WO2003059158A2

WO2003059158A2 - Method and device for generating an analysis result specific to the physiological condition of a patient

Info

Publication number: WO2003059158A2
Application number: PCT/EP2002/014797
Authority: WO
Inventors: Jörg Meier; Stefan Schaetzl; Thomas Betzl; Aleksandra Dana-Mika
Original assignee: Map Medizin-Technologie Gmbh
Priority date: 2001-12-29
Filing date: 2002-12-30
Publication date: 2003-07-24
Also published as: DE10164445A1; DE10164445B4; WO2003059158A3; AU2002360093A1; AU2002360093A8

Abstract

The invention relates to a method and a device for generating an analysis result specific to the physiological condition of a patient, which can be used in particular during the diagnosis and/or treatment of a patient exhibiting symptoms of sleep-related respiratory disorders. According to the invention, an analysis result specific to the physiological condition of a patient is generated, based on measurement signals that correlate to the respiration of said patient. Several analysis systems are used to generate analysis characteristics from said measurement signals and at least one analysis result is generated in a result-generation step based on said characteristics, by the subjection of the analysis characteristics to a linked observation.

Description

Method and device for generating a person specific to the physiological state of a person

evaluation result

The invention relates to a method and a device for generating a person-specific evaluation result with regard to the physiological state of a person, which can be used in particular in the diagnosis and / or therapy of a patient with signs of sleep-disordered breathing.

For the investigation of sleep-related respiratory disorders, so-called polysomnography devices are known which have a plurality of measurement channels for detecting physiological state variables of a patient. In order to diagnose a patient with regard to his breathing properties during a sleep phase, it is known to detect body position signals and possibly blood pressure signals and to record these together with signals that describe the respiratory activity of the patient. Those signals describing the respiratory activity of the patient are usually generated via so-called thermistors or also via pneumotachographs. The recorded signals can be displayed graphically in temporal relation to each other and evaluated within the framework of a specialist medical examination. Within the context of the therapy of sleep-related respiratory disorders, it is possible to carry out a respiratory gas pressure control on the basis of the signals indicative of the current physiological condition of a patient, so that any respiratory disorders can be counteracted by a respiratory gas pressure which is matched to the current state of the patient.

In practice, the consideration of the generated measuring signals often leads to different evaluations of the degree of severity and degree of severity of a condition of the examined person. The invention has for its object to provide a method and apparatus for generating a present example, as an electrical signal, in particular in a bit sequence existing record, present evaluation result, based on the physiological condition of the person under investigation can be assessed more appropriate than conventional manner the Case is.

According to a first aspect of the present invention, this object is achieved according to the invention by a method for generating a physiological state-specific evaluation result on the basis of measurement signals associated with the respiration of a person, wherein from the said measurement signals using at least one evaluation system , Evaluation features are generated and at least one evaluation result is generated within the framework of a result generation step based thereon, by subjecting the evaluation features to a linking consideration.

This advantageously makes it possible to obtain e.g. Automatically evaluate the amount of data describing the entire course of a patient's sleep in a standardized, repeatable manner and largely avoid misinterpretations of the generated measurement data.

According to a particular aspect of the present invention, a feature contribution contained in the evaluation features is determined within a time window that is less than a linkage window provided for the linking consideration. As a result, it becomes possible in an advantageous manner to record typical properties for each breath as part of a high-resolution view of the breath and to subject the properties of the breaths determined in this case to consideration of several breaths.

The evaluation concept according to the invention can, in the control of the pressure, be supplied to the patient by means of an overpressure respiratory system Respiratory gas application find. This makes it possible to precisely match the respiratory gas pressure on the current physiological needs of the patient without affecting the sleep process by a subjectively perceived as incorrectly high control dynamics.

The linking consideration of the breath characteristics determined for the individual breaths may extend over a time window, for example a predetermined e.g. 30 breaths or adaptively optimized number of breaths overlooked. In particular, for the assessment of the physiological condition of the patient, for example as the basis for a medical diagnosis, it is also possible to perform certain linking operations, for sleep phase-related periods, selected time periods or even the entire measurement comprehensive time window. According to a particularly preferred embodiment of the invention, based on linking operations, raw data and / or intermediate results are selected which enable a particularly reliable generation of characteristic characteristic values, in particular indices.

According to a particularly preferred embodiment of the invention, on the basis of the linking consideration, a physiological typing of the possibly existing clinical picture of the examined person takes place. On the basis of the evaluation result generated according to the invention, it is possible to adaptively optimize the control behavior of a pressure control device provided for controlling the respiratory gas pressure, so that, for example, after a period of use of several days, the control behavior of the pressure control is optimally adapted to the patient.

According to a particularly preferred embodiment of the invention, based on the linking consideration, a configuration data set is generated for configuring the breathing gas pressure regulation of a CPAP device. The configuration data record can be transmitted to the CPAP device via an interface device, or advantageously also by a mobile storage medium, for example in the form of a PCMCIA card. The configuration data set can, if necessary, with the interposition of a Adaptation procedure be modified so that this particular system characteristics of the CPAP device takes particular account.

According to a particularly preferred embodiment of the invention, the evaluation features are generated on the basis of correlation criteria, in particular statistical evaluation systems, by which, for example, similarities with previous breaths, or preferably adaptively optimized reference criteria, e.g. be evaluated for reference breaths. The correlation criteria can also be applied in particular to the first and / or second derivative of the detected respiratory gas flow. A generation of the characteristics characteristic for the respective breath can be carried out using statistical methods. The linking consideration of the properties determined for each breath can also be done using statistical methods.

On the basis of the evaluation features generated for each breath or also specific breath sequences, a feature array can be successively filled in, this feature field describing a time window at least for selected evaluation features that is at least as large as the smallest linking time window provided for interlinking the evaluation features.

The evaluation features are generated according to a particularly preferred embodiment of the invention such that among these e.g. Evaluation features containing the information on the duration of a breath and / or e.g. with regard to normal-looking breaths. On the basis of these evaluation features, it becomes possible to determine the temporal length of periods with normal respiration in the context of the linking consideration.

Furthermore, the evaluation features are advantageously also generated in such a way that they preferably also contain the information which is representative of the flow limitation of the occurrence of any flow limitation phenomena in individual breaths. Based on a By means of an evaluation characteristic, it becomes possible to describe the duration of certain properties of at least partially flow-limited respiratory sequences.

For periods in which no respiratory activity is detected, evaluation features can also be generated on the basis of which the phase length of any apnea sequences and / or characteristics characteristic of properties of this apnea phase can be generated within the context of the linking consideration. These evaluation features preferably contain information regarding the nature of the apnea phases, e.g. on whether the apnea phase should be classified as central, as obstructive or as a combination (mixed apnea phase).

Such evaluation features are also generated according to a particularly preferred embodiment of the invention for snoring phases, for phases with Cheyne-Stokes respiration and Hypoventilationphasen.

The evaluation features preferably also contain information or information from which the body position, the head position and preferably also the degree of neck rotation of the patient can be derived. Sleep-phase-specific information may already be included in the evaluation features.

The generated evaluation features are preferably stored in association with the detected breath or taking into account their temporal position. That the generated evaluation features are attributable to a defined time window - in the case of normal breathing the respective breath.

Within the context of the linking consideration, according to a particularly preferred embodiment of the invention, an index classifying the flow limitation and subsequently referred to as the flow limitation index is generated. For example, this flow limitation index can be determined based on the following evaluation rule: Calculation of the flow limitation index (FLI):

in which:

FLl, = index indicating the flow limitations per hour A = number of breaths imitated by the flow ti _m ii ⁼ time of the measuring interval in hours; the time interval results, for example, from a pressure stage. Sleeping phase, etc.

Examples:

• In one patient, 34 fixed breaths are detected in 48 minutes. The result is a FLI of 42.5.

• In one patient, 25 flow-limited breaths are detected at a CP AP pressure of 6 mbar. The compression took 10 minutes. It calculates a FL / _Λmht , _r of 150.

Alternatively, or in combination with it, it is also possible to specify a flow limitation potential characterizing indication, e.g. in the form of a quantity referred to below as the flow-limiting relation (FLR).

Calculation of the flow limitation relation (FLR):

tfl,

FLR; 100% = [%]

interval

in which:

FLR = proportion of the flow-limited inspiratory time per effectively breathed inspiration time t _K iu siimi _did i _n ⁼ inspiration time of the flow-limited breaths t inu- _r - _a ii ⁼ total inspiration time of a time interval (in hours): the time interval results, for example from a pressure stage.

Sleeping phase, etc.

Deviating from FLl, the FLR calculated in this way is influenced to a lesser extent by detected events such as apneas. example

• For a patient 11 m of total inspiratory time 2 hours and 56 minutes measured the maximum inspiratory time 2 hours and 12 minutes DTE FLR = ~ 5%

• A patient

ventilated with non-dissipated CP P-diutes There are several different periods of inspiration for the patient

At a constant CP4P pressure, a flow-limited inspiration time is present in reverse

52 minutes, the patient sleeps at a total inhalation time

3 hours and 24 minutes in jerk! age It includes a FLR \ on 255 "Xo

With the aid of the FLI and the FLR, flow-limited respiration can be brought into different dependencies and the applied therapy pressure can be evaluated

The calculations for flow limitation are made on the basis of the detected respiratory gas volume flow

Detection of the respiratory phase d h Detection of the exhalation and inspiration phase of the respiration

Determination of the inspiratory time

Evaluation of respiratory tracts as a function of the respiratory regularity by the backward collision, eg stable resp. Unstable respiration

Zeitinvertalle eg 5/10/30/60/90 min

Pressure levels eg 4/5/6/7 mbar

Sleep phases eg Wach REM NREM 1-4

Schlafqua did, for example, on the type of delektierten events z B Apnoen H \ popnoen

In general, the number of detected disturbances

Body position z B Back position Lateral position

The relationships created in this way can be entered in a table and then analyzed.

Evaluation of breathing with characteristics

Furthermore, it is possible to generate a snore index as part of the linking consideration. The signal used for determining snore sequences can be generated from the respiratory gas pressure detection device, from the engine power reference, as well as from the respiratory gas flow signal or in particular acoustic recording systems. It is also possible, in the name of the parenting authority, to generate a mouth-breath / nasal breathing index. Furthermore, it is possible to generate a sleep time index within the context of the linking consideration.

Furthermore, it is possible to generate a sleep phase index within the context of the linking consideration. It is possible in connection with the respiratory phase analysis between inspiratory

(obstruction-relevant) and expiratory (minor) snoring to distinguish. Furthermore, it is possible to generate a periodic respiratory index within the context of the linking consideration. Furthermore, it is possible to generate a breath volume index within the context of the linking consideration.

Apart from information on the body position, the evaluation features can preferably be generated on the basis of a volumetric flow measurement of the respiratory gas flow, referred to below as v-measurement. The measurement of the respiratory gas flow can be carried out at ambient pressure or under defined changed respiratory gas pressure.

At least part of the evaluation features is preferably generated taking into account the first or second derivative of the time profile of the respiratory gas flow.

According to a further aspect of the present invention, the object stated at the outset is also achieved by a device for carrying out the method described above, wherein this device comprises a measuring signal input device and a computer device for providing a plurality of evaluation systems, wherein evaluation features are generated by the evaluation signals from the said measuring signals and in the context of a result generation step based thereon, at least the evaluation result is generated by the computer device being configured in such a way that it subjects the evaluation features to a linking consideration. in the Kaπmeπ αer crrassuπg αer MimuriysicuiyKeii. ues raiierueπ aui i unuiaye vuπ indicative of the respiratory gas volume flow indicative data, it is possible to detect individual breaths as such. The beginning and end of the inhalation and expiration phases of the breath can be determined, for example, in conjunction with an examination of the first and second derivative of the respiratory gas flow signal and also taking into account the possible tidal volume. Based on these evaluation results, the duration of the breath phases, the actual volume of the breath, and the breathing pattern can be determined.

By a statistical analysis of the properties of several consecutive breaths, the current physiological state of the examined person can be further determined. Based on the extraction of features for each individual breath, raw data reduction can be achieved. From the statistical evaluation of the properties of several consecutive breaths, a distinction can be made between snoring relevant to the obstruction and obstruction-unimportant snoring. A typification of

Oscillation properties associated with snoring events can be done without a microphone device.

The occurrence of any snoring-induced oscillations can be detected on the basis of the time course of the respiratory gas pressure. Thus, it is possible, for example, to extract respiratory gas pressure oscillations caused by snoring from signals generated by corresponding respiratory gas pressure sensor device. In particular, based on frequency and amplitude analysis, e.g. Fast Fourier analysis makes it possible to classify snoring events in terms of their place of origin (soft palate, larynx ...).

As part of a statistical evaluation of the subsequent breaths, it is possible to generate a respiratory index indicative of respiratory stability. This respiratory index is preferably determined according to the following rule: Calculation of the respiratory index (AI):

AI, = index, which indicates a certain type of breathing pattern, e.g.

Unstable, stable breathing. Mouth breathing. Nasal breathing per hour indicates A = number of certain respiratory patterns eg unstable breaths ti _n i- _.T -.aii ~ Z i of the measurement interval in hours; the time interval results from eg a pressure stage. Sleeping phase, etc.

Examples:

• In one patient, 90 unstable breaths are detected in 35 minutes. The result is an AI of 154.

• In one patient, 15 breaths with specific breath patterns are taken at a CP AP pressure of 8 mbar. The pressure lasted 14 minutes. It calculates a S / _{Λ /} " _/ ," _f of 64.

Stable respiration being when the respiratory stability index is> 0.911 and unstable breathing is considered to be present when the respiratory stability index is <0.911.

Based on the linking consideration of the evaluation features, in particular the following obstructive respiratory disorders (OSA) can be recognized:

Apnea, hypopnea, river-limited breathing, stable and unstable breathing, as well as any leakage events.

A respiratory disorder is classified as an apnea event when a respiratory arrest is detected whose length exceeds a predetermined period of, for example, 10 seconds.

A hypopnea event may then be considered present if, after three breaths classified as normal, for example, at least two and a maximum of three major breaths are detected. As another criterion for this the inspiratory difference volume of the considered breath can be used.

In the breath being examined in each case, a flow limitation can be detected if the respiratory gas flow has certain plateau zones or several maxima during the inspiration phase.

Stable respiration may be considered present if the respiratory flow or respiration rate and the amplitude of the respiratory gas flow can be considered regular within a predetermined time range. Respiration may be considered to be stable, in particular, when a respiratory stability index defined for respiratory stability has an amount greater than 0.911. During stable breathing, no respiratory flows (OSA) occur.

An unstable respiration may be considered present if the aforementioned respiratory stability index has a value less than 0.911 and the respiratory flow is correspondingly irregular.

Such irregular breathing can be classified as a respiratory disorder, with such phases in the case of a breathing gas pressure control, this is done with an increased sensitivity.

Any high-frequency oscillations occurring in a pressure signal can be classified as in- or expiratory snoring in conjunction with the respiratory flow signal. The evaluation features generated with regard to the occurrence of snoring can be included in the linking consideration provided for generating the evaluation results.

On the basis of the temporal course of the nasally communicated respiratory gas volume flow system states such as, for example, variants of respiratory gas deficiencies (leakage), for example caused by mask application artifacts (mask problems) or expiratory mouth breathing, as well as permanent oral respiration, can be classified. When leakage shows the temporal course of the nasal breath gas volume flow communicated a magnitude shift relative to a reference value (eg zero line). In the case of expiratory mouth breathing between the inspiratory volume and the expiratory volume, since at least part of the gas exchange is oral. Further key features are the slope of the inspiratory / expiratory flank, the relative temporal position of the extreme values in the respective respiratory phase.

By using the generation of evaluation results on the basis of a linking consideration to control the breathing gas pressure according to the invention, it is possible to control the pressure in accordance with selected pressure control modes, e.g. first to perform in accordance with a normal mode or perform after fulfillment of predetermined criteria in accordance with a sensitive mode.

The control behavior of the pressure control device is preferably tuned in the normal mode such that any detected events or defined event chains allow an increase in pressure. As part of a so-called. Sensitive mode, the respiratory gas pressure can be successively incrementally lowered, the system can be tuned such that is responded to any, occurring at lower pressure gas pressure events with a higher regeidynamics. A change to the sensitive mode can be made dependent on several criteria, in particular depending on whether stable respiration is present (respiration stability index> 0.911).

During the normal mode, the control behavior of the pressure control device is preferably adjusted such that an increase in pressure takes place when apnea conditions, hypopnoea conditions or even flow limitations are detected. In the case of two apnea sequences with a comparatively long time duration or, for example, also in the case of three apnea phases with a shorter time duration, it is possible to successively increase the respiratory gas pressure. An increase in the respiratory gas pressure can also take place when a respiratory arrest over a predetermined period of eg 1, 2 minutes is detected. The increase of the respiratory gas pressure can be continuous or also in stages, wherein the pressure increase gradient preferably does not exceed a maximum value of 4 mbar per minute. It is possible to limit the maximum permissible pressure level to a predetermined limit of, for example, 18 mbar. In the case of detected hypopnea conditions, it is preferable to increase the pressure in comparatively small pressure stages of, for example, 1 mbar, wherein preferably the number of pressure-increasing stages is limited.

In the case of flow limitation phases, a pressure increase of 1 mbar can be initiated with a respiratory stability index of> 0.911. A decrease in pressure can occur if, within a predetermined time window of, for example, 9 minutes, stable respiration is recognized and the respiratory stability index has a value of> 0.911. In this case, a pressure reduction of, for example, 2 mbar can be allowed. It is also possible for a predetermined time periods to suppress a change in the respiratory gas pressure or to limit it to a comparatively narrow pressure change corridor. The implementation of a pressure change can be prevented, for example, if a certain combination of criteria is present in which u.a. respiration is classified as unstable and the respiratory stability index is <0.911.

Operation of the breathing gas pressure control in the sensitive mode causes an increase in pressure to occur when apnea conditions occur in accordance with a predetermined timing. It is thus possible, for example, to increase the pressure by 2 mbar if either a respiratory arrest lasting more than two minutes is recognized or two large (at least 25 see.) Or three smaller apnea states (max. 25 sea). be recognized and the breathing gas pressure is below 14 mbar. An increase in pressure by a value of 1 mbar may be induced if hypopnoetic sequences occur over a period of at least 3 minutes.

Increases in pressure by 1 mbar in each case are initiated in the sensitive mode when 4 out of 4 breaths show flow limitation features and the Respiratory stability index is> 0.911 or 5 out of 10 breaths plus limitive features and the respiratory stability index is <0.911 or even 10 of 20 breaths show flow limitation features and the respiratory stability index is also <0.911.

Decrease in pressure is preferably initiated in the Sensitv mode when stable respiration is present and the period of time for this stable respiration is at least three minutes and at the same time the respiratory stability index is> 0.911. In this case, a pressure reduction of initially 2 mbar can be initiated. In the sensitive mode, it is also preferably possible to allow no pressure change in phases or to limit the pressure change to a comparatively narrow pressure change corridor. Preferably, in particular, no pressure changes are allowed if the respiration is classified as unstable and obstruction states are detected.

A hypopnea phase may then be considered to be present if, after three normal breaths, at least two but no more than three major breaths follow. Here, there must be an inspiratory difference volume ΔV exceeding a predetermined threshold (e.g., 15% of the average tidal volume).

On the basis of the consideration according to the invention of the signal indicative of the respiratory gas flow, it is possible to detect respiratory disorders which at least initially do not require a change in the respiratory gas pressure. Such respiratory disorders may include, for example: swallowing, coughing, mouth breathing, expiratory mouth breathing, arousals and speech.

In the case of taking into account the results of the combination in the control of the respiratory gas pressure, the linking criteria are preferably chosen in such a way. Preferably, the linking conditions are coordinated such that pressure changes. Further details and features of the invention will become apparent from the following description taken in conjunction with the drawings. Show it:

Fig. 1 is a diagram for explaining the respiratory gas flow for a single breath;

Fig. 2 is a diagram describing the time course of the respiratory gas flow for several breaths;

Fig. 3 is a diagram showing the time course of the respiratory gas pressure with individual, caused by snoring pressure oscillations;

4 is a diagram illustrating the time course of the respiratory gas flow for a plurality of breaths interrupted by an apnea period;

Fig. 5 is a diagram showing the time course of the respiratory gas flow with a hypopnea event;

6 shows a diagram of the time course of the respiratory gas flow for a plurality of, in part, flow-limited breaths;

7 is a diagram for explaining the time course of the respiratory gas flow in the case of a substantially undisturbed, stable respiration;

8 is a diagram for explaining the time course of the respiratory gas flow in the case of unstable, disturbed respiration;

Fig. 9 is a diagram showing the time course of the respiratory gas flow and in the same temporal allocation to the course of the respiratory gas pressure occurring in the pressure signal oscillations caused by snoring; Fig.10 is a diagram showing the time course of the respiratory gas flow in the event of a system failure caused by, for example, mouth breathing or mask leakage;

11 shows a diagram for explaining a respiratory gas pressure change initiated in connection with the recognition and linked observation of breathing patterns;

12 shows a diagram for explaining the time profile of the respiratory gas flow and a change in the respiratory gas pressure carried out on this basis;

13 shows a diagram for explaining the time profile of the respiratory gas flow in conjunction with a respiratory gas pressure change initiated on the basis of this respiratory gas flow;

FIG. 14 shows a diagram for explaining the time profile of the respiratory gas flow in conjunction with a change in the respiratory gas pressure caused thereby; FIG.

15 shows a diagram for explaining the time course of the respiratory gas flow with hypopnoetic sequences recognized therein and a respiratory gas pressure change initiated on the basis of the recognition of these hypopnoetic sequences;

16 shows a diagram for explaining the time profile of the respiratory gas flow with flow-limited respiratory flows occurring therein and a graph for explaining the respiratory gas pressure set in this case;

17 shows a diagram for explaining the time profile of the respiratory gas flow in conjunction with the prevailing breathing gas pressure; 18 is a diagram for explaining the temporal course of the respiratory gas flow with a phase of normal respiration occurring therein, a phase of flow-limited respiration, a subsequent hypopnea phase and a phase leakage caused by a masked phase in conjunction with the prevailing respiratory gas pressure;

Fig.19 is a diagram for explaining the temporal Veriaufes the respiratory gas flow in conjunction with the prevailing here breathing gas pressure.

FIG. 20 shows a diagram for explaining the time profile of the respiratory gas flow for a normal respiration sequence and a sequence following it with additional oral respiration; FIG.

21 is a diagram for explaining the generation of the patient's physiological condition-specific evaluation result on the basis of measurement signals that are related to the respiration of the person, wherein from the measurement signals mentioned using several

Evaluation systems evaluation features are generated and at least one evaluation result is generated in a result-generating step based thereon by the evaluation features are subjected to a linking consideration.

The breath 1 shown in FIG. 1 with respect to the time course of the respiratory gas flow comprises an inspiratory phase I and an expiration phase E. The determination of the respiratory phase boundary G between the inspiration phase and the expiration phase is carried out by superimposed evaluation of several curve discussion criteria, in particular taking into account the currently prevailing respiratory pattern and the extreme values of the respiratory gas flow, the determined tidal volume and taking into account the respiratory phase periods of previous breaths. The in This figure shows the respiratory gas flow describes the respiratory gas flow in an undisturbed breath.

In Fig. 2 the course of the breathing gas flow over a longer time window is shown away. As can be seen from this illustration, the individual breaths vary in particular with regard to the minimas / maximas occurring in this case. The horizontal line 2 entered in this illustration clarifies the maximum respiratory gas flow statistically occurring for inspiration phases with the highest probability.

FIG. 3 shows the time profile of a signal indicative of the respiratory gas pressure, this signal having oscillation secrets 3a, 3b, 3c, 3d and 3e caused by snoring. The pressure fluctuations caused by snoring can be detected via a near-patient pressure detection device, for example a breathing gas pressure measuring tube. It is also possible to detect such pressure fluctuations via microphone devices or also on the basis of the power consumption of a respiratory gas delivery device.

FIG. 4 shows the time course of the respiratory gas flow for a plurality of breaths 1 interrupted by a respiratory arrest period 5. The respiratory arrest period 5, which is determined on the basis of the respiratory gas flow, has a time duration which exceeds a predetermined limit of, for example, 10 seconds and is thus classified as an apnea phase. Both the breaths detected in this presentation before respiratory arrest period 5 and the subsequent breaths show flow limitation features recorded in association with each breath.

Fig. 5 shows the time course of the respiratory gas flow with a hypopnea phase contained therein 6. The hypopnea phase 6 is considered present if after three classified as normal breaths 1 at least two but a maximum of three breaths follow their inspiratory difference volume compared to the three previous breaths a predetermined Exceeds limit. 6 shows the time course of the respiratory gas flow for several breaths, the first 4 breaths visible here showing 1 flow limitation features. These flow limitation features can be recognized in the illustrated course of the respiratory gas flow by plateaus 7 formed therein and by a plurality of local maxima 8. In the illustrated breaths, the flow limitation features occur in the inspiratory phase of the respective breath 1, respectively. The first 4 breaths 1 shown here are followed by three partially still flow-limited breaths 1 h, which are attributable to a hypopnea phase and partly also show flow limitation features.

FIG. 7 shows the course of the respiratory gas flow with a breathing period classified as stable. The respiratory gas flow, the respiratory rate and the amplitude of the respiratory gas flow are within a predetermined range which can be defined as a time range or by a number of breaths regularly, the respiratory stability moves in the case of the course of the respiratory gas flow shown here above a respiratory stability limit of 0.911. There are no respiratory disorders (OSA) during the phase of stable breathing shown here.

FIG. 8 shows the time course of the respiratory gas flow for a plurality of breaths, during which the respiratory flow is irregular and respiratory disorders (OSA) occur with individual breaths. In the embodiment shown here, the respiration stability index is below the limit of 0.911.

FIG. 9 shows the time profile of the respiratory gas flow in conjunction with a respiratory gas pressure signal. The respiratory gas pressure signal contains phase-wise high-frequency oscillations which, in the present example, can be attributed to inspiratory snoring.

FIG. 10 shows the time course of the respiratory gas flow for a plurality of breaths, the respiration being irregular in phases and starting at the time T1, for example due to mask leakage or through mouth opening caused disorder exists. From the time T1, a predetermined limit value is exceeded, which can be interpreted as an indication of a system disturbance due to mouth breathing or due to mask leaks.

The generation according to the invention of an evaluation result indicative of the physiological condition of a patient can be used to control the respiratory gas pressure in the context of positive-pressure ventilation. Such an application will be described below in conjunction with FIGS. 11 to 21. The supply of the respiratory gas to the patient is carried out using a nasally applied breathing mask which is connected via a breathing gas hose to a breathing gas source through which breathing gas is provided to a variable adjustable pressure level. This respiratory gas supply arrangement comprises a pressure detection device for generating a signal indicative of the respiratory gas pressure and a respiratory gas flow detection device for detecting a signal indicative of the respiratory gas flow. The signal indicative of the respiratory gas flow is analyzed by an evaluation device which generates evaluation features by using given evaluation systems. These evaluation features are considered linked and result in the fulfillment of predetermined connection criteria to changes in the respiratory gas pressure or information on the classification of the patient.

In the course of the signal indicative of the respiratory gas flow shown in FIG. 11, after the tenth breathing shown here, a first respiratory disturbance classified as an apnea phase occurs for a duration of approximately 15 seconds. This apnea phase is followed by a series of breaths, some of which have flow limitation features. Subsequent to these partially flow-limited breaths, a second phase of disturbed breathing, classified as apnea phase, follows, which also extends over a period of 15 seconds. After this second apnea phase follows a number (here six) of breaths, some of which have flowlimation-indicative characteristics. This breath sequence is followed by a phase of disturbed respiration which is classified here as the third apnea phase. Following this third apnea phase, three breaths of their tidal volume take an adaptively adjusted threshold exceeds and thus associated with a hypopnea phase. By the specified occurrence of said three apnea phases, the time interval of the apnea phases to each other and in view of the subsequent to the third apnea phase hypopnea phase, a linkage criterion is met, and thus an evaluation result that the previously set respiratory gas pressure rated too low and an increase in pressure by one Pressure level of 2 mbar causes. The breaths occurring after elevation of the respiratory gas pressure to a pressure of 11 mbar are further analyzed for features contained therein and linked over a larger time window.

FIG. 12 shows the course over time of the respiratory gas flow, wherein the evaluation of the detected respiratory flow signals recognizes a flow limited by respiration and successively leads to an increase of the respiratory gas pressure until a respiration which is classified as normal is carried out at predetermined time intervals.

The course of the signal which is indicative of the respiratory gas flow shown in FIG. 13 results in a first respiratory sequence being classified as a sequence of stable respiration, the condition of stable respiration continuing for a predetermined period of time causing a lowering of the respiratory gas pressure. The signals generated with this reduced respiratory gas pressure with regard to the respiratory gas pressure allow conclusions to be drawn about partially limited flow of respiration.

With regard to the flow limitation features which can be detected in the breaths, the respiratory gas pressure is increased again. However, the new respiratory gas pressure level is at least temporarily below the pressure level at which stable respiration was previously recognized.

The course of the signal indicative of the respiratory gas flow shown in FIG. 14 shows a number of apnea phases, sometimes with hypopnea phases following there. The temporal position of the apnea and the hypopnea phases leads to an evaluation result that the Regarding respiratory gas pressure classified as insufficient and causes an increase in the respiratory gas pressure.

The course of the signal which is indicative of the respiratory gas flow shown in FIG. 15 reveals three breath sequences which can be classified as hypopnea sequences. The temporal position of the hypopnea sequences to each other leads to a Auswertungsresulat that classifies the prevailing breathing gas pressure as insufficient and causes an increase in the breathing gas pressure. After the increase in the respiratory gas pressure, the course of the signal indicative of the flow of breathing gas makes it possible to detect a respiration that is classified as normal.

16 shows a sequence of the respiratory gas flow indicative signal showing flow limiting characteristics for the individual breaths, with oscillations occurring at the same time as flow limitation features appearing in the breaths in the respiratory gas pressure signal, which can be classified as inspiratory snoring.

The flow limitation features occurring in the individual breaths, in conjunction with the oscillations detected in the respiratory gas pressure signal, lead to an evaluation result which describes the prevailing respiratory gas pressure as insufficient and consequently causes an increase in the respiratory gas pressure.

The breaths detected after increasing the breathing gas pressure are classified as breaths of normal breathing.

As soon as the state of normal breathing lasts for a predetermined period of time, as shown in FIG. 17, the respiratory gas pressure can be lowered by, for example, 2 mbar. This reduced breathing gas pressure level is maintained as long as there are no flow limitation features in the individual breaths. If, even at this pressure level, respiration is classified as normal for a predetermined period of time, the respiratory gas pressure can be lowered further. After this phase of normal breathing, the breathing gas pressure can be further lowered as shown in FIG. If flow limitation features occur in the individual breaths detected during this further lowered respiratory gas pressure, then the respiratory gas pressure can be increased again on the basis of a linked analysis of the respiratory characteristics determined for the individual breaths.

FIG. 18 also shows the course of the signal indicative of the respiratory gas flow and of the respiratory gas flow in the case of a z. B. caused by mask leakage system failure. The pressure drop of the respiratory gas detected here and the increase of the respiratory gas flow occurring at the same time lead to the generation of an evaluation result which evaluates the current system state as disturbed. The system according to the invention is tuned such that in the case of a disorder classified as mask leakage, the delivery rate of the respiratory gas source is adjusted such that the respiratory gas pressure prevailing until the occurrence of the disturbance is largely maintained.

As can be seen from the illustration in FIG. 19, a system disturbance classified as mask leakage, for example as a result of temporary shifting of a respiratory mask, can be detected, for example. be rescinded after changing the head position of the patient and continue the breathing under the maintained even during system failure breathing gas pressure. Based on the signal indicative of the respiratory gas flow, as shown in FIG. 20, it can also be recognized whether oral respiration is present.

In Fig.21 the time course of an indicative of the respiratory gas flow signal S is shown. This signal is recorded, for example, as a so-called raw data signal by a pressure sensor connected to a dynamic pressure measuring point with a sampling frequency of eg 50 to 200 Hz. The raw data signal S can be determined via an approximation system 20 using approximation procedures implemented therein, eg series developments in the form of a fast Fourier analysis, an (eg) MP3 compression, Laplace Series development Binomial series development,

Correlation series development, etc. are recorded in compressed form.

The possibly compressed raw data of the signal S can be recorded within a data sequence D.

In the data sequence D, it is also possible, using several evaluation systems, to generate 21 evaluation features M, e.g. describe certain characteristics of breaths or time periods.

On the basis of the possibly compressed raw data of the signal S and / or the evaluation features M, at least one evaluation result is generated within the framework of a result generation step by subjecting the evaluation features M to a linking consideration.

In the case of using the system according to the invention for adjusting a respiratory gas pressure, one of the evaluation results may be a signal specifying, for example, the current respiratory gas pressure as appropriate, as too low or too high. As a further evaluation result, a possibly required change amount of the respiratory gas pressure can be determined. Also control parameters for the adjustment and synchronization of the respiratory gas pressure in a Bi- level pressure control can be determined as evaluation results.

The linking consideration of the evaluation features M is preferably carried out using Boolean operations, the Boolean variables Ai, A ₂ , B ... E ₂ ... from individual evaluation features M and / or by summary evaluation of the Auswerungsmerkmale M, for example

Evaluation feature groups ai, a ₂ , bi, c _{2 are} generated. The

Evaluation results can be the result of a large number of OR-linked operation systems.

On the basis of the evaluation results, raw data sets or evaluation feature sets can be selected which are used to generate the desired data Statements, such as pressure change amount, and typing indices (FLI, snore index, ...) are used.

The approximation system 20, the evaluation systems 21 and the systems for linking consideration of the evaluation features M and the preparatory generation of Boolean variables are preferably provided by a computer device configured by means of a program data record.

The evaluation results can be generated as part of a data post-processing, or used in real time - or sufficiently soon - in the setting of a breathing gas pressure or configuration of a pressure control system.

The evaluation results can be made available to a pressure control algorithm which is preferably configured in such a way that it offers at least two pressure control modes which differ in their reaction behavior in the case of a breathing gas pressure control. Thus, it is possible to operate a breathing gas pressure control system in a base mode in which certain events or a sum of events causes an increase in the breathing gas pressure.

In the context of a sensitive mode, it is possible to carry out the pressure control in such a way that it responds to events that may have been detected with a lesser delay. This sensitive mode can be adjusted in particular when the breathing gas pressure is e.g. was lowered after a phase of stable respiration (AS> 0.911).

According to the basic mode, it is preferably provided to initiate an increase in pressure when two large or three small apneas occur and the respiratory gas pressure is less than 14 mbar or a respiratory arrest is detected which exceeds a predetermined time duration of, for example, 2 minutes. In this case, a pressure increase by two mbar can be initiated. In the basic mode, an increase in pressure by 1 mbar can preferably be initiated when three hypopnoetic sequences are detected in a predetermined time sequence. Pressure increases by a pressure level of 1 mbar are preferably initiated when, with a respiratory stability index> 0.911, flow limitations occur in 5 out of 10 or even 10 out of 20 breaths.

The basic mode is further preferably tuned to cause pressure reduction when stable respiration having a respiratory stability index AS> 0.911 is present over a time period of at least 9 minutes. In this case, a pressure reduction is caused by preferably 2 mbar. In the context of the basic mode, a pressure change is suppressed, in particular, when the respiratory stability index is <0.911 and the limiting phenomena detected in the individual breaths do not exceed a predetermined severity criterion.

In the context of the sensitive mode, an increase of the respiratory gas pressure by, for example, 2 mbar is initiated when two large or three small apneas are present and the respiratory gas pressure <14 mbar. when three hypopnoetic sequences occur, the respiratory gas pressure is increased by 1 mbar.

When flow limitation features occur in the breaths studied, an increase in the respiratory gas pressure of 1 mbar is initiated when four out of four breaths have flow limitation characteristics and the respiratory stability index is> 0.911. An increase in pressure of 1 mbar is also caused when 5 out of 10 breath flow limitation features and the respiratory stability index <0.911. If 10 out of 20 breaths have flow limitation characteristics and the respiratory stability index is below a value of 0.911, in the sensitive mode there is also an increase in the respiratory gas pressure of 1 mbar.

A lowering of the respiratory gas pressure takes place in the sensitive mode already when stable respiration over a period of 3 minutes is present and the Respiratory stability index> 0.911. In this case, the breathing gas pressure can be lowered by, for example, 2 mbar

Similar to the basic mode mentioned above, even in the sensitive mode, no pressure change is caused when the respiration is classified as unstable and flow limitation characteristics are detectable in the individual breaths at a respiratory stability index <0.911.

Both in the normal mode and in the sensitive mode, it is preferably provided that events such as swallowing, coughing, mouth breathing, in particular expiratory mouth breathing, arousals and speech, at least then cause no change in respiratory gas pressure when the respiratory gas pressure is below a limit of, for example, 14 mbar.

The linking consideration, for example, can lead to pressure changes. It can also lead to the calculation of the patient-typical indices by selecting those relevant measurement data that are relevant for the respective index and were determined in a patient stage, which is highly informative. In real-time, as post-processing, or post-processing with reduced raw data volume with the reduction of the raw data volume achieved.

Claims

claims

1. A method for generating an evaluation result which is specific to the physiological state of a person on the basis of measurement signals which are related to the respiration of the person, wherein evaluation characteristics are generated from the said measurement signals using a plurality of evaluation systems and are evaluated within the framework of a result calculation based thereon. Generating step at least one evaluation result is generated by the evaluation features are subjected to a linking consideration.

2. The method according to claim 1, characterized in that a feature contribution contained predominantly in the evaluation features is generated within a generation time window that is smaller than a linking time window provided for the linking consideration.

3. The method according to claim 1 or 2, characterized in that on the basis of the linking consideration, a physiological typing is carried out.

4. The method according to at least one of claims 1 to 3, characterized in that based on the linking consideration, a configuration data set is generated to configure the breathing gas pressure control of a breathing gas supply device.

5. The method according to at least one of claims 1 to 4, characterized in that the evaluation features are generated on the basis of correlation criteria.

6. The method according to at least one of claims 1 to 5, characterized in that the evaluation features are generated on the basis of statistical evaluation procedures.

7. The method according to at least one of claims 1 to 6, characterized in that the evaluation features are generated as a feature field.

8. The method according to at least one of claims 1 to 7, characterized in that as the evaluation feature normal breathing phase lengths and / or normalatmungsbeschreibende information is generated.

9. Method according to claim 1, characterized in that flow limiting phase lengths and / or flow limiting characteristics or data records are generated as evaluation features.

10. The method according to at least one of claims 1 to 9, characterized in that as evaluation features apnea phase lengths and / or apnoe-characteristic information or data sets are generated.

11. The method according to at least one of claims 1 to 10, characterized in that as evaluation features snore phase lengths and / or snoring phase characteristic information is generated.

12. The method according to least one of claims 1 to 11, characterized in that are generated as evaluation features Cheyne-Stokes phase lengths or Cheyne-Stoke characteristic information or data sets.

13. The method according to at least one of claims 1 to 12, characterized in that as. Evaluation characteristics Hypoventilation phase lengths or hypoventilation characteristics or data sets are generated.

14. The method according to at least one of claims 1 to 13, characterized in that as evaluation features, pulmonary volume indicative information or records are generated.

15. The method according to at least one of claims 1 to 14, characterized in that are generated as evaluation features body position information or data records.

16. The method according to at least one of claims 1 to 15, characterized in that are generated as evaluation feature sleep phase characteristics or data sets.

17. The method according to at least one of claims 1 to 16, characterized in that generated evaluation features are stored by assigning their temporal position in the measurement signal acquisition period.

18. The method according to at least one of claims 1 to 17, characterized in that in the context of the linking consideration, a flow limitation index is generated.

19. The method according to at least one of claims 1 to 18, characterized in that in the context of linking an Appnoe / Hypopneaindex is generated

20. The method according to at least one of claims 1 to 19, characterized in that in the context of linking a snore index is generated.

21. The method according to at least one of claims 1 to 20, characterized in that in the context of the connecting consideration, a mouth breathing / nasal breathing index is generated.

22. The method according to at least one of claims 1 to 21, characterized in that in the context of linking a sleep time index is generated.

23. The method according to at least one of claims 1 to 22, characterized in that in the context of the linking consideration, a sleep phase index is generated.

24. The method according to at least one of claims 1 to 23, characterized in that an index relating to periodic breathing is generated in the context of the linking consideration.

25. The method according to at least one of claims 1 to 24, characterized in that a respiratory volume is determined in the context of the linking consideration.

26. The method according to at least one of claims 1 to 25, characterized in that in the context of the linking consideration, the evaluation features are considered with a weighting determined for the respective linkage.

27. The method according to at least one of claims 1 to 26, characterized in that the evaluation features are generated on the basis of a v-measurement.

28. The method according to at least one of claims 1 to 27, characterized in that at least a part of the evaluation features is generated taking into account the first and / or the second derivative of the time course of the respiratory gas flow.

29. The method according to at least one of claims 1 to 28, characterized in that in the context of the detection of the v signal, the pressure of the patient's incoming respiratory gas corresponds to the ambient pressure.

30. The method according to at least one of claims 1 to 28, characterized in that in the context of the detection v signal of the respiratory gas pressure is set to a different pressure from the ambient pressure level.

31. A device for generating a breathing result specific to the physiological state of a breathing person, based on measurement signals associated with the respiration of the person, with a measuring signal input device and a computer device for providing a plurality of evaluation systems, wherein the computer device is configured such that these are generated from the said measurement signals by the evaluation systems evaluation features and subjected to these evaluation features in the context of a resulting result generation step of a linking consideration and based on the linking consideration, an output signal or an output data set is generated or contains the evaluation result.