DE19829544C1 - Arrangement for noninvasive blood pressure measurement enables accurate and robust measurement with a minimal number of initial calibration measurements - Google Patents

Abstract

The arrangement has at least two noninvasive sensors (1,4) that measure a parameter connected with the blood flow or blood flow rate at different positions in the human blood vessel system, held on with bands (5,6). A signal processing unit determines at least two simultaneous blood flows or flow speeds at least 20 times per second from the sensor signals, the harmonic components, and a propagation factor from the phase differences. The harmonic components of the blood pressure and hence the entire blood pressure are derived from these parameters.

Description

environment

Blood pressure measurements are among the most commonly used medical devices worldwide Examination procedure. Regular blood pressure monitoring helps cardiovascular Recognize and deal with risks in good time.

In addition, continuous blood pressure monitoring has a firm place in medicine lenwert and is indispensable in the monitoring and assessment of the physiological Zu of patients in intensive care medicine, anesthesia and in pre- and aftercare rich.

State of the art

Most of the non-invasive measuring devices are based on sphygmomanometric methods according to Riva-Rocci, in which blood pressure values can only be recorded selectively, a pulse- Pulse or even a continuous measurement cannot be achieved with this. The occlusions of the Extremities can lead to tissue trauma if the measurement is too frequent Furthermore, the patient is burdened. Furthermore, the blood determined can pressure values deviate considerably from the true values.

An ideal device should therefore be as reliable as possible, non-invasive, without external pressure and can continuously determine blood pressure. The most diverse new approaches these requirements only partially meet these requirements.

In [Sec92] a method for arterial applanation tonometry is proposed. Here is used for non-invasive, continuous blood pressure measurement only partially on an artery pressure exercised. For artifact suppression, the fundamental and harmonic waves of the registered Filtered out pressure signal and thus reconstructs the pressure signal again.

A continuous process, which works without permanent external pressure, is described in [McQ93] specified. Two non-invasive ultrasonic Doppler sensors are used attached to a larger artery. The blood flow signals thus measured are then used to create a simplified mathematical-empirical model of the artery pa to characterize rametrically. The so-called resonance frequency of the Mo dells is correlated with blood pressure. A conventional cuff measurement calibrated in irregular intervals the system on blood pressure.

The object of the invention presented below is a device specify which blood pressure is accurate with as few initial calibration measurements as possible and can determine robustly. In addition, the procedure is said to be in operation cuffless measurement technology also suitable for quasi-continuous blood pressure monitoring, at which new blood pressure values can be determined per pulse.

This object is achieved with the device according to claim 1.

literature

[McQ93] McQuilkin, G., Noninvasive, non-occlusive method and apparatus which provides a continuous indication of arterial pressure and a beat-by-beat characterization of the arterial system, U.S. Patent 5,241,964, 1993

[Sec92] Seca GmbH, method and device for non-invasive continuous blood pressure measurement in humans, published application DE 41 05 447 A1, 1992

Description of the invention Theoretical basics

For non-invasive, continuous determination of blood pressure without external pressure blood pressure is derived from other, non-invasively measurable cardiovascular parameters. To In the following, relationships between the three in the propagation of pulse waves in Vascular system of humans accompanying pulse forms (pressure, current and volume pulse) be derived.

An exact mathematical description of the hemodynamic processes in the vascular system itself is extremely difficult. It is more clever to consider simplified models and then map them to the vascular system. A tried and tested model of a vascular segment is that of an elastic hose.

The most general form of the equation of motion for flowing, incompressible liquids is the Navier-Stoke equation system:

with ρ = density, = velocity of the liquid in the three spatial directions, ∇ = Nabla operator, p = pressure, _A = external force density, η = viscosity of the liquid, Δ = Laplace operator.

The terms on the left represent the inertial force densities, the local and the convective acceleration. On the right, grad p is the pressure gradient, and η. Δ corresponds to the friction force density.

To solve this partial, non-linear system of equations Eq. (1) can with regard to assumptions simplifying the artery system. The convective acceleration can be neglected when limited to sufficiently low speeds. The non-linear components can also be neglected, since the phase velocity d much greater than the axial speed component and this in turn is much larger than the radial velocity component. Likewise, the second Ab lines after z are neglected; external forces, such as gravity, and the radial pressure drop is also negligible.

Under these assumptions, the last equation from Eq. (1) can be represented in cylinder coordinates according to FIG. 1 as follows:

Due to the principle of superposition, this linearized differential equation can be solved with a harmonic separation approach from the superposition of back and forth waves:

Here are: ω = 2. π. f = angular frequency, P _k , V _k = complex amplitudes of the k. harmonic component of pressure or velocity (index H: incoming wave, index R = returning wave), and γ _k is the propagation measure or the propagation constant of the k. harmonic component.

For the sake of simplicity, the differential equation is only to be solved below for a harmonic (k = 1) and for a passing wave:

P (z, t) = P. e ^j ω ^t- γ ^z
ν _z (r, z, t) = V (r). e ^j ω ^t- γ ^z (4)

With γ = α + jβ = α + jω / c _Ph . α is the damping coating, β is the phase coating, which in turn can be expressed as the quotient of angular frequency and phase velocity c _Ph . The solution of the system of equations must meet the boundary conditions on the inner wall of the vessel R ₀ . Since the hose in the z direction is assumed to be very rigid or longitudinally fixed. Furthermore, the axial acceleration component must assume the value zero at r = 0 due to the rotational symmetry.

The time-variant inner radius consists of a static and a dynamic part:

R ₀ (z, t) = R _{0, stat} + ΔR ₀ . e ^j ω ^t- γ ^z (6)

The Bessel differential equation Eq. (2) solve:

where J _{n represent} first order _n Bessel functions:

To determine the total flow i (z, t) in the hose, the following is integrated across the cross-section:

Eq. (9) can be calculated with the rule

and the substitution

simplify too

For the back and forth waves existing in the arterial system, this results in pressure and current pulses

The time profiles of the pressure and the current pulse are sketched in FIG . It is clearly evident that according to Eq. (12) the pressure pulse from the sum, the current pulse from the difference of a back and forth wave.

If two locations z ₁ = 0 and z ₂ = 1 are now considered in accordance with FIG. 8 on a branch of the vessel, the result is f (z, t) = F (z). e ^jωt :

This can be done with the calculation rules

the pressure p ₁ can be determined as:

Conclusion

The blood pressure p _{1, total} , composed of the harmonic components p ₁ , can be determined from the harmonic components I ₁ and I _{2 of} the two current pulses i _{1, total} and i _{2, total} , the cardio-vascular parameter Z ₁ , and the degree of spread .

The mean speed pulse can also be used as an alternative to the current pulse: ν = i / πR ₀ ² . Furthermore, the pressure p ₂ at the point z ₂ can alternatively also be determined.

Measuring method:

In order to determine the blood pressure p _{1, ges} on humans, the current pulses i _{1, ges} and i _{2, ges are} measured simultaneously in at least two positions z ₁ and z _{2 of} a vascular branch at least twenty times per second, from which the harmonic components are calculated I ₁ , and I ₂ , then determines it and possibly with a third simultaneously registered current pulse i _{3, total} at a third measuring point z _3, the spreading dimension γ and can then be measured with the z. B. previously determined by calibration measurements Z _l the harmonic components p _{1 of} the blood pressure according to Eq. (14) and thus the total blood pressure p _{1, total} according to Eq. (3) determine.

According to this general description of the measurements, the calculation of blood pressure should be examined in more detail necessary determination of the individual parameters.

The harmonic components I ₁ , I ₂ and I _{3 of} the current pulses i _{1, ges} , i _{2, ges} and i _{3, ges} are easiest to calculate using a Fourier transformation.

In addition to the angular frequency ω, Z _l is dependent on the radius R ₀ , the blood density ρ and the viscosity η, where ρ and η always, R ₀ can be assumed to be constant in a first approximation. Z _l can also be assumed to be constant, empirically, intraindividually, or approximated by individual patient data such as age and gender or calibration measurements. If an even more precise determination of Z _{l is} desired, then the volume pulsation R ₀ must be taken into account, which, however, is approximately in phase with the pressure pulse. The dynamic proportion ΔR ₀ according to Eq. (6) is usually only a few percent of the static part R _0.stat . For example, on the large common carotid artery with R _0.stat = approx. 5 mm with a pressure change of approx. 70 mmHg ΔR ₀ = approx. 0.15 mm (= 3%). This influence could, for. B. by further empirical approximation or calibration measurements. Of course, the radius R _{0 can} also be measured directly, for example using the means of locally high-resolution optical tomography or using ultrasound echo methods for larger vessels.

When determining the propagation measure γ, the damping measure α and the phase measure β = ω / c _Ph must be determined. In a first approximation, lossless transmission (α = 0) can be assumed. Then Eq. (14) to:

The phase velocity c _Ph in the phase measure β is generally different for each harmonic pulse wave due to the dispersive behavior of the human vascular system. Low-frequency components of the pulse waves propagate more slowly than higher-frequency ones. If freedom from reflection is required, the frequency-dependent phase speeds can be determined by the phase difference between the individual harmonic components of the current pulses I ₁ and I ₂ . However, since there are reflections in the arterial system, this is only an approximate solution (the so-called "apparent" phase velocity). However, the "true" phase velocities can be approximated empirically or by calibration measurements. If the actual "true" phase speeds are to be determined, this requires a further simultaneous flow measurement at a third location z _{3 of} the vascular system, because in the case of reflection according to Eq. (12) To determine the three unknowns P _H , P _R and γ three flow equations (= three simultaneous flow measurements) have to be set up. The damping α can then also be calculated exactly. It is of greater importance, the larger the measuring distance l of the current pulse sensors is.

The measurement rate for blood flow determination of at least 20 Hz results from the following consideration: Human pulse rates are in the range between approx. 30 and 180 beats per minute, which corresponds to a minimum of 0.5 Hz or a maximum of 3 Hz. It can be shown that human pulses with the first harmonic components can be described adequately depending on the desired resynthesis accuracy. For example, the first 10 harmonics in resynthesis achieve a signal approximation of up to 99.5%. The relevant bandwidth of a pulse signal can thus be determined in the range from approximately 10 Hz to 30 Hz ( FIG. 3). To fulfill the sampling theorem, at least twice the bandwidth must be sampled, i.e. at least 20 Hz.

Description of the system

The technical implementation of the invention, Fig. 4 to Fig. 7 are taken. It consists of a main unit ( 1 ) with sensors ( 11 ), an extendable sensor housing ( 4 ) with further sensors ( 12 ), which is connected to the main unit ( 1 ) via a self-winding cable ( 13 ). The sensor housing ( 4 ) can be completely integrated into the main housing ( 1 ). When the sensor housing ( 4 ) is pulled out according to FIG. 6, the sensors ( 11 ) and ( 12 ) l are automatically determined by the device via the cable unwind. It is also possible to connect the sensor housing ( 4 ) to the main housing ( 1 ) wirelessly, e.g. B. by radio.

The measured and calculated biosignals (blood flow, pulse, blood pressure, spread, etc.) can be displayed graphically and / or alphanumerically on the LCD display ( 2 ). It is also possible to make settings using the keypad ( 3 ).

According to FIG. 8, the sensors ( 11 ), ( 12 ) simultaneously measure at least two blood flow signals i _{1, ges} and i _{2, ges} at two different locations 21 and 22 of a vascular branch of the body (preferably on an arterial branch), so that it can be used to determine the degree of expansion γ in addition to other parameters. The device can be used with tapes ( 5 ), ( 6 ) on the human extremities, e.g. B. the arm ( 7 ), place on superficial vessels. For example, the brachial artery ( 8 ) in the crook of the arm, the radial or ulnar artery ( 9 ) on the wrist, or the digital artery ( 10 ) on the finger. Since the phase velocities of the pulse waves are quite high (approx. 5 to 15 m / s), it is advisable not to place the sensors ( 11 ) and ( 12 ) too close to one another for higher measuring accuracy. At a distance of l = 5 cm and an average phase speed of c _ph = 10 m / s, there is a temporal phase difference of 5 ms, so that the dynamic phase differences to be detected must be resolved in the 1/10 millisecond range . If the measuring distance l increases, for example between the arteria bra chialis ( 8 ) and arteria radialis ( 9 ) to approx. 25 cm, the requirements for the temporal resolution are of course significantly lower (approx. Millisecond range) because the phase differences to be recorded have become larger are.

Doppler technology is used to obtain the blood flow signals i _{1, ges} , i _{2, ges} , i _{3, ges} . Since is according to Figure 9. And FIG. 10, either ultrasound at a well-defined frequency in the MHz range (the default value about 8 MHz for shallow vessels) or laser radiation with a well defined wavelength preferably in the near infrared range of about 700-3000 nm in the tissue irradiated .

In the case of ultrasound, the sound waves radiated at an angle α with the frequency f ₀ on moving blood particles of the speed v are scattered back with a Doppler frequency shift Δf:

Here c is the speed of sound propagation and is in human tissues approx. 1450 m / s. Thus, the ultrasound obtained on the Ultra is obtained from the backscattered ultrasound sound detector a measure of blood flow.

In the optical case, the tissue is irradiated with coherent laser light of wavelength λ ₀ . However, the light is scattered on static tissue parts due to the fact that the free path length is considerably shorter than that of ultrasound, so that the photons - if at all - hit a moving blood particle at any angle and, depending on the latter, receive a more or less large Doppler shift. Thus, a maximum Doppler frequency shift Δf _max can be detected at the photodetector only in the case of single scattering on a moving blood particle v:

n is the refractive index of tissue.

The advantages of the laser Doppler method compared to the ultrasonic Doppler method are that the light can be coupled into the tissue without contact gel (although in principle it would also be possible with ultrasound without gel, which, however, is very high hen coupling losses is connected). Secondly, you get a much higher location Liche dissolution, so that the blood flow is measured even in the smallest vessels such as capillaries that can. However, this also increases the signal bandwidth of the detected signal. The advantage of the ultrasound method is the ability to measure blood flow absolutely sen, whereas the laser Doppler technique only allows a relative measurement and thus on an initial calibration is instructed.

The received sensor signals are gem. the block diagram of Fig. 11. fed to a Ana logteil (14) and then further processed in a digital Signalverabeitungseinheit (15) white. The system also includes memory modules ( 16 ), an input / output unit ( 17 ) and a display unit ( 18 ).

In the analog section, Fig. 12. the sensors ( 11 ), ( 12 ), ( 19 ) excited ( 20 ) (As will be shown below, other sensors 1 and 2 can be used in addition to the two sensors 1 and 2. These are in the "virtual" sensor m ( 19 ) summarized.). In the case of ultrasound, this takes place via an oscillator, in the optical case with a laser diode driver circuit which ensures constant optical power and protection of the laser diode from being switched on or off. The received signals are then bandpass filtered ( 22 ) and amplified with automatic gain control (AGC) ( 23 ). It is important that the analog construction groups have constant group or phase delays, so as not to falsify the phase information of the flow signals, which would be unfavorable with regard to the phase or pulse wave speed determination. Reasonable lower limit frequencies of the bandpass are around 100 Hz in order to eliminate the high DC signal component, 50 Hz network interference and the low-frequency plethysmographic signal components in the frequency range from 1 Hz to approx. 30 Hz. They result in wall or sensor movements and, in the visual case, in the time-changing blood volumes. These optical plethysmographic components are disturbing for the flow determination, but can also contribute further information to the determination of the blood pressure. The time course of the plethysmographic signal is qualitatively linked to the volume pulse and thus to the pressure pulse.

In the case of ultrasound, the signals are demodulated after amplification ( 24 ) and band-bass filtered again. Before the AD conversion ( 27 ), a reinforcement ( 26 ) is carried out again before it is gem. Fig. 13. in the digital signal processing unit ( 15 ) who processed the.

In the optical case, demodulation is not necessary, as this is done by the photodetector Interference of frequency shifted and non-frequency shifted beams automatically he follows.

The upper limit frequency of the bandpass ( 22 ) or ( 25 ) depends on the sampling rate of the AD converter ( 15 ). It is used to suppress noise and aliasing effects. An upper limit frequency of approx. 20 kHz ( 25 ) is useful for ultrasound, and approx. 200 kHz ( 22 ) for laser beams.

In the signal processing unit ( FIG. 13), which is controlled by an overall control ( 28 ), artifacts and other disturbances are first eliminated by adaptive or parametric filters ( 29 ) and then the blood flow is determined continuously (ie at least twenty times a second) ( 30 ) . This can be done in the time domain by means of simple zero crossing counting. This method is simple and fast, but inaccurate. More precise procedures work in the frequency domain. For this purpose, spectrograms are determined from the time signal for short time intervals, which can be estimated using a Fast Fourier Transformation (FFT) or parametrically. It can be shown that the mean flow velocity can be determined via the first weighted moment in the spectral range:

The same applies in the optical case. It can also be shown here that the first weighted moment in the spectral range is proportional to the mean flow velocity:

The propagation measure γ ( 33 ) is determined on the basis of at least two simultaneously recorded and determined flow pulses i _{1, ges} and i _{2, ges} . For this purpose, the time signals i _{1, ges} and i _{2, ges} are transformed into the frequency range _, for example by means of an FFT, and the amount and phase of the harmonic components I ₁ and I ₂ are thus obtained. As explained above, the frequency-dependent phase measure β = ω / c _Ph can then be calculated by forming the phase differences between I ₁ and I ₂ . Since with this method only the "apparent" phase velocities can be determined, the "true" phase velocities and attenuations can be approximated empirically or by calibration measurements to increase the measuring accuracy. The exact calculation of the "true" phase velocities and the damping requires a further simultaneous flow measurement i _{3, total} at a third location z _{3 of} the vascular system ( 19 ). Even with this, the accuracy of measurement can be increased and the number of necessary calibration measurements can be reduced. Then the individual harmonic components of the blood pressure p ₁ ( 35 ) can be calculated with the parameter Z _l in the frequency range. Ultimately, the continuous course of the blood pressure p _{1, total} , in particular the systolic, diastolic and mean blood pressure, can thus be determined.

The cardiovascular parameter Z _l is determined depending on the desired reliability and measurement accuracy analytically according to the individual calibration ( 32 ) or statistically and empirically with individual patient data such as age and gender. In order to increase the measuring accuracy and to reduce the number of calibration measurements, the essential parameter in Z _l , the vessel radius R ₀ , can also be measured directly by means of a further non-invasive sensor ( 19 ). Local high-resolution optical tomography is suitable for determining the radii of small vessels, and the known ultrasound echo methods can be used for larger vessels.

Gravity can be used for calibration. By lifting or lowering around a well-defined height of the extremity to which the sensors are attached due to the resulting additional known static blood pressure other para meters can be determined.

To ensure that the sensors ( 11 ), ( 12 ), ( 19 ) deliver optimal signals and that the device can be easily applied to humans, the sensors can be positioned manually or automatically. In the automatic case, the sensor signals can be continuously tracked before, during or between the measurement and calculation phases, so that when the device is initially applied or the device slips, the same optimal measurement path (above the desired blood vessel) is always used. The associated control algorithm ( 31 ) is implemented in the signal processing unit ( 15 ), the hardware controls ( 21 ) are integrated in the sensors ( 11 ), ( 12 ), ( 19 ) themselves and in the analog part ( 14 ).

Several implementations are available as adaptive sensors, both in the acoustic and in the optical case. According to FIG. 14, the first possibility includes, for both cases, a mechanical tracking of the actual sensor ( 37 ) consisting of transmitting ( 38 ) and receiving piezo crystal ( 39 ) or laser diode ( 38 ) and photodetector ( 39 ), for example a movable stepper motor ( 36 ) that can travel along a spindle ( 40 ). The sensor unit ( 11 ), ( 12 ), ( 19 ) is then applied to the arm in such a way that the travel path runs across the blood vessel. The motor ( 36 ) is addressed via the hardware control ( 21 ) and regulated via the control algorithm ( 31 ).

The second option is an ultrasound or laser array according to FIG. 15. There are many individual sensors ( 41 ) (consisting of transmit and receive piezo crystal or Laserdi ode and photodetector), for example, arranged in two rows offset linearly, which can be controlled or selected via a multiplexer circuit ( 21 ). The control takes place via the control algorithm ( 31 ).

Another alternative in the optical case can be realized with light guiding devices such as optical fibers. In principle, image according to Fig. 16. There, with the help mi kromechanisch adjustable mirror (45) a virtual array is built, in which only a Sen sorelement (43) is required. The selection of a pair of light guide elements of the bundle ( 42 ) is made by rotating the mirror ( 45 ) by the angle oder and / or by driving a distance d along the optical beam axis ( 44 ), so that exactly one light supply of the strand is selected. Another possibility for the selection of a pair of light guide elements is outlined in FIG. 17. There, the beam ( 44 ) is expanded ( 47 ) with a beam expander ( 46 ), and then a pair of light guide elements is selected with an electronically controllable LCD shutter ( 48 ) by darkening ( 49 ) and non-darkening ( 50 ).

Claims (30)

1. Device for non-invasive blood pressure measurement on living beings in the

• a) simultaneously measure at least two non-invasive sensors ( 11 , 12 ) a variable linked to the blood flow or the blood flow velocity at different positions z ₁ and z ₂ on the human blood vessel system,
• b) a signal processing unit ( 15 )
  determines at least two simultaneous blood flows or blood flow velocities i _{1, ges} and i _{2, ges} at least twenty times per second on the basis of the at least two received sensor signals ( 30 ),
  the harmonic components I ₁ and I _{2 are} calculated using the blood flows or blood flow velocities i _{1, ges} and i _{2, ges} ,
  uses the phase difference of the harmonic components I ₁ and I _{2 to} estimate the spread γ ( 33 ),
  and with the four parameters I ₁ , I ₂ , γ and the individually determined cardio-vascular parameter Z _l ( 34 ) the harmonic components of the blood pressure p ₁ and thus the total blood pressure p _{1, total are} determined ( 35 ).

2. Device according to claim 1, characterized in that the sensors ( 11 ), ( 12 ) optically determine the quantities associated with the blood flows or the blood flow velocities according to the Doppler effect with coherent infrared light in the wavelength range from 700 nm to 3000 nm. 3. Apparatus according to claim 1, characterized in that the sensors ( 11 ), ( 12 ) the quantities associated with the blood flows or the blood flow velocities according to the Doppler effect acoustically with ultrasound in the frequency range from 5 MHz to 15 MHz. 4. The device according to claim 1, characterized in that in the signal processing unit ( 15 ) the blood flows or the blood flow velocities i _{1, ges} and i _{2, ges are determined} from the zero crossings of the received sensor signals ( 29 ), ( 30 ). 5. The device according to claim 1, characterized in that in the signal processing unit ( 15 ) the blood flows or the blood flow velocities i _{1, ges} and i _{2, ges are} determined by the first weighted moment of the spectrograms of the received sensor signals, the spectrograms can be estimated using a Fourier transformation, curve fitting or using parametric methods ( 29 , 30 ). 6. The device according to claim 1, characterized in that in the signal processing unit ( 15 ), the harmonic components I ₁ , and I _{2 of} the blood flows or the Blutflußge speeds i _{1, ges} and i _{2, ges} by a Fourier transformation, curve fitting or by means parametric methods can be estimated ( 30 ). 7. The device according to claim 1, characterized in that in the signal processing unit ( 15 ), the phase velocities c _{Ph of} the _degree of expansion γ ( 33 ) with the phase differences of the harmonic components I ₁ and I _{2 of} the blood flows or the blood flow velocities are estimated, and that this estimate may be improved statistically-empirically and / or by individual calibration measurements ( 32 ). 8. The device according to claim 1, characterized in that in the signal processing unit ( 15 ) in the calculation of the degree of expansion γ ( 33 ) the attenuation α is neglected, or as a statistical-empirical and / or determined by individual calibration measurements with constant ( 32 ). 9. The device according to claim 1, characterized in that at a third point z _3, a further non-invasive sensor ( 19 ) simultaneously with the sensors ( 11 , 12 ) measures a linked to the blood flow or the blood flow velocity size on the blood vessel system of the living being, in which digital signal processing unit ( 15 ) the blood flows or blood flow velocities i _{3, ges are determined} at least twenty times per second simultaneously to i _{1, ges} and i _{2, ges} ( 30 ), from which the harmonic components I ₁ , I ₂ and I ₃ are calculated in order to then exactly determine the phase velocities c _Ph and the damping α of the propagation measure γ ( 33 ). 10. The device according to claim 9, characterized in that the further sensor ( 19 ) determines the quantities associated with the blood flows or the blood flow velocities after the Doppler effect optically with coherent infrared light in the wavelength range from 700 nm to 3000 nm. 11. The device according to claim 9, characterized in that the further sensor ( 19 ) which is associated with the blood flow or blood flow velocities after the Doppler effect acoustically with ultrasound in the frequency range from 5 MHz to 15 MHz. 12. The apparatus according to claim 9, characterized in that in the signal processing unit ( 15 ) the blood flows or the blood flow velocities i _{3, ges are determined} from the zero crossings of the received sensor signal ( 29 , 30 ). 13. The apparatus according to claim 9, characterized in that in the signal processing unit ( 15 ) the blood flows or the blood flow velocities i _{3, ges is} determined by the first ge-weighted moment of the spectrograms of the received sensor signal, the spectrograms using a Fourier transformation, curve fitting or can be estimated using parametric methods ( 29 , 30 ). 14. The apparatus according to claim 9, characterized in that in the signal processing unit ( 15 ) the harmonic components I _{3 of} the blood flows or the Blutflußge speeds i _{3, ges} are estimated by a Fourier transformation, curve fitting or by means of parametric methods ( 30 ). 15. The apparatus of claim 1 or 9, characterized in that the cardio-vascular parameter Z _{l is determined} statistically-empirically with individual patient data and / or individually by calibration measurements ( 32 ). 16. The apparatus of claim 1 or 9, characterized in that the cardiovascular parameter Z _l , is determined by measuring the vessel radius R ₀ with a further non-invasive sensor ( 19 ). 17. The apparatus according to claim 16, characterized in that the non-invasive sensor ( 19 ) works with the means of optical tomography. 18. The apparatus according to claim 16, characterized in that the non-invasive sensor ( 19 ) operates according to the ultrasonic echo method. 19. The apparatus according to claim 1 or 9, characterized in that the blood pressure p _{1, total} , in particular the systolic, diastolic and mean blood pressure, quasi-continuously, so minimal for each pulse or for longer time intervals, determined ( 35 ) and with them cardiovascular parameters such as pulse, blood flow, degree of spread are displayed ( 2 ). 20. The apparatus according to claim 1 or 9, characterized in that the sensor devices ( 1 ), ( 4 ) can be applied to the human extremities with bands ( 5 ), ( 6 ), the distance l ( 13 ) of the sensors ( 11 ), ( 12 ) automatically or manually it is averaged. 21. The apparatus according to claim 1 or 9, characterized in that the sensor device ( 4 ) with the main device ( 1 ) communicates wirelessly. 22. The apparatus according to claim 1 or 9, characterized in that the sensors ( 11 , 12 , 19 ) for signal optimization are automatically and / or manually positioned on human blood via suitable blood vessels. 23. The device according to claim 22, characterized in that the positioning of the sensors ( 11 , 12 , 19 ) with a mechanical movement ( 36-40 ), which can be controlled via a digital control algorithm ( 31 ) ( 21 ). 24. The device according to claim 22, characterized in that the sensors ( 11 , 12 , 19 ) each form a matrix structure ( 41 ) which can be controlled via a digital control algorithm ( 31 ) ( 21 ). 25. The device according to claim 22, characterized in that the optical sensors ( 11 , 12 , 19 ) via light guide devices ( 42 ) are coupled to the skin. 26. The apparatus according to claim 25, characterized in that a pair of light guide elements of the bundle ( 42 ) is selected via adjustable mirrors ( 45 ). 27. The apparatus according to claim 22, characterized in that a pair of light guide elements of the bundle ( 42 ) is selected via LCD shutter ( 48 ). 28. The apparatus according to claim 1 or 9, characterized in that the calibration measurements are made by using gravity on the static blood pressure. 29. The device according to claim 1 or 9, characterized in that the calibration measurements are made using the conventional Riva-Rocci method. 30. The device according to claim 1 or 9, characterized in that the calc Cardiovascular parameters such as blood pressure, blood flow, spread and pulse and the calibration values are saved and forwarded to the outside via an interface can be given.