DE102015011390A1 - Apparatus, method and computer program for ventilating a patient - Google Patents

Abstract

Embodiments provide an apparatus, a method, and a computer program for ventilating a patient. The device (10) for ventilating a patient (20) comprises a detection device (12) for detecting an electromyographic signal of the patient and a pneumatic signal and a ventilation device (14). The device (10) further comprises a control device (16) for controlling the detection device (12) and the respiration device (14), wherein the control device (16) is designed to display based on the electromyographic signal information about a frequency-based Fatigueindex and / or to determine information about an entropy-based fatigue index. The control device (16) is further configured to determine information about a current fatigue index of the patient (20) based on the information on the frequency-based fatigue index and / or the information on the entropy-based fatigue index as a function of the pneumatic signal. The control device (16) is also designed to set the ventilation device (14) as a function of the current fatigue index.

Description

Embodiments relate to a device, a method, and a computer program for ventilating a patient, particularly but not exclusively, to a concept of ventilator control of a patient based on combined patient fatigue indexes.

In the conventional technique, various concepts are known which provide a degree of fatigue of the musculature. Muscles can tire, which is also called "fatigue" (English "fatigue"). Fatigue is by definition considered to occur when a muscle is no longer able to apply a specific force. For example, in the context of ventilation, respiratory fatigue can be problematic, especially in patients who are long ventilated artificially. An oncoming fatigue is related to various physiological effects in the muscle. Among other things, changes the conduction velocity of the electrical impulses on the muscle fibers (also English "Conduction Velocity (CV)"). In addition, other effects in connection with an impending fatigue, such. As a change in the recruitment of the muscles occur.

The physiological changes associated with the onset of fatigue can also cause changes in the electrical potentials caused by the muscles. Therefore, a Fatigue indicator can be calculated in several ways from ElectroMyoGraphic Signals (EMG). Possible ways of calculating a fatigue index from EMG are eg. Spectral estimation quantification of frequency compression: center / mean / median frequency (also mean, average or median), H / L ratio (ratio of powers in a high and a low frequency band), higher order spectral moments, wavelet based (from English "wavelet", wave train) method, estimation of signal entropy (eg English "sample entropy", sufficient entropy, or English "fuzzy approximate entropy", out of focus estimated entropy), engl. "Recurrence Quantification Analysis (RQA)", quantified recurrence detection, fractal procedures, etc.

In contrast to fatigue, which by definition is present or not, the indices derived from EMG data indicate a continuous degree of fatigue. In the following, the fatigue index is assumed to be more indicative of greater fatigue. For example, in the case of classical fatigue determination using mean frequency, a transformation can be performed. For example, in the absence of fatigue, a high mean frequency is measured and mapped to a low fatigue index. Existing fatigue results in a lower mean frequency leading to a higher fatigue index. The document US 8,256,419 B2 describes a method in which the ventilatory assistance (English "ventilatory assist") is regulated so that critical thresholds are not exceeded. The critical thresholds relate to various respiratory related features, including, but not limited to, the signal amplitude of EMG signals and fatigue indices derived therefrom. The threshold is calculated to produce fatigue when the threshold is reached. Here, an absolute threshold is used, which, however, can hardly be determined because of the individual differences in the physiology / anatomy of the subjects.

Kroworsch, Dissertation 2001 "Examination of the Frequency Spectrum of the Surface Electromyogram of the Diaphragm for the Early Detection of Muscle Fatigue", LMU Munich: Faculty of Medicine has not been able to establish an absolute limit value for fatigue in all subjects in a study with healthy volunteers. In contrast, relative changes in both center frequency and H / L ratios gave good evidence of fatigue for an individual subject.

In the publication US 20140123979 A1 it's about recognizing fatigue. For this purpose, a large number of parameters are taken into account (among many others also EMG based values). When fatigue is detected as having occurred, various steps are taken by the device. This includes warnings, hints and also switching back to a mandatory ventilation mode or a rest phase. Again, this system is essentially based on deciding whether fatigue is present or not.

• • E. J. W. Maarsingh, L. A. van Eykern, A. B. Sprikkelman, M. O. Hoekstra and W. M. C. van Aalderen, J Appl Physiol, ”Respiratory muscle activity measured with a noninvasive EMG technique: technical aspects and reproducibility”, 88: 1955–1961, 2000 ,
• • Miriam Gonzalez-Izal, Armando Malanda, Esteban Gorostiaga, and Mikel Izquierdo, ”Electromyographic models to assess muscle fatigue”, J Electromyogr Kinesiol, 22(4): 501–512, August 2012 , und
• • Dokument US 20140123979 A1 ”Systems and methods for monitoring, managing, and preventing fatigue during Ventilation”.

Further background information can be found in the following references:

• • EJW Maarsingh, LA van Eykern, AB Sprikkelman, MO Hoekstra and WMC van Aalderen, J Appl Physiol, "Respiratory muscle activity measured with a noninvasive EMG technique: technical aspects and reproducibility", 88: 1955-1961, 2000 .
• • Miriam Gonzalez-Izal, Armando Malanda, Esteban Gorostiaga, and Mikel Izquierdo, "Electromyographic Models to Assess Muscle Fatigue," J Electromyogr Kinesiol, 22 (4): 501-512, August 2012 , and
• • Document US 20140123979 A1 "Systems and methods for monitoring, managing, and.

There is therefore a need to provide an improved concept for ventilating a patient.

This need is met by embodiments of a device according to claim 1, a method according to claim 17 and a computer program according to claim 19.

Embodiments of the present invention are based on the core idea of determining a current fatigue index based on a frequency-based fatigue index and / or an entropy-based fatigue index as a function of a pneumatic signal. Ventilation support for a patient may then be adjusted based on the current fatigue index. Embodiments may continuously determine a more reliable fatigue index and use it to adjust a patient's ventilation.

Embodiments provide a device for ventilating a patient with a detection device for detecting an electromyographic signal of the patient and a pneumatic signal. The device further comprises a ventilation device and a control device, which is designed to control the detection device and the ventilation device. The control device is furthermore designed to determine information on a frequency-based fatigue index and / or information about an entropy-based fatigue index based on the electromyographic signal. The control device is further configured to determine information about a current fatigue index of the patient based on the information on the frequency-based fatigue index and / or the information on the entropy-based fatigue index as a function of the pneumatic signal. The control device is also designed to set the ventilation device as a function of the current fatigue index.

Embodiments can thus enable control / adjustment / regulation of the continuous fatigue index to a constant value or predefinable value. Embodiments may thus consider a relative change in a fatigue index / fatigue indexes. In some embodiments, accordingly, a regulation about the patient can also be carried out. The controller may then be configured to correct a difference between information about a patient-specified predetermined fatigue index and information about the current fatigue index based on the pneumatic signal and the ventilator over the patient. For example, in some embodiments, it may be regulated so that limit values are not exceeded and, while maintaining an initial fatigue level or fatigue index, may also be established without a threshold.

In some further embodiments, the detection means comprises a sensor element adapted for invasive, for example with a esophageal catheter, or surface detection of the electromyographic signal on the outer skin surface or by means of the outer skin surface of the patient. Some embodiments may thus enable noninvasive or superficial detection of the electromyographic signal. For example, the electromyographic signal may be a differential surface-detected signal that includes information about a respiratory activity of at least one patient's respiratory muscle. From the information about the respiratory activity, a setting for a ventilation support can then be selected in exemplary embodiments and, for example, a control or regulation can be set up.

In some other embodiments, the detection means may be configured to detect the pneumatic signal as a patient's respiratory cycle typical breathing signal. Under certain circumstances, exemplary embodiments can thus enable a somewhat complex or possibly natural detection of the pneumatic signal. The detection device can be designed, for example, to detect the pneumatic signal as a pressure signal or as a volume flow signal to the patient. Embodiments may thus allow the use of respiratory masks with standard components, sensors. The detection device may be formed, at least in some embodiments, to detect the pneumatic signal as an expiratory and / or inspiratory breathing signal of the patient. Depending on the further signal processing concept, exemplary embodiments may thus permit an evaluation of the patient's expiratory and / or inspiratory breathing signals.

The device may further comprise, in further embodiments, an interface for obtaining information about the patient-specified predetermined fatigue index, and the controller may be configured to obtain the information about the predetermined fatigue index over the interface from a user of the device. Exemplary embodiments may thus allow, for example, a doctor or clinical staff to specify or also to adapt a fatigue index tailored to the respective patient. In some embodiments, the interface may also serve a Adjustment mode for the fatigue index to be specified.

For example, the controller may be further configured to provide the information about the current fatigue index at a first rate, e.g. As a sampling rate, and make an adjustment of the predetermined fatigue index at a second rate, the second rate is lower than the first rate.

Some embodiments can thus allow an adjustment of the setpoint, be it manually via the interface or automated. For example, if the measured values over a certain duration or number of breathing cycles were above the default / setpoint, the predetermined fatigue index can also be tracked at least in some embodiments. In some embodiments, the control can also try to return to the old level, so depending on the controller type set minimum support. With the steady-state controller, when the condition improves, support will quickly be minimal until the patient returns to, or undercuts, the initial level. In this respect, many embodiments for a setpoint adjustment are conceivable in embodiments, wherein the possibilities of no adjustment, manual adjustment, automatic adjustment with predeterminable adjustment mode / speed, can extend to fully automatic adjustment.

In some embodiments, the controller may be configured to determine the information about the frequency-based fatigue index based on spectral estimation and quantification of frequency compression of the electromyographic signal. Embodiments may thus enable an evaluation of the information contained in the frequency spectrum of the electromyographic signal. The controller may further be configured to determine the information about the entropy-based fatigue index based on a signal entropy of the electromyographic signal. Embodiments can thus enable an evaluation of the information contained in the entropy of the electromyographic signal.

In further embodiments, the control device may be configured to detect a heart signal component based on an electromyographic signal, and to reduce or suppress an influence of the cardiac signal component in the electromyographic signal in order to obtain an intermediate signal. Embodiments can thus reduce the influence of the cardiac signal component, which under certain circumstances can also be regarded as a disturbance variable. The intermediate signal may in some embodiments be a second intermediate signal detected after a first intermediate signal and the control device may further be configured to reduce a DC component in the electromyographic signal to obtain the first intermediate signal for further processing. The control device can also be designed; to detect the heart signal component by means of the first intermediate signal.

The control device is further configured in some embodiments to determine a power signal as an envelope of the second intermediate signal and to hide and interpolate portions of the cardiac signal component in the power signal. In this respect, embodiments may provide a continuous or discrete signal adjusted or reduced by the cardiac signal component for further processing. Further, the controller may be configured to determine information about inspiratory and expiratory phases based on the pneumatic signal to determine information about an inspiratory muscle activity of the patient from the pneumatic signal in an inspiratory phase and the frequency-based and / or the entropy-based fatigue index essentially for that phase which is determined to be a phase of inspiratory muscle activity of the patient. Determination of the fatigue index can thus be based on signal components of an inspiratory phase of the muscle activity of the patient, which can promote the reliability of the fatigue indices determined in this way.

• – Ein Qualitätsmaß über das pneumatische Signal, das angibt, zu welchem Grad in dem elektromyographischen Signal Information über eine Fatigue zu erwarten ist,
• – ein Qualitätsmaß für die Herzsignalkomponente, das angibt zu welchem Grad das Leistungssignal während einer inspiratorischen Phase durch Herzsignalreste beeinflusst ist,
• – ein Qualitätsmaß für die Hüllkurve, das angibt zu welchem Grad eine Muskelaktivität des Patienten während einer inspiratorischen Phase vorliegt,
• – ein Qualitätsmaß für den frequenzbasierten Fatigueindex, das einen Grad der Zuverlässigkeit des frequenzbasierten Fatigueindex angibt, und/oder
• – ein Qualitätsmaß für den entropiebasierten Fatigueindex, das einen Grad der Zuverlässigkeit des entropiebasierten Fatigueindex angibt.

In some embodiments, the controller may be configured to determine at least two or more of the following quality measures:

• A quality measure of the pneumatic signal indicating to what extent fatigue information is to be expected in the electromyographic signal,
• A quality measure for the cardiac signal component, which indicates to what extent the power signal during an inspiratory phase is influenced by heart signal residues,
• An envelope quality measure indicating to what degree a patient's muscle activity is present during an inspiratory phase,
• A quality measure for the frequency-based fatigue index, which indicates a degree of reliability of the frequency-based fatigue index, and / or
• - A quality measure for the entropy-based fatigue index, which indicates a degree of reliability of the entropy-based fatigue index.

Embodiments may thus weight the signal processing components based on their quality or reliability, which may increase the overall quality or reliability of the particular fatigue index. The controller may be further configured to perform a combination of the frequency-based fatigue index and the entropy-based fatigue index based on at least two of the quality measures such that a higher quality measure requires a higher relative weighting in the combination. Embodiments may thus give greater weight to more reliable fatigue indexes.

Embodiments further provide a method of ventilating a patient. The method includes detecting a patient electromyographic signal and a pneumatic signal. The method further comprises determining information about a frequency-based fatigue index based on the electromyographic signal, and / or determining information about an entropy-based fatigue index based on the electromyographic signal. The method further includes determining information about a current fatigue index of the patient based on the frequency-based fatigue index information and / or the entropy-based fatigue index information. The method includes adjusting the ventilation of the patient as a function of the current fatigue index.

In some embodiments, adjusting ventilation includes adjusting a difference between information about a predetermined fatigue index matched to the patient and information about the current fatigue index about or through the ventilation.

Another embodiment is a computer program for performing at least one of the methods described above when the computer program runs on a computer, a processor, or a programmable hardware component. Another embodiment is also a digital storage medium that is machine or computer readable and that has electronically readable control signals that can cooperate with a programmable hardware component to perform one of the methods described above.

Further advantageous embodiments will be described below with reference to the embodiments illustrated in the drawings, to which embodiments are generally not limited in total. Show it:

1 an embodiment of a device for ventilating a patient;

2 a further embodiment of a device for ventilating a patient;

3 a block diagram of a control in an embodiment of a device for ventilating a patient;

4 a further embodiment of a device for ventilating a patient;

5 a block diagram of a signal processing chain in an embodiment of a device for ventilating a patient;

6 Waveforms in a signal processing chain in one embodiment of a device for ventilating a patient; and

7 a block diagram of a flowchart of a method for ventilating a patient.

Various embodiments will now be described in more detail with reference to the accompanying drawings, in which some embodiments are illustrated.

In the following description of the attached figures, which show only some exemplary embodiments, like reference characters may designate the same or similar components. Further, summary reference numerals may be used for components and objects that occur multiple times in one embodiment or in a drawing but are described together in terms of one or more features. Components or objects which are described by the same or by the same reference numerals may be the same, but possibly also different, in terms of individual, several or all features, for example their dimensions, unless otherwise explicitly or implicitly stated in the description.

Although embodiments may be modified and changed in various ways, exemplary embodiments are illustrated in the figures as examples and will be described in detail herein. It should be understood, however, that it is not intended to limit embodiments to the particular forms disclosed, but that embodiments are intended to cover all functional and / or structural modifications, equivalents and alternatives that are within the scope of the invention. Like reference numerals designate like or similar elements throughout the description of the figures.

Note that an element referred to as being "connected" or "coupled" to another element may be directly connected or coupled to the other element, or intervening elements may be present. Conversely, when an element is referred to as being "directly connected" or "directly coupled" to another element, there are no intervening elements. Other terms used to describe the relationship between elements should be interpreted in a similar manner (eg, "between" versus "directly in between,""adjacent" versus "directly adjacent," etc.).

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the embodiments. As used herein, the singular forms "a," "a," "an," and "the" are also meant to include the plural forms unless the context clearly indicates otherwise. Furthermore, it should be clarified that the printouts such. As used herein includes the presence of said features, integers, steps, operations, elements, and / or components but does not preclude the presence or addition of one or more features, integers, steps, operations, elements, components, and / or groups thereof.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly assigned to one of ordinary skill in the art to which the embodiments pertain. Furthermore, it should be clarified that expressions, e.g. For example, those defined in commonly used dictionaries are to be interpreted as having the meaning consistent with their meaning in the context of the relevant art and not to be interpreted in an idealized or overly formal sense, as long as this is so not expressly defined herein.

1 illustrates an embodiment of a device 10 for ventilation of a patient 20 , The device 10 comprises a detection device 12 for acquiring an electromyographic signal of the patient 20 and a pneumatic signal. The detection device 12 For example, it may include one or more sensors, sensor elements, electrodes or electrode pairs, such as needles, fine wire, electrodes, surface electrodes, pressure sensors, flow meters, flow sensors, gas concentration sensors, invasive esophageal catheters, etc. that provide corresponding sensor signals or information. In some embodiments, the respective sensors may optionally be coupled with appropriate evaluation and processed accordingly, z. B. amplified or filtered signals to provide.

The device 10 further comprises a ventilation device 14 , The ventilation device 14 For example, it may comprise a respirator or a respirator, to which a certain degree of ventilatory support can be specified. The ventilation device 14 may in some embodiments to a patient 20 be coupled by means of pneumatic structures, such as hoses, masks, valves, branches, etc. As the 1 further shows, the device comprises 10 In addition, a control facility 16 connected to the detection device 12 and the ventilator 14 is coupled. The control device 16 is to control the ventilator 14 and the detection device 12 educated. The control device 16 is also designed to determine information on a frequency-based fatigue index and / or information about an entropy-based fatigue index based on the electromyographic signal. The control device 16 is further configured to provide information about a patient's current fatigue index 20 based on the information on the frequency-based fatigue index and / or the information on the entropy-based fatigue index as a function of the pneumatic signal. The control device 16 is further adapted to the ventilation device 14 depending on the current fatigue index. In the following, some embodiments are described in which a combination of the fatigue indices takes place. The control device 16 is then designed to determine the frequency-based fatigue index information and the entropy-based fatigue index information based on the electromyographic signal. The control device 16 is then further trained to receive information about the patient's current fatigue index 20 based on a combination of the information on the frequency-based fatigue index and the information on the entropy-based fatigue index as a function of the pneumatic signal. In further exemplary embodiments, the determination of only one fatigue index is also conceivable, from which the current fatigue index is then determined as a function of the pneumatic signal.

In embodiments, the control device 16 correspond to any controller or processor or programmable hardware component. For example, the control device 16 also be implemented as software that is programmed for a corresponding hardware component. In that sense, the control device 16 be implemented as programmable hardware with appropriately adapted software. Any processors, such as digital signal processors (DSPs) can be used. Embodiments are not limited to a particular type of processor. There are any processors or even multiple processors to implement the control device 16 conceivable.

In a further embodiment, the control device 16 further adapted to provide a difference between information about one of the patient 20 agreed preset fatigue index and the information about the current fatigue index based on the pneumatic signal and the respiratory device 14 about the patient 20 auszuregeln. In general, embodiments can thus settings for the ventilation device 14 , Provide ventilation controls and / or ventilation controls. The 1 illustrates another embodiment that is an optional interface 18 is extended, shown with dashed lines. In this embodiment, the device further comprises the interface 18 To get information about the patient 20 includes a coordinated predetermined fatigue index. The control device 16 is then trained to get the information about the given fatigue index via the interface 18 z. B. by a user of the device 10 to obtain.

the interface 18 For example, it may correspond to one or more inputs and / or one or more outputs for receiving and / or transmitting information, such as digital bit values, analog signals, magnetic fields based on code, within a module, between modules, or between modules of different entities , the interface 18 but can also be an input interface 18 such as a control panel, a switch or rotary switch, a button, a touch-sensitive screen (English "touch screen"), etc. The interface 18 thus allows the recording, possibly also receiving or input, a piece of information about the specifiable on the patient matched Fatigueindex. Such input or specification can be done for example by a doctor or hospital staff.

In some embodiments, the predetermined fatigue index can also be tracked or adjusted, for example in the context of a setpoint tracking for a downstream ventilation control.

The electromyographic signal, or else the electromyographic derivative, can be invasive or noninvasive. In this case, for example, the EMG activity of one or more respiratory muscles or respiratory auxiliaries, z. As the diaphragm, the intercostal musculature (lat. Musculus intercostalis) and / or the respiratory aid muscles are removed or detected. Possible positions for electrodes in the non-invasive case arise z. B. from the aforementioned document from Maarsingh et al. The detection device 12 may then comprise one of the above-mentioned sensor elements, which for invasive or non-invasive, so superficial, detection of the electromyographic signal on the outer skin surface or by means of the outer skin surface of the patient 20 is trained. For example, surface electrodes or electrode pairs may be attached to the skin surface of the patient to detect the electromyographic signal. This may offer advantages over, for example, an esophageal catheter. The signal can then be detected on the outer surface of the skin, ie on the outside of the patient, outside the body openings such as mouth, nose, ears, rectum, without intubating. In a further embodiment, which will be explained in more detail below, the electromyographic signal is a differential surface-detected signal which comprises information about a respiratory activity of at least one respiratory muscle of the patient.

In a further embodiment, the detection device 12 designed to provide the pneumatic signal as a respiratory cycle typical of the human respiratory cycle of the patient 20 capture. Under such a breath signal here is a signal understood, the information about the breath, such as pressure, volume or concentrations, such as oxygen O ₂ , carbon dioxide, CO ₂ , water H ₂ O; etc. includes.

2 shows a further embodiment of a device 10 for ventilation of a patient 20 , In the embodiment of 2 is the control device 16 in two functional blocks 16a and 16b subdivided, as may occur, for example, in a hardware or software implementation. The determination or calculation of the Fatigueindex takes place in the block 16a instead, the English also. may be referred to as the "fatigue index calculation", and the determination is based on the pneumatic signal and the EMG signal. The detection device 12 Here is also divided into two parts in an EMG acquisition block 12a (also Engl. "EMG recording") for the detection of the electromyographic signal and a detection block 12b for detecting the pneumatic signal entering the ventilator 14 (also English "fan") is integrated. In the present embodiment, the detection device 12b configured to the pneumatic signal as a pressure signal or as a volume flow signal to the patient 20 to to capture. The detection of such signals can be low cost within the ventilator 14 can be done, as related sensors also from a respiratory mask or breathing tubes, which are intended for direct contact with the patient, removed, for example, to components within a ventilator, can be arranged. Compared with sensors which are arranged close to the patient, for example on the respiratory mask, on respiratory hoses or on any points or Y-pieces, savings can be achieved by exemplary embodiments. Alternatively, the EMG detection and possibly also the fatigue index determination can be arranged outside the respirator, for example within a patient-oriented EMG module. This would have the advantage of wider usability of the EMG technology, e.g. B. on different device platforms. The 2 also shows the ventilation control 16b (also called "ventilation assist controller"), based on the Fatigueindex Settings (also called "ventilatory assist" information) at the ventilation device 14 performs.

The device 10 from the embodiment of 2 Carries out a procedure for automatically adjusting the level of ventilatory assistance in assisted ventilation. 3 shows a block diagram of a control of another embodiment of the device 10 for ventilation of a patient 20 , The 3 shows the patient on the right 20 at the about the EMG acquisition 12a (EMG recording) recorded the EMG signal and the fatigue index determination 16a ("Fatigueindex recording") is forwarded. The patient 20 is to the ventilator 14 (Engl. "fan") connected, which also a capture 12b provides for the pneumatic signal. Compared to the embodiment of 2 is the ventilation control 16b ("ventilation assist controller") here divided into a collection 16b1 (English: "initial fatigue index") for the initial or predefinable fatigue index and a controller / regulator 16b2 (engl. "controller"), which is a difference between that of the fatigue index determination 16a certain fatigue index and the initial fatigue index 16b1 via the ventilation device 14 corrects.

In some embodiments, for the following fatigue index calculation, there may be a conditioned EMG signal in which artifacts not caused by the respiratory muscles have been reduced or suppressed. Such artifacts include, for. In addition, the EMG signal can be segmented to take into account inspiration and expiration differently in the following steps. This makes it possible to observe the fatigue index only for the periods when there is actually a contraction of the corresponding muscles. The fatigue index indicates the degree of fatigue. It is calculated from the prepared EMG signals. As mentioned above, different methods are available. In addition to spectral methods that quantify the frequency shift, there are various non-linear methods such. Entropy, RQA or fractal methods. An overview of known fatigue algorithms can be found z. B. in the aforementioned document from Miriam Gonzalez-Izal et al.

In the calculation of the respiratory fatigue index, in some embodiments, other peculiarities of the respiratory muscles can be considered. These include z. For example, the (temporal) cyclic contraction form, disorders by cardiac artifacts and crosstalk from adjacent muscles. The fatigue calculation is performed z. Epoch wise. In contrast to the epoch size of one second, which is widely used in the literature, in embodiments much shorter epochs can be used, for example. B. 0.05, 0.1, 0.15, 0.2 s. In epochs with major disturbances, the calculation of the fatigue index can be dispensed with. Due to the small epoch size, the raw fatigue signal in some embodiments is very noisy and can therefore be subjected to a smoothing. Smoothing may include cyclicity or periodicity of the respiratory effort, the amplitude of the EMG signal, the respiratory phase, the presence of artifacts, and the quality of the EMG signal. In a further embodiment, the detection device 12 formed to the pneumatic signal as an expiratory and / or inspiratory breathing signal of the patient 20 in order to restrict a ventilation assistance control to possibly only one of these phases, if necessary to combine values from both phases.

Some embodiments provide a system of EMG dissipation 12 . 12a , Fatigue index calculation 16 . 16a , Controller 16 . 16b . 16b2 and controlled / controlled assistive ventilation 14 in which the degree of support is regulated so that the fatigue index of the patient 20 is kept constant. In some embodiments, a calculation of fixed thresholds as the limit to the occurrence of fatigue can be dispensed with. Instead, the design, at least in some embodiments, ensures that an initial detected fatigue condition is maintained. This initial fatigue condition is the condition at the time the method / device, procedure, or control is started. In some embodiments, it is the task of the treating clinician by appropriate ventilation support to ensure that the Patient is in a steady state at this time, which is to be maintained. The initial state of fatigue should be selected by the clinician so that the ventilation situation neither leads to "true" fatigue nor to the atrophy of the musculature. As the patient's state of health hopefully changes for the better, with constant ventilation support, the state of exhaustion will become "less".

In that sense, the regulation fulfills the purpose of weaning off ventilation because the ventilatory support is continually adjusted (ie, reduced in the healthy patient). Should be the condition of a patient 20 However, once again worsening and the respiratory muscles become less efficient, the fatigue index would rise at constant load and the scheme would ensure that support from the ventilator 14 is increased accordingly. As already described in more detail above, in the embodiments of the 2 and 3 on the patient 20 by means of electrodes an EMG signal removed. Optionally, an invasive or non-invasive procedure can be used. From the resulting EMG signal a fatigue index is calculated. The ventilation assist controller 16 is the difference between the current fatigue index value and the initial fatigue index value. Depending on this difference, the built-in controller changes 16b2 the ventilatory support. A respirator 14 supports the patient 20 with the regulator 16b2 given support.

In a further embodiment, the control device 16 further configured to acquire the information about the current fatigue index at a first rate and to adjust the predetermined fatigue index at a second rate, the second rate being lower than the first rate. In this respect, the setpoint, the predefinable value, can also be adjusted with a correspondingly slower rate than that with which the control loop expires. Such an adaptation would be conceivable every few minutes, hours or days. This can be z. B. performed when the measured Fatigueindex values over a certain duration or number of breathing cycles were above the default / the setpoint. The control will then try to return to the old level, so set minimum support depending on the controller type. With the steady-state controller, when the condition improves, the support will quickly be minimal until the patient 20 returns to the initial level, or this falls below. In some embodiments, controllers can thus 16b2 are used, which deliberately tolerate a deviation in this direction, for example, without an integral control component. Other embodiments may issue a warning signal to indicate to the clinician or physician this condition of constant offset with minimal assistance. The setpoint can then be readjusted manually. In further embodiments, the target value can also be tracked automatically according to, wherein it is also conceivable that a tracking determining information is first set by the clinic staff, for example in which cycles or steps a tracking should take place. In other embodiments, one of the device 10 Proposed tracking can also be carried out only after confirmation by clinical staff.

In patients 20 Those who have been artificially ventilated for an extended period of time may use the ventilator 14 usually not easily issued. Instead, the patient becomes 20 from the ventilator 14 weaned, which also English. referred to as "weaning". This process known as "weaning" presents a major challenge in everyday clinical practice. On the one hand, it is desirable to keep the times of mechanical ventilation as short as possible and to the patient 20 to wean as early as possible. On the other hand, too early and failed weaning is detrimental to the patient 20 , In the context of weaning can support the ventilator 14 be gradually reduced. After a successful reduction, parameters such as the rapid-shallow-breathing index, which is a measure of fast and shallow breathing, are used, for example: B. expressed as the ratio of respiratory rate to respiratory volume, determined whether the patient 20 is eligible for a spontaneous breathing attempt.

The patient becomes part of the spontaneous breathing trial 20 either from the ventilator 14 disconnected or support the ventilator 14 is minimized to overcome the flow resistance in the breathing tubes and tube. The spontaneous breathing attempt may take about half to two hours. If patients persist the spontaneous breathing attempt, extubation and permanent removal of the ventilator may occur 14 be considered.

4 shows a further embodiment of a device 10 for ventilation of a patient 20 , The device shows in the middle of the control device 16 that is here as a microcontroller unit 16 implemented with a controller, a controller or a processor. The device comprises a gas mixing unit (oxygen and compressed air) 401 , Flow measurement inspiration 402 , an inspiratory valve 403 , a pressure measurement inspiration 404 , an inspiratory tube 405 , a flow measurement expiration 406 , an exhalation valve 407 , a pressure measurement expiration 408 , an expiratory tube 409 . an oxygen supply 411 , a compressed air supply 412 , an expiration outlet 413 , one to the ventilator 14 connected patients 20 and a microcontroller unit 16 that with the pressure and flow sensors 402 . 404 . 406 . 408 as well as the valves 403 . 407 and the gas mixing unit 401 is coupled.

The ventilator 14 in this embodiment includes an inspiratory (top) and an expiratory (bottom). The inspiration part represents the patient 20 the gas mixture for inhalation available. He has two gas connections 411 . 412 (for oxygen and compressed air), which can be connected to a central gas supply. Alternatively, a local gas supply unit can also be connected upstream. In the subsequent gas mixing unit 401 An air-oxygen mixture with a certain oxygen content can be provided. The gas gets through the inspiratory valve 403 provided to the patient. The inspiration valve 403 is a flow measurement 402 (Flow measurement) pre-switched. In addition, the pressure from the detection device 12 . 404 , For example by means of a pressure sensor, after the valve 403 measured.

The expiration part comprises an expiration valve 407 , a downstream flow measurement 406 (Flow measurement) and a pressure measurement 12 . 408 , As the 4 shows are the flow sensors 402 . 406 here the ventilation device 14 assigned, in other embodiments, these can also the detection device 12 be assigned. The sensors 402 . 404 . 406 . 408 and the valves 403 . 407 are with the control unit 16 coupled. The measured values from the pressure and flow sensors are sent to a microcontroller unit 16 Posted. This in turn controls the inspiratory 403 and the expiratory valve 407 , This makes it possible for the patient 20 to administer a specific gas flow (volume flow) or a specific gas pressure.

During inspiration is the expiratory valve 407 usually closed. The inspiratory part provides gas at overpressure and into the patient's lungs 20 directed. During expiration is the inspiration valve 403 closed and the exhalation valve 407 will be opened. As a result, gas from the lungs (in which there is still an overpressure due to the previous inspiration) flows through the expiration part into the environment. The microcontroller unit 16 also includes a user interface 18 where an operator can make settings and information is displayed.

• • Wenn der aktuelle Fatigueindex einen höheren Grad der Muskelermüdung indiziert, wird die Unterstützung des Patienten 20 durch das Beatmungsgerät 14 erhöht. Dadurch übernimmt das Beatmungsgerät 14. einen größeren Teil der notwendigen Atemarbeit. Die Atemmuskulatur des Patienten 20 wird entlastet und die Muskelermüdung sollte sich verringern.
• • Wenn der aktuelle Fatigueindex einen niedrigeren Grad der Muskelermüdung indiziert, wird die Unterstützung des Patienten 20 durch das Beatmungsgerät 14 verkleinert. Dadurch wird der Patient 20 einen größeren Anteil der Atemarbeit wieder selber leisten. Er kommt dem Ziel der Unabhängigkeit Von künstlicher Beatmung damit einen Schritt näher.

In the present embodiment, the current Fatigueindex with the initial or predefinable on the patient 20 matched fatigue index compared:

• • If the current fatigue index indicates a higher degree of muscle fatigue, the patient's support becomes 20 through the ventilator 14 elevated. This takes over the ventilator 14 , a greater part of the necessary work of breathing. The respiratory muscles of the patient 20 is relieved and the muscle fatigue should decrease.
• • If the current fatigue index indicates a lower level of muscle fatigue, the patient's support will be 20 through the ventilator 14 reduced. This will be the patient 20 Do a greater share of the work of breathing again. He comes one step closer to the goal of independence of artificial respiration.

• • Bei Ausführungsbeispielen mit einer volumenkontrollierten Beatmung wäre eine Veränderung des Volumens möglich.
• • Bei Ausführungsbeispielen mit einer druckkontrollierten Beatmung wäre eine Veränderung des Drucks möglich. Der Patient wird dabei beispielsweise im Modus CPAP (von engl. „Continuous Positive Airway Pressure”, kontinuierlicher positiver Atemdruck) und Druckunterstützung beatmet. Regelgröße kann in diesem Fall die Druckunterstützung (ΔPsupp, von engl. „support”, Unterstützung) sein. Für den Fall, dass im Rahmen der Fatigue-Bewertung eine erhöhte Ermüdung der Muskeln festgestellt wird, kann die Druckunterstützung erhöht werden. Im Fall einer verringerten Muskelermüdung kann die Druckunterstützung reduziert werden.
• • In Ausführungsbeispielen die andere Stellgrößen, wie etwa Konzentration im Gasgemisch, Veränderungen von ein oder mehreren Atmungsparametern (z. B. Atemfrequenz, Verhältnis der inspiratorischen und exspiratorischen Phasen, Anstiegszeit, Inspirationszeit, usw.); Veränderung des Beatmungsmodus, verwenden, kann die Regelung analog erfolgen, etwa eine Erhöhung des Sauerstoffgehalts bei größerer Ermüdung und umgekehrt.

Depending on the ventilation mode, different control variables are conceivable in exemplary embodiments:

• In embodiments with volume-controlled ventilation, a change in volume would be possible.
• • In embodiments with pressure-controlled ventilation, a change in pressure would be possible. For example, the patient is ventilated in CPAP mode (Continuous Positive Airway Pressure) and pressure support. In this case, the controlled variable can be pressure support (ΔPsupp, support). In the event that increased muscle fatigue is noted as part of the fatigue assessment, pressure support may be increased. In case of reduced muscle fatigue, the pressure support can be reduced.
• In embodiments, the other manipulated variables, such as concentration in the gas mixture, changes in one or more respiratory parameters (eg respiratory rate, ratio of inspiratory and expiratory phases, rise time, inspiratory time, etc.); Change the ventilation mode, use, the control can be analogous, such as an increase in oxygen content at greater fatigue and vice versa.

In some embodiments, a muscle fatigue ("muscle fatigue") value is derived from an electromyogram (EMG) signal of the respiratory muscles. Fatigue has occurred after a possible definition, when the muscles are no longer able to apply a certain force. Already in the run-up to this failure, however, changes are visible on the EMG signal, which indicate an imminent fatigue. One of these effects is the drop in line speed (also called conduction velocity ") of the electrical action potentials on the muscle fibers. This reduced conduction velocity is reflected, for example, in a compressed frequency spectrum of the EMG signal.

• • Die Kontraktion der Muskeln ist zyklisch entsprechend der Zyklizität/Periodizität der Atmung.
• • Die EMG Signale der betroffenen Muskeln sind nicht so gut wie z. B. am Arm abzuleiten. Das liegt daran, dass die Muskelfasern, insbesondere des Zwerchfells, nicht parallel zur Hautoberfläche liegen. Außerdem sind die EMG-Ableitungen im Brustbereich in der Regel von ElektroKardioGramm(EKG)-Artefakten kontaminiert. Darüber hinaus können die Signale von anderen Signalen, die auf EMG-Aktivitäten von nicht an der Atmung beteiligten Muskeln zurückzuführen sind, überlagert sein.

In the fatigue analysis of respiratory EMG, in comparison to the fatigue analysis of EMG signals of other muscles, such as. B. from the arm (biceps), the following challenges occur:

• • The contraction of the muscles is cyclic according to the cyclicity / periodicity of the respiration.
• • The EMG signals of the affected muscles are not as good as B. derive on the arm. This is because the muscle fibers, especially the diaphragm, are not parallel to the skin surface. In addition, chest EMG leads are usually contaminated with electrocardiogram (ECG) artifacts. In addition, the signals may be superimposed on other signals due to EMG activities of non-respiratory muscles.

To quantify muscle fatigue of the respiratory system based on EMG signals, in some embodiments the frequency compression of the EMG spectrum is evaluated. For the evaluation of the spectrum, an average frequency (also referred to as "mean" or "center frequency") can be used. In such embodiments, the control device 16 be configured to determine the information on the frequency-based Fatigueindex based on a spectral estimation and a quantification of a frequency compression of the electromyographic signal. In addition, an evaluation of the spectral estimation with other methods such. B. the moments of higher order. In further embodiments, alternatively or additionally, the control device 16 be configured to determine the information on the entropy-based fatigue index based on a signal entropy of the electromyographic signal.

Embodiments use a combination of entropy-based and spectral fatigue index when determining the fatigue index of the respiratory muscles, which can lead to increased reliability of the fatigue index compared to an exclusively spectral method. The spectral estimate is usually based on epochs of half a second in length. For large epoch sizes, the area between the heart beats may not be exploited. Partly also variable epochs are inserted between the heartbeats. However, the problem then arises that invalid signal portions of other causes are badly exploited. In the fatigue index calculation z. Epochs during inspiration and included during expiration. Embodiments may optionally make a further grading according to the strength of the contraction of the muscles.

Embodiments determine a fatigue index via a frequency-based fatigue index, an entropy-based fatigue index, or a combination of frequency-based and entropy-based fatigue index, based on the EMG signal of the respiratory muscles (preferably the diaphragm). In particular, the cyclic nature of respiratory muscular contraction may be taken into account. In addition, signal processing algorithms in embodiments may tolerate invalid time periods in the EMG raw signal. These invalid periods of time could be caused, for example, by ECG and motion artifacts.

Some embodiments may calculate different fatigue indices in parallel by different fatigue algorithms (frequency-based and entropy-based) epoch-wise. In the following, a solution with two fatigue detection algorithms (frequency-based and entropy-based) is described in detail. In principle, it is conceivable to involve additional fatigue detection algorithms (eg recurrence quantification analysis, wavelet methods, fractal methods, etc.).

The results of the different fatigue algorithms are then adaptively combined considering quality indices. The quality indices are calculated individually for each fatigue algorithm in an assigned quality value determination unit ("quality scoring unit"). In some embodiments, the quality indices are recalculated for each signal epoch, and the system is allowed to respond to the cyclic activity pattern of muscle contraction for artifacts (eg, ECG). In the following, the structure of an embodiment will be described in more detail. First, some basic aspects are described. The following section describes the individual blocks of the block diagram. 5 shows a block diagram of a signal processing chain in the embodiment of a device 10 for ventilation of a patient 20 , 6 shows the individual components of the block diagram of the 5 associated signal waveforms in the signal processing chain in the embodiment of the device 10 for ventilation of the patient 20 ,

The 5 illustrates a signal processing chain located on the left side in the detector 12 starts. This includes a raw signal acquisition 3100 (also called "raw EMG signal") for the EMG signal, a flow sensor 3150 (also Engl. "Flow") and a pressure sensor 3160 (also English "airway pressure"), which also in the ventilation device 14 can be arranged, cf. 4 , The signals thus detected become the control device 16 made available. The control device 16 includes a basic filter 3110 (also English "baseline filtering"), an ECG detector 3230 (also English "ECG detection"), a Ausbiend filter 3120 (also Engl. "Gating"), an envelope determination / calculation or detection 3220 (also Engl. "envelope calculation"), a Atemphasenbestimmung 3250 (also called "pneumatic breath phase"), a frequency-based fatigue index determination 3311 (also known as "fatigue index frequency"), an entropy-based fatigue index determination 3321 (also English: "fatigue index entropy"), quality signal determination / calculation or detectors 3221 for the envelope (also "quality signal envelope"), 3231 for the ECG signal (also "quality signal ECG") and for the respiratory phase signal 3251 (also "quality singal breath phase"), quality assessor for the frequency-based fatigue index 3312 (also English "quality scoring unit") and for the entropy-based fatigue index 3322 (also English "quality scoring unit"), an adaptive combiner 3460 (also English "adaptive combination unit") and an output 3470 (also engl. "Output combined fatigue index") for the combined fatigue index on the basis of which then a regulator 16d the ventilation device 14 regulates or controls. The 6 shows temporal waveforms respectively at the outputs of the designated there by the same reference numerals components, which also in the 5 be used.

• • Strategie ES: Die Epochen werden ohne Beachtung eines Gating-Signals (Ausblendung 3120) erstellt. Dabei wird bewusst in Kauf genommen, dass dabei Epochen entstehen, die vom Gating-Signal 3120 als ungültig gekennzeichnete Abschnitte enthalten. Für diese Epochen wird trotzdem erstmal ein Fatigueindex-Wert berechnet. Es obliegt dann der Einheit 3312, 3322 für die Berechnung der Qualitätsindizes, die betreffende Epoche entsprechend schlecht zu bewerten. Dadurch wird es der nachgelagerten Kombinationseinheit 3460 (auch engl. „adaptive combination unit”) ermöglicht, den entsprechenden Fatigue-Wert entweder wenig oder auch gar nicht zu berücksichtigen.
• • Strategie EG: Die Epochen werden so gebildet, dass sich keine durch das Gating-Signal 3120 als ungültig gekennzeichneten Signalabschnitte in den Epochen befinden. Zum Zeitpunkt der Epochenbildung werden die letzten M (wobei M für die Epochenlänge steht) als gültig gekennzeichneten Signalwerte der Epoche zugeordnet. Dabei ist es möglich, dass sich innerhalb der Epoche zeitlich nicht direkt aufeinander folgende Signalwerte befinden.
• • Strategie EP: Die Epochen werden mit variabler Länge gebildet, sodass sie den gesamten gültigen Bereich abdecken, aber innerhalb einer Epoche nur zeitlich direkt aufeinander folgende Werte beinhalten.

The fatigue algorithms 3311 . 3321 are each calculated based on epochs. The assigned Quality Scoring Unit 3312 . 3322 considers these epochs as well. The epochs have a length between 32 and 2048 ms and are in the 6 illustrated by the waveform labeled "epochs". When extracting the signal epochs, three different strategies can be pursued:

• • Strategy ES: Epochs are created without regard to a gating signal (blanking 3120). It is deliberately accepted that thereby epochs arise, the gating signal 3120 contain sections marked as invalid. Nevertheless, a fatigue index value is calculated for these epochs. It is then the unit 3312 . 3322 for the calculation of the quality indices, the respective epoch correspondingly bad to evaluate. This makes it the downstream combination unit 3460 (also called "adaptive combination unit") makes it possible to take into account the corresponding fatigue value either little or not at all.
• • Strategy EC: The epochs are formed so that no one through the gating signal 3120 signal sections marked invalid in the epochs. At the time of epoch formation, the last Ms (where M is the epoch length) are assigned as valid denoted signal values of the epoch. It is possible that within the epoch temporally not directly consecutive signal values are.
• • Strategy EP: The epochs are formed with variable length, so that they cover the entire valid range, but contain only temporally consecutive values within an epoch.

When deciding on the epoch formation strategy, the associated fatigue algorithm also plays a role. As the spectral estimate 3311 may be falsified by the recessed invalid signal portions, in some spectral estimation-based fatigue algorithms, the strategy ES or EP may be preferred. The ES strategy may then create short epochs to minimize the number of epochs that contain invalid signal values. For fatigue indices, which are more robust than recessed periods in the epochs, the EG strategy may be preferred.

In the 5 designated 3100 the raw EMG or input EMG signal. In one embodiment, it is a differential surface EMG signal that includes respiratory activity (activity of muscles involved in breathing). The input signal preferably has a sampling rate of 1024 Hz. However, larger or smaller sampling rates of the EMG signal are also possible. However, low sampling rates can lead to quality losses in fatigue detection algorithms.

In the further description all further calculations refer to the baseline filtered EMG signal 3110 , For certain fatigue indices and auxiliary signals, however, it may also be useful in special cases and other embodiments that they directly from the EMG raw signal 3100 be calculated. The baseline filter 3110 The general task is to provide a baseline of the disturbed input signal 3100 to eliminate. The baseline typically consists of a DC component (frequency equal to zero) and low-frequency components. In this respect, the control device 16 in this embodiment, to reduce a DC component in the electromyographic signal to obtain a first intermediate signal for further processing. In other embodiments, however, a further processing directly based on the electromyographic signal is conceivable. On the one hand, as a baseline filter 3110 For example, a linear high pass can be selected. The cutoff frequency can be set so that the EMG relevant frequencies can pass the filter. However, low frequency motion artifacts and low frequency components of the ECG are removed from the signal. This results in cutoff frequencies in the range of about 10 to 50 Hz. It is conceivable that the baseline filtering 3110 also adaptively responds to the ECG phase.

The baseline filter 3110 On the other hand, in further embodiments, it can be designed as a nonlinear filter in order to achieve the lowest possible signal delay times on the one hand and to change the shape of the ECG signal (or outliers) as little as possible on the other hand. This may be a problem in particular because the R-wave of the ECG usually has a triangular shape containing a DC component. This DC component is fundamentally changed to a bipolar signal by the known linear filters (in this case high-pass filters), since the R wave is subjected to a convolution with the impulse response of the high-pass filter by the linear filtering. The nonlinear filter used in a preferred embodiment is based on the use of a ranking filter. Signal filtering based on rank order filters, in contrast to sliding arithmetic averaging or based on predefined numbers of measured values, is insensitive to outliers in the signal. Ranking filters in the sense of some exemplary embodiments are, for example, extreme value filters or so-called "smoothing" filters or "filters that select a measured value" that allow the filter to pass a specific value from a predetermined number of measured values fed into the filter. that is, for example, the largest measured value, the second largest measured value, the smallest measured value, the second smallest measured value, an average value from the measured value number or a next larger or next smaller value to the mean value from the measured value number.

Such filters are also available as the following filter types "1-out-of-3 filter", "1-out-of-5 filter", "1-out-of-n filter", "minimum value-out-5 filter", "maximum Value-out-of-5 filter "," minimum-out-of-n filter "," maximum-out-n-filter "," average out-of-n-filter "are known and widely used in the field of image processing. The filter types described are mentioned in the context of embodiments only in an exemplary list, which are in no way limiting to the name of a ranking filter, but rather are modifications and combinations of these types in the sense of exemplary embodiments. Characteristic of such a ranking filter is that a measured value can pass through the filter unchanged, unlike other filters in which multiple readings result in a new value at the filter output, such as arithmetic mean filtering. Forming the difference with the unfiltered signal produces a signal that still contains these outliers but still removes the baseline. Outliers in the form of the R wave of the ECG signal are thereby preserved in their shape, but with reduced height and length. This reduction in height and length depends on the delay time of the baseline filter 3110 together. The signal thus generated, which in the following can also be referred to as the first intermediate signal, then becomes the ECG detector 3230 and the blanking filter 3120 provided.

As part of the ECG Detection 3230 ECG signal artifacts are detected that are not detected by the previous baseline filtering 3110 could be eliminated. In embodiments, the control device 16 be designed to detect based on the electromyographic signal or based on the first intermediate signal, a heart signal component, and in the electromyographic signal or in the first intermediate signal to reduce or hide an influence of the cardiac signal component to obtain a second intermediate signal as a result. This can be done in the currently considered embodiment in the blanking filter 3120 happen. In the exemplary embodiment shown, this intermediate signal is a second intermediate signal detected according to the above-described first intermediate signal. In summary, it can be stated that the control device 16 is here designed to detect a heart signal component based on the electromyographic signal, and to reduce or suppress an influence of the cardiac signal component in the electromyographic signal. This signal processing can be accomplished, for example, via the first intermediate signal. The control device 16 is then configured to reduce a DC component in the electromyographic signal in order to obtain the first intermediate signal for further processing and to detect the cardiac signal component by means of the first intermediate signal. As a result, the second intermediate signal is then obtained in which the influence of the cardiac signal component is reduced.

The heart signal may correspond to an electrocardiogram and have at least one QRS complex. For example, in embodiments, the influence of the at least one QRS complex in the first intermediate signal can be reduced. In the in the 5 and 6 illustrated embodiment, mainly the QRS complex is reduced. For detection z. B. the Pan Tompkins Algorithm, cf Pan, Jiapu, Tompkins, Willis J., "A Real-Time QRS Detection Algorithm, "Biomedical Engineering, IEEE Transactions on, vol. BME-32, no. 3, pp. 230, 236, March 1985 ,

The Pan-Tompkins algorithm outputs a pulse stream. The temporal position of the QRS complex is marked by a pulse, cf. 3230 in 6 , For the block ECG Detection 3230 Each spike is replaced by a square wave signal. This rectangle is 1, starts about 20 to 50 ms before the spike, and ends 50 to 90 ms after the spike. Outside the rectangles caused by the spikes, the signal is 0.

In the blanking or gating filter 3120 signal portions of the baseline filtered EMG signal are out of block 3110 set to the value "invalid", for which the block 3230 ECG artifacts. This block is also called QRS gating 3120 designates and supplies the second intermediate signal. The signal of the ECG detector 3230 is further a quality signal detector or a determination / calculation 3231 , also called Quality Signal ECG, where a quality measure is calculated for each epoch, which expresses the extent to which the epoch is left over by ECG residues and thus gated 3120 is charged. It is considered whether and if so how much the epoch is burdened by invalid signal sections. If there is no load, the value is 1. The greater the strain of gating 3120 , the more the value goes to zero.

The second intermediate signal or the output signal of the blanking filter 3120 then becomes an envelope detection / determination / calculation 3220 fed. In this block the EMG signal magnitude is determined. In addition, the RMS (of narrow: "root mean square") method is preferably used. There is the option of pre-whitening the signal prior to application of the RMS method (prefiltering to obtain as white a noise as possible, ie equally distributed in the frequency domain), cf. Merletti, R. & Parker, PA (Eds.) Electromyography: physiology, engineering, and noninvasive applications Wiley-Interscience [et al], 2004, XXII, 494 p , The resulting signal 3220 is also referred to as an envelope or Envelope. On by gating 3120 as invalid determined signal sections, in one embodiment in the RMS power signal, simply draw a straight line between the adjacent values to interpolate the signal from the envelope signal.

The control device 16 is then configured to determine a power signal as an envelope of the second intermediate signal and to hide / reduce and interpolate portions of the cardiac signal component in the power signal. From the envelope signal from block 3220 is in the subsequent quality signal detector or the determination / calculation 3221 , also English "Quality signal envelope detector", a signal with the value range 0 to 1 formed. Here, 1 stands for a large magnitude with a lot of muscle activity and zero for no EMG activity (apart from the background noise).

In the flow sensor 3150 can be a pneumatic gas flow / gas flow from the ventilator 14 be measured as a pneumatic signal. This is the flow / flow in the patient 20 meant. In the event that no flow sensor is available directly at the mouthpiece (as in 4 ) allows the patient flow from the ventilator 14 calculated from the inspiratory and expiratory flow. In addition, from the ventilator 14 still corrections such. B. for the dynamic behavior of the hose system and for leakage compensation. In the pressure sensor 3160 , also English "Airway pressure sensor", can be a pressure difference between gas pressure on the patient's mouthpiece 20 and the ambient pressure. If no direct pressure sensor is available on the mouthpiece, it may be removed from the ventilator 14 calculated using the internal expiratory and inspiratory flow sensors, cf. 4 , Similar to flow / flow, correction algorithms (tube system, tube, leackage, etc.) can also be used here. The control device 16 is then configured to determine information about inspiratory and expiratory phases based on the pneumatic signal to obtain information about an inspiratory muscle activity of the patient from the pneumatic signal in an inspiratory phase 20 to determine. The frequency-based and / or entropy-based fatigue index may then be determined essentially for that phase which is considered to be a phase of inspiratory muscle activity of the patient 20 is determined.

In the respiratory phase detector 3250 , also English "Pneumatic breath phase", are the output signals of the sensors 3150 and 3160 processed. The respiratory phase is calculated from the pneumatic data. The signal is 1 during inspiration. Otherwise, it has the value 0. It can also be considered the information whether it is a mandatory hub. As a result, mandatory strokes can be excluded from the fatigue calculation. Different ways of detecting the inhalation are dealt with in the following publication:
Robert Goulet, R .; Hees, D .; Kacmarek, RM Pressure vs. Flow Triggering During Pressure Support Ventilation, CHEST 1997; 111: 1649-53 http://journal.publications.chestnet.org/pdfaccess.ashx?ResourceID=2090985&PDFSource=13 , Algorithms for Cycling Off (End of Inspiration) can be found in:
Gentile, MA Cycling of the Mechanical Ventilator Breath, Respir Care 2011, 56 (1): 52-57 http://services.aarc.org/source/DownloadDocument/Downloaddocs/01.11.0052.PDF . 3251 , Quality Signal Breath Phase. From the signal from block 3250 becomes a quality signal in the subsequent quality signal detector or the quality calculation / determination 3251 educated. The quality signal is calculated from the mean value of the respiratory phase, also engl. "Breath phase", within the appropriate epoch.

The frequency-based fatigue index is epoch-wise in the block 3311 calculated. As part of this block 3311 the frequency compression is evaluated. First, a spectral estimate of the corresponding epoch is calculated. The spectral estimation can be done both non-parametrically (as in the above-mentioned literature) by means of the fast Fourier transformation (also Fast Fourier transformation (FFT)) and by means of parametric methods (eg Burg algorithm). In the second step the frequency compression is quantified. In addition to the classical center frequency (also called "mean" or "central frequency"), a quotient of spectral moments is preferably used for this purpose. This quotient is known from the following literature: Dimitrov, GV, Arabadzhiev, TI, Mileva, KN, Bowtell, JL, Crichton, N. & Dimitrova, NA, Muscle fatigue during dynamic contractions assessed by new spectral indexes Medicine & Science in Sports & Exercise, 2006, 11, 1971-1979 ,

• • Das Qualitätssignal EKG 3231, auch engl. „Quality Signal ECG” (QS_ECG), sofern die entsprechende Signalepoche ECG Artefakte aufweist, ist dieser Qualitätswert stark abgesenkt oder sogar Null. Dies betrifft insbesondere die durch die vorhergehende Baslinefilterung 3310 nicht herausfilterbaren Anteile des EKG Signals. Dies können hauptsächlich die Anteile sein, deren Frequenzbereich innerhalb des EMG-Frequenzbereiches liegt. In der Regel ist es hauptsächlich der QRS Komplex. Es ist denkbar, dass (abweichend von der weiter unten angegebenen Berechnungsformel für Block 3312) bei geringem QS_ECG der kombinierte Quality Score auf einen niedrigeren Wert oder sogar Null abgewertet wird.
• • Das Qualitätssignal der Einhüllenden 3221, Quality Signal Envelope (QS_ENV), ist ein Maß für die Stärke der Muskelkontraktion. Da die Fatigue/Ermüdung hauptsächlich während der aktiven Kontraktionsphasen zu erwarten ist, sorgen kleine Envelope-Werte für einen niedrigen, hohe Envelope-Werte für einen hohen Qualitätswert. Sehr große Envelope-Signale können auf Störungen bzw. nicht mit der regulären Atmungsanstrengung zusammenhängende Kontraktionen (z. B. Husten, Bewegungen) hinweisen. Diese sehr großen Envelope-Werte können somit wieder für einen niedrigen Qualitätswert sorgen. Die Zuordnung eines Envelope-Wertes auf einen bestimmten Qualitätsfaktor kann dabei adaptiv aus den zurückliegenden Atemzügen erfolgen. Dies ermöglicht es dem System in diesem Ausführungsbeispiel sich auf eine langsam verändernde Atemanstrengung einzustellen.
• • Das Atemphasensignal 3251 (QS_BreathPhase) wird ebenfalls dazu benutzt Atemphasen, bei denen keine Fatigue/Ermüdung im EMG-Signal zu erwarten ist, mit einem niedrigen Qualitätswert auszuzeichnen. Die Kontrolleinrichtung 16 ist in dem Ausführungsbeispiel ausgebildet, um ein Qualitätsmaß 3251 QS_BreathPhase, Phase über das pneumatische Signal 3250 zu bestimmen, das angibt, zu welchem Grad in dem EMG-Signal 3100 Information über den Fatigue/die Ermüdung zu erwarten ist. Im einfachsten Fall unterscheidet das Atemphasensignal zwischen Einatmung und Ausatmung, Es ist aber auch denkbar, dass Einatmung und Ausatmung in weitere Teilphasen unterteilt werden. Außerdem können unplausible Atemphasen zu einer Abwertung führen. Es kann dabei auch die Information berücksichtigt werden, ob es sich um einen mandatorischen Hub handelt. Dadurch können mandatorische Hübe aus der Fatigue-Berechnung ausgeschlossen werden. Ein mandatorischer Hub ist ein Atemhub der zwangsweise vom Beatmungsgerät 14 ausgelöst wurde. In diesem Fall ist der Atemhub nicht durch eine Atemanstrengung des Patienten 20 ausgelöst, sondern durch interne Prozesse im Beatmungsgerät 14 (z. B. Zeitüberschreitung einer Maximalzeit seit dem letzten Einatmen).

In the quality evaluation blocks 3312 . 3322 also "quality scoring unit (frequency / entropy)", a quality index is formed parallel to each fatigue algorithm. As part of this quality assessment unit 3312 . 3322 A measure of the quality or the trustworthiness of a fatigue value from its particular epoch is calculated using a specific algorithm. The initial value of the function is a continuous value between 0 and 1. Zero means that the fatigue value is untrustworthy and therefore not used at all. 1 means that the associated fatigue value is classified as highly trustworthy and should be included in the following calculations with a high weight or maximum proportion. The following values are included in the calculation of the quality value in the present exemplary embodiment:

• • The ECG quality signal 3231 , also English "Quality Signal ECG" (QS _ECG ), if the corresponding signal epoch has ECG artifacts, this quality value is greatly lowered or even zero. This applies in particular to the previous baseline filtering 3310 non-filterable portions of the ECG signal. This may mainly be the portions whose frequency range is within the EMG frequency range. As a rule, it is mainly the QRS complex. It is conceivable that (deviating from the calculation formula given below for block 3312 ) At low QS _ECG the combined Quality Score is devalued to a lower value or even zero.
• • The quality signal of the envelope 3221 , Quality Signal Envelope (QS _ENV ), is a measure of the strength of muscle contraction. Since fatigue / fatigue is most likely to occur during active contraction phases, small envelope values provide low, high envelope values for high quality value. Very large envelope signals may indicate disturbances or contractions (eg, coughing, movement) that are not related to regular respiratory effort. These very large envelope values can thus again ensure a low quality value. The assignment of an envelope value to a specific quality factor can be done adaptively from the previous breaths. This allows the system in this embodiment to adjust to a slowly varying breath effort.
• • The respiratory phase signal 3251 (QS _BreathPhase ) is also used to _label low-quality respiratory phases that do not expect fatigue / fatigue in the EMG signal. The control device 16 is formed in the embodiment to a quality measure 3251 QS _BreathPhase , phase via the pneumatic signal 3250 indicating to what degree in the EMG signal 3100 Information about fatigue / fatigue is expected. In the simplest case, the respiratory phase signal distinguishes between inhalation and exhalation, but it is also conceivable that inhalation and exhalation are subdivided into further subphases. In addition, implausible breathing phases can lead to a devaluation. It can also take into account the information as to whether it is a mandatory hub. As a result, mandatory strokes can be excluded from the fatigue calculation. A mandatory stroke is a breath of forced release from the ventilator 14 was triggered. In this case, the breath is not due to a respiratory effort of the patient 20 but triggered by internal processes in the ventilator 14 (eg timeout of a maximum time since the last inhalation).

• – ein Qualitätsmaß für die Herzsignalkomponente, das angibt zu welchem Grad das Leistungssignal während einer inspiratorischen Phase durch Herzsignalreste beeinflusst ist,
• – ein Qualitätsmaß für das pneumatische Signal, das angibt, zu welchem Grad in dem elektromyographischen Signal Information über einen Fatigue zu erwarten ist,
• – ein Qualitätsmaß für die Hüllkurve, das angibt zu welchem Grad eine Muskelaktivität des Patienten 20 während einer inspiratorischen Phase vorliegt,
• – ein Qualitätsmaß für den frequenzbasierten Fatigueindex, das einen Grad der Zuverlässigkeit des frequenzbasierten Fatigueindex angibt, und
• – ein Qualitätsmaß für den entropiebasierten Fatigueindex zu bestimmen, das einen Grad der Zuverlässigkeit des entropiebasierten Fatigueindex angibt.

Generally, the control device 16 be designed to determine at least two or more of the following quality measures,

• A quality measure for the cardiac signal component, which indicates to what extent the power signal during an inspiratory phase is influenced by heart signal residues,
• A quality measure of the pneumatic signal indicating to what extent fatigue information is to be expected in the electromyographic signal,
• A quality measure of the envelope that indicates to what degree a muscle activity of the patient 20 during an inspiratory phase,
• A quality measure for the frequency-based fatigue index indicating a degree of reliability of the frequency-based fatigue index, and
• - To determine a quality measure for the entropy-based fatigue index, which indicates a degree of reliability of the entropy-based fatigue index.

The control device 16 is formed in the embodiment shown to a quality measure 3231 QS _ECG for the cardiac signal component 3230 This determines to what degree the power signal during an inspiratory phase is affected by heart signal residues. The control device 16 is further adapted to a quality measure 3221 QS _ENV for the envelope 3220 This determines to what degree a muscle activity of the patient 20 during an inspiratory phase. The control device 16 is also formed in an embodiment to a quality measure 3312 for the frequency-based fatigue index 3311 to determine the degree of reliability of the frequency-based fatigue index 3311 indicating, and the control device 16 is trained to be a quality measure 3322 QS _Entropy for the entropy Fatigueindex 3321 to determine the degree of reliability of the entropy-based fatigue index 3321 indicates.

For the block 3312 Preferably, the following implementation is proposed in one embodiment: QS _Freq = β · (QS _ECG ) ³ · QS _ENV · QS _{Breath Phase} .

Here, β is a constant which is the weighting of the two fatigue algorithms in the later combination in the adaptive combination block 3460 and the forgetting factor in the accompanying smoothing determined. The weighting results from the relative ratio to another constant γ, which is introduced below for the entropy-based entropy index. The smoothing is based on the absolute value. The sum of β and γ should be less than or equal to one. Since the smoothing should be adapted to the epoch size, the sum of β and γ of the epoch size is also chosen to be appropriate.

Preferably, in block 3312 in another embodiment, the respective fatigue index value 3311 used to make the quality value plausible. In addition, in the combined quality score QS _Freq still included (not shown in the new block diagram) artifact detection and quality _measures for the underlying EMG signal with flow. The artifact detection detects disturbed sections (apart from ECG artifacts) in the signal. This can z. Motion artifacts, or coughing. The recognition of the artifacts can, for. B. amplitude based. In addition, artifacts can be detected by the absence of a cyclic respiratory motion. An artifact may also be present if the ventilator 14 performs a maneuver, or due to the pneumatic signals detects a non-ordinary situation. The quality measures for the underlying EMG signal include, in particular, measures that examine to what extent the spectrum has EMG-typical properties. Possibilities of an evaluation of the EMG signal are z. In Sinderby, C., Lindström, L. & Grassino, AE, "Automatic assessment of electromyogram quality", J. Appl. Physiol., 1995, 79, 1803-1815 , described.

In block 3321 , also English "Fatigueindex entropy", the entropy-based fatigue index is calculated epochwise. The entropy method investigates the extent to which the consequences of signal values are repeated in a certain time epoch. In particular, a "fuzzy approximate entropy" offers the advantage that a possible offset is subtracted when assessing equality of the signal sequences. This is especially important if a part has been cut / reduced / hidden within an epoch and the signal has a different offset. In one exemplary embodiment, the fuzzy approximate entropy is preferably used for this purpose. Further information can be found, for example, in Xie, H. -B., Guo, J.-Y. & Zheng, Y. -P. "Fuzzy approximate entropy analysis of chaotic and natural complex systems: detecting muscle fatigue using electromyography signals", Ann. Biomed. Eng., Springer, 2010, 38, 1483-1496 ,

In the block 3322 (Quality Scoring Unit (Entropy)) is essentially the equivalent of Block 3312 implemented for entropy-based fatigue detection. Therefore, it is described here in brief only. For the block 3322 In one embodiment, the following implementation is preferably proposed: QS _Entropy = γ · QA · QA _{ENV Breath phase}

In this case, γ is a constant which is the weighting of the two fatigue indices or algorithms in the later combination in the adaptive combination block 3460 as well as the forgetting factor at the accompanying smoothing determined. Preferably also in block 3322 still the respective fatigue index value 3321 used to make the quality value plausible. In addition, the description of Block also applies to this block 3312 ,

• • der frequenzbasierte Fatigueindex aus Block 3311 (im Folgenden als FIF bezeichnet),
• • der frequenzbasierte Quality Score (QS_Freq) aus Block 3312 (im Folgenden als QSF bezeichnet, wobei hier die oben erwähnte Plausibilisierung durch den frequenzbasierten Fatigueindex einfließen kann),
• • der entropiebasierte Fatigueindex aus Block 3321 (im Folgenden als FIE bezeichnet), und
• • der entropiebasierte Quality Score (QS_Entropy) aus Block 3322 (im Folgenden als QSE bezeichnet, wobei hier die oben erwähnte Plausibilisierung durch den entropiebasierten Fatigueindex einfließen kann).

In the adaptive combination block 3460 , also English "Adaptive combination unit", the smoothing to remove the noise is performed. The control device 16 Thus, it is configured to perform a combination of the frequency-based fatigue index and the entropy-based fatigue index based on at least two of the quality measures such that a higher quality measure requires a higher relative weighting in the combination. The following signals flow into this block as an input:

• • the frequency-based fatigue index from block 3311 (hereinafter referred to as FIF),
• • the Frequency-based Quality Score (QS _Freq ) from Block 3312 (hereinafter referred to as QSF, where the above-mentioned plausibility check can be incorporated by the frequency-based fatigue index),
• • the entropy-based fatigue index from Block 3321 (hereinafter referred to as FIE), and
• • The _{Entropy-based} Quality Score (QS _Entropy ) from Block 3322 (hereinafter referred to as QSE, in which case the above-mentioned plausibility can be incorporated by the entropy-based fatigue index).

The combined fatigue value is calculated in this embodiment by means of the following formula: FIC _n = QSFn _b * TF (FIF _n ) + QSE _n * TF (FIE _n ) + (1-QSF _n -QSE _n ) * FIC _n-1 .

TF denotes a linear transformation that maps the specific output range of each fatigue algorithm (frequency, entropy) to the range 0 to 1. 0 stands for no fatigue and 1 for the greatest possible fatigue. The subscript n designates the current sample. The subscript n-1 denotes the previous sample. The sample rate corresponds to the sample rate of the epochs. The factors QSF _n , QSE _n , (1-QSF _n -QSE _n ) model a combination of forgetting factor and weighting based on the quality measures, with the above-mentioned factors β and γ entering into QSF and QSE as explained above.

By using the quality indices for calculating the weight, the adaptive combination fulfills the task of considering the cyclic nature of the contraction. During a period of severe muscle contraction, the current fatigue raw signal value is weighted more heavily than in phases of weak muscle contraction. The output block 3470 , also English "Output combined fatigueindex", outputs the combined fatigue indices. In this case, a low-pass filter can be used to reduce any remaining oscillations in the respiratory rate.

In summary, for one embodiment, it can be described that the raw EMG signal 3100 via the baseline filter algorithm 3110 freed from DC voltage component. Based on this, the QRS complex is detected by ECG detection (block 3230 ). The result is the baseline filtered and QRS gated EMG signal ( 3120 ). The gate function is called QRS gate ( 3230 ). From the derived signal ( 3120 ), an RMS power envelope ( 3220 - Envelope) determine. Below the signal 3250 in the 6 show the periods of time, which are referred to as epochs.

From the baseline filtered signal at the point where the invalid values have been inserted by the QRS gating, the RMS power signal from the envelope signal simply draws a straight line between the adjacent values to interpolate the signal.

The Envelope-Calculation (preferably envelope RMS Power) takes place in block 3220 ,

The ECG detection (QRS complex) takes place in block 3230 , This is preferably done by the Pan Tompkins algorithm. The Pan-Tompkins algorithm outputs a pulse stream. The temporal position of the QRS complex is marked by a pulse. For the block ECG Detection 3230 Each spike is replaced by a square wave signal. This rectangle is 1, starts about 20 to 50 ms before the spike, and ends 50 to 90 ms after the spike. Outside the rectangles caused by the spikes, the signal is 0.

The flow signal as a volume flow signal is out of block 3150 used, in 3160 the pressure signal is used. At least one of the two pneumatic signals off 3150 or 3160 is used to in 3250 to determine the phase of inspiratory or expiratory activity. Preferably, in the block 3250 detects whether, after an expiratory phase, the volume flow signal briefly decreases, indicating a general inspiratory activity of the patient, so that the following periods of the following respiratory cycle may be included in the fatigue calculation. If there is no inspiratory muscle activity, so that the volume flow does not decrease, one can assume that there is no inspiratory muscle activity at all then preferably the following periods of the following respiratory cycle are not included in the fatigue calculation (mandatory stroke).

From the EMG raw signal are in the blocks 3311 and 3321 calculates two fatigue values based on frequency compression and entropy. The raw fatigue index signal corresponds to block 3311 , Another signal corresponding to 3321 , For every block 3312 respectively. 3322 Separate overlay rules can still be specified to superimpose the Envelope-Calculation, ECG-Detection, and Pneumatic-Breath-Phase values in each case in such a way that they result in meaningful evaluation values based on the respective fatigue index value 3311 or 3321 are applicable. For the block 3312 Preferably, the following implementation is proposed: QS _Freq = β · (QS _ECG ) ³ · QS _ENV · QS _{Breath Phase} .

For the block 3322 Preferably, the following implementation is proposed: QS _Entropy = γ · QA · QA _{ENV Breath phase.}

Β and γ are constants that determine the weighting of the two fatigue algorithms in the later combination as well as the forgetting factor in the accompanying smoothing.

Preferably, in blocks 3322 respectively. 3312 the respective fatigue index value was used to make the quality value plausible. The resulting value is in the block 3470 output.

7 shows a block diagram of a flowchart of a method for ventilating a patient 20 , The method includes detection 22 an electromyographic signal of the patient and a pneumatic signal. The method further comprises determining 24 information about a frequency-based fatigue index based on the electromyographic signal and / or a determination 26 information about an entropy-based fatigue index based on the electromyographic signal. The method further comprises determining 28 information about a current fatigue index of the patient based on the information on the frequency-based fatigue index and / or the information on the entropy-based fatigue index. The method further comprises adjusting 30 the ventilation of the patient 20 depending on the current fatigue index.

In a further embodiment, the adjusting comprises 30 a balancing of a difference between information about one on the patient 20 agreed predetermined fatigue index and the information on the current fatigue index on ventilation.

Another embodiment is a program or computer program having program code for performing at least one of the above-described methods when the program code is executed on a computer, a processor or a programmable hardware component.

Some embodiments do not perform the control in response to a threshold, but it is waited until the patient 20 is in an acceptable state with an acceptable initial fatigue value, and the regulatory process is started to maintain the initial fatigue value. Some embodiments may avoid the use of an esophageal catheter wherein electrodes at approximately equidistant intervals are used to detect voltage potentials to detect breathability. Some embodiments utilize electrodes attached to the outer surface of the skin and thus may dispense with the insertion of an esophageal catheter. Upon activation of the control - for example by a physician - the condition of the patient may be maintained in the future. Prescribing one on the patient 20 Adjusted fatigue may prevent the use of a threshold that is difficult for all types of patients to determine.

For the most part, ventilator patients are initially in a relatively depleted state, and embodiments may provide some breathing assistance by the respirator 14 provide. The goal is the patients to the own activity of the respiration too (their, which is called as "Weaning process".) The goal is thus that in the future the patient with the same state of exhaustion of the respiratory musculature, thus same fatigue value, successively takes over the respiratory activity and At the same time the actual respiratory support through the respirator becomes less and less, so that ideally at a later time at the end of the therapy the patient breathes independently.

In embodiments, pneumatic signals, such as pressure or volumetric flow signals, or also gas values, such as CO ₂ or O ₂ concentration, are used to counteract fatigue of the respiratory activity of the patient. In some embodiments, at least two electrodes are present for receiving the EMG signal as the differential potential between the electrodes. The block diagram of the 5 The algorithm shown is performed for a single such signal. Preferably, in the presence of a plurality of pairs of electrodes, a plurality of EMG signals can be evaluated and then later in parallel evaluation lead to respective fatigue values, which can then be superimposed or averaged.

A segmentation of the EMG signal may be omitted in some embodiments, since a value from the envelope calculation takes into account the corresponding phase as inspiratory or expiratory. The baseline filtered signal is optionally evaluated epoch-by-week, with one epoch preferably being 0.1 to 1.0 seconds long. The sampling rate is preferably 1024 Hz. Thus, one epoch preferably corresponds to 102 to 1024 samples. The epoch elements overlap. For each epoch a single fatigue value can be calculated as a raw fatigue index. The above consideration of larger perturbations during the epochs can be done as artifact detection, but is in the block diagram of FIG 5 not explicitly shown as this is only preferred.

The features disclosed in the above description, the claims and the drawings may be important both individually and in any combination for the realization of embodiments in their various embodiments and - unless otherwise stated in the description - are combined with each other as desired.

Although some aspects have been described in the context of a device, it will be understood that these aspects also constitute a description of the corresponding method, so that a block or a component of a device is also to be understood as a corresponding method step or as a feature of a method step. Similarly, aspects described in connection with or as a method step also represent a description of a corresponding block or detail or feature of a corresponding device.

Depending on particular implementation requirements, embodiments of the invention may be implemented in hardware or in software. The implementation may be performed using a digital storage medium, such as a floppy disk, a DVD, a Blu-Ray Disc, a CD, a ROM, a PROM, an EPROM, an EEPROM or FLASH memory, a hard disk, or other magnetic disk or optical memory are stored on the electronically readable control signals, which can cooperate with a programmable hardware component or cooperate such that the respective method is performed.

A programmable hardware component may be integrated by a processor, a central processing unit (CPU), a graphics processing unit (GPU), a computer, a computer system, an application-specific integrated circuit (ASIC) Circuit (IC = Integrated Circuit), a system on chip (SOC) system, a programmable logic element or a field programmable gate array with a microprocessor (FPGA = Field Programmable Gate Array) may be formed.

The digital storage medium may therefore be machine or computer readable. Thus, some embodiments include a data carrier having electronically readable control signals capable of interacting with a programmable computer system or programmable hardware component such that one of the methods described herein is performed. One embodiment is thus a data carrier (or a digital storage medium or a computer readable medium) on which the program is recorded for performing any of the methods described herein.

In general, embodiments of the present invention may be implemented as a program, firmware, computer program, or computer program product having program code or data, the program code or data operative to perform one of the methods when the program resides on a processor or a processor programmable hardware component expires. The program code or the data can also be stored, for example, on a machine-readable carrier or data carrier. The program code or the data may be present, inter alia, as source code, machine code or bytecode as well as other intermediate code.

Another embodiment is further a data stream, a signal sequence, or a sequence of signals that represents the program for performing any of the methods described herein. The data stream, the signal sequence or the sequence of signals can be configured, for example, to be transferred via a data communication connection, for example via the Internet or another network. Embodiments are also data representing signal sequences that are suitable for transmission over a network or a data communication connection, the data representing the program.

For example, a program according to one embodiment may implement one of the methods during its execution by, for example, reading or writing one or more data into memory locations, optionally switching operations or other operations in transistor structures, amplifier structures, or other electrical, optical, magnetic or caused by another operating principle working components. Accordingly, by reading a memory location, data, values, sensor values or other information can be detected, determined or measured by a program. A program can therefore acquire, determine or measure quantities, values, measured variables and other information by reading from one or more storage locations, as well as effect, initiate or execute an action by writing to one or more storage locations and control other devices, machines and components ,

The embodiments described above are merely illustrative of the principles of the present invention. It will be understood that modifications and variations of the arrangements and details described herein will be apparent to others of ordinary skill in the art. Therefore, it is intended that the invention be limited only by the scope of the appended claims and not by the specific details presented in the description and explanation of the embodiments herein.

QUOTES INCLUDE IN THE DESCRIPTION

This list of the documents listed by the applicant has been generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Cited patent literature

• US 8256419 B2 [0004]
• US 20140123979 A1 [0006, 0007]

Cited non-patent literature

• EJW Maarsingh, LA van Eykern, AB Sprikkelman, MO Hoekstra and WMC van Aalderen, J Appl Physiol, "Respiratory muscle activity measured with a noninvasive EMG technique: technical aspects and reproducibility", 88: 1955-1961, 2000 [0007]
• Miriam Gonzalez-Izal, Armando Malanda, Esteban Gorostiaga, and Mikel Izquierdo, "Electromyographic Models to Assess Muscle Fatigue," J Electromyogr Kinesiol, 22 (4): 501-512, August 2012 [0007]
• Maarsingh et al. [0046]
• Miriam Gonzalez-Izal et al. [0050]
• Pan Tompkins Algorithm, cf Pan, Jiapu, Tompkins, Willis J., "A Real-Time QRS Detection Algorithm," Biomedical Engineering, IEEE Transactions on, vol. BME-32, no. 3, pp. 230, 236, March 1985 [0078]
• Merletti, R. & Parker, PA (Eds.) Electromyography: physiology, engineering, and noninvasive applications Wiley-Interscience [et al], 2004, XXII, 494 S [0081]
• Robert Goulet, R .; Hees, D .; Kacmarek, RM Pressure vs. Flow Triggering During Pressure Support Ventilation, CHEST 1997; 111: 1649-53 http://journal.publications.chestnet.org/pdfaccess.ashx?ResourceID=2090985&PDFSource=13 [0084]
• Gentile, MA Cycling of the Mechanical Ventilator Breath, Respir Care 2011, 56 (1): 52-57 http://services.aarc.org/source/DownloadDocument/Downloaddocs/01.11.0052.PDF [0084]
• Dimitrov, GV, Arabadzhiev, TI, Mileva, KN, Bowtell, JL, Crichton, N. & Dimitrova, NA, Muscle fatigue during dynamic contractions assessed by new spectral indexes Medicine & Science in Sports & Exercise, 2006, 11, 1971-1979 [0085]
• Sinderby, C., Lindström, L. & Grassino, AE, "Automatic assessment of electromyogram quality", J. Appl. Physiol., 1995, 79, 1803-1815 [0091]
• Xie, H. -B., Guo, J.-Y. & Zheng, Y. -P. "Fuzzy approximate entropy analysis of chaotic and natural complex systems: detecting muscle fatigue using electromyography signals", Ann. Biomed. Eng., Springer, 2010, 38, 1483-1496. [0092]

Claims (19)

Contraption ( 10 ) for the ventilation of a patient ( 20 ), with a detection device ( 12 ) for detecting an electromyographic signal of the patient and a pneumatic signal; a ventilation device ( 14 ); and a control device ( 16 ) for controlling the detection device ( 12 ) and the ventilation device ( 14 ), the control device ( 16 ) is designed to determine based on the electromyographic signal information about a frequency-based fatigue index and / or an information on an entropy-based fatigue index, wherein the control device ( 16 ) is further adapted to provide information about a current fatigue index of the patient ( 20 ) based on the information on the frequency-based fatigue index and / or the information on the entropy-based fatigue index as a function of the pneumatic signal, and wherein the control device ( 16 ) is further configured, the ventilation device ( 14 ) depending on the current fatigue index. Device according to claim 1, wherein the control device ( 16 ) is further adapted to detect a difference between information about one of the patients ( 20 ) and the information about the current fatigue index by means of the ventilation device ( 14 ) about the patient ( 20 ) to correct. Contraption ( 10 ) according to one of the preceding claims, wherein the detection device ( 12 ) comprises a sensor element for the invasive or superficial detection of the electromyographic signal on the outer skin surface or by means of the outer skin surface of the patient ( 20 ) is trained. Contraption ( 10 ) according to claim 3, wherein the electromyographic signal is a differential surface detected signal comprising information about a respiratory activity of at least one respiratory muscle of the patient. Contraption ( 10 ) according to one of the preceding claims, wherein the detection device ( 12 ) is designed to detect the pneumatic signal as a human respiratory cycle typical respiratory signal of the patient ( 20 ) capture. Contraption ( 10 ) according to claim 5, wherein the detection device ( 12 ) is adapted to the pneumatic signal as a pressure signal or as a volume flow signal to the patient ( 26 ) capture. Contraption ( 10 ) according to one of the preceding claims, wherein the detection device ( 12 ) is adapted to the pneumatic signal as expiratory and / or inspiratory breathing signal of the patient ( 20 ) capture. Contraption ( 10 ) according to any one of the preceding claims, further comprising an interface ( 18 ) for obtaining the information about the patient ( 20 ), and whereby the control body ( 16 ) is adapted to transmit the information about the given fatigue index via the interface ( 18 ) by a user of the device ( 20 ) to obtain. Contraption ( 10 ) according to one of the preceding claims, wherein the control device ( 16 ) is adapted to determine the frequency-based fatigue index information based on spectral estimation and quantification of frequency compression of the electromyographic signal. Contraption ( 10 ) according to one of the preceding claims, wherein the control device ( 16 ) is configured to determine the information about the entropy-based fatigue index based on a signal entropy of the electromyographic signal. Device according to one of the preceding claims, wherein the control device ( 16 ) is configured to detect a cardiac signal component based on the electromyographic signal, and to reduce or suppress an influence of the cardiac signal component in the electromyographic signal to obtain an intermediate signal. Contraption ( 10 ) according to claim 11, wherein the intermediate signal is a second intermediate signal, and wherein the control device ( 16 ) is further configured to reduce a DC component in the electromyographic signal to obtain a first intermediate signal for further processing, and wherein the control device ( 16 ) is further configured to detect the heart signal component by means of the first intermediate signal. Contraption ( 10 ) according to claim 12, wherein the control device ( 16 ) is designed to determine a power signal as an envelope of the second intermediate signal and to hide and interpolate portions of the cardiac signal component in the power signal. Contraption ( 10 ) according to claim 13, wherein the control device ( 16 ) is configured to receive information based on the pneumatic signal determine inspiratory and expiratory phases to obtain information about an inspiratory muscle activity of the patient from the pneumatic signal in an inspiratory phase ( 20 ) and to determine the frequency-based and / or the entropy-based fatigue index essentially for the phase which is the phase of inspiratory muscle activity of the patient ( 20 ) is determined. Contraption ( 10 ) according to claim 14, wherein the control device ( 16 ) is configured to determine at least two or more of the following quality measures: a quality measure of the cardiac signal component indicating the degree to which the power signal is affected by cardiac arrest during an inspiratory phase, a quality measure of the pneumatic signal indicative of which degree of fatigue information is to be expected in the electromyographic signal, an envelope quality measure indicating the degree of muscle activity of the patient ( 20 ) during an inspiratory phase, - a quality measure for the frequency-based fatigue index, which indicates a degree of reliability of the frequency-based fatigue index, and - a quality measure for the entropy-based fatigue index, which indicates a degree of reliability of the entropy-based fatigue index. Contraption ( 10 ) according to claim 15, wherein the control device ( 16 ) is adapted to perform a combination of the frequency-based fatigue index and the entropy-based fatigue index based on at least two of the quality measures so that a higher quality measure is higher. relative weighting in the combination conditionally. Method for ventilating a patient ( 20 ), with capture ( 22 ) an electromyographic signal of the patient and a pneumatic signal; Determine ( 24 ) information about a frequency-based fatigue index based on the electromyographic signal; and / or determining ( 26 ) information about an entropy-based fatigue index based on the electromyographic signal; Determine ( 28 ) information about a patient's current fatigue index based on the frequency-based fatigue index information and / or the entropy-based fatigue index information; and setting ( 30 ) of the patient's ventilation ( 20 ) depending on the current fatigue index. Method according to claim 17, wherein said setting ( 30 ) comprises balancing a difference between information about a given patient-specified fatigue index and information about the current fatigue index about ventilation. Program comprising a program code for carrying out at least one of the methods according to one of claims 17 or 18, when the program code is executed on a computer, a processor or a programmable hardware component.

Images