Method and apparatus for disposing artifacts in multi-energy computed tomography

Abstract

A computer-implemented method for automatically estimating and / or correcting errors caused by artifacts in a multi-energy computed tomography result dataset related to at least one target material, wherein at least two three-dimensional energy datasets of an imaging region of a patient are acquired using an x-ray imaging device with different x-ray energy spectra, and the result dataset is determined by evaluating the energy datasets using material-specific energy dependencies of x-ray attenuation, wherein the following steps are performed to estimate the error caused by the artifacts: - determining at least one first sub-region (2) of the imaging region, the at least one first sub-region being free of the target material and containing at least one second material, specifically exactly one second material, the at least one second material having a known material-specific energy dependency of x-ray attenuation, - for each first sub-region (2), comparing the image values of each voxel of the energy datasets by taking into account the known energy dependency to determine a deviation value indicating an artifact, - for at least a part of at least one remaining second sub-region (5) of the imaging region, calculating an estimated deviation value by interpolating and / or extrapolating from the determined deviation values in the first sub-region (2), wherein the estimated deviation value is used as the estimated error caused by the artifact.

Description

Technical Field

[0001] The present invention relates to a computer-implemented method for automatically estimating and / or correcting errors caused by artifacts in multi-energy computed tomography result data sets related to at least one target material, wherein at least two three-dimensional energy data sets of an imaging region of a patient are acquired using an x-ray imaging device (specifically, a computed tomography device) using different x-ray energy spectra, and the result data set is determined by evaluating the energy data sets using material-specific energy correlations of x-ray attenuation. The present invention also relates to an x-ray imaging device, a computer program, and an electronically readable storage medium. Background Art

[0002] Multi-energy computed tomography (multi-energy CT), sometimes also referred to as "spectral imaging" or "spectral CT", is a technique for energy-resolved x-ray imaging (specifically, in medicine). This imaging technique utilizes different energy correlations of x-ray attenuation, for example, in order to provide quantitative image data and / or reduce image artifacts through so-called material decomposition. Dual-energy CT (i.e., imaging at two energy levels) is one of the most widely used variants of this technique. However, due to the emergence of photon-counting x-ray detectors, the potential to measure at a larger number of energy spectra (specifically, simultaneously) has been provided.

[0003] Specifically, multi-energy imaging data (which can also be referred to as "spectral data") allows the calculation of material-specific images (commonly referred to as material maps) by, for example, applying a material decomposition algorithm to the acquired energy data sets, each energy data set being associated with a different energy spectrum and thus with a different energy level. In the case of tumor / lesion characterization, tumor response assessment, and / or other specific tasks in medicine, the ability to extract iodine images / iodine maps from multi-energy data and / or the ability to quantify the iodine distribution in tissue as a result data set opens up great potential.

[0004] In known approaches, energy data sets are acquired at different x-ray energy spectra (i.e., different energy levels). Then, the energy data sets are evaluated to determine the result data set, for example, the result data set being a material map and / or other values (specifically, quantitative values). In this evaluation, known material-specific energy correlations of x-ray attenuation for the material are used. For example, the x-ray attenuation of two materials (e.g., iodine and calcium) can differ not only in the energy spectra centered at 140 kV and the energy spectra centered at 80 kV, but each material can also have its own characteristic change pattern from 80 kV to 140 kV. Thus, the change in the measured attenuation values between 80 kV and 140 kV (and / or other energy spectra) will enable the distinction of such two materials (e.g., iodine and calcium).

[0005] However, for reliable diagnosis, the reading physician must be able to trust the quantitative information provided in the result dataset. On the other hand, similar to single-energy computed tomography images, the main-spectrum computed tomography image data (i.e., the energy dataset) may contain image artifacts caused by, for example, acquisition techniques, patient size, and / or patient motion. During the material decomposition process or other evaluation processes, the errors induced by such artifacts are amplified and propagated, such that incorrect image values, specifically, HU values or density values, may be determined for the result dataset. Currently, it is impossible for the reading physician and / or (generally) the user to identify whether the image values of the result dataset are correct or are affected by artifacts. If the user has doubts about the evaluation results, they will request additional scans and / or different imaging modalities. Summary of the Invention

[0006] An object of the present invention is to provide information about image artifacts in a result dataset and / or to provide a result dataset with improved quality regarding artifacts.

[0007] This object is achieved by providing a computer-implemented method, an x-ray imaging device, a computer program, and an electronically readable storage medium according to the independent claims. Advantageous embodiments are described in the dependent claims.

[0008] In a computer-implemented method for automatically estimating and / or correcting errors caused by image artifacts in a multi-energy computed tomography result dataset related to at least one target material, wherein at least two three-dimensional energy datasets of an imaging region of a patient are acquired using an x-ray imaging device (specifically, a computed tomography device) with different x-ray energy spectra, and the result dataset is determined by evaluating the energy datasets using material-specific energy correlations of x-ray attenuation, according to the present invention, the following steps are performed to estimate the error caused by artifacts:

[0009] - Determine at least one first sub-region of the imaging region that does not contain the target material and contains at least one (specifically, exactly one) second material having a known material-specific energy correlation of x-ray attenuation,

[0010] - For each first sub-region, compare the image values of each voxel of the energy datasets by considering the known energy correlation to determine a deviation value indicating an artifact,

[0011] - For at least a part of at least one remaining second sub-region of the imaging region, calculate an estimated deviation value by interpolating and / or extrapolating from the determined deviation values in the at least one first sub-region, where the estimated deviation value is used as the estimated error caused by the artifact.

[0012] In the course of the present invention, it has been found that most of the artifacts occurring in multi-energy computed tomography have low spatial frequencies. If such basic artifacts of the energy datasets are different, artifacts (so-called "difference artifacts" related to at least two energy datasets) derived when evaluating the differences between the energy values of different energy datasets are generated, and these derived artifacts also have low spatial frequencies and are highly correlated when evaluating the x-ray attenuation changes between different energy spectra. The present invention is concerned with these derived artifacts or difference artifacts in order to derive their artifact intensity.

[0013] Due to the low spatial frequencies, if an estimate of the artifact intensity can be determined for a large enough part of the imaging region, these artifact intensities described by the deviation values in the present invention can be extrapolated and / or interpolated such that conclusions can be drawn about the artifact intensity in other (second) sub-regions of the imaging region. Specifically, in medical imaging, regions without a target material (e.g., iodine or bone) are usually included in the imaging region and can be identified as the first sub-region. Specifically, the present invention proposes to detect and quantify image artifacts (specifically, image artifacts with low spatial frequencies) relevant for multi-energy evaluation by analyzing a first sub-region of the imaging region, which first sub-region does not contain the target material and / or other materials with strong energy correlations. Suitable second materials in the first sub-region can include, for example, air surrounding the patient and inside the patient's hollow organs, and other anatomical sub-regions can also be used, for example, anatomical sub-regions containing subcutaneous fat and / or muscle as the second material. In a specific example, in the absence of any artifacts, the energy values of air (i.e., the image values of the energy datasets) (specifically, HU values) should be the same for all energy datasets at different energy spectra / energy levels. Since the basic artifacts affect the energy datasets in different ways, a difference in the energy values indicates the presence of image artifacts. The higher the difference, the stronger the artifacts likely to be related to multi-energy evaluation. By evaluating any deviations found during the evaluation process (specifically, the material decomposition process) and the error propagation within the evaluation process, the error of the resulting dataset caused by such variations of the basic artifacts can be estimated and provided as additional information together with the result values (i.e., the image values of the resulting dataset) to increase the confidence in the result values. Additionally and / or alternatively, the derived information about the artifact intensity can be used (e.g., by using a correction algorithm) to apply a corresponding correction to the energy datasets such that the resulting dataset is also corrected and at least the error caused by the image artifacts is reduced.

[0014] That is, in a preferred embodiment of the present invention, before determining the result value of the result data set, at least one energy value in the energy value (that is, the image value of the energy data set) is corrected according to the estimated deviation value; and / or in addition to the result data set, an error data set is determined, and the error data set shows the spatially resolved deviation value and / or the error propagated through the evaluation process; and / or according to the deviation value and / or the error value, the result data set is modified. Specifically, the result data set is colored. For example, the deviation value can be provided as additional information by adding corresponding coloring to the result data set. For example, areas with higher artifact intensity are indicated in red, and areas with low artifact intensity are indicated in green, where different colors and shades can be used between them to provide high-resolution additional information. However, the error information can also be provided as an additional data set, specifically, an error data set showing the spatially resolved deviation value and / or the propagated error value. Of course, the deviation value can also be propagated through the evaluation process so that the error propagation effect is also covered by the resulting error value, and the resulting error value can also be provided as an error data set. However, preferably, the deviation value (specifically, the deviation value in the sub-region containing the target material) is used for the correction of the energy data set and thus for the correction of the result data set.

[0015] In an embodiment, at least one first sub-region is determined by dividing the energy data set at least with respect to at least one second material, where specifically, threshold segmentation and / or an anatomical atlas are used. Such techniques have been used in the prior art for other applications and thus do not need to be described in detail here.

[0016] As already explained, preferably, at least one sub-region containing a second material having at least substantially constant spectral behavior is used as at least one first sub-region. Specifically, the expected change in the energy value between the spectra used for the second material can be lower than a threshold. For example, the threshold can be selected as one-fifth of the pre-determined expected artifact-induced change. For example, in spiral CT, it has been observed that artifacts have an intensity of, for example, 50 HU to 80 HU, so that, for example, 65 HU can be selected as the basis for evaluation. On the other hand, for the commonly used x-ray spectra (e.g., 80 kV and 140 kV), the HU value only changes by about 10, which will be far lower than 20% of 65. In other words, preferably, a second material can be selected that does not show a measurable change in the HU value or, compared with the expected intensity of the image artifact, the change in its HU value is negligible.

[0017] However, it should be noted that a second material with a greater correlation to the energy spectrum change can also be selected. As long as this correlation is known, for example, at least one reference energy value can be calculated from the measured energy values in one of the energy datasets in the energy dataset using the known correlation. Other measured energy values from other energy datasets can be compared with the corresponding energy values to determine the deviation value.

[0018] However, it is preferred to use a second material whose energy value does not change with the energy spectrum or only changes in a negligible manner. For example, air, because the energy values of a pair of two energy datasets can be directly compared to determine the deviation value. For example, for air, the expected HU value for each energy spectrum is -1000 HU.

[0019] Therefore, in a specifically preferred embodiment, air can be used as the second material, and in addition, adipose tissue and / or muscle tissue can be used. In the case where air is the second material, at least one sub-region can include a region outside the patient and / or at least one hollow organ containing air, specifically, the trachea and / or the intestine and / or at least one lung. Considering the fact that air also exists in the patient's body, at least for some imaging regions, together with the air sub-region outside the patient, a first sub-region within the patient can additionally be selected, thereby providing additional sampling points for the deviation value and thus improving the quality of extrapolation and / or interpolation.

[0020] Regarding extrapolation and / or interpolation, the most recent nearest neighbor interpolation and / or inverse distance weighting and / or spline interpolation can be employed. Since the artifacts mainly discussed here occur at low spatial frequencies, the intensity also changes slowly, such that simple interpolation and / or extrapolation techniques already provide good results regarding the deviation value.

[0021] In a specific example, the energy dataset can be acquired by spiral computed tomography, where during the acquisition, the acquisition device including the x-ray source and the x-ray detector also moves in the z direction (along the rotation axis / along the longitudinal direction of the patient) relative to the patient / imaging region. Therefore, since structures whose shape or position changes along the z-axis will be in different positions for different projection images used in the reconstruction of the energy dataset, artifacts of such structures may appear. Specifically, increasing the pitch may worsen these artifacts. Such spiral artifacts generally have low spatial frequencies and can thus be adequately described by interpolating and / or extrapolating from certain sampling points.

[0022] Other examples of artifacts that can be detected and estimated and / or corrected using the method according to the present invention can for example include scatter artifacts and / or undersampling artifacts.

[0023] It should be noted that the present invention provides a very easy-to-implement estimation process that yields good results. Based on an easily distinguishable first sub-region (e.g., a sub-region containing air, where the deviation value can be easily determined by a simple comparison of energy values (i.e., the image values of the energy data set)), a simple extrapolation and / or interpolation approach allows the determination of the deviation values of at least one second sub-region (specifically, containing the target material). These deviation values are directly related to the artifact intensity at the corresponding positions. However, in some embodiments, the approach of the present invention can also form the basis for a more complex process.

[0024] For example, in an embodiment, interpolation and / or extrapolation can be performed by considering at least one expected characteristic of at least one of the at least one artifact (specifically, the expected spatial frequency or the expected spatial frequency range of the artifact). In some cases, for example, in the case of helical artifacts, certain acquisition parameters of the x-ray imaging device are directly related to the expected spatial frequency of the image artifact. For example, in helical CT, when a pitch of about 3 is used, the minimum and maximum values of the helical artifact typically follow each other every 6 cm. However, using such information as a constraint during interpolation and / or extrapolation can suppress and / or disregard other artifacts, such that such use of existing knowledge should preferably be extensive enough to cover all forms of expected artifacts.

[0025] In some cases, interpolation and / or extrapolation can include the use of artificial intelligence algorithms. Such artificial intelligence algorithms (which can also be referred to as trained functions) can be trained using ground truth information about the artifacts in the training data set (e.g., by adopting a deep learning approach). The artificial intelligence algorithm can, for example, include a neural network. However, it is difficult to determine the ground truth information about the artifacts, specifically, for image artifacts with low spatial frequencies. Specifically, the patient has to be imaged at least twice, thereby increasing the radiation dose received by the patient, which would be undesirable. In addition, using artificial intelligence would require more computing power, while the simpler approach explained above has already yielded good results.

[0026] In this regard, it should be noted that in the case of dual-energy computed tomography, only two energy data sets can be calculated, and thus one deviation value can be calculated. However, in the case of more than two energy data sets, the deviation values for each pair of energy data sets can be determined. In one embodiment, a reference energy data set can be determined, and the deviation values of all other energy data sets with respect to the reference energy data set can be determined, and the deviation values can be used for the corresponding relative correction. However, in other approaches, the deviation values for each pair can be calculated, and the deviation values can be propagated through error propagation to calculate the error value with respect to the resulting data set. Further, statistical methods can be employed to calculate the total error value and / or determine the total correction.

[0027] The method of the present invention can be employed, for example, with respect to iodine measurement. That is, iodine can be used as the target material, where, for example, an iodine material map can be determined as at least one result data set among at least one result data set. The iodine material map can be quantitative or qualitative.

[0028] The present invention also relates to an x-ray imaging device, specifically, a computed tomography device. The x-ray imaging device includes at least one acquisition device and a control device. The acquisition device includes an x-ray source and an x-ray detector. The control device is configured to execute the method according to the present invention. All features and discussions regarding the method of the present invention similarly apply to the x-ray imaging device according to the present invention, such that the same advantages are achieved. In this way, additional information and / or correction can be directly provided at the x-ray imaging device.

[0029] For example, a computer program according to the present invention can be directly loaded into the storage device of the control device of the x-ray imaging device and includes program means for executing the steps of the method according to the present invention when the computer program is executed in the control device of the x-ray imaging device. The computer program according to the present invention can be stored on an electronically readable storage medium according to the present invention. The electronically readable storage medium thus includes electronically readable control information stored thereon, where the control information includes at least one computer program according to the present invention and is configured such that: if the electronically readable storage medium is used in the control device of the x-ray imaging device, the method according to the present invention is executed.

[0030] In summary, the present invention allows obtaining the level of image artifacts in a spectral image by specifically comparing a first sub-region with invariant energy spectra in an energy data set (e.g., a first sub-region of air or fat). Specifically, when performing a quantitative evaluation, in addition to the result value, the user also receives information about the expected error caused by these artifacts. Since the accuracy of the result value is known to the user (specifically, a physician), this allows for improved diagnosis. The quantitative information about the level of image artifacts in different sub-regions allows for corresponding correction of these image artifacts in the energy data set and / or directly in the result data set.

[0031] By calculating the deviation value, the present invention focuses on information related to the multi-energy CT evaluation process, that is, the "differential artifact" related to evaluating the change from one energy spectrum to another. Such an artifact is the result of a basic artifact having different intensities in at least two energy data sets among at least two energy data sets. Therefore, the artifact intensity described by the deviation value is understood to be related to the difference between the energy data sets, which is exactly the information used in multi-energy CT evaluation (e.g., material decomposition). BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Other objects and features of the present invention will become apparent from the following detailed description considered in conjunction with the accompanying drawings. However, the drawings are only schematic diagrams designed for illustrative purposes and do not limit the present invention. The drawings show:

[0033] Figure 1 is a flowchart of an embodiment of a method according to the present invention;

[0034] Figure 2 is a schematic diagram of an energy image in an energy data set;

[0035] Figure 3 is an exemplary view of a result image in a result data set;

[0036] Figure 4 is a computed tomography apparatus according to the present invention; and

[0037] Figure 5 is Figure 4 the functional structure of a control device of the computed tomography apparatus. Detailed Description

[0038] Figure 1 is a flowchart of an embodiment of a method according to the present invention. In this embodiment, dual-energy computed tomography of a patient is to be performed to determine the quantitative distribution of iodine, for example, for tumor or lesion characterization. Specifically, iodine is the target material to be examined.

[0039] In step S1, two energy data sets are acquired at different x-ray energy spectra. For example, one spectrum in the spectrum can be centered at about 80 kV (low energy), and the other spectrum can be centered at 140 kV (high energy). To acquire a three-dimensional energy data set, a computed tomography apparatus is used such that a helical trajectory of the x-ray source is achieved. In other words, helical CT is employed, where the patient table carrying the patient moves through the gantry during the rotation of the acquisition assembly. However, helical CT may cause certain image artifacts, such as helical artifacts. These artifacts may vary between the energy data sets, resulting in distortion of the effects induced by different x-ray attenuations for different energies.

[0040] The method described herein allows for the estimation of the intensity of such helical artifacts associated with dual-energy CT evaluation, as well as the intensity of other artifacts of low spatial frequency (such as scatter artifacts) that cause different effects in different energy data sets.

[0041] In step S1, as is known, projection images of the imaging region of the patient are acquired, and three-dimensional energy data sets are reconstructed for each energy spectrum.

[0042] In step S2, at least one first sub-region of the imaging region is determined, which at least one first sub-region does not contain the target material (in this case iodine) and contains exactly one second material (in this case air), and the exactly one second material has a known material-specific energy correlation of x-ray attenuation (in this case, invariant), which means that the HU value of air should be the same for each energy data set. A segmentation algorithm (e.g., threshold-based segmentation) can be used, specifically in combination with information from an anatomical atlas, to determine the first sub-region. Of course, other approaches known in the art can also be employed.

[0043] Figure 2 An exemplary two-dimensional energy image of one of the energy data sets in the energy data set is schematically shown, e.g., an MPR image. In this case, the energy image 1 shows a part of three first sub-regions 2 containing air, i.e., the regions outside the patient (left and right first sub-regions 2) and the region including the trachea 3 and the lungs 4.

[0044] For these first sub-regions containing air as the second material, since the attenuation behavior of air is invariant with respect to different energy spectra, for each voxel, the energy values (i.e., the image values of the energy data set) of each data set are compared to determine the deviation value. Since the energy value of air should be the same in each energy data set of the energy data set in the absence of artifacts, these deviation values indicate the presence of artifacts relevant to the evaluation. For example, the deviation value can be determined by calculating the difference between the energy value of a voxel in the low-energy data set and the energy value of the voxel in the high-energy data set.

[0045] This comparison and determination of the deviation value occur in step S3 (see Figure 1 ). According to Figure 2 , it can be seen that the deviation values over the entire range of the energy data set are now known, such that there are a large number of sampling points even within the patient. This allows, in step S4, interpolation and / or extrapolation from the deviation values determined in step S3 to calculate the estimated deviation value of the second sub-region 5 in step S4. As shown in Figure 2 , the second sub-region 5 can be a region of interest (ROI) where iodine is expected to be present and the iodine concentration is to be determined. However, all the remaining parts of the imaging region can also be covered by at least one second sub-region 5.

[0046] The process of interpolation and / or extrapolation is indicated by the arrow 6 in Figure 2 .

[0047] The determined estimated deviation values can be used to correct the energy data set. However, they can also be used as error values indicating the intensity of image artifacts or as starting points for error propagation, as will be further discussed below.

[0048] In step S5, as is known in the prior art, an (optionally corrected) energy data set is used to derive a result data set, which in this embodiment is a material map of iodine as the target material and / or a quantitative distribution of iodine. A material decomposition approach can be employed. In parallel with this evaluation, an error propagation process may be performed to propagate the error described by the estimated deviation values in the second sub-region 5 through the evaluation process to determine at least one result data set.

[0049] In any case, in step S6, the voxel-by-voxel error values derived from the error propagation process (specifically, the voxel-by-voxel error values provided as an error data set) or the estimated deviation values as error values in other variants can be used to modify the result data set for display. For example, as Figure 3 shown, the result image 7 of the result data set can use color coding within the second sub-region 5 (as indicated for Figure 3 regions 8, 9) to indicate the reliability of the determined iodine distribution. For example, green can indicate high reliability, while red can indicate low reliability, that is, large error values.

[0050] Of course, other ways can also be used to display the estimated artifact errors. In a variant, it may also be useful to still display information about the estimated errors even when the energy data set has been corrected.

[0051] Figure 4 A schematic diagram of a computed tomography device 10 according to the present invention is shown. The computed tomography device 10 includes a gantry 11 in which an acquisition device 12 is rotatably mounted. The acquisition device 12 includes an x-ray source 13 and an x-ray detector 14. A patient table 15 can be used to position a patient inside the opening 16 of the gantry 11 so that an imaging region can be imaged. Using spiral CT, specifically, by continuously moving the table 15 during the acquisition of projection images, a larger imaging region can be imaged.

[0052] The computed tomography device 10 further includes a control device 17 for controlling the computed tomography device 10, and the control device 17 is also configured to perform the steps of the method according to the present invention.

[0053] Figure 5 The functional structure of the control device 17 is shown. In principle, as is known from the prior art, the control device 17 includes an acquisition unit 18 for acquiring an image data set, specifically, also acquiring an energy data set for multi-energy CT (spectral CT). That is, the acquisition unit 18 is configured to perform S1.

[0054] In the determination unit 19, as described with respect to step S2, the first sub-region 2 can be determined. The comparison unit 20 compares the energy values of the energy data set according to step S3 to determine the deviation values. These deviation values can be used by the calculation unit 21 to: determine the estimated deviation value by interpolating and / or extrapolating from the determined deviation values in at least one second sub-region 5. In some embodiments, the calculation unit 21 may also be adapted to correct the energy data set according to the estimated deviation value.

[0055] In the evaluation unit 22, as discussed with respect to step S5, the result data set can be determined. In some embodiments, the evaluation unit 22 is also configured to perform an error propagation process to determine an error data set with the propagated error values.

[0056] This error data set or the error data set including the estimated deviation value can be used in the display unit 23 to modify the result data set for display, or to display the error estimation information in the error data set in addition to the result data set.

[0057] It should be noted that the functional units described above can be implemented by software and / or hardware in the processor of the control device 17, which also includes a storage device 24. Specifically, the functional units can be implemented in the form of a computer program according to the present invention.

[0058] Although the present invention has been described in detail with reference to the preferred embodiments, the present invention is not limited to the disclosed examples, and those skilled in the art can derive other variations from the disclosed examples without departing from the scope of the present invention.

Claims

1. A computer-implemented method for automatically estimating and / or correcting errors caused by artifacts in a multi-energy computed tomography result dataset related to at least one target material, wherein at least two three-dimensional energy datasets of an imaging region of a patient are acquired using an x-ray imaging device with different x-ray energy spectra, and the result dataset is determined by evaluating the energy datasets using material-specific energy correlations of x-ray attenuation, characterized in that, The following steps are performed to estimate the error caused by artifacts: - Determine at least one first sub-region (2) of the imaging region, the at least one first sub-region (2) being free of the target material and containing at least one second material having a known material-specific energy correlation of x-ray attenuation, - For each first sub-region (2), compare the image values of each voxel of the energy data set by taking into account the known material-specific energy correlation to determine a deviation value indicative of an artifact, - For at least a part of at least one remaining second sub-region (5) of the imaging region, calculate an estimated deviation value by interpolation and / or extrapolation from the determined deviation values in the first sub-region (2), wherein the estimated deviation value is used as an estimated error caused by the artifact.

2. The method according to claim 1, wherein Before determining the result value of the result data set, at least one energy value among the energy values is corrected according to the estimated deviation value; and / or in addition to the result data set, an error data set is determined, the error data set showing spatially resolved deviation values and / or the error propagated by evaluating the energy data set using the material-specific energy correlation of x-ray attenuation; and / or the result data set is modified according to the deviation value.

3. The method according to claim 1 or 2, characterized in that, The first sub-region (2) is determined by segmenting the energy data set at least with respect to the at least one second material.

4. The method according to claim 1 or 2, characterized in that, At least one sub-region (2) containing a second material having an energy spectrum invariant behavior is used as the at least one first sub-region (2).

5. The method according to claim 1 or 2, characterized in that, Air and / or adipose tissue is used as the second material.

6. The method according to claim 5, wherein In the case where air is the second material, the at least one first sub-region (2) includes a region outside the patient and / or at least one hollow organ containing air.

7. The method according to any one of claims 1, 2, and 6, characterized in that, For interpolation and / or extrapolation, nearest neighbor interpolation and / or inverse distance weighting and / or spline interpolation are employed.

8. The method according to any one of claims 1, 2, and 6, characterized in that The energy data set is acquired by helical computed tomography, wherein the artifact includes a helical artifact.

9. The method according to any one of claims 1, 2, and 6, characterized in that, The interpolation and / or the extrapolation are performed by taking into account at least one expected characteristic of at least one of the artifacts.

10. The method according to any one of claims 1, 2 and 6, characterized in that Iodine is used as the target material, and / or at least one material map is determined as the result data set.

11. The method according to claim 1, wherein the at least one second material is exactly one second material.

12. The method according to claim 2, wherein the result data set is colored according to the deviation value.

13. The method according to claim 1 or 2, wherein the first sub-region (2) is determined by segmenting the energy data set at least with respect to the at least one second material, wherein threshold segmentation and / or an anatomical atlas are used.

14. The method according to claim 4, wherein the expected change in the image values between the energy spectra used for the second material is below a threshold.

15. The method according to claim 4, wherein the expected change in the image values between the energy spectra used for the second material is less than one fifth of a pre-determined expected artifact-induced change.

16. The method according to claim 6, wherein the at least one hollow organ comprising air comprises the trachea (3) and / or the intestine and / or at least one lung (4).

17. The method according to claim 9, wherein the at least one expected characteristic of the at least one of the artifacts is the expected spatial frequency or the expected spatial frequency range of the artifact.

18. An x-ray imaging device comprising at least one acquisition device (12) and a control device (17), the at least one acquisition device (12) comprising an x-ray source (13) and an x-ray detector (14), the control device (17) being configured to perform the method according to one of the preceding claims.

19. The x-ray imaging device according to claim 18, wherein the x-ray imaging device is a computed tomography device (10).

20. A computer program which, when executed on a control device (17) of an x-ray imaging device, performs the steps of the method according to any one of claims 1 to 17.

21. An electronically readable storage medium on which is stored the computer program according to claim 20.

Images