METHOD AND APPARATUS FOR THE RECORDING AND RECONSTRUCTION OF A FOUR-DIMENSIONAL DYNAMIC MAGNETIC RESONANCE IMAGE DATA RECORD

Abstract

In a method and apparatus for acquiring and reconstructing a four-dimensional dynamic magnetic resonance (MR) image data record, MR data are continuously acquired by radial scanning of an examination region along radial k-space lines, and a dynamic region of the examination region, in which said dynamic procedure is relevant, is determined, as well as a non-dynamic region, which is not relevant to the dynamic procedure. Static image data are reconstructed from all of the acquired MR data, and image data therein originating from the non-dynamic region are marked and are then not used for reconstructing a dynamic image data record for the dynamic region.

Claims

1. A method for acquiring and reconstructing a four-dimensional dynamic magnetic resonance (MR) image data record of an examination region of a patient, said method comprising: from a computer, operating an MR data acquisition scanner so as to execute a dynamic MR data acquisition procedure wherein MR data of an examination region of a patient are continuously acquired, and entered into a memory organized as k-space, by radial scanning of k-space along respective radial k-space lines; in said computer, during respective individual time segments of a recording period during which said MR data are acquired in said dynamic data acquisition procedure, reconstructing respective MR images from at least some of the acquired MR data respectively in said k-space lines into which said MR data were entered during the respective time segment; in said computer, dividing said examination region spatially into a dynamic region, in which said dynamic data acquisition procedure takes place, and a non-dynamic region outside of said dynamic region, and reconstructing a static MR image data from all of the acquired MR data, and marking, in said static MR image data record, image data that originate from said non-dynamic region, and eliminating the marked image data from said MR data by inverse transformation of the marked image data; and in said computer, reconstructing MR images for said different time segments from MR data from which said marked image data have been eliminated, so as to compile a dynamic image data record in said dynamic region.

2. A method as claimed in claim 1 comprising using image data in said static image data record for the MR images of the dynamic image data record in said non-dynamic region.

3. A method as claimed in claim 1 comprising determining said dynamic region and said non-dynamic region by evaluating the static image data record, by localization and marking of at least some anatomical features of the patient that are unrelated to the dynamic data acquisition procedure.

4. A method as claimed in claim 1 comprising determining said dynamic region and said non-dynamic region by evaluating a position of a saturation region in an MR data acquisition sequence that is used to acquire said MR data.

5. A method as claimed in claim 1 comprising removing said image data that were inverse transformed into k-space, of said non-dynamic region, by subtraction.

6. A method as claimed in claim 5 comprising removing said inverse-transformed image data in k-space after implementing a Cartesian gridding, so that said image data are in a Cartesian grid, and recalculating the MR data in said Cartesian grid.

7. A method as claimed in claim 5 comprising removing the inverse-transformed image data in k-space at scanning points used in said radial scanning of k-space.

8. A method as claimed in claim 1 comprising spacing successively scanned radial lines in k-space apart from each other by the golden angle.

9. A method as claimed in claim 1 comprising administering a contrast agent to the patient before executing said dynamic data acquisition procedure and acquiring said MR data from the liver of the patient, as said examination region.

10. A method as claimed in claim 1 comprising determining said non-dynamic region to be a region of the patient consisting of the arms of the patient.

11. A magnetic resonance (MR) apparatus comprising: an MR data acquisition scanner; a computer configured to operate said MR data acquisition scanner so as to execute a dynamic MR data acquisition procedure wherein MR data of an examination region of a patient are continuously acquired, and entered into a memory organized as k-space, by radial scanning of k-space along respective radial k-space lines; said computer being configured to reconstruct, during respective individual time segments of a recording period during which said MR data are acquired in said dynamic data acquisition procedure, respective MR images from at least some of the acquired MR data respectively in said k-space lines into which said MR data were entered during the respective time segment; said computer being configured to divide said examination region spatially into a dynamic region, in which said dynamic data acquisition procedure takes place, and a non-dynamic region outside of said dynamic region, and to reconstruct a static MR image data from all of the acquired MR data, and to mark, in said static MR image data record, image data that originate from said non-dynamic region, and to eliminate the marked image data from said MR data by inverse transformation of the marked image data; and said computer being configured to reconstruct MR images for said different time segments from MR data from which said marked image data have been eliminated, so as to compile a dynamic image data record in said dynamic region.

12. A non-transitory, computer-readable data storage medium encoded with programming instructions, said storage medium being loaded into a computer of a magnetic resonance (MR) apparatus, and said programming instructions causing said computer to: operate said MR apparatus so as to execute a dynamic MR data acquisition procedure wherein MR data of an examination region of a patient are continuously acquired, and entered into a memory organized as k-space, by radial scanning of k-space along respective radial k-space lines; during respective individual time segments of a recording period during which said MR data are acquired in said dynamic data acquisition procedure, reconstruct respective MR images from at least some of the acquired MR data respectively in said k-space lines into which said MR data were entered during the respective time segment; divide said examination region spatially into a dynamic region, in which said dynamic data acquisition procedure takes place, and a non-dynamic region outside of said dynamic region, and reconstruct a static MR image data from all of the acquired MR data, and mark, in said static MR image data record, image data that originate from said non-dynamic region, and eliminate the marked image data from said MR data by inverse transformation of the marked image data; and reconstruct MR images for said different time segments from MR data from which said marked image data have been eliminated, so as to compile a dynamic image data record in said dynamic region.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0019] FIG. 1 shows a cross-section through an examination region of a patient.

[0020] FIG. 2 is a schematic diagram for the continuous recording of radial k-space lines in the golden angle model.

[0021] FIG. 3 is a flowchart of an exemplary embodiment of the method according to the invention.

[0022] FIG. 4 schematically illustrates a magnetic resonance device according to the invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0023] The following section describes an exemplary embodiment of the present invention regarding the determination of a dynamic image data record of a patient, which is intended to show liver perfusion. In order to be able to map the liver perfusion as a dynamic procedure in the dynamic magnetic resonance imaging, the patient is administered with a contrast agent and the course of the contrast agent concentration in the liver of the patient is measured.

[0024] In this context, FIG. 1 shows, as an example, the examination region 1 that arises during the radial scanning used here, which is substantially the entire patient 2 (shown in a cross-section). For the dynamic procedure that is to be implemented, only the torso 3 with the liver 4 is visible and relevant; no dynamic procedure to be measured takes place in the arms 5, which lie in the outer regions in which the homogeneity of the magnetic fields of the magnetic resonance device is already limited. The examination region 1 therefore can be divided into a dynamic region 6 and a non-dynamic region 7, which here is formed by the outer regions that include the arms 5.

[0025] In the present case a GRASP technique is used for the continuous recording of magnetic resonance data, in order to measure the dynamic procedure. Radial scanning of k-space takes place, wherein a golden angle model is used, as indicated in FIG. 2. Here, the first radial k-space line 8 to be measured runs through k-space center 9, as can be seen under I. Subsequently, see II, a second radial k-space line 10 is measured, which is rotated by a golden angle 11, here 111.25, with respect to the radial k-space line 8. Accordingly, in turn, the third k-space line 12, see III, is measured with a rotation by the golden angle 11 from the direction of the second radial k-space line 10, and so on.

[0026] This results in an excellent uniform coverage of k-space even when extracting only a few of the radial k-space lines 8, 10, 12, as is necessary in order to map dynamic procedures with a higher temporal resolution.

[0027] This means that, in order to define the individual magnetic resonance images of the dynamic image data record, magnetic resonance data lying within a time segment with certain few k-space lines 8, 10, 12 are always singled out in order to reconstruct the corresponding magnetic resonance image. There is therefore an underscanning, in which streaking artifacts may arise. The causes thereof mostly lie, as has been recognized, outside the relevant dynamic region 6, i.e. in the non-dynamic region 7, and are therefore produced by the arms 5 in particular.

[0028] In the method according to the invention, image data are identified that are to be allocated to the non-dynamic region, so as to transform that data back into k-space, and there to eliminate that data from the magnetic resonance data. Therefore, when reconstruction of magnetic resonance images takes place, fewer streaking artifacts, even from few k-space lines 8, 10, 12 occur, because the causes thereof have for the most part been removed.

[0029] FIG. 3 shows a flowchart of an exemplary embodiment of the method according to the invention.

[0030] Here, in a step S1 during the dynamic procedure, i.e. the contrast agent throughflow in this case, magnetic resonance data are continuously recorded by radial scanning along k-space lines 8, 10, 12 rotated with respect to one another by the golden angle.

[0031] Following this, the reconstruction phase begins, wherein initially a reconstruction of a static image data record takes place in step S2, from all the recorded magnetic resonance data of all k-space lines 8, 10, 12.

[0032] There is therefore no underscanning, and streaking artifacts are improbable. In a step S3, image data of the non-dynamic region 7 are identified in the static image data record. Two approaches, which can be used individually or in combination, are possible. One possibility is to already deduce the position of the non-dynamic regions 7 from the positioning of saturation regions, in which a fat saturation technique or another saturation technique (local saturators) has been applied during the recording in step S1. Another possibility is to evaluate the static image data record so as to identify the location of anatomical features, for example the arms 5, in which the dynamic procedure does not take place (or only with a significant reduction).

[0033] In a step S4, the image data of the non-dynamic region 7 are inverse-transformed into k-space, wherein steps of a gridding reconstruction and corresponding filtering operations are implemented, in order to subsequently achieve a removal of the correction data, which has been obtained in this manner, from the magnetic resonance data by subtracting the correction data from the magnetic resonance data. This can take place in k-space at the actual radial scanning points, but also on a Cartesian grid (after a regridding).

[0034] In a step S5, the magnetic resonance images of the dynamic image data record are then determined by a reconstruction from the modified magnetic resonance data of step S4 taking place in the dynamic region 6. For the non-dynamic region 7, in which the dynamic procedure to be mapped does not take place in any case, use is made of the image data of the static image data record in order to achieve an orientation. Of course, it is also possible to restrict the magnetic resonance images to the dynamic region 6.

[0035] FIG. 4 shows a magnetic resonance apparatus 13 according to the invention. This has a scanner 14 with a basic field magnet 14 which defines a patient aperture 15, into which a patient can be introduced by a patient bed (not shown). A radio-frequency (RF) coil arrangement and a gradient coil arrangement (not shown) surround the patient aperture 15.

[0036] The operation of the magnetic resonance apparatus 13 is controlled by a control computer 16, which is configured to perform the method according to the invention. To this end, the control computer 16 have suitable reconstruction processors, a marking processor and a correction processor.

[0037] Although modifications and changes may be suggested by those skilled in the art, it is the intention of the Applicant to embody within the patent warranted hereon all changes and modifications as reasonably and properly come within the scope of the Applicant's contribution to the art.