PRESSURE MEASUREMENT IN THE EXTRACORPOREAL BLOOD CIRCUIT

Abstract

A device and method for calibrating a first pressure sensor. The method includes: a) recursive analysis and forecasting of at least one correction function for finding a correction signal for the correction of a drift signal with the aid of a corresponding pressure reference signal, which is measured by the first pressure reference sensor, at constant internal pressure and at constant internal tube temperature; b) first calibration of a force signal, measured by the first pressure sensor and corrected using the correction signal, with the pressure reference signal, which is measured by the first pressure reference sensor, prior to an active use of the tube; and c) second calibration of the force signal, measured by the first pressure sensor and corrected using the correction signal, with the pressure reference signal, which is measured by a second pressure reference sensor during an active use of the tube.

Claims

1. A method for calibrating a first pressure sensor or force sensor which measures a first pressure of a liquid present inside a dialyzer tube in an extracorporeal circuit, in the form of a force signal, wherein the first pressure or force sensor abuts directly on the dialyzer tube and is integrated into or inserted into a first clamping device, the method comprising the following steps: a) regression analysis and prediction of at least one tube parameter independent correction function for finding a correction signal for correcting a drift signal achieved by clamping the dialyzer tube within the first clamping device using a corresponding force or pressure reference signal measured by a first force or pressure reference sensor at a constant internal tube pressure and a constant internal tube temperature; b) first calibrating of the pressure or force signal, measured by the first pressure or force sensor and corrected using the correction signal, with the force or pressure reference signal measured by the first force or pressure reference sensor before connecting the dialyzer tube to a patient; and c) second calibrating of the pressure or force signal, measured by the first pressure or force sensor and corrected using the correction function, with the force or pressure reference signal measured by a second force or pressure reference sensor when the dialyzer tube is connected to the patient.

2. The method for calibration according to claim 1, wherein, in addition to the first pressure, a second pressure is measured with a second force sensor or pressure sensor integrated in a second clamping device and corrected; and the pressure or force signal of the second pressure or force sensor is calibrated in the first calibration and in the second calibration with the force or pressure reference signal measured with the second force or pressure reference sensor.

3. The method for calibration according to claim 1, wherein the dialyzer tube comprises an arterial portion and a venous portion, and the first and/or second pressure sensor or force sensor and, before of the dialyzer tube is connected to the patient, the first force or pressure reference sensor, are arranged at the arterial portion, while the second force or pressure reference sensor comprises a common pressure sensor arranged at the venous portion, wherein the second force or pressure reference sensor is not integrated into a clamping device and does not generate a drift signal.

4. The method for calibration according to claim 1, wherein the first and the second force or pressure reference sensor each have a higher measurement accuracy than the corresponding pressure sensor.

5. The method for calibration according to claim 2, wherein for calibration of the second force or pressure sensor and for calibration of the first and the second force or pressure sensor the internal tube pressure is matched, via a bypass circuit in the arterial portion of the dialyzer tube, with the internal tube pressure in the venous portion of the dialyzer tube.

6. The method for calibration according to claim 1, wherein the constant internal tube pressure is obtained before the dialyzer tube is connected to the patient by adjusting a pumping ratio between a first pump and a second pump.

7. The method for calibration according to claim 1, wherein the drift signal is or corresponds to the reset force of the dialyzer tube in a clamped state.

8. The method for calibration according to claim 1, wherein the first and/or second pressure or force signal is converted into a pressure signal by the corresponding force or pressure reference signal via a linear recursion.

9. The method for calibration according to claim 4, wherein two pressure or force sensors are used, wherein the first pressure or force sensor is arranged at an inlet opening of the first pump, and wherein the second pressure or force sensor is arranged at an outlet opening of the first pump.

10. A device comprising an extracorporeal circuit, at least one pressure or force sensor integrated in a clamping device, for internal tube pressure measurement in a fluid-filled dialyzer tube with arterial portion and venous portion, at least one force or pressure reference sensor provided for referencing a pressure or force signal output by this pressure sensor or force sensor and which is not designed as a clamping device and comprising at least a first pump and a second pump and which is provided to apply the method for calibrating the pressure or force signal of the at least one pressure or force sensor by a reference signal of the at least one force or pressure reference sensor.

Description

BRIEF DESCRIPTION OF THE DRAWING FIGURES

[0030] FIG. 1 shows a diagram that illustrates the pressure course in a clamping device over time by way of example;

[0031] FIG. 2A shows a clamping device in which a tube is clamped;

[0032] FIG. 2B shows a diagram showing the course of a drift signal compared to a reference signal;

[0033] FIG. 3A shows the front of a dialysis machine in a state before active use of the tube;

[0034] FIG. 3B shows an alternative arrangement of two sensors;

[0035] FIG. 4 shows a diagram showing the drift behavior of a tube in a closed clamping device;

[0036] FIG. 5 shows a diagram showing the course of the drift signal at a constant internal tube pressure and the correction function representing the drift behavior of the tube;

[0037] FIG. 6 shows a diagram showing graphically the determination of the pressure signal from the drift signal/force signal;

[0038] FIG. 7 shows the front of a dialysis machine in active use of the tube with a patient connected to the machine;

[0039] FIG. 8 shows a diagram showing an example of the pressure course at a first pressure sensor with time when the method is performed;

[0040] FIG. 9 shows a diagram showing the pressure courses of a conventional pressure sensor, of a first pressure reference sensor and of a first pump;

[0041] FIG. 10A shows a diagram showing the pressure courses of two conventional pressure sensors and of a second pressure reference sensor recorded simultaneously with the course curves of FIG. 10A;

[0042] FIG. 10B shows a diagram showing the time course of a (simulated) blood flow rate of a patient and a patient blood pressure;

[0043] FIG. 11A shows a diagram showing a deviation between a calculated pressure signal and a reference signal due to a temperature drift;

[0044] FIG. 11B shows a diagram related to the diagram in FIG. 11A showing the temperature course of a fluid in the tube at the first and second pressure sensors;

[0045] FIG. 12A shows a diagram showing the deviation of the pressure signal of the second pressure reference sensor from the corresponding pressure reference signal and showing a linear correction signal;

[0046] FIG. 12B shows a diagram related to the diagram of FIG. 12A showing the courses of the pressure signal corrected with the linear correction signal and of the pressure reference signal;

[0047] FIG. 13A shows a diagram showing the deviation of the pressure signal of the second pressure reference sensor from the corresponding pressure reference signal and a polynomized correction signal; and

[0048] FIG. 13B shows a diagram which is related to the diagram of FIG. 13A and shows the courses of the pressure signal corrected with the polynomized correction signal and of the pressure reference signal.

DETAILED DESCRIPTION

[0049] Embodiments of the present disclosure are described below based on the accompanying figures. It should be noted that the figures shown are exemplary only and are not limiting.

First Embodiment

Overall Method

[0050] FIG. 1 shows an example of the pressure course in millimeters of mercury (mmHg) of a pressure sensor (here the PBE pressure sensor, which is explained in more detail below) with time t in seconds (s).

[0051] In phase 1, a tube is inserted into two clamping devices, into each of which at least one pressure sensor is integrated and measures a pressure in the form of a force signal, at a dialysis machine and the tube is filled with a fluid. The tube system is filled by varying the flow pump speed of at least one pump. In this phase, leak tests are also carried out on the machine and tube.

[0052] In phase 2, the pressure in the tube is kept constant. After a short settling phase, the already described step a) is performed for regression analysis and prediction of at least one correction function for finding a correction signal for correcting the drift signal using a corresponding pressure reference signal. In step a), a correction function f is determined as a function of time t with two constants a.sub.0, b for the viscoelastic behavior of the tube before therapy. The two constants a.sub.0 and b are determined using a mathematical method referred to hereinafter as ‘fitting’. This method is explained below. With the aid of this function, a drift signal in the form of a force signal of the respective pressure sensor is determined.

[0053] In phase 3, a pressure drop can be detected, which is used to correct the force signal using the correction function determined in step a). Furthermore, the corrected force signal of the pressure sensor is converted into a corrected pressure signal using the correspondingly provided pressure reference signal of a pressure reference sensor via a linear relationship between reference signal and corrected force signal. The pressure reference sensor is a conventional pressure sensor. Phase 3 thus shows the sequence of step b), which takes place before the start of therapy. In step b), the corrected force signal is calibrated with the pressure reference signal of the corresponding pressure reference sensor before the therapy. Calibration may also be performed using a second constant pressure level.

[0054] A conventional pressure sensor is a pressure sensor that is not integrated in a clamping device and in which no reset force influences the pressure signal. With a conventional pressure sensor, an internal tube pressure is determined, for example, via a T-piece or a pressure pod or something similar (described above).

[0055] In phase 4, after a predetermined time after the start of therapy, preferably after 5 minutes, the newly corrected force signal is calibrated again with the pressure reference signal of the corresponding pressure reference sensor. In phase 4, step c) is thus carried out.

[0056] In phase 5, the course of therapy is illustrated during which the PBE pressure is largely constant. It can be seen that the corrected PBE pressure signal and the reference signal are superimposed, which means that the correction of the pressure signal is sufficient and also works with varying pressure (see time interval between approx. 2600-2700 s).

[0057] The clamping device, which serves as a pressure sensor, the structure of a machine for which the method according to the invention is used, and steps a) to c) are described in detail and by way of example below. Furthermore, alternative embodiments of the invention are given below.

Background

[0058] FIG. 2A shows a (first) tube 1 whose internal pressure can be measured by a force sensor 2. For this purpose, the tube 1 is clamped in a clamping device 3. This (clamping device 3) clamps the tube 1 and an expansion or contraction of the tube 1 is transmitted to the force sensor 2 via a force-transmission means 4. The change in force that the force sensor 2 can measure is proportional to the change in internal pressure in the tube 1.

[0059] FIG. 2B shows a graph which displays the pressure course of a drift signal and the pressure course of a reference signal over time. The pressure signal/drift signal shown is the course of a pressure in the tube 1 clamped into the clamping device 3 at time t0. The reference pressure has a value of 0 mmHg (ambient pressure) over the entire time curve shown. The pressure signal shows a pressure increase at time t0 and then a logarithmically decreasing pressure course, which can be explained by the reset force of the tube. This drift signal has to be subtracted from the actual measurement signal, so that the absolute pressure equals the reference pressure.

Set-Up for Carrying Out the Method According to the Invention

[0060] FIG. 3A shows the front of a dialysis machine 6, to which a (first) tube 1 is attached, the internal pressure of which is to be measured at various points. The dialysis machine 6 has an extracorporeal circuit. The tube 1 has an arterial portion/branch 1a and a venous portion 1b. A first substituate port SP1 connects the arterial portion 1a of the tube 1 to the machine 6, and a second substituate port SP2 connects the venous portion 1b of the tube 1 to the machine 6. In the embodiment shown, the tube 1 is not connected to a patient, that is, the tube is not in active use and therefore in a state before therapy. Therefore, the tube 1 is not filled with blood but with another fluid, which is a substitute (electrolyte fluid/eloat).

[0061] The fluid is first conveyed into the arterial tube portion 1a via a dialysate input flow pump FPE, which is located outside the front of the dialysis machine 6. Before the fluid reaches the front of the dialysis machine 6, a first pressure reference sensor PHOP measures the internal tube pressure or takes a reading of the internal tube pressure. The pressure reference sensor PHOP is thus also arranged at the substituate port SP and is an additional pressure sensor compared to conventional dialysis machines 6. After the fluid enters the front of the dialysis machine 6, it first passes the arterial tube clamp SAKA, which is usually open. Then the fluid passes the first clamping device, which is also called PA pressure sensor or first pressure sensor PA, and thus the first force sensor. The first clamping device is integrated in the front side of the dialysis machine 6. The PA sensor measures the pressure in the arterial portion 1a of the tube 1. The pressure reference sensor PHOP can be used for referencing the first pressure sensor PA, since it has a higher measuring accuracy than the first pressure sensor PA.

[0062] The fluid then reaches the first pump, a blood pump BP, which continues to convey the fluid. Finally, the fluid passes a second clamping device, also called a PBE pressure sensor or second pressure sensor PBE, and thus the second force sensor. The PBE pressure sensor measures the dialysate inlet pressure at a point downstream of the blood pump BP in the flow direction of the medium in the tube. Behind the PBE pressure sensor, the fluid may pass through a dialyzer 8.

[0063] However, in the case of a bypass circuit via the bypass 10, it is also possible that the fluid does not flow through the dialyzer, but bypasses it. The venous tube portion 1b is located behind the dialyzer/bypass in the direction of fluid flow. At a point downstream of the dialyzer/bypass and upstream of an air trap/deaerator 12, where air trapped in the fluid is removed from the fluid, the fluid in the venous tube portion 1b passes a conventional pressure transducer, which is referred to as a PV measurement point. The conventional pressure transducer may be, for example, a T-piece or a pressure pod.

[0064] After the PV measuring point, the fluid passes through the deaerator 12, then through an air detector 14 and finally through a venous tube clamp SAKV, which is normally open. The venous and arterial tube clamp SAKV, SAKA are only closed in the event of a fault and block the patient access during therapy. Such an error may be, for example, that the air detector detects an amount of air greater than a certain threshold. After the fluid has passed through the venous tube clamp SAKV, it flows out via the substitute port SP2 with the aid of a pump output of a dialysate output flow pump FPA, which is arranged outside the front of the dialysis machine 6.

[0065] Furthermore, FIG. 3A shows that the dialysis machine 6 is connected to a CPU having a first computer portion, a second computer portion, and a third computer portion. Here, the CPU can control the dialysate input flow pump FPE, the dialysate output flow pump FPA, the pressure reference sensor PHOP, the pressure reference sensor PV, the first pressure sensor PA, the second pressure sensor PBE, the blood pump BP, the arterial tube clamp SAKA, and the venous tube clamp SAKV.

[0066] FIG. 3B shows the blood pump BP and an alternative arrangement of the first and second pressure sensors PA and PBE. In this case, the first (arterial) pressure sensor PA is located directly at the blood inlet of the blood pump BP and the second (dialyzer inlet) pressure sensor PBE is located directly at the blood outlet of the blood pump BP. In this case, both pressure sensors PA and PBE are integrated into the blood pump BP and the tube material, the temperature and the insertion time of the tube for both pressure sensors PA, PBE are identical. The expected drift behavior is also identical for both pressure sensors PA, PBE. Thus, the differential pressure P.sub.PBE−P.sub.A of both pressure sensors can be determined without correction of the drift, i.e. without calibration. The differential pressure P.sub.PBE−P.sub.A should correspond to the difference of the corrected pressures P.sub.PBE_korr−P.sub.A_korr during the entire therapy. By comparing the two differential pressures (uncorrected with corrected difference), the correctness of the correction function, which is described later, can be assessed. If the differential pressures differ from each other by more than a predetermined amount, recalibration of the system is advisable.

Step a)

[0067] Similar to FIG. 2B, FIG. 4 shows the signal course of the clamped tube after closing the clamping device. However, the force signal of the first pressure sensor PA is shown here as an example in the form of voltage values in the unit volt (V) over time tin seconds (s). At time t0 (t0=0 s), the clamping device is closed. The signal course already known from FIG. 2B is shown for a period of time before and during therapy. The decrease in the signal course can be explained by the viscoelastic properties of the tube material, which influences the pressure transmission between the force sensor and the fluid in the tube. The elastic portion of the tube generates a reset force. The viscous part of the tube leads to a slow, irreversible deformation of the tube. This tube deformation causes the reset force to decrease and thus also the force with which the tube presses against the force sensor. The reset force course shown in FIG. 4 is also referred to as a drift signal. In order to be able to represent the force signal or pressure signal of the force sensor as a signal that depends only on the internal tube pressure, it is useful to determine the drift signal in order to be able to calculate/eliminate it from the measured force signal or pressure signal, i.e. to subtract the drift signal from the measured force signal.

[0068] The viscoelastic behavior of the tube generally follows equation (1), so that the drift signal can be described as a mathematical correction function, which thus follows the following equation:

f(t)=a.sub.0.Math..sup.−b   (1)

[0069] Here, t is the time and a.sub.0 and b are unknown constants. f(t) has the unit V, since the force signal is output as a voltage value.

[0070] For further use of the equation, it is useful to determine the constants a.sub.0 and b by fitting. For this purpose, it is necessary to generate a constant internal tube pressure by adjusting the pumping ratio of the pumps BP and FPE or FPA. Furthermore, a constant internal tube temperature is required. The substitute flowing through the tube is preheated to 36°, so that the internal tube temperature is also constant. Furthermore, the force signal of a pressure sensor, here of the first pressure sensor PA, is determined in a measurement. However, the force signal of another pressure sensor, such as the second pressure sensor PBE, can also be used. The course of the force signal of the first pressure sensor PA over time can be seen in FIG. 5.

[0071] In FIG. 5, the force signal is shown as a voltage with the unit volt [V] as a function of time t in seconds [s]. In the time range from 0 s to approx. 1200 s, during the so-called ‘priming’, a state exists before active use of the tube, i.e. before therapy. Before therapy, no patient is connected to the dialysis machine 6 and in this case the dialysis machine 6 is configured as described and shown in connection with FIG. 3A. In the time range starting at about 1200 s, the voltage signal during therapy is shown, also referred to as ‘therapy’. During therapy, a patient is connected to the dialysis machine 6, which is then configured as described and shown in connection with FIG. 7.

[0072] In order to be able to determine the constants a.sub.0 and b using this force signal, the respective signal values of the force signal f(t=t1) and f(t=t2) are determined in the fitting at two specific times t1 and t2 in the range before the therapy. In the example shown in FIG. 5, the time points t1=600 s and t2=800 s were selected. This results in the following system of equations of formulas (2) and (3), which have to be solved to obtain the constants a.sub.0 and b:

I f(t=t1)=a0.Math.t1.sup.31 b   (2)

II f(t=t2)=a0.Math.t2.sup.−b   (3)

[0073] From equation I, i.e. formula (2), a.sub.0 can be represented as follows:

a.sub.0=f(t=1).Math.t1.sup.b   (4)

[0074] Substituting a.sub.0 in the form, shown in formula (4), into formula (3) gives b in the form shown in formula (5), in which it now depends only on the known times t1, t2 and the corresponding signal values of the force signal f(t=t1) and f(t=t2) and can thus be calculated:

b=ln(f(t=t2))−ln(f(t=t1))/(ln(t1)−ln(t2))   (5)

[0075] After determining the value of b, this value can be inserted into formula (4), so that the value of the constant a.sub.0 is obtained, and equation (1) represents the viscoelastic behavior of the tube used. With the times t1 and t2 selected above and the voltage values applicable in this experiment, the value of a.sub.0 is 1.19 and the value of b is 0.03.

[0076] Step a) is shown here as an example for the PA pressure sensor and is performed analogously for the PBE pressure sensor.

Step b)

[0077] Next, the measured force signal is to be corrected and to be converted into a pressure signal using the appropriate pressure reference signal. This is done in a phase before the therapy and when the internal tube pressure changes, for example when it drops, or via a second constant pressure level with a different pressure compared to the first level. The internal tube pressure drops, for example, when the tube is disconnected from the substitute ports SP1, SP2 in preparation for therapy. A second pressure level can be set by a different pumping ratio of blood pump and flow pump. Step b) is carried out as an example for the first pressure sensor PA.

[0078] First, the drift signal f(t) with the calculated values for a.sub.0 and b (value of f(t) obtained from equation (1)) is subtracted from the force signal P.sub.S_gem measured with the first pressure sensor PA, which is output as a voltage value (in V). Thus, the corrected force signal P.sub.S_Korr follows the following equation (formula (6)):

P.sub.S_Korr=P.sub.S_gem−f(t)   (6)

[0079] Then, using the first pressure reference sensor PHOP, pressure reference values P.sub.PHOP are acquired in the unit mmHg and are plotted over the associated voltage value P.sub.S_Korr in [V] in a diagram, which is shown as an example in FIG. 6.

[0080] In the diagram in FIG. 6, the measured pressure reference values P.sub.PHOP on the y-axis are plotted as points above the calculated voltage values P.sub.S_korr on the x-axis, and a linear progression/course is shown for these points. Matching this, a straight line A is determined mathematically, which runs best through these points and thus represents the relationship between voltage values P.sub.S_korr and the pressure correction values P.sub.A_Korr calculated from them. This relationship can be stated mathematically as follows in the form of formula (7):

P.sub.A_Korr=m.Math.P.sub.S_korr+t   (7)

[0081] Here, m is the slope of the line, which is also called the scaling value, and t is the pressure reference value at which the line intersects the y-axis and which is also called the offset value. In the example shown, the scaling value is 5278 mmHg/V and the offset value is 23 mmHg.

[0082] This means that the corrected and thus correct pressure signal P.sub.A_korr of the first clamping device, i.e. of the first pressure sensor PA, is known before therapy and step b) is completed.

[0083] Step b) for the PBE pressure sensor is analogous to the procedure shown here for the PA pressure sensor, but here the PV pressure reference sensor is used as reference instead of the PHOP pressure reference sensor.

[0084] Since, for example, the internal tube pressure values and/or the internal tube temperatures may change from a state before therapy to a state during therapy, it is recommended that the correction for the measured force signal of the first and/or second pressure sensor is performed repeatedly during therapy.

Step c)

[0085] During therapy, the setup shown in FIG. 2A changes as shown in FIG. 7. In FIG. 7, it can be seen that the arterial and venous tube portions 1a and 1b are connected to the patient. In this case, the patient's heart replaces the dialysate inflow and outflow pumps. An (arm) artery, which is connected to the arterial tube portion 1a, and an (arm) vein of the patient, which is connected to the venous tube portion 1b, are connected to each other via an artificial connection 16, in particular via a patient shunt. As a result, the same blood pressure and the same blood flow values are present in the vein and artery (can also be referred to as blood vessels in general terms) of the patient. By setting a bypass on the dialyzer, the same blood pressures and the same blood flow values are also present in the arterial and venous tube portion 1a, 1b and thus in the entire system consisting of tube and patient veins. For experiments, it is conceivable to simulate an experimental patient circuit which has a water pump, a heated water bath and a counterpressure valve.

[0086] The principle for calibration and referencing is for step c) equal/identical as for step b). Once again, the measured force signal of the pressure sensor is corrected by the correction signal found in step a), and the corrected pressure signal can be calculated from the relationship between the corrected pressure signal and the corrected force signal known from step b) (cf. formula (6) with the scaling value and offset value determined in step b)).

[0087] Step c) may be performed for the first and second PA, PBE pressure sensors. However, here the conventional PV pressure reference sensor serves as the pressure reference sensor for both the PBE pressure sensor and the PA pressure sensor to allow simultaneous referencing of the PA pressure signal and the PBE pressure signal.

[0088] To simplify matters, it is possible to reference only the pressure signal of the second pressure sensor PBE using the pressure reference sensor PV by switching the dialyzer flow to the bypass. Due to the production-related identical properties of the tube at the position of the PA clamping device and at the position of the PBE clamping device, the correction function found for the PBE pressure signal can be applied to the PA pressure signal, even if this method is not as precise as determining a correction function for the PA pressure signal and the PBE pressure signal, respectively.

Result of the Method According to the Invention

[0089] After filtering and scaling the corrected pressure signal, this signal can be compared with the corresponding, directly measured pressure reference signal. This comparison is shown in FIG. 8 as an example for the corrected pressure signal P.sub.A_korr and the corresponding pressure reference signal P.sub.PHOP.

[0090] The diagram in FIG. 8 shows that the course of the calculated pressure correction signal P.sub.A_Korr and the course of the directly measured pressure reference signal P.sub.PHOP are congruent. This means that the mathematical correction function from (1) together with the calculated values for a.sub.0 and b can eliminate the drift course, but this only works if the internal tube pressure and the internal tube temperature are constant. If, for example, the internal tube pressure varies when the second voltage value f(t2) is measured, the correction function is not suitable for describing the course of the drift signal.

Referenceability

[0091] In the following, FIG. 9 and FIG. 10A show the pressure courses of the conventionally used pressure sensors for measuring the arterial pressure P.sub.A and the dialyzer inlet pressure P.sub.PBE in comparison with the pressure course of the corresponding pressure reference sensors PHOP and PV, respectively.

[0092] In FIG. 9, the pressure course of a conventional first pressure sensor PA_herk for measuring arterial pressure is compared with the pressure course of the first pressure reference sensor PHOP before therapy. The pressure courses in mmHg are shown in the course with time t in s. Furthermore, the pressure course of the blood pump BP is shown, which repeatedly shows a pressure of 0. The blood pump BP is repeatedly stopped to generate constant pressure values of the PA_herk sensor and of the PHOP sensor.

[0093] It can be seen that the curves of the two pressure signals of the PA_herk pressure sensor and of the PHOP pressure reference sensor are similar and run parallel to each other, here with a parallel shift of the curves/a pressure difference of approx. 20 mmHg. The difference between the two pressure signal curves is due to the height differences of the first pressure sensor PA_herk and the first pressure reference sensor PHOP. In this configuration example, the first pressure reference sensor PHOP is mounted higher than the first pressure sensor PA (see FIG. 2A).

[0094] Although the pressure difference between the conventional pressure sensor and the pressure reference sensor has to be taken into account, the comparison in FIG. 9 shows that the pressure reference sensor is suitable as a reference sensor for arterial pressure.

[0095] FIG. 10A shows the pressure courses of the first conventional pressure sensor PA_herk for measuring the arterial pressure and of a second conventional pressure sensor PBE_herk for measuring the dialyzer inlet pressure in comparison with the pressure course of the second pressure reference sensor PV in mmHg with time tin s during therapy. FIG. 10B shows a diagram depicting the time courses of the blood pump flow in ml/min and of the simulated patient pressure in mmHg, which were recorded simultaneously with the pressure values from FIG. 10A.

[0096] In the ranges in which the blood pump flow is 0, i.e. at the times when the blood pump is stopped, the pressure signals of the sensors PA_herk, PBE_herk and PV match each other and are constant. In these constant pressure ranges, the pressure courses of the two pressure sensors PA_herk and PBE_herk are essentially congruent and there is a pressure difference to the pressure course of the pressure reference sensor PV, which in this case is about 20 mmHg and can be explained by the height difference between the pressure sensors PA_herk, PBE_herk and the pressure reference sensor PV.

[0097] Again, although the pressure difference between the conventional pressure sensor and the pressure reference sensor needs to be considered, the comparison in FIG. 9 shows that the second pressure reference sensor is suitable as a reference sensor for arterial pressure and for dialysate inlet pressure during therapy.

Temperature Drift

[0098] So far, the above description has assumed a constant internal tube pressure and a constant internal tube temperature. However, between step b) and step c), i.e. between the phase before therapy and the phase during therapy, there may be temperature differences in the filled tube which can lead to linear deviations between the reference sensor value and the pressure sensor value.

[0099] Such a deviation is exemplarily shown for the PBE pressure sensor and the PV pressure reference sensor in the diagram of FIG. 11A. FIG. 11B shows the corresponding temperature course at the measuring points of the PBE pressure sensor and the PV pressure reference sensor. In order to be able to determine the temperature at these two measuring points, it is necessary to integrate a temperature sensor into the PBE clamping device and/or into the PA clamping device. Here, calibration before therapy (up to approx. 900 s) takes place at T1, for example at 35.8° C., while calibration during therapy takes place at T2, for example at 37.2° C. Due to the temperature difference ΔT (=T2−T1) before and during therapy, which is here 1.4° C., the calculated PBE pressure signal shown in FIG. 11B does not follow the PV reference signal during therapy, but deviates linearly from it. Investigations have shown that the deviation between pressure signal and reference signal is linearly proportional to the temperature deviation before and during therapy. The pressure signal can be corrected using an empirically determined correction function.

[0100] FIG. 12A shows the pressure deviation between pressure signal and reference signal and a straight line B over time tin s found for it. In the example, the straight line equation has the following form, which is shown in formula (8):

PBE.sub.Signal=−0.0064182.Math.ΔT+3.4282   (8)

[0101] FIG. 12B shows the PBE pressure signal corrected using the determined formula (8), which is now again shown to be congruent with the PV reference signal.

[0102] As an alternative to a linear relationship between the deviation between pressure signal and reference signal and the temperature deviation before and during therapy, a polynomial relationship may also exist. However, more computing power is required to calculate the corresponding formula, even though such a relationship can represent the deviation more accurately than a linear one.

[0103] FIG. 13A shows, like FIG. 12A, the pressure deviation between pressure signal and reference signal and a polynomial C calculated for it over time t in seconds. In the example, the PBE pressure signal follows the polynomial deviation shown in formula (9):

PBE.sub.Signal=2.687.Math.10.sup.−6.Math.ΔT.sup.2−0.018486+15.8826   (9)

[0104] FIG. 13A shows the PBE pressure signal corrected using the determined formula (9), which is now again shown to be congruent with the PV reference signal.

Second Embodiment

[0105] The second embodiment is similar to the first embodiment, therefore only the differences to the first embodiment are elaborated below.

[0106] The reference pressure measurement of the pressure signals PBE and PA in step c) (during therapy) may also be performed using the venous tube clamp SAKV and the arterial tube clamp SAKA as an alternative to the method described in the first embodiment.

[0107] For this purpose, the dialyzer flow is switched into the bypass, as in the first configuration example, and the tube clamps SAKV and SAKA are closed. A pressure-tight connection is created in the tube. The blood pump BP is stopped. However, due to the delay, the blood pump BP continues to rotate for a short time after the stop, so that a negative pressure builds up in the arterial tube portion and a positive pressure builds up in the venous tube portion, which have the same pressure ratio to each other over time. The PBE pressure signal is calibrated using the PV pressure reference sensor. The PA pressure signal can also be calibrated using the PV pressure reference sensor.

[0108] Since the patient is cut off from the extracorporeal circuit due to the closed tube clamps SAKV and SAKA, this alternative version can be performed in step c) independently of the patient. However, in this case, the blood no longer circulates in the extracorporeal circuit, so that the blood can clot in the circulation and the temperature may drop. However, the coagulation of the blood and its temperature drop depend on the duration of the blood pump stop, which should therefore be as short as possible.