MEDICAL DEVICE FOR DETERMINING THE POSTURE OF PATIENT

Abstract

The present invention relates to a method for determining the posture of a patient. The method comprises the steps of: initiating (50, 52) a patient posture determining session by performing an electrical bio-impedance measurement session in at least one of a number of different electrode configurations in order to measure an impedance value for the at least one configuration; obtaining reference impedance values (54) stored in advance for the at least one configuration and for at least one potential posture of the patient; comparing (54) the measured impedance value for the at least one configuration with corresponding stored reference impedance values for at least one potential posture of the patient; and determining (56) the present posture of the patient by using results from the comparison between measured impedance values and the stored reference impedance values. Furthermore, the invention relates to a medical device for determining the posture of a patient and a computer readable medium comprising instructions for bringing a computer to perform the inventive method.

Description

TECHNICAL FIELD

The present invention generally relates to implantable medical devices, such as cardiac pacemakers and implantable cardioverter/defibrillators, and in particular to a method, a medical device, a computer program product and a computer readable medium for determining the posture of a patient.

BACKGROUND OF THE INVENTION

A severe problem associated with measurement of, inter alia, the blood pressure of the chambers, contractility, endocardial acceleration, blood flow, coronary blood flow, the sinus rate, the electrical bio-impedance, such as the thoracic impedance and the cardiogenic impedance is that the accurateness and reliability, and, hence the repeatability, of the obtained signals are greatly affected by factors like the body position of the patient, patient activity levels, heart rate frequency, etc. In particular, it has been found that the body position of the patient is of major importance with regard to the blood pressure of the chambers, contractility, endocardial acceleration, blood flow, coronary blood flow, the sinus rate, the electrical bio-impedance, such as the thoracic impedance and the cardiogenic impedance, etc. Repeatable measurements of such parameters are of a great value for identifying changes of many different conditions in the body of a patient. For example, for many algorithms it is important to know the position or posture of the patient to be able to compare obtained data to data recorded previously, for example at pulmonary edema monitoring, measurements of electrical bio-impedance (e.g. cardiac impedance) in order to calculate trends on how a patient's disease progresses or on how a certain condition progresses in a accurate and reliable way. Electrical bio-impedance signals have, for example, been found to constitute an effective measure for identifying changes of many different conditions in the body of a patient, such as incipient pulmonary edema and the progression of pulmonary edema due to CHF, i.e. the accumulation of fluids in the lung-region associated with pulmonary edema affects the thoracic impedance, or more specifically the DC impedance level, since the resistivity of the lung changes in accordance with a change of the ratio of fluid to air.

Clinical and pre-clinical measurements have, in fact, shown that the electrical bio-impedance is dependent to a significant extent also regarding different positions even when the patient is lying down, for example, whether the patient is lying on a side or is lying on the back. Regarding impedance measurements, a major reason for this is that the measurement depends on the measurement vector, i.e. the vector between the nodes that the current is applied between and the vector the voltage is measured between. When the body shifts position, these vectors will change since the gravity will influence, for example, tissue between the nodes and how it moves. This significant posture dependence applies both when looking at the morphology of the impedance signal in general and the DC-levels of the impedance signal. Pre-clinical trials on sheep have shown that the DC-level of the impedance may vary as much as 25-40% depending on the subjects posture: This is true for both the thoracic impedance, for example, measured over the lungs (for example between a case of an implantable device and an electrode placed within the heart) and the cardiogenic impedance, for example the DC impedance measured within the heart (for example RV-RA or RV-LV).

Accordingly, there is a need of a method and medical device that are capable of determining the posture of a patient in an accurate and reliable way.

BRIEF DESCRIPTION OF THE INVENTION

Thus, an object of the present invention is to provide a method and medical device that are capable of determining the present posture of a patient in an accurate and reliable way.

This and other objects are achieved according to the present invention by providing a method, medical devices, and a computer readable medium having the features defined in the independent claim. Preferable embodiments of the invention are characterised by the dependent claims.

In the context of this application, the term “impedance” refers to the DC component of the impedance. The measured impedance consists of a DC component and an AC component, where the DC component is the baseline around which the AC component fluctuates. The DC component reflects the amount of tissue and fluids that are located between the measuring points that the impedance is measured in-between and the AC components reflects how respiration and cardiac activity influence the impedance signal. Hence, the term “impedance value” corresponds to the DC impedance level.

For the purpose of clarity, the term “intra thoracic impedance” refers to an impedance measurement over the thorax by using an implantable medical device, i.e. an impedance measurement where the impedance measurement vector spans over the thorax.

Moreover, in order to clarify, the term “cardiogenic impedance” is defined as the impedance or resistance variation that origins from cardiac contractions.

According to an aspect of the present invention, there is provided a method for determining the posture of a patient. The method comprises the steps of: initiating a patient posture determining session by performing an electrical bio-impedance measurement session in at least one of a number of different electrode configurations in order to measure an impedance value, i.e. the DC impedance level, for the at least one configuration; obtaining or reading reference impedance values stored in advance for the at least one configuration and for at least one potential posture of the patient; comparing the measured impedance value for the at least one configuration with corresponding stored reference impedance values for at least one potential posture of the patient; and determining the present posture of the patient by using results from the comparison between measured impedance values and the stored reference impedance values.

According to a second aspect of the present invention, there is provided a medical device for determining the posture of a patient, the device being connectable to a patient in a number of different electrode configurations. The medical device according to the present invention comprises means for measuring an electrical bio-impedance at each of the number of electrode configurations, wherein the means for measuring an impedance is adapted to, upon receiving a triggering signal, initiate a patient posture determining session by performing an impedance measurement session in at least one of the electrode configurations in order to measure an impedance value for the at least one configuration; and processing means including:

means for obtaining or reading reference impedance values stored in advance in a storing means for the at least one configuration and for at least one potential posture of the patient; means for comparing the measured impedance values with corresponding stored reference impedance values for at least one potential posture of the patient; and means for determining the present posture of the patient by using results from the comparison between measured impedance values and the stored reference impedance values.

According to a third aspect of the present invention, there is provided a computer program product, which when executed on a computer, performs steps in accordance with the first aspect of the present invention.

According to a further aspect of the present invention, there is provided a computer readable medium comprising instructions for bringing a computer to perform a method according to the first aspect of the present invention.

Thus, the invention is based on the insight that the significant impedance variance between different postures of a subject and different electrode configurations for performing the impedance measurements within the subject in fact can be used to determine the present posture of the subject. The invention utilizes the findings of clinical and pre-clinical measurements showing that the electrical bio-impedance is posture dependent to a great extent also regarding different positions or postures even when the patient is lying down, for example, whether the patient is lying on a side or is lying on the back. According to the invention, the present impedance, which preferably is the DC impedance, of the patient is measured in at least one electrode configuration and compared with corresponding reference impedance values of a reference impedance matrix. The result from the comparison is used to determined the present posture of the patient. Consequently, variations of the DC impedance values of the patient are used to identify the present posture.

This invention provides several advantages. For example, one advantage is that reliable and accurate determinations of the posture of the patient can be achieved. These, in turn, can be used to, for example, obtain repeatable measurements of parameters such as blood pressure of the chambers, contractility, endocardial acceleration, blood flow, coronary blood flow, the sinus rate, the electrical bio-impedance, such as the thoracic impedance and the cardiogenic impedance. Repeatable measurements of such parameters are of a great value for identifying changes of many different conditions in the body of a patient. For example, for many algorithms it is important to know the position or posture of the patient to be able to compare obtained data to data recorded previously, for example at pulmonary edema monitoring, measurements of electrical bio-impedance (e.g. cardiac impedance) in order to calculate trends on how a patient's disease progresses or on how a certain condition progresses in a accurate and reliable way. Electrical bio-impedance signals have, for example, been found to constitute an effective measure for identifying changes of many different conditions in the body of a patient, such as incipient pulmonary edema and the progression of pulmonary edema due to CHF. Another advantage is that the method and device for determining the posture of the patient is customized to the specific patient. A conventional position sensor can not be adapted to a specific patient in such a convenient and efficient way. Yet another advantage is that the need of an additional position sensor in the implantable device is removed.

According to an embodiment of the present invention, at least one reference impedance value is obtained for each of a predetermined number of electrode configurations and each of a predetermined number of postures, which at least one value is stored the reference value matrix, thus containing reference impedance values for each electrode configuration and each potential posture. This training of the device may be performed at a doctors office, where the patient, under the doctor's supervision, is instructed to change his or her posture to a number of predetermined distinct postures. In each posture an impedance value is measured for each electrode configuration and is stored as a reference impedance value in the reference matrix. Alternatively, the patient may perform this training in accordance with, for example, written instructions, and, hence, the assistance of a doctor or any hospital stuff is not required in this case.

In another embodiment of the present invention, it is checked whether a difference between the measured impedance value for the at least one configuration and the corresponding stored reference impedance values is within a predetermined interval; and if the difference is identified to be outside the interval at a predetermined number of posture determining sessions, the reference impedance values are updated by using the measured impedance values. Preferably, the measured impedance values from the latest posture determining session are used to replace the former reference values in order to update the reference matrix. In another embodiment, a mean value of the impedance values of the predetermined number of posture determining sessions is calculated and used to update the reference matrix. In yet another embodiment, a weighted average value of the impedance values of the predetermined number of posture determining sessions is calculated and used to update the reference matrix. In still another embodiment, a moving average of the impedance values of the predetermined number of posture determining sessions is calculated and used to update the reference matrix.

In a preferred embodiment of the present invention, each impedance value is calculated as a mean value of the measured impedance signal over a predetermined period of time. Alternatively, the breathing rate of the patient is sensed. This sensed rate can be used, for example, to calculate each measured impedance value as a mean value of the measured impedance signal over a predetermined number of breaths. In another embodiment, the measurements are synchronized with the breathing cycle of the patient and each impedance value is calculated as a mean value of the measured impedance signal over a predetermined number of breathing cycles. Hence, the accuracy of the impedance signals can be further increased.

According to embodiments of the present invention, a deviation factor is calculated for the at least one potential posture by using corresponding measured impedance values and stored reference impedance values, which deviation factor for the at least one potential posture is used to determine the present posture. According to one embodiment, the deviation factor is the square difference sum of the measured impedance values and corresponding stored reference impedance values for at least one potential posture and the potential posture of the at least one potential postures having the lowest square difference sum is identified. This posture is determined to be the actual or present posture of the patient.

In another embodiment of the present invention, it is checked whether the square difference for said at least one potential posture is within a predetermined interval and the present posture of patient is determined to be the potential posture having a square difference sum within the predetermined interval. That is, during the comparison procedure is checked whether each respective measured impedance value is within the interval for each respective reference value of each potential posture. The potential posture where all measured impedance values is within respective interval is determined to be the present posture of the patient.

According to an alternative, a heart rate of the patient is sensed, it is determined whether the heart rate is within a predetermined heart rate interval; and the patient posture determining session is initiated if said sensed heart rate is within said predetermined heart rate interval. Consequently, the positions determination is not performed if the heart rate has increased above a predetermined upper limit of the interval or if the heart rate has decreased below a predetermined lower limit since it may indicate, for example, that the patient is moving too much or suffering from an infection, which, in turn, may affect the impedance measurements. Moreover, the procedure for determining the posture of the patient can be customized and adjusted to the needs and life pattern of the specific patient. Hence, the accuracy and reliability of the position determination can be further improved.

In an alternative embodiment of the present invention, an activity level of the patient is sensed, it is determined whether the sensed activity level is below a predetermined activity level; and the patient posture determining session is initiated if said sensed activity level is below said predetermined activity level. Thus, the positions or posture determination is only performed if the patient is not moving overly much since such an event may affect the measurements. This feature also allows a posture detection procedure adjusted to the specific patient. Thereby, the accuracy and reliability of the posture determination can be further improved.

In yet another embodiment of the present invention, a body temperature of the patient is sensed, it is determined whether the temperature is within a predetermined temperature level interval; and the patient posture determining session is initiated if said sensed body temperature is within said predetermined temperature level interval. Thus, it is assured that the body temperature of the patient is within a predetermined range during the posture determination. This entails the procedure for determining the posture of the patient can be customized to a higher extent. Thereby, the accuracy and reliability of the position determination can be further improved since an increased or decreased temperature may affect the measurements

Alternatively, signals related to heart pumping activity of the patient are sensed, the cardiac output based on the signals is determined, it is determined whether the cardiac output is within a predetermined cardiac output level interval; and the patient posture determining session is initiated if the sensed cardiac output is within the predetermined cardiac output level interval. Moreover, the procedure for determining the posture of the patient can be customized and adjusted to the needs and life pattern of the specific patient. Thereby, the accuracy and reliability of the position determination can be improved.

Preferably, an updating indication signal indicating that the reference impedance values require an updating is initiated if a posture not have been determined in a patient posture determining session during a predetermined period of time. This may be the case if the patient has lost or gained weight, especially around the thorax since the new/lost tissue will cause changes in the pressure the tissue surrounding the heart will put on the heart in different positions and, thus, affecting the DC levels of the impedance.

In an alternative embodiment, the patient posture determining session can be initiated at a certain point of time during the day. In other words, it is possible to correlate the patient posture determining session with a suitable point of time during the day, for example, during the night when the activity of patient may be lower.

As realized by the person skilled in the art, the methods of the present invention, as well as preferred embodiments thereof, are suitable to realize as a computer program or a computer readable medium.

These and other advantages with, and aspects of, the present invention will become apparent from the following detailed description including exemplifying embodiments and from the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

In the following detailed description, reference will be made to the accompanying drawings, of which:

FIG. 1 is schematic diagram showing a medical device implanted in a patient in which device the present invention can be implemented.

FIG. 2 is block diagram of the primary functional components of a first embodiment of the medical device according to the present invention.

FIG. 3 is a flow chart illustrating the steps in accordance with one embodiment of the present invention for determining the posture of a patient.

FIG. 4 is block diagram of the primary functional components of a further embodiment of the medical device according to the present invention.

FIG. 5 is block diagram of the primary functional components of a another embodiment of the medical device according to the present invention.

FIG. 6 is a schematic diagram showing a classifying network in accordance with the present invention.

FIG. 7 a is a diagram showing the normalized impedance value vectors and their associated body postures used in a neural network in accordance with the present invention.

FIG. 7 b is a diagram showing the classifying of a new impedance value vector in a neural network in accordance with the present invention.

DETAILED DESCRIPTION OF THE INVENTION

With reference to FIG. 1, there is shown a schematic diagram of a medical device implanted in a patient in which device the present invention can be implemented. As seen, this embodiment of the present invention is shown in the context of a pacemaker 2 implanted in a patient (not shown). The pacemaker 2 comprises a housing being hermetically sealed and biological inert. Normally, the housing is conductive and may, thus, serve as an electrode. One or more pacemaker leads, where only two are shown in FIG. 1 namely a ventricular lead 6a and an atrial lead 6b, are electrically coupled to the pacemaker 2 in a conventional manner. The leads 6a, 6b extend into the heart 8 via a vein 10 of the patient. One or more conductive electrodes for receiving electrical cardiac signals and/or for delivering electrical pacing to the heart 8 are arranged near the distal ends of the leads 6a, 6b. As the skilled man in the art realizes, the leads 6a, 6b may be implanted with its distal end located in either the atrium or ventricle of the heart 8, or in the coronary sinus or in the great cardiac vein, or they may be in form of epicardial leads attached directly at the epicardium

With reference now to FIG. 2, the configuration including the primary components of an embodiment of the present invention will be described. The illustrated embodiment comprises an implantable medical device 20, such as the pacemaker shown in FIG. 1, and leads 26a and 26b, of the same type as the leads 6a and 6b shown in FIG. 1. The leads 26a, 26b may be unipolar or bipolar, and may include any of the passive or active fixation means known in the art for fixation of the lead to the cardiac tissue. As an example, the lead distal tip (not shown) may include a tined tip or a fixation helix. The leads 26a, 26b comprises one or more electrodes (as described with reference to FIG. 1), such as a tip electrode or a ring electrode, arranged to, inter alia, measure the impedance or transmit pacing pulses for causing depolarization of cardiac tissue adjacent to the electrode(-s) generated by a pace pulse generator 25 under influence of a control circuit 27. Preferably, the medical device 20 is connectable to the patient by means of a number of different electrode configurations including RV-tip (i.e. the distal electrode in a bipolar lead located in right ventricle) and RV-ring (i.e. the proximal electrode in a bipolar lead located in right ventricle), LV-tip (i.e. the distal electrode in a bipolar lead located in left ventricle) and RV-tip (i.e. the distal electrode in a bipolar lead located in right ventricle), LV-tip (i.e. the distal electrode in a bipolar lead located in left ventricle) and RA-tip (i.e. the distal electrode in a bipolar lead located in right atrium), VC-tip (i.e. the distal electrode in a bipolar lead located in vena cava) and RA-tip (i.e. a distal electrode in a bipolar lead located in right atrium), or RA-tip (i.e. a distal electrode in a bipolar lead located in right atrium) and RV-tip (i.e. a distal electrode in a bipolar lead located in right ventricle) for excitation currents and RV-tip (i.e. the distal electrode in a bipolar lead located in right ventricle) and RV-ring (i.e. the proximal electrode in a bipolar lead located in right ventricle), LV-ring (i.e. the proximal electrode in a bipolar lead located in left ventricle) and RV-ring (i.e. the proximal electrode in a bipolar lead located in right ventricle), LV-ring (i.e. the proximal electrode in a bipolar lead located in left ventricle) and RA-ring (i.e. the proximal electrode in a bipolar lead located in right atrium), VC-coil (i.e. the conductor in a bipolar lead having a helical configuration located in vena cava) and RA-ring (i.e. the proximal electrode in a bipolar lead located in right atrium), or RA-ring (i.e. the proximal electrode in a bipolar lead located in right atrium) and RV-ring (i.e. the proximal electrode in a bipolar lead located in right ventricle) for voltages. Of course, as the skilled man realizes, there are other conceivable configurations that can be used.

The control circuit 27 controls pace pulse parameters such as output voltage and pulse duration. Moreover, an impedance circuit 29 is arranged to carry out the impedance measurements. The impedance circuit 29 is arranged to apply the excitation current pulses between a first electrode of the above mentioned electrode configurations and second electrode of the configurations. The impedance circuit 29 is also arranged to measure the impedance in the tissues between the first and second electrode to the excitation current pulse of the configuration. Further, the impedance circuit 29 is coupled to a microprocessor 30, where processing of the measured impedance values can be performed. In an embodiment where the cardiac component of the electrical bio-impedance is measured, the impedance circuit 29 is arranged to apply an excitation current pulse between a first electrode and a second electrode arranged within the heart of the patient and to measure the impedance in the tissues between the first and second electrode to the excitation current pulse. By means of the different electrode configurations a number of different impedance measurements can be performed. For example, the excitation current can be measured between RV-tip and RV-ring, LV-tip and RV-tip, LV-tip and RA-tip, VC-tip and RA-tip, or RA-tip and RV-tip, and, correspondingly, the voltage between RV-tip and RV-ring, LV-ring and RV-ring, LV-ring and RA-ring, VC-coil and RA-ring, or RA-ring and RV-ring. See above for definitions of the abbreviations. Of course, as the skilled man realizes, there are other conceivable configurations that can be used.

The impedance sensing circuit 29 is controlled by the microprocessor 30 and the control circuit 27. The control circuit 27 acts under influence of the microprocessor 30. A storage means 31 is connected to the control circuit 27 and the microprocessor 30, which storage means 31 may include a random access memory (RAM) and/or a non-volatile memory such as a read-only memory (ROM). The implantable medical device 20 is powered by a battery 33, which supplies electrical power to all electrical active components of the medical device 20. Data contained in the storage means 31 can be transferred to a programmer (not shown) via a programmer interface (not shown) for use in analyzing system conditions, patient information, calculation of surrogate parameters such as systolic and diastolic slopes, the pre-ejection period, or left ventricular ejection time and changing pacing conditions, etc. Furthermore, data can be transferred from the programmer to the storage means 31.

Measured impedance values are obtained by the microprocessor 30 from the impedance measurement means 29 and stored in the storage means 31. The microprocessor may be adapted to normalize the measured impedance values before they are used for posture determination. The medical device 20 further comprises comparing means 35 adapted to compare impedance values obtained by means of the impedance circuit 29 via one of, some of, or all of the above-mentioned electrode configurations with corresponding stored reference impedance values, which reference impedance values are stored in a reference value matrix in the storage means 31. The reference impedance values may be created in a training session, which will be described in detail below, or may be obtained from a programmer as standard values.

Moreover, determining means 37 is adapted to determine a present posture of the patient by using the results from the comparison between the measured impedance values and corresponding stored reference impedance values. In a preferred embodiment, the comparing means 35 and the determining means 37 are integrated into the microprocessor 30. However, the comparing means 35 and the determining means 37 may also, as the skilled man realize, be implemented as external from the microprocessor 30 but connected to it. Furthermore, the comparing means 35 and determining means 37 may be realized by means of software or hardware.

The medical device 20 may also comprise alarm means 38 adapted to send an updating indication signal indicating that the reference impedance values require an updating if a posture have not been determined in a patient posture determining session during a predetermined period of time. That is, the microprocessor 30 monitors whether a posture is detected in each posture determining session and if, if a posture have not been determined in a patient posture determining session during a predetermined period of time, it sends a triggering command to the alarm means 38. The alarm means 38 may be a vibrator causing the device to vibrate or it may be adapted to deliver a beeping sound in order to alert the patient of the situation. Furthermore, an alarm signal can, for example, also or instead be sent to the programmer (not shown) via the programmer interface (not shown). In a preferred embodiment, the alarm means 38 is integrated into the microprocessor 30.

According to an embodiment of the present invention, see FIG. 4, the medical device 20′ comprises means for sensing a breathing rate of the patient 39 connected to the microprocessor 30. The means for sensing a breathing rate of the patient 39 can be adapted to filtering the impedance signal values by means of a band-pass filter adapted to the breathing rate.

The impedance measurement sessions can be executed over a predetermined number of breaths, and the impedance values can be calculated as a mean value over a predetermined number of breaths in accordance with:

$\begin{array}{cc}\mathrm{DC\_impedance}\mathrm{\_level}=\frac{1}{k}\sum _{n=1}^kZ0\left(n\right)& \left(\mathrm{eq}.1\right)\end{array}$

Alternatively, the reference impedance values are calculated as mean values of the measured impedance signal over a predetermined number of breathing cycles according to equation 1, for example, 2-10 cycles. In yet another embodiment, each reference value is calculated as a mean value of the measured impedance signal over a predetermined period of time, for example, about 16 seconds.

The medical device 20′ may also include means for sensing a heart rate 41 connected to the microprocessor 30. The microprocessor 31 is adapted to determine whether the sensed heart rate is within a predetermined heart rate interval, for example, a heart rate between 60-90. If it is found that the sensed heart rate is within the predetermined heart rate interval, the processing means 30 is adapted to bring the impedance measurement means 29 to initiate a patient posture determining session.

Furthermore, the medical device 20′ may include means for sensing an activity level of said patient 43. The microprocessor 30 is adapted to determine whether the sensed activity level is below a predetermined activity level, for example, an activity lower than slow walk. If it is found that the sensed activity level is within the predetermined activity level interval, the processing means 30 is adapted to bring the impedance measurement means 29 to initiate a patient posture determining session.

In addition, the medical device 20′ may include means for sensing a body temperature of said patient 45, for example, a thermistor or other temperature sensor located in the vicinity of the heart connected to the microprocessor 30. The microprocessor 30 is adapted to determine whether the temperature is within a predetermined temperature level interval, for example, a temperature within the interval 36.5-37.5° C. and bring the impedance measurement means 29 to initiate the patient posture determining session if the sensed body temperature is within said predetermined temperature level interval.

According to a further embodiment, the medical device 20 may comprise means for sensing signals related to heart pumping activity of the patient (not shown) connected to the microprocessor 30, which may be adapted to determine the cardiac output based on the obtained signals, to determine whether the cardiac output is within a predetermined cardiac output level interval, and to bring the impedance measurement means 29 to initiate the patient posture determining session if the sensed cardiac output is within the predetermined cardiac output level interval.

As the skilled man realize, none, only one of, some of or all of the following features: the means for sensing a breathing rate of the patient 39, the means for sensing a heart rate 41, the activity level sensing means 43, temperature sensing means 45, or the means for sensing signals related to heart pumping activity of the patient may be included in the medical device. That is, criteria from one, some, all the following features: the means for sensing a breathing rate of the patient 39, the means for sensing a heart rate 41, the activity level sensing means 43, temperature sensing means 45 can be combined in order to find appropriate conditions for initiating the patient posture determining session.

The reference value matrix discussed above, contains reference DC-level impedance values for a number of electrode configurations and for a number of different postures of the patient. An example of such a matrix is shown below in table 1. As can be seen, the matrix contains reference values for the electrode configurations config 1 to config x for the postures supine, prone, sitting, and standing. But, as the skilled man realizes, there are other postures than can be included, for example, lying on the right side or the left side.

TABLE 1
Posture	Config 1	Config 2	Config 3	Config x
Supine	S1	S2	S3	Sx
Prone	P1	P2	P3	Px
Sitting	Si1	Si2	Si3	Six
Standing	St1	St2	St3	Stx

As discussed above, the impedance is very posture sensitive and pre-clinical trials on sheep has shown that the DC-level of the impedance may vary 25-40% depending on the subject's position. This is true for both the DC-impedance measured over the lungs (e.g. between the case of the device and an electrode within the heart) and for the DC-impedance measured within the heart (e.g. the RV-RA or RV-LV). Hence, the reference values shown in table 1 will vary significantly between different postures for an electrode configuration and between different configurations for the same posture.

According to a preferred embodiment, these reference values S1, S2, . . . , Stx are created during a training session. For example, this session can be performed at a doctors office, where the patient is located in the four positions and an impedance measurement sessions is performed for each configuration Config 1 to Config x in each position. That is, the patient is first instructed to position his or her body in the first posture, for example, supine. Then, an impedance measurement session is initiated in a first electrode configuration, for example, RV-tip-RV-ring for the excitation current and RV-tip-RV-ring for the voltage in order to measure the reference impedance values for the first configuration. Thereafter, an impedance measurement session is initiated in a second electrode configuration, for example, LV-tip-RV-tip for the excitation current and LV-ring-RV-ring for the voltage in order to measure the reference impedance values for the second configuration. This will be repeated for all electrode configurations. After the reference impedance values has been measured for each electrode configuration for the first posture, in this exempla supine, the patient is instructed the change his or hers posture to the following, in this example, prone. Thereafter, the procedure is repeated in accordance with the above given description. Accordingly, the medical device is provided with reference impedance values customized for the specific patient, thereby increasing the accuracy and reliability of the position determination.

The measured impedance values are stored in the reference value matrix. Preferably, each impedance value is calculated as the mean value of the impedance signal over a predetermined number of breaths in accordance with equation 1, see above.

Alternatively, the reference impedance values are calculated as mean values of the measured impedance signal over a predetermined number of breathing cycles according to equation 1, for example, 2-10 cycles. In yet another embodiment, each reference value is calculated as a mean value of the measured impedance signal over a predetermined period of time, for example, about 16 seconds.

According to a preferred embodiment of the present invention, the reference DC impedance values (S1 . . . Sx, P1 . . . Px, Si1 . . . . Six, St1 . . . Stx) are normalized, which is performed within each DC impedance configuration. In a first alternative normalization procedure, each reference value is normalized with the sum of all the impedance values of the configuration according to the following:

$S1\mathrm{norm}=S1/Z1\mathrm{norm},Z1\mathrm{norm}=\left(S1+P1+\mathrm{Si}1+\mathrm{St}1\right);$ $S2\mathrm{norm}=S2/Z2\mathrm{norm},Z2\mathrm{norm}=(S2+P2+\mathrm{Si}2+\mathrm{St}2);$ $⋮$ $\mathrm{St}3\mathrm{norm}=\mathrm{St}3/\mathrm{Zt}3\mathrm{norm},\mathrm{Zt}3\mathrm{norm}=\left(S3+P3+\mathrm{Si}3+\mathrm{St}3\right);$ $\mathrm{and}$ $\mathrm{Stx}\mathrm{norm}=\mathrm{Stx}/\mathrm{Ztx}\mathrm{norm},\mathrm{Ztxnorm}=\left(\mathrm{Sx},\mathrm{Px},\mathrm{Six},\mathrm{Stx}\right)$

That is, each configuration is normalized such that the sum of the normalized DC impedance reference values for each configuration is 1.

In another alternative normalization procedure the measurements for each configuration is divided with the largest DC impedance value in the configuration according to the following:

$S1\mathrm{norm}=S1/Z1\mathrm{norm},Z1\mathrm{norm}=\max \left(S1+P1+\mathrm{Si}1+\mathrm{St}1\right);$ $S2\mathrm{norm}=S2/Z2\mathrm{norm},Z2\mathrm{norm}=\max (S2+P2+\mathrm{Si}2+\mathrm{St}2);$ $⋮$ $\mathrm{St}3\mathrm{norm}=\mathrm{St}3/\mathrm{Zt}3\mathrm{norm},\mathrm{Zt}3\mathrm{norm}=\max \left(S3+P3+\mathrm{Si}3+\mathrm{St}3\right);$ $\mathrm{and}$ $\mathrm{Stx}\mathrm{norm}=\mathrm{Stx}/\mathrm{Ztx}\mathrm{norm},\mathrm{Ztxnorm}=\max \left(\mathrm{Sx},\mathrm{Px},\mathrm{Six},\mathrm{Stx}\right)$

Thus, irrespective which normalization procedure being used, the reference matrix will now contain normalized reference values for the electrode configurations config 1 to config x for the postures supine, prone, sitting, and standing.

TABLE 2
Posture	Config 1	Config 2	Config 3	Config x
Supine	S1norm	S2norm	S3norm	Sxnorm
Prone	P1norm	P2norm	P3norm	Pxnorm
Sitting	Si1norm	Si2norm	Si3norm	Sixnorm
Standing	St1norm	St2norm	St3norm	Stxnorm
Norm. factor	Z1norm	Z2norm	Z3norm	Z4norm

At initiation of the training session for finding the reference values, a configuration verification procedure is performed in order to check whether a specific configuration is qualified. That is, whether a specific configuration suitable to use. According to a first embodiment, this is performed in accordance with the following. The examples disclosed hereinafter are described with reference to the normalized values, but, as the skilled man realize, the non-normalized values can of course be used instead. First, it is checked that the difference between the DC impedance values for different postures are larger that a predetermined value Q. This predetermined value Q can be programmed in advance to the training sessions or can be determined manually at initiating of the procedure. Hence, the following check is performed for each configuration y:

min(|Synorm-Pynorm|,|Synorm-Siynorm|,|Synorm-Stynorm|, . . . ,|Stynorm-Siynorm|)>Q  (eqt. 2)

According to an alternative procedure, the sum of all differences is calculated and, thereafter, it is checked whether the sum is larger than a predetermined value R. This predetermined value R can be programmed in advance to the training sessions or can be determined manually at initiation of the procedure. Hence, the following check is performed for each configuration y:

sum(|Synorm-Pynorm|,|Synorm-Siynorm|,|Synorm-Stynorm|, . . . ,|Stynorm-Siynorm|)>R  (eqt. 3)

If the condition in either equation 2 or 3 is not satisfied, the specific configuration is judged not to discriminate different postures of the patient in a sufficient way and should therefore be excluded from use when determining a posture of the patient.

According to further embodiments, an identifying procedure is performed during the training session in order to identify which configurations that should be used in combination with each other to provide a sufficient separation between DC impedance reference values from each other for the different postures. In a one embodiment, two configurations is checked first and the square difference is calculated for each postures DC impedance reference values into the other postures in accordance with the following:

$\mathrm{MSE\_SP}\mathrm{\_false}={\left(S1\mathrm{norm}-\frac{P1}{Z1\mathrm{norm}}\right)}^2+{\left(S2\mathrm{norm}-\frac{P2}{Z2\mathrm{norm}}\right)}^2$ $\mathrm{MSE\_SiP}\mathrm{\_false}={\left(\mathrm{Si}1\mathrm{norm}-\frac{P1}{Z1\mathrm{norm}}\right)}^2+{\left(\mathrm{Si}2\mathrm{norm}-\frac{P2}{Z2\mathrm{norm}}\right)}^2$ $⋮$ $\mathrm{MSE\_SSt}\mathrm{\_false}={\left(S1\mathrm{norm}-\frac{\mathrm{St}1}{Z1\mathrm{norm}}\right)}^2+{\left(S2\mathrm{norm}-\frac{\mathrm{St}2}{Z2\mathrm{norm}}\right)}^2$ $⋮$

If all MSE_false values are larger than a predetermined value (which may be programmable), the two configurations are considered to be sufficient. If not all MSE_false values are larger than the predetermined value, two other configurations are tested. This is repeated until either a combination being good enough has been identified or all combinations has been tested. If all combination of two configurations has been tested without identifying a suitable combination, the above mentioned MSE_false values are calculated for three configurations instead of for two. This is repeated until either a combination (of three configurations) being good enough has been identified or all combinations has been tested. The above-mentioned procedure can be repeated for four, five combinations and so on until a suitable combination of configurations is found.

According to an alternative embodiment, the MSE_false values are calculated in accordance with the description above, but the procedure starts with all configurations and then removes one configuration at a time and checks the impact on the MSE_false values. For example, in the example described above with reference to table 1 and 2, the procedure starts with 4 configurations and one configuration is removed and the MSE_false values are calculated. This is repeated for all four configurations, i.e. each of the four configurations is removed one at a time and the MSE_false is calculated. The configuration having the least impact on the MSE_false values is determined to be the least important configuration and is removed from the test. This is then repeated for the three remaining configurations and so on.

Instead of performing a training session as described above, impedance reference values can be obtained from a look-up table containing standard values, which may be normalized in accordance with the above given description For example, these standard values can be transferred to the device 20 wirelessly from a programmer.

Referring now to FIG. 3, a high-level description of the method for determining the posture of a patient according to the present invention will be given. First, at step 50, a patient posture determining session is initiated by performing an electrical bio-impedance measurement session in the different electrode configurations in order to measure an impedance value for the configuration. Preferably, impedance measurement sessions are initiated in several configurations and the microprocessor 30 is adapted to send a control instruction to the impedance circuit 29 containing information regarding which electrode configuration that is to be used in, if several are utilized, in which order they should be used. Preferably, each impedance value is calculated as the mean value of the impedance signal over a predetermined number of breaths in accordance with equation 1. Alternatively, the reference impedance values are calculated as mean values of the measured impedance signal over a predetermined number of breathing cycles according to equation 1, for example, 2-10 cycles. In yet another embodiment, each reference value is calculated as a mean value of the measured impedance signal over a predetermined period of time, for example, about 16 seconds. The measured impedance values for the x used configurations newZ1, newZ2, newZ3, newZx are then stored temporarily in storage means 31.

Thereafter, at step 52, the corresponding reference impedance values, shown in table 1 or table 2, for the postures supine, prone, sitting and standing are collected from the reference matrix in the storage means together with the present impedance values. Then, at step 54, the measured impedance values are compared in the comparing means 35 with corresponding reference impedance values. A deviation factor is determined for each position and according to a preferred embodiment, the square error sum for each potential position or posture is calculated in accordance with the following equations 4-7:

$\begin{array}{cc}\mathrm{MSE\_S}={\left(\frac{S1-\mathrm{newZ}1}{S1}\right)}^2+{\left(\frac{S2-\mathrm{newZ}2}{S2}\right)}^2+\dots +{\left(\frac{\mathrm{Sx}-\mathrm{newZx}}{\mathrm{Sx}}\right)}^2& \left(\mathrm{eqt}.4\right)\\ \mathrm{MSE\_P}={\left(\frac{P1-\mathrm{newZ}1}{P1}\right)}^2+{\left(\frac{P2-\mathrm{newZ}2}{P2}\right)}^2+\dots +{\left(\frac{\mathrm{Px}-\mathrm{newZx}}{\mathrm{Px}}\right)}^2& \left(\mathrm{eqt}.5\right)\\ \mathrm{MSE\_Si}={\left(\frac{\mathrm{Si}1-\mathrm{newZ}1}{\mathrm{Si}1}\right)}^2+{\left(\frac{\mathrm{Si}2-\mathrm{newZ}2}{\mathrm{Si}2}\right)}^2+\dots +{\left(\frac{\mathrm{Six}-\mathrm{newZx}}{\mathrm{Six}}\right)}^2& \left(\mathrm{eqt}.6\right)\\ \mathrm{MSE\_St}={\left(\frac{\mathrm{St}1-\mathrm{newZ}1}{\mathrm{St}1}\right)}^2+{\left(\frac{\mathrm{St}2-\mathrm{newZ}2}{\mathrm{St}2}\right)}^2+\dots +{\left(\frac{\mathrm{Stx}-\mathrm{newZx}}{\mathrm{Stx}}\right)}^2& \left(\mathrm{eqt}.7\right)\end{array}$

If, instead the normalized values found in table 2, the deviation factor is determined for each position in accordance with the following equations 8-11:

$\begin{array}{cc}\mathrm{MSE\_S}={\left(S1\mathrm{norm}-\frac{\mathrm{newZ}1}{Z1\mathrm{norm}}\right)}^2+{\left(S2\mathrm{norm}-\frac{\mathrm{newZ}2}{Z2\mathrm{norm}}\right)}^2+\dots +{\left(\mathrm{Sxnorm}-\frac{\mathrm{newZx}}{\mathrm{Zxnorm}}\right)}^2& \left(\mathrm{eqt}.8\right)\\ \mathrm{MSE\_P}={\left(P1\mathrm{norm}-\frac{\mathrm{newZ}1}{Z1\mathrm{norm}}\right)}^2+{\left(P2\mathrm{norm}-\frac{\mathrm{newZ}2}{Z2\mathrm{norm}}\right)}^2+\dots +{\left(\mathrm{Pxnorm}-\frac{\mathrm{newZx}}{\mathrm{Zxnorm}}\right)}^2& \left(\mathrm{eqt}.9\right)\\ \mathrm{MSE\_Si}={\left(\mathrm{Si}1-\frac{\mathrm{newZ}1}{Z1\mathrm{norm}}\right)}^2+{\left(\mathrm{Si}2-\frac{\mathrm{newZ}2}{Z2\mathrm{norm}}\right)}^2+\dots +{\left(\mathrm{Sxnorm}-\frac{\mathrm{newZx}}{\mathrm{Zxnorm}}\right)}^2& \left(\mathrm{eqt}.10\right)\\ \mathrm{MSE\_St}={\left(\mathrm{St}1\mathrm{norm}-\frac{\mathrm{newZ}1}{Z1\mathrm{norm}}\right)}^2+{\left(\mathrm{St}2\mathrm{norm}-\frac{\mathrm{newZ}2}{\mathrm{St}2}\right)}^2+\dots +{\left(\mathrm{Stxnorm}-\frac{\mathrm{newZx}}{\mathrm{Stx}}\right)}^2& \left(\mathrm{eqt}.11\right)\end{array}$

Finally, at step 56, the present posture of the patient is determined in the determining means 37 by using the results from the comparison between measured impedance values and the stored reference impedance values and the subject, i.e. the patient, is determined to be in the position having the lowest square error sum. In the preferred embodiment, where the square difference sum for each potential position or posture was calculated, the patient is determined to be in the position or posture that has the lowest square difference sum.

In another embodiment of the present invention, an interval or range is set for each reference value. During the comparison procedure is checked whether each respective measured impedance value is within the interval for each respective reference value of each potential posture. The potential posture where all measured impedance values is within respective interval is determined to be the present posture of the patient.

According to yet another embodiment of the present invention, the sum of the differences is calculated according to the following equations 12-15:

SUM_—S=S1−newZ1+S2−newZ2+S3−newZ3+ . . . +Sx−newZx  (eqt. 12)

SUM_—P=P1−newZ1+P2−newZ2+P3−newZ3+ . . . +Px−newZx  (eqt. 13)

SUM_—Si=Si1−newZ1+Si2−newZ2+Si3−newZ3+ . . . +Six−newZx  (eqt. 14)

SUM_—St=St1−newZ1+St2−newZ2+St3−newZ3+ . . . +Stx−newZx  (eqt. 15)

The potential posture having the lowest total sum is then determined to be the actual posture of the patient. Of course, these sums can also be normalized.

With reference now to FIGS. 5, 6, 7a and 7b, a further embodiment of the present invention will be discussed. Turning first to FIG. 5, the medical device according to this embodiment will be described, where parts and/or means similar to parts and/or means shown in FIGS. 2 and 4 will be denoted with the same reference numerals and the description thereof will be omitted in the following. The medical device 60 comprises a classifying network 64, which utilizes a Neural network or a Fuzzy logic structure. The design and general function of a Neural network or a Fuzzy logic classifying network is well known for the man skilled within the art and descriptions thereof are, therefore, omitted. During the training session, measured impedance values from the different electrode configurations are obtained by the microprocessor 30 from the impedance measurement means 29, stored in the storage means 31, and normalized in the processor 30. The classifying network 64 is adapted to operate in a learn mode and in a classifying mode. In the learn mode the classifying network 64 is trained to recognize a known or specific body posture and in the classifying mode the network 64 identifies a body posture by means of measured impedance values obtained during a body posture determining session. The output of the classifying network is preferably binary and a logical 1 may indicate a classified body posture other outputs are at a logical 0. In FIG. 6, a schematic diagram of such a classifying network in shown. It should be noted that such a network can be implemented in software as well as in hardware, as shown in FIGS. 5 and 6. As can be seen, the classifying network 64 is adapted to receive a vector containing the normalized impedance values from the different configurations config. 1, config. 2, config. 3 and config. x at respective inputs 66a-66d. In addition, the network 64 is adapted to output binary outputs at outputs 68a-68d in terms of classified body posture supine, prone, sitting or standing, respectively. Furthermore, the network 64 is adapted to receive, via input 69, an learn mode activation signal from the processor 30 activating the learn mode of the network 64, and, via input 70, a classifying mode activation signal from the processor 30 activating the classifying mode.

Hereinafter, an illustrating example of the function of the network 64, for example, a probalistic neural network will be discussed with reference to FIGS. 7a and 7b, respectively. In this illustrative example, only two configurations are used to discriminate between three different positions or postures, but, of course, an arbitrary number of electrode configurations can be used and there are also a large number of other postures that are of interest to identify in addition to the above-mentioned.

During the training session, which may take place in a doctors office as outlined above, the following set of impedance values is obtained (measured in ohms), see table 3.

TABLE 3
Posture	Config. 1 (Ohms)	Config. 2 (Ohms)	Target class
Supine	706	410	1
Left side	748	386	2
Prone	687	326	3

The measure impedance values are normalized, for example, to the maximum impedance value of respective configuration and the normalized values will have a value in the range 0 to 1, see table 4 below.

	TABLE 4
	Posture	Config. 1	Config. 2	Target class
	Supine	0.9439	1	1
	Left side	1	0.9415	2
	Prone	0.9184	0.7951	3

The training set, shown in table 4, is then used to train the network to the three output target classes 1, 2, and 3 in accordance with conventional practice within the art. Preferably, the output is coded as a binary one of P code, where P is the number of classified body postures, see table 5 below.

TABLE 5
Posture	Config. 1	Config. 2	Target class	Binary code
Supine	0.9439	1	1	0 0 1
Left side	1	0.9415	2	0 1 0
Prone	0.9184	0.7951	3	1 0 0

Thus, for example, a binary output of 0 0 1 will indicate that the patient is in supine. This is convenient, for example, in a situation where a certain body posture has to be identified for measurement of a physical parameter by the implanted device, e.g. measurement of the pre-ejection period in the supine body posture using an implantable pressure sensor. In FIG. 7a, a schematic diagram illustrating the three normalized impedance value vectors of table 4 and 5, i.e. the calibration values, and their associated body postures in the learn mode is shown. At a body posture determining session, the classifying mode, as described above, is initiated. The new measured impedance value vector is normalized and compared with the calibration values as shown in FIG. 7b. As can be seen in this example, the new classified impedance value vector is positioned as indicated in FIG. 7b and the body posture is accordingly classified to be prone. The classifying is performed by assuming that the present position is the position being closest to the measured impedance value vector in FIG. 7b.

Although an exemplary embodiment of the present invention has been shown and described, it will be apparent to those having ordinary skill in the art that a number of changes, modifications, or alterations to the inventions as described herein may be made. Thus, it is to be understood that the above description of the invention and the accompanying drawings is to be regarded as a non-limiting example thereof and that the scope of protection is defined by the appended patent claims.

Claims

1. A method for determining the posture of a patient characterized by the steps of initiating (50, 52) a patient posture determining session by performing an electrical bio-impedance measurement session in at least one of a number of different electrode configurations in order to measure an impedance value for said at least one configuration;obtaining (54) reference impedance values stored in advance for said at least one configuration and for at least one potential posture of said patient;comparing (54) said measured impedance value for said at least one configuration with corresponding stored reference impedance values for at least one potential posture of said patient; anddetermining (56) the present posture of the patient by using results from said comparison between measured impedance values and said stored reference impedance values.

2. The method according to claim 1, further comprising the steps of obtaining reference impedance values, wherein at least one reference impedance value is obtained for each of a predetermined number of electrode configurations and each of a predetermined number of postures; andstoring said reference impedance values in a reference value matrix, said matrix containing reference impedance values for each electrode configuration and each potential posture.

3. The method according to claim 1, further comprising the steps of: checking whether a difference between said measured impedance value for said at least one configuration and said corresponding stored reference impedance values is within a predetermined interval; andif said difference is outside said interval at a predetermined number of posture determining sessions, updating said stored reference impedance values by using said measured impedance values of said predetermined number of posture determining sessions.

4. The method according to claim 2, wherein the step of measuring reference impedance values comprises the steps of: sensing the breathing rate of the patient; andperforming an impedance measurement session for each reference impedance value, wherein each value is calculated as a mean value of the measured impedance signal over a predetermined number of breaths.

5. The method according to claim 2, wherein the step of measuring reference impedance values comprises the steps of: sensing the breathing rate of the patient; andperforming an impedance measurement session for each reference impedance value, said measurement being synchronized with the breathing of said patient, wherein each value is calculated as a mean value of the measured impedance signal over a predetermined number of breathing cycles.

6. The method according to claim 2, wherein the step of measuring reference impedance values comprises the step of: performing an impedance measurement session for each reference impedance value, wherein each value is calculated as a mean value of the measured impedance signal over a predetermined period of time.

7. The method according to any one of claims 1-6, wherein the step of initiating comprises the steps of: sensing the breathing rate of the patient, andcalculating said impedance value for said at least one configuration as a mean value of the measured impedance signal over a predetermined number of breaths.

8. The method according to any one of claims 1-6, wherein the step of initiating comprises the steps of sensing the breathing rate of the patient,synchronizing said impedance measurement session with the breathing of said patient; andcalculating said impedance value for said at least one configuration as a mean value of the measured impedance signal over a predetermined number of breathing cycles.

9. The method according to any one of claims 1-6, wherein the step of initiating comprises the steps of: calculating said impedance value for said at least one configuration as a mean value of the measured impedance signal over a predetermined period of time.

10. The method according to any one of preceding claims, wherein the step of comparing comprises the step of calculating a deviation factor for said at least one potential posture using said corresponding measured impedance values and stored reference impedance values, wherein said deviation factor for said at least one potential posture is used to determine the present posture.

11. The method according to claim 4, wherein the step of calculating a deviation factor comprises the step of: calculating a square difference sum for said at least one potential posture using said measured impedance values and stored reference impedance values, wherein said deviation factor corresponds to said square difference sum.

12. The method according to claim 5, wherein the step of determining the present posture comprises the steps of: identifying the potential posture of said at least one potential postures having the lowest square difference sum; anddetermining the present posture of the patient as the potential posture having the lowest square difference.

13. The method according to claim 5, wherein the step of determining the present posture comprises the steps of checking whether said square difference for said at least one potential posture is within a predetermined interval; anddetermining the present posture of the patient to the potential posture having a square difference sum within said predetermined interval.

14. The method according to any one of preceding claims, further comprising the steps of: sensing a heart rate of said patient;determining whether said heart rate is within a predetermined heart rate interval; andinitiating said patient posture determining session if said sensed heart rate is within said predetermined heart rate interval.

15. The method according to any one of preceding claims, further comprising the steps of sensing an activity level of said patient;determining whether said activity level is below a predetermined activity level; andinitiating said patient posture determining session if said sensed activity level is below said predetermined activity level.

16. The method according to any one of preceding claims, further comprising the steps of sensing a body temperature of said patient; anddetermining whether said temperature is within a predetermined temperature level interval; andinitiating said patient posture determining session if said sensed body temperature is within said predetermined temperature level interval.

17. The method according to any one of preceding claims, further comprising the steps of: sensing signals related to heart pumping activity of said patient;determining cardiac output based on said signals;determining whether said cardiac output is within a predetermined cardiac output level interval; andinitiating said patient posture determining session if said sensed cardiac output is within said predetermined cardiac output level interval.

18. The method according to any one of preceding claims, further comprising the step of: sending an updating indication signal indicating that said reference impedance values require an updating if a posture not have been determined in a patient posture determining session during a predetermined period of time.

19. A medical device for determining the posture of a patient, said device being connectable to a patient in a number of different electrode configurations, characterized by means for measuring an electrical bio-impedance (29) at each of said number of electrode configurations, wherein said means for measuring an impedance (29) is adapted to, upon receiving a triggering signal, initiate a patient posture determining session by performing an impedance measurement session in at least one of said electrode configurations in order to measure an impedance value for said at least one configuration; and processing means (30) including: means for obtaining (30) reference impedance values stored in advance in a storing means (31) for said at least one configuration and for at least one potential posture of said patient;means for comparing (35) said measured impedance values with corresponding stored reference impedance values for at least one potential posture of said patient; andmeans for determining (37) the present posture of the patient by using results from said comparison between measured impedance values and said stored reference impedance values.

20. The device according to claim 19, wherein said processing means is adapted to bring said impedance measuring means to measure at least one reference impedance value for each of a predetermined number of electrode configurations and each of a predetermined number of postures, said measured reference impedance values being stored in a reference value matrix in said storing means, wherein said matrix contains reference impedance values for each electrode configuration and each potential posture

21. The device according to claim 19, wherein said processing means further comprises means for checking whether a difference between said measured impedance value for said at least one configuration and said corresponding stored reference impedance values is within a predetermined interval; and wherein said processing means is adapted to, if said difference is found to be outside said interval at a predetermined number of posture determining sessions, update said stored reference impedance values by using said measured impedance values of said predetermined number of posture determining sessions.

22. The device according to any one of preceding claim 19-21, further comprising: means for sensing the breathing rate of the patient (39).

23. The device according to claim 22, wherein processing means is adapted to bring said impedance measurement means to perform an impedance measurement session for each reference impedance value, wherein each value is calculated as a mean value of the measured impedance signal over a predetermined number of breaths.

24. The device according to claim 22, wherein said processing means is adapted to bring said impedance measurement means to perform an impedance measurement session for each reference impedance value synchronized with the breathing of said patient, wherein each value is calculated as a mean value of the measured impedance signal over a predetermined number of breathing cycles.

25. The device according to claim 19, wherein said processing means is adapted to bring said impedance measurement means to perform an impedance measurement session for each reference impedance value, wherein each value is calculated as a mean value of the measured impedance signal over a predetermined period of time.

26. The device according to any one of claims 22-25, wherein said processing means is adapted to bring said impedance measurement means to perform an impedance measurement session for each impedance value, said impedance value for said at least one configuration being calculated as a mean value of the measured impedance signal over a predetermined number of breaths.

27. The device according to any one of claims 22-25, wherein said processing means is adapted to bring said impedance measurement means to synchronize said impedance measurement sessions with the breathing of said patient, said impedance value for said at least one configuration being calculated as a mean value of the measured impedance signal over a predetermined number of breathing cycles.

28. The device according to any one of claims 19-25, wherein said processing means is adapted to bring said impedance measurement means to perform an impedance measurement session for each impedance value, said impedance value for said at least one configuration being calculated as a mean value of the measured impedance signal over a predetermined period of time.

29. The device according to any one of preceding claims 19-28, wherein said processing means further comprises means for calculating a deviation factor for said at least one potential posture using said corresponding measured impedance values and stored reference impedance values, wherein said deviation factor for said at least one potential posture is used to determine the present posture.

30. The device according to claim 29, wherein said calculating means is adapted to calculate a square difference sum for said at least one potential posture using said measured impedance values and stored reference impedance values, wherein said deviation factor corresponds to said square difference sum.

31. The device according to claim 30, wherein said determining means is adapted to identify the potential posture of said at least one potential postures having the lowest square difference sum; anddetermine the present posture of patient as the potential posture having the lowest square difference.

32. The device according to claim 30, wherein said determining means is adapted to check whether said square difference for said at least one potential posture is within a predetermined interval; and determine the present posture of patient to the potential posture having a square difference sum within said predetermined interval.

33. The device according to any one of preceding claims 19-32, further comprising means for sensing a heart rate of said patient (41); and wherein said processing means is adapted to determine whether said heart rate is within a predetermined heart rate interval; and bring said impedance measurement means to initiate said patient posture determining session if said sensed heart rate is within said predetermined heart rate interval.

34. The method according to any one of preceding claims 19-33, further comprising means for sensing an activity level of said patient; and wherein said processing means is adapted to determine whether said activity level is below a predetermined activity level;and bring said impedance measurement means to initiate said patient posture determining session if said sensed activity level is below said predetermined activity level.

35. The device according to any one of preceding claims 19-34, further comprising means for sensing a body temperature of said patient; and wherein said processing means is adapted to determine whether said temperature is within a predetermined temperature level interval; and bring said impedance measurement means to initiate said patient posture determining session if said sensed body temperature is within said predetermined temperature level interval.

36. The device according to any one of preceding claims 19-35, further comprising: means for sensing signals related to heart pumping activity of said patient; and wherein said processing means is adapted to determine the cardiac output based on said signals;determine whether said cardiac output is within a predetermined cardiac output level interval; and bring said impedance measurement means to initiate said patient posture determining session if said sensed cardiac output is within said predetermined cardiac output level interval.

37. The method according to any one of preceding claims 19-36, further comprising alarm means being adapted to send an updating indication signal indicating that said reference impedance values require an updating if a posture not have been determined in a patient posture determining session during a predetermined period of time.

38. A computer program product, which when executed on a computer, performs steps in accordance with any one of claim 1-18

39. Computer readable medium comprising instructions for bringing a computer to perform a method according to any one of the preceding claims 1-18.