EP2007281A1

EP2007281A1 - Device and process for magnetic resonance elastography (mre)

Info

Publication number: EP2007281A1
Application number: EP07724354A
Authority: EP
Inventors: Ingolf Sack; Jürgen BRAUN; Jens Rump
Original assignee: Charite Universitaetsmedizin Berlin
Current assignee: Charite Universitaetsmedizin Berlin
Priority date: 2006-04-13
Filing date: 2007-04-12
Publication date: 2008-12-31
Also published as: WO2007118710A1; DE102006037160B4; US20110006767A1; DE102006037160A1; US8305076B2

Abstract

The invention relates to a device to produce mechanical oscillations in a research object using magnetic resonance elastography (MRE) with a membrane that can be set into periodic motion (312) and a transmission element (4, 400) for the transmission of periodic motion of the membrane (312) onto the research object (1), whereby the membrane (312) is connected to the transmission element (4, 400) by means of a mounting medium (311, 320, 321, 322) in such a way that periodic motion of the membrane is transmitted over the mounting medium (311, 320, 321, 322) to the transmission element(4, 400). A second aspect relates to the invention of a device with a step motor to generate periodic motion and a transmission element for the transmission of the periodic motion of the step motor to the research object in order to generate mechanical oscillation therein. The invention also relates to a procedure for the magnetic resonant elastographic determination of biomechanical characteristics of tissue with steps: c) excitation of mechanical oscillations using a frequency of fv in which n is the tissue to be researched; d) proof of mechanical oscillation within the tissue using magnetic resonance, whereby a periodic magnetic field with a repeat time of TR≤1/fv is used on the tissue.

Description

Device and method for magnetic resonance elastography (MRE)

description

The invention relates to a device for generating mechanical oscillations in an examination object in magnetic resonance elastography (MRE) according to the preamble of claim 1 and to a method for magnetic resonance elastographic determination of biomechanical properties of tissue according to claim 23.

With the help of magnetic resonance elastography (MRE) it is possible to record biomechanical properties of biological tissues. The method can be compared with a "tactile palpation", with which not only near-surface but also deep-lying tissues or organs with good transillumination and spatial resolution can be quantified in terms of biomechanical properties comparable to classical palpation, but without limitation to near-surface tissue and objectively measurable The MRE is based on the principles of magnetic resonance tomography (MRI), but in the MRE additional periodic particle deflections have to be generated in the tissue to be examined by mechanical excitation by magnetic preparation of the nuclear spins and motion-sensitive phase contrast techniques.

The resulting grayscale images have a typical wave character, which reflects the periodic tissue distortions (shear waves) caused by the mechanical excitation. From the detected distortions, elastic characteristics of the tissue, such as the shear modulus, the Young modulus, the modulus of compression, or the Poisson ratio can be calculated. The measurement of the distortion in all spatial directions allows the complete quantification of the elastic parameters taking into account their directional dependence.

Studies of the frequency and amplitude dependence of the periodic tissue distortions thus provide information on viscoelastic properties of tissues. Nonlinear stress-strain relationships, which also have a high potential for characterizing tissue properties, are determined by the measurement of At the same time as the excitation frequency, harmonics of the tissue distortion are accessible.

The number of observable shear waves in the tissue to be examined depends on its elasticity properties and the frequency of the mechanical excitation. So far, the MRE uses three different types of mechanical excitation units to create deflections in tissues. These are:

1) excitation units based on the elongation of piezoelectric crystals,

2) excitation units which are based on a movement deflection of current-carrying coils moving in the magnetic field of the tomograph (electromechanical excitation),

3) Passive, pressure-activated excitation units that are driven by pneumatic lines.

The known devices have a number of limitations and disadvantages associated with the previously available excitation methods. For example:

1) Piezoelectric method:

- High voltage: Piezoelectric crystals are operated with high voltage amplifiers with up to 1 kV voltage. This is problematic for the application of patient examinations and involves a tremendous amount of security. - Complex mechanics: since the deflection amplitudes are also in the range of 200 microns from piezo stacks with a length of 200 mm, deflections must be realized with levers to allow deflections of the order of 1 mm.

- Image disturbances: due to the lever deflection, springs, and an approximately 250 mm long aluminum sleeve, in which the piezo crystals are pre-stressed, problems arise with picture disturbances. These are caused by the various metal parts necessary for the construction, which lead to distortions of the magnetic field and thus to image disturbances whose extent makes the evaluation of the image material partially impossible. - Positioning: by the length of the finished excitation unit of usually

250 mm and a max. The diameter of the tomograph of 60 cm, the positioning of the excitation unit depending on the object of investigation is sometimes only partially possible. 2) Electromechanical excitation units:

Positionability: The main disadvantage of the electromechanical excitation units is the limitation to predetermined coil orientations in the magnetic field, since otherwise the magnetic fields necessary for the movement can not be induced by the applied alternating voltage. As a result, either the applicability is limited or complicated mechanical deflection mechanisms are necessary. Image disturbances: the use of aluminum for improved heat dissipation and enameled copper wire in turn results in image cancellations, for the avoidance of which the coil of the mechanical excitation unit must be at least 2 coil diameters away from the examination subject. In addition, due to the magnetic fields induced therein, image disturbances occur as a function of the alternating voltage increasing with the deflection. This often prevents the generation of larger, reliably evaluable MRE imagery more advantageous, deflection amplitudes.

- Image interference due to power supply cable: To supply current and voltage to electromechanical excitation units, appropriate wiring is required that extends into the center of the magnet of the scanner.

Although shielded cables are used that lead through a filter plate into the examination room, electromagnetic interference that is captured by the cables and that in turn leads to image artifacts, can not always be ruled out. - For the different organs / tissues to be examined, special excitation units optimally adapted to the respective requirements have to be developed.

In the MRE, the excited mechanical tissue oscillations (shear waves) are detected by means of magnetic resonance imaging (MRI). For this, the movements of the particles in the tissue must be coded magnetically, which takes place in the prior art by means of synchronously oscillating magnetic field gradients, i. the acoustic excitation to generate the vibrations and the magnetic coding take place with identical frequencies.

Oscillating motion encoding gradients (MEG) can be incorporated into almost any MRI acquisition technique. Therefore, today MRE recording techniques are available based on the principles of spin-echo, Gradient echo techniques are based. Especially interesting are fast EPI (echo planar imaging) or SSFP (steady-state free precession) acquisition techniques.

The present invention has for its object to provide a way to perform MRE examinations more efficient than before.

This object is achieved by the device having the features according to claim 1 and by the method having the features according to claim 23. Preferred and particularly advantageous developments of the invention are given in the subclaims.

Thereafter, a device for generating mechanical vibrations in an examination subject for magnetic resonance elastography (MRE) is provided. The device comprises: a membrane displaceable in periodic movements and a transmission element for transmitting periodic movements of the membrane to the examination subject. The membrane is connected to the transmission element via fastening means in such a way that periodic movements of the membrane are forwarded via the fastening means to the transmission element in order to generate mechanical vibrations in the object to be examined.

The coupling of the membrane movements thus takes place directly via the mechanical connection of the membrane and the transmission element and not via an existing between the membrane and the transmission element, gaseous medium. The membrane vibrations are therefore not transmitted via pressure fluctuations (for example, an air-filled cavity) but directly via the fastening means for attaching the transmission element to the diaphragm on the transmission element.

The present invention is an acousto-mechanical excitation unit for the MRE that allows for trouble-free, direct, and precisely localizable transmission of high-power vibrations into any body tissue or organs. Interactions with the electronics or the receiving components of a magnetic resonance tomograph for the detection of tissue oscillations do not occur.

The fastening means are in principle arbitrary, provided that on the one hand they ensure a stable connection between the diaphragm and the transmission element and, on the other hand, that the membrane vibrations are transmitted to the transmission element (as far as possible). poor) forward. For this example, elements for producing a screw or rivet connection can be used.

In a preferred variant, the fastening means have a carrier plate centrally fastened to the membrane by means of an adhesive, which is connected to the transmission element.

Compared with the use of piezoelectric, electromechanical and pneumatic excitation units, the device according to the invention offers, for example, the following advantages:

- No high voltage technology, it eliminates complex measures to protect patients and personnel from dangerous high voltage.

No image interference due to moving or stationary metallic components in or near the MR detection coil and the magnet of the MR tomograph.

- No picture interference due to induced additional magnetic fields.

- Unlimited frequency range for mechanical excitation in MRE applications.

- Efficient and precisely localized power transmission. - Increased patient comfort and ease of use, as depending on the application specifically different positions are stimulated and no complete excitation units need to be fixed to the body.

In a further preferred embodiment of the device according to the invention, a surface of the membrane is connected to the transmission element, e.g. the surface of the membrane pointing in the direction of the examination object. In particular, the connection may be cohesively, e.g. by means of adhesive. Particularly advantageous is the use of a tube-shaped or rod-shaped transmission element, which is coupled with one end to the membrane surface. The transmission element can run in a straight line; but it can also have a curved course. It is crucial that the transmission element excites a sufficient vibration for the MRE in the examination subject (person or tissue sample).

The transmission element may be, for example, a solid, rigid rod or a tube (rigid or flexible). Furthermore, a tube filled with a medium (gas or liquid) may also be provided. Such a filled tube is, for example, gas-tight or liquid-tightly connected to the membrane surface, so that the membrane movements be transferred directly to the gas or liquid and forwarded to the object under investigation.

Particularly preferably, the end of the transmission element is connected to a central portion of the membrane. When using a round diaphragm (e.g., a speaker diaphragm), the end of the rod or tubular transmission member is connected to it at the center of the diaphragm. Here, the membrane may additionally have a centrically arranged stabilizing sleeve for transmitting power. In addition, a joint may be provided which is connected to the transmission element on one side and to the membrane on the other side, possibly via the stabilizing sleeve.

In an advantageous embodiment of the invention, the transmission element is flexible. In particular, a flexibly formed transmission element can run in a curved manner. Flexible in this context does not necessarily mean "elastic." The material of the transmission element is in any case chosen so that the membrane vibrations are forwarded as possible without damping.

It is particularly advantageous if, in addition to the first diaphragm connected to the transmission element, a second diaphragm is arranged and the first diaphragm can be set in motion by movements of the second diaphragm. The second membrane is here an active membrane, the first a passive membrane. Between the two membranes, there may be an air-filled cavity through which movements of the active membrane are transferred to the passive membrane; For example, the two membranes may be mounted on opposite side walls of a common housing that encloses the air filled cavity.

Furthermore, it is advantageously provided that, on its side facing the examination subject, the transmission element has a rod-shaped or tubular excitation element which extends at an angle to its main extension direction. As a result, an efficient coupling into the examination subject (eg a reclining person) is made possible. By "main direction of extension" is meant the extension of the transmission element in one direction from the membrane to the examination subject In a rod-shaped transmission element, its end not connected to the membrane bends towards the examination object, for example by means of a joint. Carbon tubes, carbon rods, or rigid connections that can be made by solid materials provide a direct, compression-free power transfer option. This concept can also be implemented by the use of hydraulic equipment using all appropriate incompressible or slightly compressible fluids. The power transmission is based here generally on the transmission of a periodic force by a hydraulic fluid (for example, by a hydraulic piston) to a connected by means of a non-elastic and non-magnetic pressure hose stretchable membrane. This membrane can be made from any stretchable material such as elastomers, rubber and plastics.

The method according to the invention for the magnetic resonance elastographic determination of biomechanical properties of tissue comprises the steps of: a) exciting mechanical vibrations with a frequency f _v in the tissue to be examined; b) Detection of the mechanical vibrations in the tissue by means of magnetic resonance, wherein a periodic magnetic field with a repetition time TR≤1 / f _{v is} applied to the tissue.

As already described above, the MRE comprises, on the one hand, the stimulation of mechanical vibrations in the tissue to be examined and, on the other hand, the detection of these oscillations via a magnetic resonance method. Biomechanical properties of the tissue can then be determined by means of characteristic quantities (for example wavelength) by means of the magnetic resonance method of proven vibrations in the tissue (shear waves) (elasticity modulus, viscosity).

The periodic magnetic field used in the context of magnetic resonance to detect the vibrations in the tissue is advantageously an MRE sequence having subsequences that repeat with the repetition time TR. The repeating MRE subsequences thus represent the periodic magnetic field applied to the tissue.

As already mentioned above, an MRE sequence comprises a recording sequence of the conventional MRT (MRT sequence). Such an MRT sequence usually has a radio signal (RF signal) for deflecting the spins and pulsed magnetic fields of different duration for the preparation of the spins. The radio signal and the magnetic fields are repeated in accordance with the repetition time TR of the MRT sequence. Magnetic resonance can be used to record (ie depict) the tissue and the vibrations excited in the tissue. The production of such a recording requires one or more recording steps, each recording step comprising the application of an MRE subsequence and thus having a time duration corresponding to the repetition time TR of the MRE subsequence.

A recording can consist of one image line or several image lines, with one MRE sequence producing one image line or several image lines of the image. It should be noted that the generation of a recording can also take place by means of an MRE sequence which has no periodic subsequences (but only a single MRE sequence).

The magnetic resonance method for detecting the oscillations is thus based on a conventional MRI method and, in addition to the usual MRI magnetic field sequence, comprises a magnetic field sequence (motion coding sequence) for coding (ie "visualizing") the tissue oscillations The motion coding sequence may be, for example, a time-dependent gradient field (MEG) The MRT method (in which the motion coding sequence is embedded) is basically arbitrary as mentioned and is selected with respect to the desired application, eg EPI or SSFP methods However, MRI methods are known per se, so they will not be described in detail here.

Decisive for the MRT method according to the invention is that the repetition time TR of the periodic magnetic field (eg of the MRE subsequences) is less than or equal to the reciprocal of the excitation frequency f _v for the generation of mechanical oscillations in the tissue.

In the conventional MRE, the excitation frequency for the tissue oscillations f _{v is} identical to the frequency f _{g of} the motion coding sequence for encoding the vibration (eg the MEG frequency in the case of a gradient field). Therefore, in the conventional MRE, the repetition time TR is always greater than 1 / f _v . Compared with the conventional MRE, the MRE method according to the invention (with TR≤1 / f _v ) offers two decisive advantages:

1) Shortening the coding time, ie acceleration of MRE recordings 2) Reduction of the excitation frequency and thus increase of the penetration depth of the shear waves, respectively the effective phase signal for soft and viscous materials.

Point 1) allows the use of fast steady-state recording techniques (SSFP) with a TR in the range of 5 ms. This enables MRE recordings with the SSFP-typical good SNR (signal-to-noise ratio) in less than a second. At the same time, point 2) enables excellent illumination of the tissue to be examined by means of shear waves.

The combination of both points allows for example:

- Complete MRE examinations on skeletal muscle with recording of multiple, time-resolved wave images within 50 seconds.

- the recording of 16 time-resolved wave images of the human liver with excellent SNR within 4 breathing pauses - the recording of shear waves in the human heart, leading for the first time

Application of MRE to measure myocardial elasticities allowed.

Thus, the method according to the invention (also referred to as fractionated MRE) generally opens up the possibility of performing clinically relevant examinations on patients, without the considerable temporal burdens associated with conventional MRI and which automatically lead to acousto-mechanical stress of the examination subjects.

With the method according to the invention, only part of a movement cycle of the vibrations excited in the tissue is magnetically coded, whereby a phase difference signal (detected by means of the magnetic resonance for visualization of tissue oscillations) becomes correspondingly smaller. However, as has been shown, this is overcompensated for soft and viscous materials with short transverse relaxation times. In addition, very short acquisition times are achievable by the fractionated MRE, as they are required in the presence of a blood stream or heart movements, so that the method of the invention allows in vivo investigations of myocardial elasticities.

The detection of tissue oscillations according to the invention relates primarily to the detection of the first harmonic (with the frequency f _v ) of tissue oscillations which are generated by a periodic excitation at the frequency f _v . However, it is also possible in principle to detect higher harmonics of the vibration excited in the tissue. For a better understanding of the present invention, the relationship between decay of the MRI signal and attenuation of shear waves in viscous materials using a motion-encoding gradient will be described below. Since both physical quantities (MRI signal, shear wave attenuation) are linked in the MRE, the derivation of this relationship is referred to herein as the elastography equation.

As a first step, an analytical solution for the coded phase signal in the MRE should be specified. This phase signal carries the information of the particle deflection in the tissue. It is measured as a difference signal, i. two phase signals with inverse motion contrast are subtracted from each other, so that only motion information is retained.

It starts from spins in a steady state of mechanical vibrations. The spins are said to oscillate harmonically with a polarization u and phase 2πf _v t + θ, where θ is a constant phase offset that depends on the onset of vibrational excitation and the location of the vibrating particles relative to the vibrator.

The harmonic-dependent phase accumulated during the n-th TR can be calculated as follows:

mtr

Ψ _n = Yg \ - \) TR G (t) sin {2τf _v t + θ) dt

Where γ is the gyromagnetic ratio for protons. The gradient G is the sum of the motion-encoding gradients G _ME and all the imaging gradients contributing to φ _n .

In the following, a periodic, stationary state of the spin phase φ = φ "= φ _{n + N is} assumed. In addition, it is assumed that harmonic gradients are used for motion coding, ie G _ME has the form sin (2 ^ t). The number of gradient cycles (n _g ) is equal to the shortest possible G _M E at a given frequency f _g . No relaxation is considered. This results from the above equation for the phase for the maximum achievable by the motion-encoding gradient phase shift φ: 1

θ = π (\ - q)

In a MRE experiment in which two pictures are taken (either with a changing G _ME or inverse wave amplitudes), the phase difference is twice as large.

The above equation for φ is given for the case that only a single MEG cycle is used. For the classical case fv = f g ^ D converges to the limit γGu / f ". For the case q ≠ 1, the coded phase difference is smaller, i. the signal in the MRE experiment is, as mentioned above, weaker than in the classical case. However, for a correct estimation of the coded signal, the effective deflection u inside the organ to be examined must be taken into account. This can be stated as follows for a distance X from the vibration source if the distance is one wavelength (X = c / fv, with c: velocity of the shear waves in the tissue):

respectively.

μ and η denote the shear modulus and shear viscosity, respectively. U ₀ is the displacement at the location X = O.

The above relationship is a good approximation for the range of viscoelastic parameters studied with the MRE, such as liver or skeletal muscle with μ = 5 kPa, η = 2 Pas, f _v = 100 Hz and n _v = 1. Typical characteristics for a MEG-FeId are: variable gradient direction in all spatial directions with (currently) maximum Amplitude of 35 mT / m; f _g maximum up to 800 Hz for 35 mT / m; trapezoidal-bipolar gradient form with n _g = 1. As base sequence (without motion coding): balanced SSFP with a TR of about 3-4 ms.

In a further advantageous development of the method according to the invention, a period length of the vibrations excited in the tissue corresponds to the repetition time TR of the applied magnetic field or to an integral multiple of the repetition time TR.

In addition, it is preferably provided that the excitation of the mechanical vibrations is synchronized with the magnetic field. The synchronization of TR and fv is important so that the signal phase correlates with an identical phase of motion in each step of the recording (ie in each TR). If the above equation for f _v is disregarded, coherences occur between the signal and the motion phase, which disturb the image reconstruction.

In order to combine short repetition times TR and low oscillation frequencies (f _v ), the number n _v of oscillation periods during a repetition period TR must become fractional, ie:

n _v = f _v -TR <\

This results in f _v ≦ 1 / TR ≦ 1 / (TR _min + 1 / f _g ) <f _g , where TR _{min is} the minimum TR of the sequence without motion coding. A measure of the synchronization between the motion encoding and the vibration is given by the ratio q:

In the conventional MRE q = 1, since here f _{v is} not bound to TR but to f _g . In contrast, with n _v <1 and TR _min > 0 q is always a fraction of one.

It is particularly advantageous if, in order to detect the vibrations generated in the tissue, a first bipolar gradient field is applied to the tissue and a first data record characterizing the phase of the oscillations in the tissue is generated, and subsequently - Applied a second bipolar gradient field on the tissue and a second, the phase of the oscillations in the tissue characterizing record is generated.

Images that represent the phase distribution in the tissue can be generated from the respective data records, which makes it possible to draw conclusions about viscoelastic sizes of the tissue and thus to pathological changes in the tissue. Preferably, data of the first and of the second data set assigned to the same section of the tissue are subtracted from one another and the result is stored in a further data set (from which, for example, image information can be obtained).

One way to generate two images is to produce two consecutive images with different gradient fields, wherein the first gradient field is temporally inverse to the second gradient field, so that the first gradient field at a certain time measured from the beginning of a period of the gradient field Maximum, while the second gradient field at this time has a minimum.

A further possibility is particularly suitable in the method according to the invention if a period of the vibration excited in the tissue consists of a plurality of time intervals corresponding respectively to the repetition time TR of the applied magnetic field and in each case during each of the time intervals by detecting the magnetic field emanating from the tissue Time interval associated (the phase of oscillations in the tissue characteristic) record is generated.

For example, a period of the tissue-excited oscillation may be composed of a first and a second time interval each having a length TR, wherein a) a first data record associated with the first time interval is generated; b) generating a second record associated with the second time interval, wherein the data of the first and second records are respectively associated with particular portions of the fabric; and c) data of the first and the second data set assigned to the same section of the tissue are subtracted from each other. This method is particularly advantageous because the polarity of the gradient field does not have to be changed to take several images, but the same gradient field is used for each image.

In a particularly advantageous development, the stimulation of mechanical vibrations in the tissue takes place by means of the device according to the invention described above. This provides an MRE process that is quick and easy to perform. It is noted, however, that the use of such an excitation device is indeed advantageous, but this is not mandatory for the implementation of the method according to the invention.

The generation of the tissue oscillations can take place in addition to loudspeakers by all apparatuses which directly or indirectly generate a periodic movement. In a second aspect, the invention includes a device for generating mechanical oscillations in an examination subject in magnetic resonance elastography (MRE)

a stepper motor for generating a periodic motion and

- A transmission element for transmitting the periodic movement of the stepping motor to the examination object to generate in this mechanical vibrations.

With a stepper motor, the high accuracy and stability of the oscillatory motion required for synchronization with the motion encoding gradients can be ensured in a simple manner. The connection to a transmission element can be done in the case of rigid materials (pipes, rods, rods), for example by means of eccentric discs. In addition, a hydraulic system for vibration generation may be provided, e.g. for coupling to a transmission element, an eccentric can be attached to a hydraulic piston or to other components of the hydraulic system.

The invention will be explained in more detail below with reference to exemplary embodiments with reference to the figures. Show it:

1A is a side view of an MRE system with a Schwingungserzeu- generating device;

FIG. 1B shows a spatial representation of an MRE system; FIG. 2A is a detailed view of a first embodiment variant of the vibration generating device;

2B shows a second embodiment variant of the vibration generating device;

3 shows a third embodiment of the vibration generating device;

Fig. 4 is a side view of an end piece of a transmission element for

Generation of shear waves in the brain;

Fig. 5A ₁ 5B is a side and front view of a variable end of a transmission element;

6 shows a representation by means of MRE generated shear waves in a liver;

FIG. 7 is an evaluation of the MRE examination shown in FIG. 6; FIG.

8A, 8B show the principal time course of mechanical excitation and motion coding in the MRE;

9A, 9B a determination of a phase difference signal for different excitation frequencies;

10A, 10B show the course of the spin phase for different magnetic field gradients;

11A, 11B show a representation of a calculated phase-to-noise ratio PNR for different repetition times;

Fig. 12 is a time chart of an in vivo myocardial MRE;

FIGS. 13A, 13B show a result of a bSSFP-MRE measurement on a human bicep;

Fig. 14 shows a result of an in vivo bSSFP-MRE experiment on a human liver;

FIG. 15 transmembrane waves recorded by MRE in an intraventricular septum; FIG. FIG. 16 is an exemplary illustration of tissue shear waves in the MRE. FIG.

FIG. 1A shows a patient 1 as the subject of an MRE examination in a clinical tomograph 2. The tomograph 2 has a magnet 21 for generating a static and a periodic magnetic field in a tissue of the patient 1 to be examined. The magnet 21 has a receiving opening 211 for receiving the patient 1.

In the magnet 21 all systems are housed, which are necessary for the generation of the static see magnetic field and the periodic magnetic field (for spatial resolution and energetic excitation of the hydrogen nuclei). It is a self-contained superconducting coil 212, coils 213 orthogonal in FIG

Spaces for spatial resolution briefly switchable magnetic field gradients (also for generating a Bewegungsungskodiergradienten) and a main excitation coil 214 for the polarization of the hydrogen nuclei.

The homogeneity of all the magnetic fields generated with the different coils 212-214 is decisive for the image quality. Disruptions of these magnetic fields by, for example, metallic objects must therefore be carefully excluded.

The stimulation of mechanical vibrations in the tissue to be examined, which is necessary for an MRE examination, is generated by a movement source 3, which is arranged at a distance from the tomograph 2. The movement source 3 comprises a loudspeaker diaphragm 312 which can be set in periodic oscillations.

The movement of the loudspeaker diaphragm 312 is transmitted to the patient 1 via a transmission element connected to a joint 311 in the form of a tube 4 with an optimized ratio of bending stiffness and weight. In this case, one end 41 of the tube 4 is coupled via the joint 311 directly to a surface of the loudspeaker diaphragm 312 so that the oscillations are transmitted directly to the tube via the joint 311 without involving an intermediate gaseous medium.

By the joint 311 and the possible bending of the pipe 4 (and thus of the power transmission) relative to the primary deflection direction of the loudspeaker diaphragm 312, a flexible positioning of the tube 4 on the patient in the region of the tissue to be examined is made possible. These for optimal coupling mechanical Vibrations in the tissue to be examined necessary flexibility is additionally supported by the movable support frame 32.

The movement source 3 is modular. Thus, the loudspeaker diaphragm 312 is part of a loudspeaker 31 which is mounted in an airtight housing 33. The housing 33 has standardized dimensions and forms, with the loudspeaker 31, a loudspeaker module 35 which is exchangeably mounted on a non-magnetic support frame 32. With this structure of the movement source 3, it is possible to easily and quickly replace one speaker module with another, so that different speakers (depending on the required excitation frequency) can be used depending on the kind of the tissue to be examined.

The support frame 32 is stabilized by sand weights, but easily movable on mounted on its underside rollers 34. As a result of the fact that the movement source 3 is set up remotely from the tomograph 2, mutual interactions and disturbances between the loudspeaker magnet with all magnetic fields necessary for the MRI examination are excluded.

For controlling the loudspeaker diaphragm 31 is a function generator 5 which is electrically connected via an amplifier 6 to the loudspeaker diaphragm 31. The function generator 5 is controlled by a control computer 7, which in turn is part of a control of the scanner 2.

The following are some key figures of a device according to the embodiment described above:

Impedance of the speaker: 2 - 16 ohms max. Hub speaker diaphragm: 1 - 60 mm (no load, depending on the speaker used). Power: 100 W - 2 kW continuous load (depending on the speaker used).

Materials: all but the speaker magnets used in both modules

Parts made of non-magnetic materials.

Support frame: wood, non-ferrous metals, PVC, sand

Motion generation modules: wood, magnesium, PVC, stainless steel, epoxy resin Joint: plastic, rubber

Transmission tube: carbon fiber

Transfer heads: wood, PVC, nylon, carbon fibers It should be understood that the values are exemplary only and may be readily changed as required by a particular application of the device.

FIG. 1B shows a similar MRE tomograph as FIG. 1A. In a tomograph 2 there is a patient 1. A movement source 3 for generating mechanical oscillations in a tissue of the patient 1 is installed at a distance d from the tomograph 2. The distance in this embodiment is more than 2.5 m, to safely exclude a disturbing interaction of the speaker magnet with the magnetic fields of the scanner. It is understood that this distance may vary depending on the design of the source of motion and the tomograph, that is also less than said 2.5 m may be.

Figure 2A shows in detail a loudspeaker module 35 for use in a motion source as described in Figures 1A, 1B. The loudspeaker module 35 has a doubly mounted loudspeaker 31, which is arranged in a closed housing 33. The speaker 31 has a speaker diaphragm 312 and for generating a periodic diaphragm movement in comparison with a conventional speaker unchanged drive from a magnet 313 and an induction coil 314. To control the speaker 31 electrical connections 315 are also present over which the speaker with a Control unit (not shown) can be connected.

However, the speaker 31 is modified in terms of the conditions required for a direct transmission of vibration movements. Thus, an additional stabilization sleeve 316 is integrated for power transmission over as large an area as possible, which is advantageous for high stability and robustness against the forces occurring. A transmission element embodied as a rod 400 is connected to the loudspeaker diaphragm 312 via fastening means in the form of adhesive 320, a plastic carrier plate 321 and a threaded journal 322.

The loudspeaker diaphragm 312 is adhesively bonded with the adhesive 320 together with the plastic carrier plate 321 over a large area and into the loudspeaker 31 in a gap-filling manner. The joint 311 is anchored to the plastic carrier plate 321 via the threaded pin 322 in the carrier plate 321 and in the adhesive 320.

The connection via the fastening means described is designed so that vibrations of the membrane directly via the fastening means (adhesive, carrier plate, Joint) are passed into the transmission rod, without the vibrations would be previously passed through a gaseous medium such as air.

An end 41 of the rod 400 for the vibration transmission is fixed to a condyle 3111. The rod 400 has at its opposite end of the joint 311 403 a variable, easily replaceable tail 44, with which the vibration movements are transmitted by direct contact in the object to be examined. The end piece 44 is fixed to the rod 400 with screw connections and easily replaceable. The shape of the end pieces can thereby be optimized variably to the respective examination object. Designs range here from simple ball head over bite timbers to drive direction reversal or power deflection on special downstream transmission mechanisms (see Fig. 4, 5).

FIG. 2B shows a further embodiment of the device according to the invention. As in FIG. 2A, a loudspeaker module 35 has a loudspeaker 31 arranged in a housing 33. A hinge 311 is connected to a portion 3112 via an adhesive 320 to a diaphragm 312 of the speaker 31. The connection takes place via a first threaded pin 322, which projects into the adhesive 320 with one end. The joint 311 is connected at its side facing away from the speaker diaphragm 312 with a pipe 4 as a transmission element. In contrast to the loudspeaker module of FIG. 4, the connection of the joint 311 with the tube 4 takes place via a second threaded pin 323.

With the help of the joint 311, it is possible to see the tube 4 in an angle range α between + - to tilt 10 °. This allows a more flexible positioning of the loudspeaker module 35 with respect to a tomograph (not shown in Fig. 2B, see Fig. 1B). The distance d of the loudspeaker module 35 to the tomograph can be 2-4 m.

FIG. 3 illustrates in detail a further embodiment of the device, namely a loudspeaker module 35, which is designed for an indirect generation of a periodic movement of a loudspeaker diaphragm. The loudspeaker module 35 consists of a closed housing 33 with electrical connections 315 and a doubly mounted first loudspeaker 31 with a loudspeaker diaphragm 312, which is coupled to a transmission element in the form of a tube 4. In this speaker module, however, the speaker diaphragm 312 is a passive diaphragm because the speaker 31 does not have its own drive. The (passive) loudspeaker diaphragm 312 is driven by a second loudspeaker 30 with a (active) loudspeaker diaphragm, which generates periodic pressure fluctuations in the overall closed system (housing 33), followed by the airtight passive diaphragm 312.

The transmission of force from the passive membrane 312 via the tube 4 and a transfer head 44 to a patient (not shown) takes place in the manner already described. The passive membrane 312 has, like the membranes of FIG. 2A ₁ 2B, an additionally integrated stabilizing sleeve 316, a plastic carrier plate 321 fixed with adhesive 320 and a screwed joint 311. The tube 4 and the transfer head 44 can be used universally for all modules.

FIG. 4 shows an end piece 45 for transmitting mechanical vibrations into a brain parenchyma. Special emphasis was placed on a simple application for patient examinations and short preparation times for this component. Individual adjustments, such as those required for bite timbers, are thereby eliminated.

The mechanical vibrations are transmitted from a transfer tube 4 via a positive fit 441 and screw 442 via a joint head 453 in the end piece 45. The central component of the end piece 45 is a sturdy plastic tube 454 in which the head 11 of a person to be examined rests. At the provided with windows 455 plastic tube 454, a roller 456 is attached to the joint head 453, with which the horizontal primary movement of the transfer tube 4 is converted into an up and down movement of the head 11. The roller 456 is slidably mounted to move the end piece 45 relative to a support point 357 of the head 11. This is necessary to ensure the maximum possible excitation of the tissue to be examined. The transmission of motion into the head becomes minimal when the fulcrum of the roller and the support point 357 of the head 11 are superimposed.

Figures 5A, 5B show different views of an end piece 46 for variably transmitting mechanical vibrations into a plurality of tissues. The linear movement of a transfer tube 4 is transmitted via a positive fit 441 and screw 442 via a ball head 461 on an excitation rod 462 of the tail 46. The excitation rod 462 terminates in a transfer head 465, which is also variable in shape and size and for generating vibrations in direct contact with the body surface of a patient (not shown).

The excitation rod 462 is mounted in a plastic ball joint 463 via a bearing unit 467, wherein the bearing unit 467 is connected to a horizontal attachment 468 of a positioning frame 464. The horizontal mount 468 is connected to a base plate 470 via a vertical spacer bar 469.

In order to ensure maximum variability, there are the following possibilities for changing the position of the transfer header:

- Vertical fine adjustment of the excitation rod 462 by a clamping screw 466 on the ball joint 463th

- Horizontal fine adjustment of the excitation rod 462 by displaceability of the bearing unit 467 relative to the horizontal mounting 468. - Vertical coarse adjustment of the bearing unit 467 by different length spacer rods 469th

Horizontal coarse adjustment by displaceability of the spacer rods 469 on the base plate 470.

Figure 6 shows a series of images of a tissue (liver) generated by means of an MRE examination using the vibration generating device according to the invention. The partial image a) shows that vibrations were coupled via a transmission rod 4 into the liver 12 of a patient 1. The vibrations were detected by means of magnetic resonance and a gray value image was generated, whereby the gray values correspond to different phase states of the mechanical vibrations excited in the liver (in the form of shear waves). Here, for various slices 13 (numbered 1-11), such images were generated by the subject's liver 12 (the partial images of Fig. 6 each show an MRE image associated with one of these slices).

The bright regions 14 of each gray value image represent minima of the shear waves. The wavelength of the shear waves excited in the tissue can thus be determined from the gray scale images, for example, by determining the distance between two adjacent bright spots 14. In turn, viscoelastic characteristics of the tissue can be determined from the wavelength. This is shown in FIG. 7 for the respective sections (x-axis, # 1-11) of FIG. 6. The shear modulus is plotted on the y-axis so that each slice # 1-11 is assigned a shear modulus through the liver. If a value falls from the Normal range (as in the case of sections # 5-8), this is an indication of a pathological change in the respective tissue section.

FIG. 8A shows the principle of a classical MRE experiment, according to which mechanical vibrations with a frequency f _{v are} excited in a tissue. The motion sensitization (motion coding) takes place via bipolar movement gradients GM _E (with frequency f _g ), which form part of a magnetic field sequence for detecting the tissue oscillations. The MRI sequence has a repetition time TR. In the conventional MRE method shown in FIG. 8A, the frequency f _{g of} the movement gradient G _ME corresponds to the frequency f _{v of} the mechanical vibrations, so that

In two successive experiments A ₁ B, the polarity of the movement gradient G _M E is reversed in order to obtain a phase contrast Δφ, which depends only on the coded motion (the generated mechanical vibration). For this purpose, the data generated in each case in experiments A, B (characterizing the phase of the oscillations) are processed with one another, for example subtracted from one another.

FIG. 8B shows the principle of the method according to the invention, in which

Unlike the classical method of FIG. 8A, f <-. In Fig. 8B three comparable ^v TR different vibration modes for the polarization vector u (the mechanical oscillation) are shown. The vibration modes are characterized in that the number n _v during a repetition period TR different fractions of the value one has: n _v = 1 (f _v = 1 / TR) ₁ n _v = ^_1/2 (f _v = 1 / (2TR) ) and n _v = 1/4 (f _v = 1 / (4TR)). The star designates a trigger for the wave generator (to synchronize the mechanical vibrations with the MRI sequence). For n _v = 1, two separate experiments with opposite G _ME polarities must be performed to calculate Δφ (see Figure 8A). For n _v = V ₂ and n _v = VA, the phase images for subtraction are taken by interleaving (Δφ = A - B and 2Δφ = A - C + i (B - D), respectively).

Fig. 9A shows a phase difference signal Δ ^ in the MRE for various tissues (muscle and liver tissue A ₁ B). The motion encoding gradient MEG has a frequency f _g = 150 Hz. It can be clearly seen that the maxima of the two curves are far be reached below the MEG frequency f _g . This means that the use of low- rer excitation frequencies for generating tissue oscillations (f _v <f _g ) is advantageous. In particular, a loss in the phase signal by encoding only a fraction of a period of oscillation (as in the method according to the invention) is thereby at least compensated.

FIG. 9B also shows the dependence of the phase difference signal Δφ as a function of the excitation frequency f _v for different encoding frequencies f _g . Again, it can be seen that the maximum achievable phase difference lies at excitation frequencies f _v below f _g .

FIGS. 10A, 10B illustrate the principle of detecting tissue oscillations by means of magnetic resonance. Fig. 10A illustrates a portion of spin phase φ _n due to a phase encode gradient, Fig. 10B a portion of spin phase φ _n due to a (bipolar) motion encoding gradient. The phase-encoding gradient decreases linearly with the line number n, while the bipolar motion-encoding gradient has a constant amplitude. Three different modes of vibration are considered: With n _v = 1 (circular symbols), a phase shift occurs which is proportional to the magnitude of the phase encode gradient. With n _v = ¹ A (square symbols) results in a periodically changing phase shift, which is exploited to assign odd or even line numbers to different images A, B.

The dashed curve corresponds to n _v = 0.9, ie a frequency mismatch between 1 / TR and f _v . This results in a "beat" (or interference) of φ "with a period TR / 0.1. The dotted line represents the spin phase of static particles.

FIGS. 11A, 11B show a calculated phase-to-noise ratio (PNR) for two viscoelastic materials with different T ₂ * relaxation times in order to obtain muscle tissue (FIG. 11A with μ = 5000 kPa, FIG. η = 2 Pas, T ₂ * = 17 ms) or liver tissue (Fig. 11 B with μ = 2000 kPa, η = 4 Pas, T ₂ * = 9 ms). For comparison, a dashed curve represents a classical one MRE experiment, according to which the oscillation frequency f _v and the coding frequency f _g coincide, ie f _v = f _g = 1 / TR (TR _min = 0). The open circles indicate a respective optimal repetition time TR ', which results from the PNR:

Further simulation parameters are: | u _o | = 100 μm; | G _ME | = 20 mT / m (muscle tissue), 35 mT / m (liver); SNR ₀ = 11; TR _min = 3.4 ms (muscle) 3.2 ms (liver).

FIG. 12 represents a time protocol of an in vivo myocardial MR elastography. Due to the high proportion of internal movements, it is necessary here to select a low coding ratio (q << 0.1, n _v =%). In addition, the phase encoding of k-space is divided into 8 segments with 9 lines, which are recorded within 8 cardiac cycles. As a result, 4x4 nested (interleaved) phase recordings are formed, which are assembled into 4 complex phase difference images associated respectively with the cardiac phases © to ® in FIG.

Figures 13A, 13B show a result of bSSFP-MRE measurement on a human bicep using excitation frequency for shear waves of 120 Hz (Fig. 13A) and 75 Hz (Fig. 13B) and n _v = 1, respectively Images show the amplitude of the shear waves (translated into gray values), the lower images the phase difference. In contrast to Fig. 13B, no edge artifacts (banding artifacts) occur in Fig. 13A due to the smaller amplitude of the motion coding and the shorter repetition time (μ ₃ = 50 μm; q = 0.6). FIG. 13B shows voids and phase wrapping artefacts (excitation amplitude U ₃ = 150 μm, q = 0.75). The PNR of the measurement of FIG. 13B is approximately 2.3 times higher than the PNR of the measurement of FIG. 13A, due to the extension of the motion encoding gradient from 5 to 10 ms and the larger excitation amplitude.

The measurement (as well as the measurements described below) was carried out with a 1, 5 T scanner. The bSSFP sequence is a conventional sequence to which the described motion encoding gradient is added (trapezoidal, variable direction bipolar, with frequency and amplitude between phase encode and read out gradient). The oscillation phase was coupled to the MR image acquisition by triggering the wave generator at the start of every Nth TR cycle (N = number of nested images); see. 8A, 8B.

20 waveforms of alternating amplitude of the motion encoding gradient were taken to obtain 10 Δφ images by complex image subtraction. Bernstein MA, King KF, Zhou XJ. Handbook of MRI pulse sequences. Burlington: Elsevier

Academic Press; 2004. "The wave trigger was shifted 10 times to traversing offset θ from π / 5 to 2π. The data acquisition was repeated every 2.5 s, giving a measurement time of 50 s. Other sequence parameters are: | G _ME | = 20 mT / m along cutting direction; FoV = 250 mm; 128x128 matrix; Cutting thickness: 5 mm; a = ± 50 °; Coronal image plane through a longitudinal axis of the muscle.

The data evaluation for deriving elastic parameters was performed by applying a group velocity inversion method as described in Papazoglou S, Rump J, Klatt D, Hamhaber U, Braun J, Sack I. Group velocity inversion in MR elastography on skeletal muscles. In: Proceeding of the 14th Annual Meeting of ISMRM. Seattle. Of 2006.

Figure 14 shows the result of an in vivo bSSFP-MRE experiment on a human liver. The image labeled a) was taken in a plane (with respect to the human body) without applying a motion encoding gradient, ie, TR = TR _min = 3.2 ms. The images labeled b), c) of FIG. 14 show bSSFP-MRE images on the same section as image a), but a motion encoding gradient was applied to the liver with f _v = 51 Hz and n _v =! 4, TR = 9.85 ms (partial image b)) and f _v = 76 Hz and n _v = 1, TR = 13.2 ms (partial c)).

The upper images each represent the MRE amplitude. There is a large decrease in signal intensity due to T ₂ ^* dephasing due to long echo times. Despite the low SNR (signal-to-noise ratio), especially in the images c), the PNR within the liver is still high enough to reconstruct elastic parameters by inversion of the wave equation. The PNR in the images b) is larger than in the images c), although the frequency mismatch of motion encoding and (mechanical) vibrations is greater (q = 0.34 in b) than q = 0.76 in c)).

The data obtained were processed by extracting the first harmonic from the Fourier space at f _v (see Sinkus R, Lorenzen J, Schrader D, Lorenzen M, Dargatz M, Wood D. High-resolution tensor MR elastography for breast tumor detection) Phys. Med. Biol. 2000; 45 (6): 1649-1664). The complex data was entered into a linear inversion program based on an algebraic inversion of the wave equation with viscosity (see Catheline S, Gennisson JL, Delon G, Fink M ₁ Sinkus R, Abouelkaram S, CuMoIi J. Measuring of viscoelastic properties of homogeneous-ous soft solid using transient elastography: an inverse problem approach J Acoust Soc Am 2004; 116 (6): 3734-3741). The viscosity of the liver was estimated by means of the wave velocity dispersion function at f _v = 51 and 76 Hz with Voigt's viscoelastic model; see. Lai WM, Rubin D, Krempl E. Introduction to Continuum Mechanics: Butterworth Heynemann Ltd .; 1994. 570 p.

In the images of Figure 15, transmural waves in the intraventricular septum (IVS) of a subject externally excited by 50 Hz vibrations of the thorax are shown. The images labeled a) are Cine-bSSFP-MRI images for anatomical comparison. The location of the IVS is indicated by the closed line. In this area, the myocardium extends within an image plane of 5 mm thickness. The time resolution of the images in a) is about 40 ms.

The phase difference images of the images b) with about 180 ms time resolution show the externally induced vibrations in the IVS. The oscillations were coded using bipolar motion-encoding gradients at 500 Hz along the cutting plane. This speaker movements were transmitted via a transmission rod to the chest. The vibration intensity was adjusted according to the subjective feeling of the subject.

The maximum amplitude of oscillation was estimated to be 40μm since accurate calculation due to the contribution of the image gradients to Δφ would require complete knowledge of all the components of the deflection polarization u. The arrows K indicate the propagation direction of the shear waves.

FIGS. 16a-d show the principle of the MRE evaluation. Shear waves are coupled into a tissue and detected by magnetic resonance (Figure 16a, b). The bright regions in Figures 16a, b represent regions of maximum amplitude of the shear waves. A computer-implemented evaluation algorithm may be used to filter the MRE data prior to imaging. For example, a threshold for the vibration amplitudes to be represented (ie, the amplitude of the shear waves) may be set to filter out amplitudes that are below the threshold. The amplitude threshold value in FIG. 16b is higher than in FIG. 16a, for which reason the shear wave illustrated "breaks off" in comparison to FIG. 16b after a shorter path. Figs 16c-d show measurements from MRE-specific values for the shear modulus μ and viscosity η at various excitation frequencies f _v. The determination of these quantities takes place by evaluating the representation of the shear waves in the tissue (as in FIG. 16a, b), ie the amplitude and the wavelength of the shear waves.

Representative parameters of the in vivo fractional MRE on biceps, liver and IVS are summarized in Table 1. With q = 0.6, 0.33, and 0.1, the MRE motion encoding from biceps to heart becomes more and more "fractional." The proposed oscillation frequencies may not always be possible because they depend on TR _min , which in turn is determined by the image resolution. The PNR was calculated using the equation described above and with the following parameters: TR _min : 3.5 ms, SNR ₀ = 11, | G _ME | = 35 mT / m, | u _o | = 100 μm; Muscle: T ₂ ^* 17 ms, μ = 5000 Pa, η = 2 Pas; Liver: T ₂ ^* = 9 ms, μ = 2000 Pa, η = 4 Pas.

The parameter μ summarizes the elastic moduli determined in the described measurements for liver and muscle or is taken from the literature (Kanai H. Propagation of spontaneously actuated pulsed vibration in human heart wall and in vivo viscoelasticity estimation.) IEEE Trans Ultrason Ferroelectr Freq Control 2005; 52 (11): 1931-1942 (ultrasound) and Wen H, Bennett E, Epstein N, Plehn J. Magnetic resonance imaging assessment of myocardial elastic modulus and viscosity using displacement imaging and phase-contrast velocity mapping 54 (3): 538-548 (MRI)).

( ^* ) The total scan time is an estimate for planar MRE with 8 θ shifts and a matrix size of 128 (biceps, liver) and 64 (heart), respectively. (**) In the case of the heart, a single scan produces two images Aφ (θ) and Δ #> (0 + 9O °) versus 2Δφ = A-C + i (B-D) as described above. LIST OF REFERENCE NUMBERS

1 patient

11 head

13 cut

14 bright spot

2 tomograph

21 magnet

211 receiving opening

212 superconducting coil

213 coil

214 main excitation coil

3 movement source

31 speakers

311 joint

3111 condyle

3112 joint section

312 loudspeaker diaphragm

315 electrical connections

316 stabilization sleeve

32 support frame

320 adhesive

321 plastic carrier plate

322 threaded pins

323 threaded pin

33 housing

34 castors 5 loudspeaker module 57 support point 0 active loudspeaker 01 active loudspeaker diaphragm

Pipe 1 end 41 positive fit 42 fitting 4 end piece 400 pole

403 end

45 tail

453 condyle

454 plastic pipe

455 windows

456 roll

46 tail

461 ball head

462 excitation rod

463 ball joint

464 positioning rack

465 transmission head

466 clamping screw

467 storage unit

468 Horizontal mounting

469 distance bar

470 base plate

5 function generator

6 amplifiers

7 MRT control

Claims

claims

1. A device for generating mechanical vibrations in an examination subject in magnetic resonance elastography (MRE) with - a displaceable in periodic movements membrane (312) and

- A transmission element (4, 400) for transmitting periodic movements of the membrane (312) on the examination object (1), wherein

- The membrane (312) with the transmission element (4, 400) via fastening means (311, 320, 321, 322) is connected such that periodic movements of the membrane via the fastening means (311, 320, 321, 322) to the transmission element ( 4, 400) are forwarded in order to generate mechanical vibrations in the examination object (1).

2. Apparatus according to claim 2, characterized in that a surface of the membrane (312) with the transmission element (4, 400) is connected.

3. Device according to claim 1 or 2, characterized in that the membrane

(312) is integrally connected to the transmission element (4, 400).

4. Apparatus according to claim 3, characterized in that the cohesive connection by means of adhesive (320).

5. Apparatus according to claim 1 or 2, characterized in that the membrane

(312) is positively connected to the transmission element (4, 400).

6. Device according to one of the preceding claims, characterized in that the fastening means (311, 320, 321, 322) comprise a with an adhesive (320) on the membrane (312) centrally mounted carrier plate (321) connected to the transmission element ( 4, 400) is connected.

7. The device according to claim 6, characterized in that the fastening means (311, 320, 321, 322) elements for producing a screw or rivet connection between the transmission element (4, 400) and the membrane (312).

8. Device according to one of the preceding claims, characterized by a tube-shaped or rod-shaped transmission element (4, 400).

9. Apparatus according to claim 8, characterized in that the membrane with one end (41) of the tubular or rod-shaped transmission element (4, 400) is connected.

10. The device according to claim 9, characterized in that the end (41) of the

Transmission element (4, 400) is connected to a central portion of the membrane (312).

11. Device according to one of the preceding claims, characterized in that the membrane (312) for transmitting force has a centrally arranged stabilizing sleeve (316).

12. Device according to one of the preceding claims, characterized in that the transmission element (4, 400) via a joint (311) is connected to the membrane.

13. Device according to one of the preceding claims, characterized in that the transmission element (4, 400) is rigid.

14. Device according to one of claims 1 to 12, characterized in that the transmission element (4, 400) is designed to be flexible.

15. Device according to one of the preceding claims, characterized in that the transmission element (4) comprises a tube filled with a liquid.

16. Device according to one of the preceding claims, characterized in that the membrane (312) is a speaker diaphragm.

17. Device according to one of the preceding claims, characterized in that in addition to the with the transmission element (4, 400) connected to the membrane (312) (passive membrane), a further membrane (301) (active membrane) is arranged, by their movements with the transmission element (4, 400) connected membrane (312) is set in motion.

18. The device according to claim 17, characterized in that between the two membranes (301, 312) an air-filled cavity is arranged over the loading movements of the active membrane (301) on the passive membrane (312) are transferred.

19. Device according to one of the preceding claims, characterized in that the transmission element (4, 400) on its the examination object (1) facing side has an angled to its main extension direction extending rod or tubular excitation element (462).

20. Device for generating mechanical oscillations in an examination subject in magnetic resonance elastography (MRE) with

a stepper motor for generating a periodic motion and

- A transmission element (4, 400) for transmitting the periodic movement of the stepping motor to the examination object (1) to generate mechanical vibrations in this.

21. The apparatus according to claim 20, characterized in that the stepper motor via an eccentric disc with the transmission element (4, 400) is coupled so that the periodic movement of the stepping motor causes a periodic reciprocation of the transmission element (4, 400).

22. MRE device comprising a device according to one of the preceding claims.

23. A method for magnetic resonance elastographic determination of biomechanical properties of tissue, comprising the steps of: a) exciting mechanical vibrations with a frequency f _v in the tissue to be examined; b) Detection of the mechanical vibrations in the tissue by means of magnetic resonance, wherein a periodic magnetic field with a repetition time TR≤1 / f _{v is} applied to the tissue.

24. The method according to claim 23, characterized in that a periodic

Magnetic field with a repetition time TR <1 / f _{v is} applied to the tissue.

25. The method according to claim 23 or 24, characterized in that the periodic magnetic field is part of a recording sequence of a magnetic resonance elastography

Procedure (MRE sequence), which has the repetition time TR.

26. The method according to any one of claims 23 to 25, characterized in that the

Magnetic resonance for detecting the mechanical vibrations comprises applying a magnetic resonance (MRE) sequence comprising MRE subsequences repeating with the repetition time TR.

27. The method according to claim 25 or 26, characterized in that the MRE

Sequence comprises a recording sequence of a magnetic resonance imaging method (MRI sequence).

28. The method according to claim 27, characterized in that the MRT sequence comprises magnetic fields of different duration and amplitude and at least one radio signal.

29. The method according to claim 25, characterized in that the MRE sequence comprises an oscillating magnetic field for encoding the mechanical vibrations excited in the tissue.

30. Method according to claim 29, characterized in that the magnetic field for coding the mechanical vibrations excited in the tissue is a time-dependent gradient field (MEG).

31. The method according to any one of claims 23 to 30, characterized in that the

Detecting the vibrations excited in the tissue by means of magnetic resonance comprises generating a visual image of the tissue and the vibrations excited in the tissue.

32. Method according to claim 26, characterized in that the production of a single recording requires one or more recording steps, wherein a recording step respectively comprises the application of an MRE subsequence, such that the time duration of a recording step of the repetition time TR of the MRE

Subsequence corresponds.

33. The method according to claim 32, characterized in that the recording is composed of one or more picture lines and a recording step consists of generating one or more picture lines of the picture signal.

34. Method according to claim 23, characterized in that a period length of the vibrations excited in the tissue corresponds to the repetition time TR or is an integer multiple of the repetition time TR.

35. The method according to claim 29, characterized in that the excitation of the mechanical oscillations is synchronized with the oscillating magnetic field for encoding the stimulated in the tissue mechanical vibrations.

36. The method according to claim 30, characterized in that the gradient field (MEG) during a period, the magnetic field direction changes (bipolar gradient field).

37. The method according to claim 36, characterized in that the magnetic resonance for detecting the vibrations generated in the tissue - a first bipolar gradient field applied to the tissue and a first, the phase of the oscillations in the tissue characterizing record is generated, and then

- Applied a second bipolar gradient field on the tissue and a second, the phase of oscillations in the tissue characterizing record is created.

38. Method according to claim 37, characterized in that the first gradient field is temporally inverse to the second gradient field, so that the first gradient field has a maximum at a specific time measured from the beginning of a period of the gradient field, while the second gradient field at this time Minimum.

39. The method according to claim 37 or 38, characterized in that the first and the second data set contain numerical data, which are each associated with a certain section of the tissue.

40. The method according to claim 39, characterized in that in each case the same

Section of the tissue associated data of the first and second data set are subtracted from each other and the result is stored in another record.

41. Method according to claim 23, characterized in that a period of the vibration excited in the tissue consists of a plurality of time intervals of a length corresponding to the repetition duration TR and in each case during one of the time intervals by detecting tissue-induced vibration. magnetic field associated with the respective time interval, the phase of

Vibrations in the tissue characterizing record is generated.

42. The method according to claim 41, characterized in that a period of the vibration excited in the tissue is composed of a first and a second time interval, each with a length TR, and

- a first time interval associated first, the phase of oscillations in the tissue characterizing record is generated;

- A second time interval associated second, the phase of the oscillations in the tissue characterizing record is generated, wherein the data of the first and the second data set are respectively assigned to certain sections of the tissue; and

- Each of the same section of the tissue associated data of the first and second data set are subtracted from each other.

43. The method according to any one of claims 37 to 42, characterized in that by means of the data contained in the data sets viscoelastic parameters (μ, η) of the tissue are determined.

44. The method according to any one of claims 37 to 42, characterized in that an image signal for displaying the tissue is generated in dependence on the data of the data records.

45. The method according to any one of claims 23 to 44, characterized in that the

Stimulating mechanical vibrations in the tissue according to step a) by means of a device according to one of claims 1 to 21 takes place.

46. The method according to any one of claims 23 to 45, characterized in that the

Procedural steps are formulated as program code with which the method is executable on a computer.