiration We.ins elastic work done at the end of an inspiration resistive (viscous) work resistive power resistive work done at the end of an inspiration resistive work done at the end of an inspiration Diaphragm of the Pa client P 20201001LU 02.09.2020

Claims (15)

-1/10- Patent claims (0108081 1. A method for automatically calculating a target ventilation frequency (fset) for a ventilator (1), wherein a fluid connection is established or can be established between the lungs (Lu) of a patient (P) to be artificially ventilated and the ventilator (1), the The ventilator (1) is capable of performing ventilation strokes into the fluid connection as a function of the calculated target ventilation rate (fset), and the method comprises the steps of: - a measure of a target volume flow into the lungs (Lu) and out of the lungs (Lu) of the patient (P), in particular a required alveolar minute volume (V'a req) or a required proximal minute volume (V'P,req), - depending on measured values of at least one parameter of the patient (P), a lung time constant (1st ) is determined for the patient's lungs (P), - the volume (Vp) of a dead space in the fluid connection is determined, this dead space between the ventilator (1) and a n an area of the patient’s (P) lungs (Lu) suitable for gas exchange, target frequency (fsetmand) for the mandatory ventilation of the patient (P) by the ventilator (1) is calculated, ideal spontaneous breathing frequency (fspon) for the spontaneous breathing of the patient (P) is calculated, 20201001LU 09/02/2020 -2/10- this is a frequency with which the patient (P) achieves the desired volume flow (Va req, V'p req) solely through spontaneous breathing 12108081 — a measure (Nspon/Nges) for the intensity of the spontaneous breathing of the patient (P) is determined and — the target respiratory rate (fsetr) is calculated as the weighted mean of the mandatory target rate (fsetmand) and the ideal spontaneous respiratory rate (fspon), with the weighting (SML) of the determined measure being calculated during the averaging (Nspon / Nges) for spontaneous breathing. 2. The method according to claim 1, characterized in that the step of calculating the mandatory setpoint frequency (fsetmand) comprises the steps that - a required inhalation proportion (D1) is specified, the inhalation proportion (D1) being related to the time proportion of an inhalation process in an entire ventilation process and/or correlates with the ratio between the duration (Ti) of an inspiration process and the duration of a breathing process (Te), — an upper limit (fset,mandmax) for the mandatory target rate ( fset,mand) is calculated, an ideal mandatory target rate (fset,mand,ia) is calculated depending on the predetermined required inspiration fraction (D1), and — the mandatory target rate (fsetmand) is calculated using the ideal mandatory target rate (fsetmand,id) SO is that it is less than or equal to the upper bound (fset,mand,max), preferred as a minimum of ideal mandatoris cher setpoint frequency (fsetmand.id) and upper limit (fset,mand,max). 20201001LU 02.09.2020 -3/10- 3. The method according to claim 2, LU102081, characterized in that the step of calculating the ideal mandatory setpoint frequency (fsetmand,id) includes the steps that - a lung model is specified which approximates the pneumatic behavior of the lungs (Lu) of a human being (P) describes and - the ideal mandatory setpoint frequency (fsetmanaia) additionally as a function of a resistive power (Wr) which, according to the lung model, results from mandatory ventilation with this frequency (fsetmanaid) and an elastic power (W 'c), which results from the mandatory ventilation according to the lung model with a mandatory ventilation with this frequency (fsetmandid). 4. The method according to claim 3, characterized in that a power factor (1 + a) is specified, which is greater than or equal to 1 and preferably less than 1.2, the ideal mandatory target frequency (fsetmand, id) is calculated such that at of mandatory ventilation with this frequency (fset,mand,ia) the ratio of — the resistive power (Wr) that results from the mandatory ventilation according to the lung model, and — the elastic power (W'c) that results from the lung model the mandatory ventilation results, 20201001LU 02.09.2020 -4/10- is equal to the specified power factor (1+a). LU102031 5. The method according to claim 4 or claim 4, characterized in that a constant is calculated in an initialization phase using the lung model and the optional power factor (1+a) and in a subsequent usage phase the ideal mandatory target frequency (fsetmand,id) is used — the inspiration fraction (D1), — the determined lung time constant (1), and — the constant is calculated. 6. Method according to one of Claims 2 to 5, characterized in that the ideal mandatory reference frequency (fsetmand,id) is calculated in such a way that it is greater the greater the predetermined required inhalation fraction (D1). 7. The method according to any one of claims 2 to 6, characterized in that the upper limit (fset mand max) for the mandatory target frequency (fsetmand) additionally using the desired volume flow (V'a req) and / or the determined dead space volume (Vp) is calculated, preferably using the quotient of the target volume flow (V'A,req) and the dead space volume (Vo). 20201001LU 02.09.2020 -5/10- 8. The method according to any one of the preceding claims, LU102031, characterized in that the mandatory setpoint frequency (fsetmand) for mandatory ventilation is calculated in such a way that it is greater than or equal to the ideal spontaneous breathing frequency (fspon) for spontaneous breathing. 9. The method according to any one of the preceding claims, characterized in that as a measure (Nspon / Nges) for the intensity of the patient's spontaneous breathing (P) is determined, which proportion of the mechanical work (WVOBPat) or the mechanical power, which the spontaneous Patient's (P) respiration during an inspiration breath contributes to the total work done (WOBpat+WOBvent) during the inspiration breath. 10. The method according to any one of the preceding claims, characterized in that depending on - the desired volume flow (V'A,req, V'P req), - the determined dead space volume (Vp) and - the calculated target ventilation frequency (fset ) a target tidal volume (VolTiaset) is calculated, which is a value for the volume that the ventilator (1) should feed into the fluid connection during a ventilation stroke, and the ventilator (1) is controlled with the control aim that at least one ventilation stroke the generated tidal volume (Volrid,actual) is equal to the calculated target tidal volume (VolTia,set). 20201001LU 02.09.2020 -6/10- LU102031 11. The method according to any one of the preceding claims, characterized in that depending on the determined lung time constant (1) a required ramp time (ir set) is determined for at least one ventilation stroke, that is the period of time that elapses at the beginning of a ventilation stroke until the maximum pressure has been reached, with which the ventilator (1) performs the ventilation stroke, and the ventilator (1) is controlled with the control aim that with at least one ventilation stroke, preferably every ventilation stroke, the ramp time achieved (frist) is equal to the calculated required ramp time (tr set) is. 12. Signal processing unit (10), which is designed to automatically calculate a setpoint ventilation frequency (fset) for a ventilator (1), with a fluid connection between the lungs (Lu) of a patient (P) to be artificially ventilated and the ventilator (1) can be manufactured or manufactured, the ventilator (1) being designed to carry out ventilation strokes into the fluid connection as a function of the calculated target ventilation frequency (fse), the signal processing unit (10) comprising a data memory or having at least intermittent read access to a data memory, a measure of a desired volume flow into the lungs (Lu) of the patient (P), in particular a required alveolar minute volume (V'A req) or a required proximal minute volume (V'P,rea), being stored in the data memory, wherein the signal processing unit (10) is designed to 20201001LU 02.09.2020 -7/10- — receiving measured values from at least one patient sensor (2, 3, 7) for 10106091, the patient sensor (2, 3, 7) being able to measure at least one parameter of the patient (P), — dependent determining a lung time constant (1) for the patient's lungs (P) from the measured values of the or at least one patient sensor (2, 3, 7), - determining the volume (Vp) of a dead space in the fluid connection, this dead space occurs between the ventilator (1) and the lungs (Lu) of the patient (P), - depending on the stored target volume flow (V'A req, V'Preq), the determined lung time constant (1) and the determined dead space volume ( Vo) to calculate a mandatory setpoint frequency (fset mana) for the mandatory ventilation of the patient (P) by the ventilator (1), - depending on the desired volume flow (V'a), the determined lung time constant (T) and the determined dead space volume (Vp) an ideal spontaneous breathing frequency (fspon) for d To calculate the spontaneous respiration of the patient (P), that is a frequency with which the patient (P) achieves the desired volume flow (V'A,rea, V'P,req) solely through spontaneous respiration - once (nspon / Nges) for the intensity of the patient's spontaneous breathing (P) and — calculate the target respiratory rate (fse) as the weighted mean of the mandatory target rate (fsetmand) and the ideal spontaneous respiratory rate (fspon), with the averaging the weighting (SML) depends on the determined measure (nspon / Nges) for spontaneous respiration. 13. Respirator (1) for artificial respiration of a patient (P), 20201001LU 09/02/2020 -8/10- wherein a fluid connection between the lungs (Lu) of a patient (P) to be artificially -/102081 ventilated and the ventilator (1) can be established or established, the ventilator (1) at least temporarily having at least one patient sensor (2, 3, 7) which is capable of measuring at least one parameter of the patient (P), the ventilator (1) comprising a data memory or having at least intermittent read access to a data memory, the data memory being stored once for a desired volume flow into the lungs (Lu) of the patient (P), in particular a required alveolar minute volume (V'A req) or a required proximal minute volume (V'P,rea), is stored, the ventilator (1) being configured for this is, - to calculate a set ventilation rate (fset) and - to perform ventilation strokes into the fluid connection depending on the calculated set ventilation rate (fset), the ventilator (1) is designed to - receive measured values from at least one patient sensor (2, 3, 7), the patient sensor (2, 3, 7) being able to measure at least one parameter of the patient (P), - to determine a lung time constant (1) for the patient's lungs (P) depending on the measured values of the or at least one patient sensor (2, 3, 7), - to determine the volume (Vo) of a dead space in the fluid connection, this dead space occurs between the ventilator (1) and the lungs (Lu) of the patient (P), — depending on the stored target volume flow (Va req, V'Preq), the determined lung time constant (1) and the determined dead space volume (Vb ) a mandatory target frequency (fset,mand) for the 20201001LU 02.09.2020 -9/10- to calculate the mandatory ventilation of the patient (P) by the ventilator LU102031 (1), — depending on the desired volume flow (V'a), the determined lung time constant (1) and the determined dead space volume (Vp). to calculate the ideal spontaneous breathing frequency (fspon) for the spontaneous breathing of the patient (P), that is a frequency with which the patient (P) achieves the desired volume flow (Va req, V'P req) solely through spontaneous breathing, - to determine a measure (nspon/Nges) for the intensity of the patient's spontaneous breathing (P) and — to calculate the target ventilation rate (fset) as the weighted mean of the mandatory target rate (fsetmand) and the ideal spontaneous ventilation rate (fspon), with the averaging, the weighting (SML) depends on the determined measure (nspon / Nges) for spontaneous respiration. 14. Computer program which can be executed on a signal processing unit (10) and when executed on the signal processing unit (10) when the signal processing unit (10) receives measured values from at least one patient sensor (2, 3, 7), causes that the signal processing unit performs a method according to any one of claims 1 to 11. 15. Signal sequence, which can be received and executed by a signal processing unit (10) and 20201001LU 02.09.2020 - 10/10 - when executed on the signal processing unit (10), when the LU102031 signal processing unit (10) receives readings from at least one patient sensor (2, 3, 7), the signal processing unit causes a method according to any one of the claims 1 to 11 performed. 20201001LU 02.09.2020