Motion correction in magnetic resonance imaging

DE502017016539D1Active Publication Date: 2024-11-21SIEMENS HEALTHINEERS AG
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE502017016539
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Filing Date
2017-05-17
Publication Date
2024-11-21
Estimated Expiration
2037-05-17

AI Technical Summary

Technical Problem

Current magnetic resonance imaging (MRI) techniques struggle to accurately reconstruct dynamic image data from moving objects, leading to movement artifacts due to the assumption that changes in the imaged area between different times have insignificant effects, which is not valid with greater dynamics.

Method used

A method that acquires raw data in a time-dependent manner, assigns it to different movement parameters, and uses deformation fields based on multiple time-dependent movement parameters to correct and iteratively reconstruct image data, improving accuracy and reducing artifacts.

Benefits of technology

This approach allows for improved correction of movement artifacts and enhanced image quality even with strong undersampling, effectively addressing the limitations of existing MRI reconstruction methods by accounting for multiple movement parameters and iteratively refining deformation fields.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a method for reconstructing dynamic image data. Furthermore, the invention also relates to an image reconstruction device. The invention further relates to a magnetic resonance system.

[0002] Modern imaging techniques are frequently used to generate two- or three-dimensional image data or time series of image data, which can be used to visualize an object under investigation. Imaging systems based on magnetic resonance measurement, particularly of nuclear spins, known as magnetic resonance tomography (MRI) scanners, have become established and proven successful in a wide range of applications. In this type of image acquisition, a static background magnetic field B0, which serves to initially align and homogenize the magnetic dipoles under investigation, is usually superimposed with a rapidly switched magnetic field, the so-called gradient field, to achieve spatial resolution of the imaging signal. To determine the material properties of an object under investigation, dephasing or...The relaxation time after a deflection of the magnetization from the initial alignment is determined, so that different material-specific relaxation mechanisms or relaxation times can be identified.

[0003] The deflection is usually achieved by a number of RF pulses, and the spatial resolution relies on a time-defined manipulation of the deflected magnetization using the gradient field in a so-called measurement sequence or control sequence. This sequence defines a precise timing of RF pulses, changes in the gradient field (by emitting a switching sequence of gradient pulses), and the acquisition of measured values. Besides relaxation, there are several other mechanisms for contrast generation, such as flux measurement and diffusion imaging.

[0004] Typically, an intermediate step is used to correlate measured magnetization—from which the aforementioned material properties can be derived—with a spatial coordinate of the measured magnetization in the space where the object under investigation is located. In this intermediate step, acquired raw magnetic resonance data are arranged at readout points in so-called "k-space," where the coordinates of k-space are encoded as a function of the gradient field. The magnitude of the magnetization (especially the transverse magnetization, determined in a plane perpendicular to the previously described fundamental magnetic field) at a specific location of the object under investigation can be determined from the k-space data using a Fourier transform. In other words, the k-space data (magnitude and phase) are required to calculate the signal strength and, if applicable, its phase in space.

[0005] Magnetic resonance imaging (MRI) is also used to acquire images of moving areas under investigation. Examples include MR angiography and dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), which are described in examples in German publications DE 102009036237 A1 and US 8195275 B2. During raw data acquisition, only undersampled raw data is captured at certain time points. Therefore, to reconstruct image data from a specific time point, data from other time points must also be used, but these data were acquired when the area under investigation was in a different state of motion.To be able to use raw data acquired at different times together, it is conventionally assumed that changes in the area being imaged between the individual points in time have only a negligible effect on the image. However, this assumption no longer holds true with greater dynamics in the image area, so that motion artifacts appear in the reconstructed images.

[0006] Approaches to incorporating motion into image reconstruction are described in the literature. For example, applying deformation fields can achieve higher consistency between raw data acquired at different times and the image data reconstructed from them. Some approaches consider motion during iterative image reconstruction. However, these approaches typically only consider one dimension, either time or the different states of motion. For instance, in time-resolved imaging, motion fields are calculated during the first image reconstruction and then included in a subsequent second reconstruction.

[0007] In M. Salman Asif et al: "Motion-adaptive spatio-temporal regularization for accelerated dynamic MRI", MAGNETIC RESONANCE IN MEDICINE, Vol. 70, No. 3, November 6, 2012 (2012-11-06), pages 800-812, XP055159307, ISSN: 0740-3194, DOI: 10.1002 / mrm.24524) a reconstruction within the framework of MR imaging is described, in which spatially constant deformation fields are used.

[0008] In DE 10 2013 205 832 A1, an MR imaging procedure with image reconstruction is described, whereby deformation fields are determined depending on the respiratory movement of a patient for image correction.

[0009] It is therefore an object of the present invention to develop a method for reconstructing dynamic MR image data with which improved correction of motion artifacts is possible.

[0010] This problem is solved by a method for reconstructing dynamic image data according to claim 1, an image reconstruction device according to claim 8 and by a magnetic resonance imaging system according to claim 9.

[0011] In the inventive method for reconstructing dynamic image data, raw data from a study area are acquired in a time-dependent manner, wherein at least a portion of the raw data is assigned different values ​​of motion parameters. In this context, dynamic image data refers to image data of a moving object that is recorded over a predetermined period. The term "motion parameters" refers to quantities that influence a state of motion. Examples include time, respiratory state, or cardiac activity. "Time-dependent acquisition" refers to the fact that different time points are assigned to individual k-space points of the raw data, and a set of consecutively measured k-space points is summarized at the calculated time points.

[0012] To obtain information about assigning raw data to individual motion states, additional Navigator raw data can be acquired and reconstructed into Navigator image data. The Navigator image data identifies the motion states of the area under investigation. Based on these now-known motion states and the acquired raw data, raw data can then be assigned to individual motion states; that is, raw data ordered according to motion state are generated.

[0013] Based on the acquired raw data, initial time-dependent image data are reconstructed. This means that image data for several different time points are reconstructed. Based on this initial image data, deformation fields are then determined as a function of at least two time-dependent motion parameters. Deformation fields provide information about the displacement of a point within an area of ​​investigation as a function of these motion parameters, for example, time and respiratory or cardiac activity. Finally, corrected image data are generated based on the deformation fields of the raw data and the initial image data.

[0014] Deformation fields are used to register portions of the initial image data, each corresponding to a different motion parameter, to obtain more image information for each parameter value. Based on this more precise image information, improved deformation fields can then be generated, for example, through iteration or iterative image reconstruction. These improved fields can then be used to reconstruct even more accurate, i.e., "corrected," image data. Therefore, even with significant undersampling, as is common in magnetic resonance imaging of moving objects, image data with improved accuracy and / or reduced artifacts can be generated.

[0015] The image reconstruction device according to the invention comprises an input interface for receiving time-dependent acquired raw data from a test area, wherein at least a portion of the raw data is assigned to different values ​​of motion parameters. The image reconstruction device also includes a reconstruction unit for reconstructing initial time-dependent image data based on the acquired raw data. Furthermore, the image reconstruction device according to the invention comprises a deformation field determination unit for determining deformation fields based on the initial image data as a function of at least two time-dependent motion parameters, and a correction unit for generating corrected image data based on the deformation fields, the raw data, and the initial image data.

[0016] The magnetic resonance imaging system according to the invention comprises a control unit designed to control the magnetic resonance imaging system using the method according to the invention. For this purpose, the magnetic resonance imaging system according to the invention includes the image reconstruction device according to the invention.

[0017] The essential components of the image reconstruction device according to the invention can be predominantly implemented as software components. This applies in particular to the reconstruction unit, the deformation field determination unit, and the correction unit. However, these components can also be partially implemented as software-supported hardware, such as FPGAs or similar devices, especially when particularly fast calculations are required. Similarly, the necessary interfaces, for example, when only data transfer from other software components is required, can be implemented as software interfaces. Alternatively, they can be implemented as hardware interfaces controlled by suitable software.

[0018] In particular, the image reconstruction device according to the invention can be part of a user terminal or a control unit of a magnetic resonance imaging system. A largely software-based implementation has the advantage that even previously used image reconstruction devices or control units can be easily retrofitted by a software update to operate in the manner of the invention. In this respect, the problem is also solved by a corresponding computer program product with a computer program that can be directly loaded into a memory device of a control unit of a magnetic resonance imaging system, with program sections to execute all steps of the method according to the invention when the program is executed in the control unit. In addition to the computer program, such a computer program product may optionally include additional components such as, for example,Documentation and / or additional components may include hardware components such as hardware keys (dongles, etc.) for using the software.

[0019] For transport to the control unit and / or for storage on or in the control unit, a computer-readable medium, such as a memory stick, a hard drive, or another portable or permanently installed data carrier, can be used, on which the program sections of the computer program that can be read and executed by a computing unit of the control unit are stored. The computing unit can, for example, comprise one or more cooperating microprocessors or the like. The computing unit can, for example, be part of a terminal or the control unit of the magnetic resonance imaging system. The dependent claims and the following description each contain particularly advantageous embodiments and further developments of the invention. In particular, the claims of one claim category can also be further developed analogously to the dependent claims of another claim category.Furthermore, within the scope of the invention, the various features of different embodiments and claims can also be combined to form new embodiments, provided that the resulting object falls within the scope of protection of the invention as defined by the claims.

[0020] In one embodiment of the inventive method, the corrected image data is generated based on an iterative reconstruction, in which the deformation fields are used to generate corrected raw data. Based on the deformation fields, additional, previously unavailable raw data is generated and / or existing raw data is assigned to different values ​​of the motion parameters. The additionally generated raw data, which is available for one or more sets of motion parameter values, can be used to assign raw data that was incorrectly assigned to motion states to the correct motion states. In this way, the image quality can be improved and / or motion artifacts can be reduced. Deformation fields are also determined via optimization. A known deformation field can be used as a starting point.Typically, one simply sets the number of iterations for optimizing the deformation fields. Therefore, "determining" deformation fields should also include optimizing them. Deformation fields and image data can also be optimized simultaneously or in a single optimization process.

[0021] In one embodiment of the method according to the invention, the acquisition of the raw data comprises a Cartesian, potentially incomplete scan. In this context, incomplete scanning is understood to mean undersampling. Such scanning can be particularly advantageous, for example, in liver imaging, since in such imaging the image representations are oriented in the head-to-toe direction or in the anterior-posterior direction.

[0022] In a preferred embodiment of the method according to the invention, the determination of deformation fields comprises the registration of image data with at least partially different values ​​of the time-dependent motion parameters. Displacements in the investigation area are determined using these registrations. In this way, image information that is assigned to specific values ​​of motion parameters can also be assigned to other values, thus improving the image quality for individual tuples of motion parameter values.

[0023] According to the invention, the at least two different time-dependent motion parameters are time and a breathing state or time and a cardiac motion state.

[0024] Advantageously, the method according to the invention takes several motion parameters into account. For example, raw data or image data, which are assigned to the same state of motion but to different times, can be combined using the deformation fields, thus improving the image quality.

[0025] According to the invention, raw data is synthesized to generate corrected image data based on deformation fields for different values ​​of at least two time-dependent motion parameters. Using the deformation fields, image data can be assigned to corrected positions. Based on this corrected assignment, synthesized raw data can then be generated, which in turn can be used to reconstruct corrected image data.

[0026] According to the invention, the acquired raw data are reassigned to modified values ​​of motion parameters based on the synthesized raw data. For example, by comparing the synthesized raw data with the acquired raw data, incorrect assignments of the acquired raw data can be detected, and the acquired raw data can then be assigned to the correct motion states. In this way, effects based on deformation, which lead to incorrect assignments of raw data to individual motion states, can be compensated for. Incorrect assignments can also arise from inaccurate measurement of motion states, for example, in one-dimensional navigation. These errors can also be corrected by a subsequent correction during the assignment of the raw data to individual motion states.

[0027] Preferably, the assignment to modified values ​​of motion parameters depends on whether the states assigned to the values ​​are similarly far apart. In other words, for each measurement point, the determined raw data is compared to the different synthetic measurement points / raw data using a norm, distance measure, or correlation, and raw data is assigned to the motion state that is closer or has a higher correlation. Thus, it is determined which synthesized raw data points are closest to the measured raw data points in terms of distance measure or correlation.

[0028] In one variant of the inventive method, the reassignment of the raw data is first carried out on the basis of a subset of the acquired raw data, preferably in a fully sampled direction, for example in the readout direction, and subsequently a reconstruction of the corrected image data is carried out on the basis of the entire raw data.

[0029] In the second reconstruction, the assignment between raw data and motion states, determined based on the subset, is then carried out. No further state assignments are changed at this stage. This approach advantageously saves time and data capacity. For example, an initial reconstruction of image data can be performed solely on the basis of raw data acquired in the readout direction. This raw data is then used to determine deformation fields, which in turn are used to correct the assignment of the acquired raw data.

[0030] In one embodiment of the method according to the invention, raw data that differ in at least one of the at least two time-dependent motion parameters are discarded, and the corrected image data are generated only as a function of a subset of the at least two time-dependent motion parameters. Advantageously, the reconstruction time can be shortened due to the reduced data processing effort.

[0031] In a fully scanned direction, orthogonal disks or disk-shaped layers can be reconstructed independently of each other. Therefore, to reduce the computational effort, the number of disks or layers can be reduced, thus decreasing the reconstructed image volume.

[0032] The invention is explained in more detail below with reference to the accompanying figures and exemplary embodiments. The figures show: FIG 1 a flowchart illustrating a method for reconstructing dynamic image data that does not include all the features of the method defined in claim 1, FIG 2 a flowchart illustrating a method for reconstructing dynamic image data according to an embodiment of the invention, FIG 3 Block diagram illustrating an image reconstruction device according to an embodiment of the invention, FIG 4 a schematic representation of a magnetic resonance imaging system according to an embodiment of the invention.

[0033] In FIG 1 A flowchart 100 is shown, illustrating a method for reconstructing dynamic image data that does not include all the features of the method defined in claim 1.

[0034] In step 1.I, raw data RD is initially acquired from a study area. During the acquisition of the raw data RD, the study area behaves dynamically; that is, at least part of the raw data RD is assigned different values ​​of motion parameters BP t. In step 1.II, the first time-dependent image data BD1 is reconstructed based on the acquired raw data RD.

[0035] In step 1.III, deformation fields DF are determined based on the first image data BD1. The deformation fields DF depend on at least two time-dependent motion parameters, for example, time t and the respiratory movement state AZ.

[0036] In step 1.IV, synthetic raw data RD S is generated based on the first image data and the deformation fields DF.

[0037] Furthermore, in step 1.V, corrected image data BD k is reconstructed based on the synthetic raw data RD s. In step 1.VI, it is determined whether the corrected image data BD k meets a quality criterion (crit). For example, the corrected image data BD k is checked for artifacts. If the quality criterion has not yet been met, which occurs in FIG 1 If marked with "n", the process returns to step 1.III, in which deformation fields DF are generated again, this time based on the corrected image data BD k. The procedure then continues with steps 1.IV to 1.VI. If, in step 1.VI, it was determined that the quality criterion crit for the corrected images BD k was met, which is the case in FIG 1 If marked with "j", the process proceeds to step 1.VII, in which the last generated image data BD k is adopted as final image data BD.

[0038] In FIG 2 Figure 200 shows a flowchart illustrating a method for reconstructing dynamic image data according to an embodiment of the invention.

[0039] In step 2.I, raw magnetic resonance (MR) data (RD) are first acquired from a patient's field of view (FOV). Additionally, in step 2.II, raw navigator data (NV-RD) are acquired and reconstructed into navigator image data (NV-BD). The navigator image data (NV-BD) characterizes a motion state of the FoV being imaged. Based on these known motion states and the raw MR data, in step 2.III, raw MR data are assigned to individual motion states, thus generating raw data (RD 1) ordered by motion state.

[0040] In step 2.IV, image data BD is reconstructed based on the raw data RD 1, which is ordered according to motion states. Raw data from different time points but with the same motion states are combined. Furthermore, in step 2.V, a complete set of raw data with additional synthesized RD s in k-space is generated for each motion state n and each time point t. These synthesized RD s replace the raw data missing due to undersampling during raw data acquisition. In step 2.VI, the acquired raw data RD is then reassigned to the individual motion states by comparing the assignment of the acquired raw data RD or the ordered raw data RD 1 with the assignment of the synthesized raw data RD s. Corrected or reordered raw data RD k are generated as needed. In step 2.VII, it is checked whether a termination criterion crit is met.In the event that the termination criterion has not yet been met, which is in . FIG 2 If marked with "n", the process returns to step 2.IV and reconstructs image data BD based on the raw data RD k corrected in step 2.VI. Steps 2.V and 2.VI are then executed again. If a termination criterion crit is met in step 2.VII, for example, if a counter index has reached a maximum number of iterations, the process terminates the step. FIG 2 If marked with "j", proceed to step 2.VIII.

[0041] In step 2.VIII, image data BD n,t are reconstructed based on the corrected raw data RD k determined in step 2.VI, which are ordered according to motion state n and time t. Images of different motion states are generated using deformation fields DF. In other words, the deformation fields DF are used to register image data from different motion states against each other, thus ensuring a sufficient data basis for each motion state even if individual motion states are undersampled. For example, a matrix representing the identity can be used as the initial deformation field DF 0. Subsequently, in step 2.IX, deformation fields DF n,t between different motion states and different times are determined based on the reconstructed image data BD n,t. If a termination criterion crit is not met, which occurs in FIG 2 If the step is marked with "n", then step 2.X returns to step 2.VIII and reconstructs image data BD n,t ordered by motion state and time, this time applying the newly generated deformation fields DF n,t between different motion states n and different times t. If a termination criterion crit for the iteration is met in step 2.X, which occurs in FIG 2 If marked with "j", the process proceeds to step 2.XI, in which the last reconstructed image data BD n,t are output as final image data BD.

[0042] In FIG 3 An image reconstruction device 30 according to an embodiment of the invention is illustrated. The image reconstruction device 30 comprises an input interface 31, which is configured to receive time-dependent acquired raw data RD. At least a portion of the raw data RD is assigned to different values ​​of motion parameters BP t. The raw data RD is transmitted to a reconstruction unit 32, which reconstructs initial time-dependent image data BD 1 based on the acquired raw data RD. The image data BD 1 is then passed on to a deformation field determination unit 33, which determines deformation fields DF based on the initial image data BD 1 as a function of at least two time-dependent motion parameters BP t.The determined deformation fields DF as well as the received raw data RD and the reconstructed image data BD 1 are transmitted to a correction unit 34, which determines corrected image data BD k on the basis of the deformation fields DF, the raw data and the first image data BD1.

[0043] The correction unit 34 comprises a synthesis unit 34a, which generates synthesized raw data RD s based on the deformation fields DF, the raw data RD, and the image data BD1. Also part of the correction unit 34 is an image data reconstruction unit 34b, which reconstructs corrected image data BD k based on the synthesized raw data RD s. The corrected image data BD k is transmitted to a verification unit 34c, which is configured to check the corrected image data BD k for the occurrence of artifacts.If the inspected image data BD k does not yet meet a predetermined quality criterion, the corrected image data BD k is transmitted back to the deformation field determination unit 33, which generates corrected deformation fields DF based on the corrected image data BD k. These are then transmitted, together with the corrected image data BD k, to the correction unit 34, which generates further corrected image data BD k. If these meet the aforementioned quality criterion, the most recently generated corrected image data BD k are designated as the final image data BD and output via an output interface 35.

[0044] FIG 4 Figure 1 shows a schematic representation of a magnetic resonance imaging system 1 according to the invention. This system comprises the actual magnetic resonance scanner 2 with a measurement chamber 8 or patient tunnel located therein. A bed 7 can be moved into this patient tunnel 8, so that a patient O lying on it can be positioned at a specific location within the magnetic resonance scanner 2 relative to the magnet system and radio frequency system arranged therein during an examination, or can be moved between different positions during a measurement.

[0045] Essential components of the magnetic resonance scanner 2 are a base field magnet 3, a gradient system 4 with gradient coils for applying arbitrary magnetic field gradients in the x, y, and z directions, and a whole-body radio frequency coil 5. Magnetic resonance signals induced in the object under investigation O can be received via the whole-body coil 5, which is also typically used to transmit the radio frequency signals for inducing the magnetic resonance signals. However, these signals are usually received using local coils 6 placed, for example, on or under the object under investigation O. All these components are generally known to those skilled in the art and are therefore described in [reference to relevant document / reference]. FIG 4 Only a rough schematic representation.

[0046] The whole-body high-frequency coil 5 can, for example, be in the form of a so-called birdcage antenna and have a number N of individual antenna rods, which can be controlled separately as individual transmission channels S1, ..., SN by a control unit 10; i.e., the magnetic resonance imaging system 1 is a pTX-capable system. However, it is expressly pointed out that the method according to the invention is also applicable to conventional magnetic resonance imaging devices with only one transmission channel.

[0047] The control unit 10 can be a control computer, which can also consist of a large number of individual computers – possibly even spatially separated and interconnected via suitable bus systems, cables, or the like. This control unit 10 is connected to a terminal 20 via a terminal interface 17, through which an operator can control the entire system 1. In this case, this terminal 20 has a computer 21 with a keyboard 28, one or more monitors 27, and other input devices such as a mouse or the like, so that a graphical user interface is available to the operator.

[0048] The control unit 10 includes, among other things, a gradient control unit 11, which in turn can consist of several subcomponents. The individual gradient coils are connected to control signals SGx, SGy, SGz via this gradient control unit 11. These are gradient pulses that are applied during a measurement at precisely defined temporal positions and with a precisely defined temporal profile in order to scan the object under investigation O and the associated k-space, preferably in individual layers SL, according to a control sequence AS.

[0049] The control unit 10 also includes a high-frequency transmitter / receiver unit 12. This RF transmitter / receiver unit 12 also consists of several subcomponents to transmit high-frequency pulses separately and in parallel to the individual transmission channels S1, ... SN, i.e., in this case, to the individually controllable antenna elements of the body coil 5. Magnetic resonance signals can also be received via the transmitter / receiver unit 12. In this embodiment, however, this is done using the local coils 6. The raw data RD received by these local coils 6 are read out and processed by an RF receiver unit 13. The magnetic resonance signals received by this receiver or by the whole-body coil 5 via the RF transmitter / receiver unit 12 are transferred as raw data RD to an image reconstruction device 30 according to the invention, which uses this data to process the information associated with the FIG 3The described method reconstructs the image data BD and stores it in a memory 16 and / or transmits it to the terminal 20 via the interface 17 so that the operator can view it. The image data BD can also be stored and / or displayed and evaluated at other locations via a network NW. If the local coils 6 have a suitable switching unit, they can also be connected to an RF transmit / receive unit 12 in order to use the local coils for transmitting, especially in pTX operation.

[0050] The gradient control 11, the RF transmit / receive unit 12, and the receiver unit 13 for the local coils 6 are each coordinated and controlled by a measurement control unit 15. This unit ensures, through appropriate commands, that a desired gradient pulse train GP is transmitted by suitable gradient control signals SGx, SGy, SGz, and simultaneously controls the RF transmit / receive unit 12 so that a multi-channel pulse train MP is transmitted, i.e., that the appropriate high-frequency pulses are simultaneously applied to the individual transmitting rods of the whole-body coil 5 on the individual transmitting channels S1, ... SN. Furthermore, it must be ensured that the magnetic resonance signals at the local coils 6 are read out and processed by the RF receiver unit 13 at the appropriate time, and any signals at the whole body coil 5 are read out and processed by the RF transmitter / receiver unit 12.The measurement control unit 15 outputs the corresponding signals, in particular the multi-channel pulse train MP to the high-frequency transmitter / receiver unit 12 and the gradient pulse train GP to the gradient control unit 11, according to a predefined control protocol P. This control protocol P contains all control data that must be set during a measurement according to a predefined control sequence AS.

[0051] Typically, a large number of control protocols P for various measurements are stored in memory 16. These can be selected and, if necessary, modified by the operator via terminal 20 to obtain a suitable control protocol P for the desired measurement, which the measurement control unit 15 can then use. Furthermore, the operator can also retrieve control protocols P via a network NW, for example, from a manufacturer of the magnetic resonance system, and then modify and use them as needed.

[0052] The basic procedure of such a magnetic resonance measurement and the aforementioned control components are known to those skilled in the art, so they will not be discussed in detail here. Furthermore, such a magnetic resonance scanner 2 and the associated control unit 10 can include a multitude of other components, which are also not explained in detail here. It should be noted that the magnetic resonance scanner 2 can also be constructed differently, for example with a laterally open patient chamber, and that, in principle, the high-frequency whole-body coil does not necessarily have to be configured as a birdcage antenna.

[0053] Finally, it should be noted once again that the methods and devices described above are merely preferred embodiments of the invention and that the invention can be varied by a person skilled in the art without departing from the scope of the invention as defined by the claims. The method and the reconstruction device were primarily explained using a magnetic resonance imaging (MRI) system for acquiring medical image data. However, the invention is not limited to applications in the medical field; it can also be applied, in principle, to MRI systems for acquiring dynamic images for other purposes. For the sake of completeness, it should also be noted that the use of the indefinite articles "a" or "an" does not preclude the possibility that the features in question may be present multiple times.Likewise, the term "unit" does not preclude the possibility that it consists of several components, which may also be spatially distributed.

Claims

1. Method for reconstructing dynamic image data, having the following steps: - time-dependent acquisition of magnetic resonance signals as raw data (RD) from an examination region (FoV) of a patient, wherein at least some of the raw data (RD) is assigned to different values of at least two movement parameters (BPt), wherein the at least two time-dependent movement parameters are the time and a respiratory condition or the time and a heart movement condition, - reconstruction of first time-dependent image data (BD1), based on the acquired raw data (RD), - ascertainment of deformation fields (DF) based on first image data (BD1) as a function of the at least two time-dependent movement parameters (BPt), - generation of corrected image data (BDk) based on deformation fields (DF), the raw data (RD) and the first image data (BD1), wherein for the generation of corrected image data (BDk) based on deformation fields (DF), raw data (RDs) synthesised for different values of the at least two time-dependent movement parameters (BPt) is ascertained and raw data (RD) acquired on the basis of synthesised raw data (RDs) is reallocated to changed values of movement parameters (BPt).

2. Method according to claim 1, wherein the generation of corrected image data (BDk) takes place on the basis of an iterative reconstruction in which the deformation fields (DF) are used to generate corrected raw data.

3. Method according to claim 1 or 2, wherein the time-dependent acquisition of raw data (RD) comprises a Cartesian scanning of raw data.

4. Method according to one of claims 1 to 3, wherein the determination of deformation fields (DF) comprises a registering of image data with at least partially different values of the time-dependent movement parameters (BPt).

5. Method according to one of the preceding claims, wherein the assignment to changed values of movement parameters (BPt) takes place as a function of which synthesised raw data points (RDs) are closest to the measured raw data points (RD) in a distance measurement or by means of correlation.

6. Method according to claim 5, wherein the reassignment of the raw data (RD) is first performed on a subset of the acquired raw data (RD), preferably in a fully sampled direction, and then the corrected image data (BDk) is reconstructed on the basis of all the raw data (RD).

7. Method according to one of the preceding claims, wherein raw data (RD) which differs in at least one of the at least two time-dependent movement parameters (BPt) is discarded and the corrected image data (BDk) is only generated as a function of a subset of the at least two time-dependent movement parameters (BPt).

8. Image reconstruction apparatus (30) having: - an input interface (31), which is designed for receiving magnetic resonance signals acquired in a time-dependent manner as raw data (RD) from an examination region (FoV) of a patient, wherein at least some of the raw data (RD) is associated with different values of at least two movement parameters (BPt), wherein the at least two time-dependent movement parameters are the time and a respiratory condition or the time and a heart movement condition, - a reconstruction unit (32), which is designed for reconstructing first time-dependent image data (BD1) based on acquired raw data (RD), - a deformation field determination unit (33), which is designed for determining deformation fields (DF) based on first image data (BD1) as a function of at least two time-dependent movement parameters (BPt), - a correction unit (34), which is designed for generating corrected image data (BDk) based on deformation fields (DF), raw data (RD) and first image data (BD1), wherein the correction unit (34) is designed, for the generation of corrected image data (BDk) based on deformation fields (DF), to ascertain raw data (RDS) synthesised for different values of the at least two time-dependent movement parameters (BPt) and to reallocate raw data (RD) acquired on the basis of synthesised raw data (RDs) to changed values of movement parameters (BPt).

9. Magnetic resonance imaging system (1), comprising an image reconstruction apparatus (30) according to claim 8.

10. Computer program product with a computer program which can be loaded directly into a storage device of a control device (10) of a magnetic resonance imaging system (1), with program sections to perform all the steps of the method according to one of claims 1 to 7 when the computer program is executed in the control device (10) of the magnetic resonance imaging system (1).

11. Computer-readable medium on which program sections readable and executable by a processing unit are stored to perform all the steps of the method according to one of claims 1 to 7 when the program sections are executed by the processing unit.