Image quality of a magnetic resonance image dataset

1. A method for improving an image quality of a three-dimensional magnetic resonance image dataset recorded with a magnetic resonance device, the method comprising:

deriving material parameters locally-resolved for a magnetic resonance device from at least one correction image dataset recorded with a modality other than the magnetic resonance device, registered with the magnetic resonance image dataset, showing at least partly a same recording region as the magnetic resonance image dataset;

establishing a virtual magnetic resonance comparison dataset in a simulation using the material parameters;

determining, as a function of a comparison between the magnetic resonance image dataset and the virtual magnetic resonance comparison dataset, at least one measure parameter describing an image quality improvement measure to be applied in a k-space; and

applying the image quality improvement measure with the at least one measure parameter relating to the magnetic resonance image dataset.

2. The method of claim 1 , wherein the at least one correction image dataset is an x-ray image dataset.

3. The method of claim 1 , wherein the correction image dataset is recorded with a same image recording device as the magnetic resonance image dataset.

4. The method of claim 1 , wherein, to establish the material parameters, a material or material compound is established for each image element of the correction image dataset, and/or

one or more of proton densities, at least one relaxation constant, or spin characteristics are used as the material parameters.

5. The method of claim 1 , wherein a magnetization underlying the simulation is established from the material parameters using Bloch equations.

6. The method of claim 1 , wherein at least one recording parameter of the magnetic resonance image dataset, at least one operating parameter of the magnetic resonance device, or the at least one recording parameter and the at least one operating parameter are taken into account as part of the simulation, and/or

wherein idealized recording conditions are used to establish the magnetic resonance comparison dataset.

7. The method of claim 1 , wherein a k-space having different characteristics to the k-space of the magnetic resonance image dataset is used for the magnetic resonance comparison dataset.

8. The method of claim 7 , wherein the k-space having the different characteristics is an expanded k-space having another Nyquist frequency.

9. The method of claim 1 , wherein the measure parameters are established iteratively in an optimization method.

10. The method of claim 9 , wherein the optimization method is related to a minimization of a difference between the magnetic resonance image dataset and the magnetic resonance comparison dataset.

11. The method of claim 9 , wherein a measure parameter describing a frequency band containing noise is established within the optimization method, wherein a lowpass filter related to the frequency band is applied as the image quality improvement measure.

12. The method of claim 1 , wherein, when the magnetic resonance comparison dataset includes points in the k-space not contained in the magnetic resonance image dataset, the magnetic resonance image dataset is supplemented as the image quality improvement measure by at least a part of these points, and

wherein the measure parameters describe k-space areas in which the supplementing is to be done, and the comparison is carried out in a local space.

13. The method of claim 12 , wherein, for frequency ranges above a Nyquist frequency of the magnetic resonance image dataset not covered by the supplementation, portions of the magnetic resonance comparison dataset to be added are multiplied beforehand by an attenuation function.

14. The method of claim 1 , wherein magnetic resonance image data for which there is a deviation of the magnetic resonance image dataset from the magnetic resonance comparison dataset fulfilling a deviation criterion is identified as part of the comparison carried out in the k-space,

wherein the identified magnetic resonance image data is corrected as a function of the corresponding magnetic resonance comparison data.

15. The method of claim 14 , wherein a measure parameter describing an interpolation kernel allowing a calculation of replacement data for the identified magnetic resonance image data from the magnetic resonance image data adjacent to the identified magnetic resonance image data is established, and

wherein, as the image quality improvement measure, the replacement data replaces the identified magnetic resonance image data.

16. The method of claim 15 , wherein the measure parameter describing the interpolation kernel is established in an optimization method.

17. The method of claim 1 , wherein, in the comparison carried out in the k-space, a measure parameter defining a convolution kernel describing a distortion in the k-space is established, and

wherein, as the image quality improvement measure, the convolution kernel is applied to the magnetic resonance image dataset.

18. The method of claim 1 , wherein the at least one measure parameter is only established and the image quality improvement measures are only carried out on a target portion of the magnetic resonance image dataset able to be selected by a user, determined automatically, or selected by the user and determined automatically.

19. The method of claim 1 , wherein a two-dimensional projection image dataset is used as the correction image dataset,

wherein, for carrying out the comparison, an integration comparison dataset integrated into projection directions according to a recording geometry of the projection image dataset is established as the magnetic resonance comparison dataset and is compared with an integration image dataset established from the magnetic resonance image dataset.

20. A computing device comprising:

a control device of a combination-imaging device for magnetic resonance imaging and for at least one further modality,

wherein the computing device is configured to:

derive material parameters locally-resolved for a magnetic resonance device from at least one correction image dataset recorded with the at least one further modality, registered with a magnetic resonance image dataset, showing at least partly a same recording region as the magnetic resonance image dataset;

establish a virtual magnetic resonance comparison dataset in a simulation using the material parameters;

determining, as a function of a comparison between the magnetic resonance image dataset and the virtual magnetic resonance comparison dataset, at least one measure parameter describing an image quality improvement measure to be applied in a k-space; and

applying the image quality improvement measure with the at least one measure parameter relating to the magnetic resonance image dataset.

21. A non-transitory computer readable medium storing a computer program, with at least one processor, when executed on a computing device, cause the computing device to at least perform:

derive material parameters locally-resolved for a magnetic resonance device from at least one correction image dataset with at least one further modality, registered with a magnetic resonance image dataset, showing at least partly a same recording region as the magnetic resonance image dataset;

establish a virtual magnetic resonance comparison dataset in a simulation using the material parameters;

determining, as a function of a comparison between the magnetic resonance image dataset and the virtual magnetic resonance comparison dataset at least one measure parameter describing an image quality improvement measure to be applied in a k-space; and

applying the image quality improvement measure with the at least one measure parameter relating to the magnetic resonance image dataset.