Method and apparatus for testing the image reconstruction of an imaging X-ray system

Abstract

The invention relates to a method for testing the image reconstruction of an imaging X-ray system (1) and / or of algorithms for processing image recordings, comprising the steps: - Provision of digital source images (A) acquired using X-rays with at least two exposure energies, - Specifying at least two base materials (M1, M2) and a plurality of attenuation values ​​(W) for these base materials (M1, M2) for the absorption energies and for various other energies, - Calculating simulated image values ​​(Y) for at least some of the different energies from image values ​​(X) of the original images (A) and the attenuation values ​​(W) for the recording energies and for the energies in question, - Output of result images (P) with the simulated image values ​​(Y) for testing or validating an image reconstruction of the imaging X-ray system (1) and / or of algorithms for processing image recordings. Furthermore, the invention comprises a device, a control unit and an imaging X-ray system.

Description

[0001] The invention relates to a method and a device for testing the image reconstruction of an imaging X-ray system and / or algorithms for processing image recordings, a control device for an imaging X-ray system and an imaging X-ray system.

[0002] For testing medical imaging systems, such as CT scanners, or for verifying algorithms, such as mono-kV, virtual non-contrast, bone removal, etc., spectral data of an object are required. This spectral data can be obtained, for example, by taking two measurements with different X-ray spectra, ideally simultaneously. Alternatively, the data can be obtained through simulation, such as using Monte Carlo methods.

[0003] The problem here is that both recording and simulating this spectral data is comparatively complex. Recording also involves the additional cost of radiation exposure. It should also be noted that very large datasets are required for tests or verifications to provide the appropriate test data for each application. In particular, applications requiring data for different recording energies, e.g., kV spectra from 70 kV to 150 kV in 10 kV increments, necessitate a complete dataset for each individual energy level.

[0004] Up to now, large amounts of data have been stored or long testing times have been required on the systems to test algorithms, applications, workflows and parameter combinations.

[0005] It is an object of the present invention to provide a method and a device for testing the image reconstruction of an X-ray imaging system, a control device for an X-ray imaging system, and an X-ray imaging system itself, with which the disadvantages described above are avoided. In particular, it is an object of the invention to generate the necessary data in a simple manner using a small number of provided spectral data sets.

[0006] This problem is solved by a method according to claim 1, a device according to claim 10, a control device according to claim 11 and an imaging X-ray system according to claim 12.

[0007] A method according to the invention serves to test the image reconstruction of an X-ray imaging system and / or algorithms for processing image data. It comprises the following steps: - Provision of digital source images acquired using X-rays with at least two exposure energies, - Specifying at least two base materials and a plurality of attenuation values ​​for these base materials for the absorption energies and for various other energies, preferably wherein attenuation values ​​for the same energies are specified for each base material, - Calculating simulated image values ​​for given energies from image values ​​of the original images and the attenuation values ​​for the recording energies and for the energies in question, - Output of result images with the simulated image values ​​for testing or validating an image reconstruction of the imaging X-ray system and / or algorithms for processing image recordings.

[0008] When acquiring CT images, grayscale images are typically created. The gray value of a pixel or voxel is a measure of the absorption of the corresponding area in reality. For the calculation of the CT images, a CT scan typically involves acquiring a large number of projection images for the raw data. These images represent the attenuation of an X-ray beam along a line from the beam focus to the detector element. From these, a 3D volume is then calculated, with each voxel having a value that corresponds to the attenuation of the radiation within that voxel.

[0009] In the present case, where multiple CT images (at least two) are acquired simultaneously at different energies, their voxels can be assigned multiple values, each corresponding to the attenuation at that voxel at the different energies. With two acquisition energies, each voxel V can therefore have a first image value µ1 and a second image value µ2, which can be combined into a single vector. However, it is also possible to have two separate images representing the images acquired at the different energies.

[0010] The image value in each voxel depends on the concentration and properties (e.g., electron density) of the substances at that location. Since different substances exhibit different attenuation at different energies, substances with different absorption energies can be resolved or separated from each other.

[0011] This is state of the art and is described, for example, in "Optimal Selection of Base Materials for Accurate Dual-Energy Computed Tomography: Comparison Between the Alvarez-Macovski Method and Dira" by Maria Magnusson et al. If, for example, one assumes that two substances ("base materials") with concentrations c are present in a voxel of the body... A and c B If two predominant materials are, for example, water (which also represents soft tissue) and bone, then the image values ​​of a voxel with the attenuation A of one base material and the attenuation B of the other base material at the two energies (indicated by the indices 1 and 2) can be calculated using the following formula: (μ1μ2)=(A1B1A2B2)(cAcB)

[0012] If water is used as the second base material, then the images are often standardized in such a way that its absolute weakening B Wasser = µ Wis constant over all energies. Using the spectral data, the concentration of water and a second base material can then be determined for each volume element (RE Alvarez and A. Macovski, “Energy-selective reconstructions in x-ray computerised tomography,” Phys. Med. Biol. 21, 733 (1976)). The attenuation of the first base material can then also be expressed as a relative attenuation S = B / A. Formula (1) could then be rearranged as follows: (μ1μ2)=μW(1S11S2)(cAcW)

[0013] However, it should be noted that in patient images the image values ​​µ1 and µ2 are known and the concentrations c A and c B unknown. Rearranging the formula to solve for the concentrations yields: (cAcB)=1A1B2−A2B1(B2−B1−A2A1)(μ1μ2) or with relative weaknesses (cAcB)=1 / μW1S2−S1(S2−S1−11)(μ1μ2)

[0014] For other energies, formula (1) or (1a) would yield different image values ​​µ'1 and µ'2 from corresponding attenuations A' and B' or S' at the two energies (indices 1 and 2) for the base materials: (μ'1μ'2)=(A'1B'1A'2B'2)(cAcB) or with relative weaknesses (μ'1μ'2)=μW(1S'11S'2)(cAcB)

[0015] Although the densities are unknown, formula (2) can be substituted into formula (3). After matrix multiplication, we obtain: (μ'1μ'2)=1A1B2−A2B1(A'1B2−B'1A2B'1A1−A'1B1A'2B2−A2B'2B'2A1−A'2B1)(μ1μ2)

[0016] In formulas (2a) and (3a), the absolute weakening of water µ cancels out. W away, since this is defined as 1000 HU in the Hounsfield scale, independent of the energy spectrum in the CT images (CT number = 1000 (µ - µ)). W ) / (µ W - µ Luft The result is: (μ'1μ'2)=1S2−S1(S2−S'1S'1−S1S2−S'2S'2−S1)(μ1μ2)

[0017] The attenuation values ​​A, B, S and A', B', S' are known or at least determinable for the respective energies. Therefore, image values ​​µ1 and µ2 determined (by a dual-energy CT scan) can be used to determine image values ​​µ'1 and µ'2 for other energies. However, it is not always necessary to determine pairs of values ​​for new image values. If one of the two image values ​​is "held," e.g., µ2 (this image value is also used for µ'2), then the following results for µ'1: μ'1=(A'1B2−B'1A2)μ1+(B'1A1−A'1B1)μ2A1B2−A2B1 or with relative weaknesses μ'1=(S2−S'1)μ1+(S'1−S1)μ2S2−S1

[0018] Thus, with predefined attenuation values ​​for different energies and the image values ​​µ1 and µ2 determined by measurement, intermediate values ​​for a wide variety of energies can be calculated, which represent an estimate for these energies and can be used for checking CT evaluation programs or reconstruction methods.

[0019] Digital source images are provided, which can be 2D projection images or 3D CT images. Since the method can be advantageously used in radiography, CT, fluoroscopy, or tomosynthesis, source images corresponding to these procedures can be provided. These can be images of real objects or simulations of virtual objects. The following discussion will focus on CT as an example, but this does not imply that other imaging methods are not also suitable.

[0020] It is important that these initial images were acquired using X-rays with at least two exposure energies ("dual energy" or "multi-energy"). Ideally, the initial images should have been acquired at the same time, or at least with a stationary subject. They must depict the exact same scene; otherwise, the pixels will not represent the corresponding points of the subject. For stationary objects, the exposure times are essentially irrelevant, but for moving subjects, the initial images should have been acquired at the same time.

[0021] The source images can be several separate images showing the subject at different energies. Alternatively, they can be a single image where pixels are represented as vectors with different values ​​for different energies. Essentially, the only requirement is that the corresponding image coordinates of the source images include values ​​for at least two energies. More energies are possible, but not strictly necessary.

[0022] To perform the procedure, at least two base materials must be specified. Water is a suitable base material, as its values ​​are normalized to be constant in most CT images. For other imaging methods, the most suitable base materials can be selected. It is advisable not to specify more base materials than the number of acquisition energies used.

[0023] For each of the specified base materials, a number of attenuation values ​​are given. These values ​​indicate how X-rays passing through a comparable volume of the base material are attenuated. The attenuation values ​​are specified at least for the acquisition energies and for various other energies. These other energies should be those for which further image datasets are to be created. However, this is not strictly necessary, because attenuation values ​​can also be extrapolated, as described in more detail below. It is also preferred that attenuation values ​​for the same energies are specified for each base material, although this too is not strictly necessary (also for the reason stated above).

[0024] The attenuation values ​​are known in the prior art and can be in the form of a list for each base material, in the form of a table (with the columns base material and energy) or in the form of a function or graph (energy versus attenuation value) for each base material.

[0025] It is important here that for the energies for which images are to be created, attenuation values ​​for all specified (relevant) base materials are known or can at least be determined.

[0026] The digital images now provide the image values ​​µ1 and µ2 for each pixel at the two energies, as well as the attenuation values ​​for different energies. These attenuation values ​​can be absolute attenuation values ​​A, B, or relative attenuation values ​​S. These relative attenuation values ​​should be chosen relative to an absolute attenuation value of another base material. This absolute attenuation value should be constant for all energies so that it can be removed from the final calculation (see formulas 4a and 5a).

[0027] Simulated image values ​​for various other energies can now be calculated from the image values ​​of the original images and the attenuation values ​​for the acquisition energies and for the energies in question, particularly according to formulas 4 and / or 5 or 4a and / or 5a given above. For example, simulated image values ​​for energies between 70 keV and 150 keV can be calculated in regular or irregular energy steps. It is preferable to determine which energies the respective test or verification requires and specify these energies. The calculation is very simple, at least with the formulas given above, and can be performed quickly with an averagely powerful computer, so that no large amounts of data need to be stored. The original images and the attenuation values ​​are sufficient for this.

[0028] Result images are then generated with the simulated image values ​​to test the image reconstruction of the X-ray imaging system and / or algorithms for processing image acquisitions, e.g., algorithms for the virtual removal of contrast agent or bone from the image data. The result images can correspond to the form of the original images, e.g., 3D CT images, or be raw data, especially projection images, so that a reconstruction can be tested. The following cases are preferred: 1) The source images and the result images are both raw data. 2) The source images and the result images are both reconstructed data. 3) The source images are reconstructed data; intermediate images are generated from the simulated pixels, from which raw data is then generated as the result data.

[0029] The resulting images are then output, primarily for testing purposes or to validate an algorithm. In particular, the resulting images can serve as a basis for validating the results of a machine learning algorithm within a supervised training environment.

[0030] A device according to the invention serves to test the image reconstruction of an X-ray imaging system and / or algorithms for processing image data. It comprises the following components: - a data interface designed to receive digital output images acquired using X-rays with at least two exposure energies, - a storage unit designed to specify at least two base materials and a plurality of attenuation values ​​for these base materials for the absorption energies and for various other energies, - a calculation unit designed to calculate simulated image values ​​for given energies from image values ​​of the source images and the attenuation values ​​for the recording energies and for the energies in question, - a data interface designed to output result images with the simulated image values ​​for testing or validating an image reconstruction of the imaging X-ray system and / or algorithms for processing image recordings.

[0031] The function of the device's components has already been described. The device is preferably designed for carrying out a method according to the invention.

[0032] The invention is particularly advantageous in the field of computed tomography (CT), as it allows the simulation of spectral data for any energy. For this purpose, a volume dataset containing spectral information is preferably provided as the input image, e.g., acquired with a Naetom Alpha with two energy thresholds, whereby the energy-dependent attenuation of the photons is determined using an X-ray spectrum. The spectral data are therefore consistent for each volume element, since the spectral data were acquired simultaneously.

[0033] The resulting images can then be used to test and validate the product components CT reconstruction and CT applications. The data is preferably generated in a CT scanner, depending on the selected parameters (e.g., kV), or in a computing unit that simulates the CT scanner. The simulation then replaces scans and generates the measurement data from spectral image volumes. This invention now makes it possible to better simulate the kV dependence of the data. It should be noted that this can also be applied to all other X-ray devices, particularly in tomosynthesis, fluoroscopy, or radiography.

[0034] A significant advantage of the invention is that energy simulation is faster than more complex simulations using Monte Carlo methods or measurements on CT scanners. Furthermore, the data volume is very small, as essentially only one spectral dataset is required for all energy levels. A single spectral CT image volume can also be used to simulate images for conventional detectors, such as projection images for radiography applications. In addition, the resulting images offer better image quality than previous simulation methods and are therefore closer to actual measured data. Spectral algorithms can thus be tested easily and consistently (same object at exactly the same position) (e.g., bone removal, virtual non-contrast, etc.). The method according to the invention can be extended to any number of spectra. Furthermore, any concentration can be simulated for each voxel, for example, by...The concentration of the material changes over time, thus simulating the influx of a contrast agent.

[0035] A control device according to the invention for an imaging X-ray system comprises a device according to the invention. Alternatively or additionally, it is designed to carry out the method according to the invention.

[0036] An X-ray imaging system according to the invention comprises a control device according to the invention.

[0037] The invention can be implemented, in particular, in the form of a computer unit with suitable software. The computer unit can, for example, comprise one or more cooperating microprocessors or the like. In particular, it can be implemented in the form of suitable software program components within the computer unit. A largely software-based implementation has the advantage that even previously used computer units can be easily retrofitted by a software or firmware update to operate according to 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 computer unit, containing program sections to execute all steps of the method according to the invention when the program is run in the computer unit.In addition to the computer program itself, such a computer program product may include additional components such as documentation and / or additional components, including hardware components such as hardware keys (dongles, etc.) for using the software.

[0038] For transport to the computer unit and / or for storage on or in the computer unit, a computer-readable medium, such as a memory stick, a hard drive or other 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 computer unit are stored.

[0039] Further, particularly advantageous embodiments and developments of the invention result from the dependent claims and the following description, wherein the claims of one claim category may also be further developed analogously to the claims and description parts of another claim category and, in particular, individual features of different embodiments or variants may be combined to form new embodiments or variants.

[0040] It is preferred that in one embodiment of the method, the attenuation values ​​for one base material, preferably water, are absolute attenuation values, and for the other base material, they are relative attenuation values ​​to the absolute attenuation values. This is particularly advantageous for CT applications, since the absolute attenuation is typically constant over different energies, and thus the absolute attenuation of water cancels out of the formulas (see formulas 4a and 5a). It is particularly preferred that an absolute attenuation value for one base material µ A is and the absolute attenuation value for the other base material µ B is and the relative attenuation value for the other base material S = µ B / µ A or generally g(µ B ) / f(µ A) with the functions f and g. For example, f or g can represent a multiplication by a constant factor or be an energy-dependent function. For example, S = µ B / 1000µ A be.

[0041] According to a preferred embodiment of the method, the simulated image values ​​are calculated using relative attenuation values. This makes the calculation simple and fast. However, a calculation using absolute attenuation values ​​is also not particularly complex, as formulas 4 and 5 above demonstrate.

[0042] It is preferred that in one embodiment of the method the calculation of the simulated weakening values ​​for the base material is carried out according to one of the formulas of the group: (μ'1μ'2)=1A1B2−A2B1(A'1B2−B'1A2B'1A1−A'1B1A'2B2−A2B'2B'2A1−A'2B1)(μ1μ2), (μ'1μ'2)=1S2−S1(S2−S'1S'1−S1S2−S'2S'2−S1)(μ1μ2), μ'1=(A'1B2−B'1A2)μ1+(B'1A1−A'1B1)μ2A1B2−A2B1 or μ'1=(S2−S'1)μ1+(S'1−S1)μ2S2−S1, This is done using the formulas 4, 4a, 5, and 5a explained above. The factors are the image value µ and the simulated image value µ', the absolute attenuation values ​​A and B and the relative attenuation value S for the recording energies, and the absolute attenuation values ​​A' and B' and the relative attenuation value S' for the energies, where the index 1 represents the first energy and the index 2 represents the second energy.

[0043] Alternatively, a density map with the densities c can also be generated using the image values ​​µ1 and µ1. A and c B The simulated weakening values ​​for the base material are then preferably calculated using one of the formulas from the group: (μ'1μ'2)=(A'1B'1A'2B'2)(cAcB),(μ'1μ'2)=μW(1S'11S'2)(cAcB)and μ'1=(A'1cA(t)+B'1cB), which correspond to formulas 3 or 3a or can be derived from them.

[0044] According to a preferred method, the source images are 2D projection images (e.g., CT raw data) and corresponding 2D projection images are created as result images using the simulated image values.

[0045] According to a preferred method, the source images are 3D images (e.g. reconstructed CT images) and the simulated image values ​​are calculated in a three-dimensional space.

[0046] From the simulated image values, simulated 3D images corresponding to the original images are then generated as result images.

[0047] According to a preferred method, the source images are 3D images (e.g., reconstructed CT images), and the simulated image values ​​are calculated in a three-dimensional space. From these simulated image values, simulated 3D images corresponding to the source images are generated as intermediate images. From these intermediate images, 2D projection images are then created as final images. Their shape preferably corresponds to the shape of the raw data of the source images. However, data from other X-ray systems can also be simulated; for example, projection images for radiography can be generated from the intermediate images. The term "correspond" here means that the subject and, if applicable, the viewing angle are the same, and only the image values ​​vary according to the different energy.

[0048] It is preferred that, in one embodiment of the method, simulated image values ​​are calculated for several energies, preferably at equidistant energy intervals, particularly in steps between 1 kV and 30 kV. However, arbitrary irregular energy levels can also be selected.

[0049] It is preferred that in one embodiment of the method, the procedure is carried out for several time points at the same energies, and the concentration values ​​are varied according to a predefined schedule for different time points. This allows, for example, the simulation of the onset of a contrast agent.

[0050] This is preferably achieved by determining the concentration of the base materials (e.g., water and iodine) for each voxel of the measured data set in formula 1 (or 1a). For the simulation of the contrast agent influx, the concentration of one base material (e.g., iodine) is then changed in formula (3) or (3a) by replacing the time-constant value c with a different value. A a time-varying value c A (t) is used, i.e.: (μ'1(t)μ'2(t))=(A'1B'1A'2B'2)(cA(t)cB)or(μ'1(t)μ'2(t))=μW(1S'11S'2)=μW(1S'11S'2)(cA(t)cB)

[0051] Let c be, for example, A the concentration of iodine, then c A (t) the concentration of iodine depending on the time course of an influx of a contrast agent or a time-varying concentration of contrast agent. The concentration of the other base material c BIn this example, it remains constant, but it could also change in such a way that the sum of the two concentrations always remains constant. This then results in a time-varying µ'1(t) and µ'2(t) in formulas (3) and (3a), respectively. However, µ'1(t) can also be calculated for a single energy using the formula: μ'1(t)=(A'1cA(t)+B'1cB)

[0052] According to a preferred method, simulated CT raw data (i.e., a CT scan dataset) are generated using the simulated image values, and this raw data is then reconstructed into CT images by the X-ray imaging system in question as part of testing the image reconstruction. The reconstructed CT images are then examined. It is preferred that energies desired for image reconstruction are specified, and attenuation values ​​for the base materials are specified for these energies.

[0053] It is preferred that, in one embodiment of the method, if no attenuation values ​​are available for a desired energy for a number of base materials, these values ​​are calculated from attenuation values ​​for energetically adjacent energies. This is preferably done based on an averaging calculation or from a graph fitted to the specified attenuation values.

[0054] The method is preferably used for training AI-based methods (AI: "Artificial Intelligence"). Artificial intelligence is based on the principle of machine learning and is generally implemented using a learning algorithm that has been trained accordingly. The English term "machine learning" is frequently used for machine-based learning, and this also includes the principle of "deep learning." The method according to the invention is very well suited for creating a large number of datasets for verifying or validating results within the framework of supervised learning.

[0055] Preferably, components of the invention are provided as a "cloud service." Such a cloud service serves to process data, particularly using artificial intelligence, but can also be a service based on conventional algorithms or a service where human evaluation takes place in the background. Generally, a cloud service (hereinafter also referred to simply as "cloud") is an IT infrastructure in which, for example, storage space or computing power and / or application software is provided via a network. Communication between the user and the cloud takes place via data interfaces and / or data transmission protocols. In the present case, it is particularly preferred that the cloud service provides both computing power and application software.

[0056] In a preferred method, data obtained within the scope of the invention is provided to the cloud service via the network. This cloud service comprises a computing system that typically does not include the user's local computer. The method can be implemented using a command structure within a network. The data processed in the cloud is subsequently sent back to the user's local computer via the network.

[0057] The invention is explained in more detail below with reference to the accompanying figures and exemplary embodiments. The same components are designated with identical reference numerals in the various figures. The figures are generally not to scale. They show: Fig. 1 a rough schematic representation of a CT system with a device according to the invention, Fig. 2 a block diagram illustrating the process of a method according to the invention, Fig. 3 a preferred test setup, Fig. 4 a result in accordance with the state of the art, Fig. 5 a result according to the invention.

[0058] Fig. Figure 1 shows a computed tomography system (CT system 1) as an example of an X-ray system 1 with a radiation detector 4 and an X-ray source 5. The X-ray source 5 is configured to expose the radiation detector 4 with X-rays. The CT system 1 shown comprises a gantry 2 with a rotor 3. The rotor 3 includes the X-ray source 5 and the radiation detector 4, which is configured to detect X-rays.

[0059] The rotor 3 is rotatable about the axis of rotation 8. A patient 6 is positioned on the patient table L and can be moved along the axis of rotation 8 through the gantry 2. The processing unit 9 is provided for controlling the CT system 1 and / or for generating an image data set based on signals detected by the radiation detector 4.

[0060] Typically, a (raw) X-ray image dataset of the patient 6 is acquired from a variety of angular directions using the radiation detector 4. Subsequently, based on the (raw) X-ray image dataset, an image dataset (source images A) can be reconstructed using a mathematical procedure, for example, including a filtered backprojection or an iterative reconstruction method.

[0061] The processing unit 9 serves here as a control unit 9 for controlling the CT system 1. An input device 10 and an output device 11 are connected to this processing unit 9. The input device 10 and the output device 11 can, for example, enable interaction by a user or the display of a generated image data set.

[0062] The control unit 9 comprises a device 12 for testing the image reconstruction of an X-ray imaging system. The device 12 comprises a data interface 13, a storage unit 14, and a processing unit 15.

[0063] The data interface 13 is used to receive digital output images A acquired using X-rays with at least two exposure energies. These are, for example, 2D projection images or 3D CT images.

[0064] The storage unit 14 serves to specify at least two base materials M1, M2 and a plurality of attenuation values ​​W for these base materials M1, M2 for the absorption energies and for various other energies. These are stored in the storage unit 14 and can be retrieved from it.

[0065] The calculation unit 15 is used to calculate simulated image values ​​Y for given energies from image values ​​X of the original images A and the attenuation values ​​W for the recording energies and for the energies in question.

[0066] The data interface 13 is used to output result images P with the simulated image values ​​Y for testing or validating an image reconstruction of the imaging X-ray system 1 and / or algorithms for processing image recordings.

[0067] Fig. Figure 2 shows a method for testing the image reconstruction of an imaging X-ray system.

[0068] In step I, digital source images A, acquired using X-rays with at least two exposure energies, are provided. These are, for example, CT images.

[0069] In step II, at least two base materials M1 and M2, and a plurality of attenuation values ​​W for these base materials M1 and M2 for the absorption energies and for various other energies, are specified. For the sake of simplicity, water is chosen as one base material M1, and for the other base material M2, relative attenuation values ​​W to the absolute attenuation values ​​W of water are chosen.

[0070] In step III, simulated image values ​​Y for given energies are calculated from the image values ​​X of the original images A and the attenuation values ​​W for the acquisition energies and for the energies in question. The calculation is preferably performed using the relative attenuation values ​​W.

[0071] In step IV, result images P with the simulated image values ​​Y are output for testing or validating an image reconstruction of the imaging X-ray system 1 and / or of algorithms for processing image acquisitions. In this example, the initial images A are 3D images, and the simulated image values ​​Y are calculated as an intermediate image in three-dimensional space. From these intermediate images, 2D projection images P are then created as result images E, which serve as raw CT data.

[0072] This method is preferably used to calculate simulated image values ​​Y for several energies, preferably at equidistant energy intervals, in particular in steps between 1 kV and 30 kV.

[0073] Fig. Figure 3 shows a preferred test setup. The inventive method is essentially carried out in the simulator 16 at the top. This simulator 16 can, for example, be controlled by a control unit 9 as shown in Figure 3. Fig. Figure 1 shows a device 12 according to the invention. The raw data R generated by the simulator 16 from the initial images A are fed to the product 17 to be tested, which here comprises a reconstruction unit 18 and an evaluation unit 19. A reconstruction can then be tested there, whereby the result is already known within the framework of the simulation and can be used for verification. The evaluation for a wide variety of energies can also be validated, since the initial image A is known.

[0074] Product 17 can specify energies (left arrow pointing upwards) that are desired within the framework of image reconstruction, and attenuation values ​​W for the base materials M1, M2 can be specified for these energies for the process.

[0075] Fig. Figure 4 shows a result according to the state of the art. Here, the goal of the simulation is to test a dual-energy algorithm, "Monoenergetic," which, instead of the usual CT values ​​(attenuation for an entire spectrum), represents the attenuation values ​​for a single energy (here, 70 keV). These attenuation values ​​of the dual-energy algorithm "Monoenergetic" are independent of the image spectrum (e.g., independent of the scan energy of 120 kV or 140 kV). The left image shows the result of the "Monoenergetic" algorithm on the measured data set (here, 120 kV), while the middle image shows the result of the "Monoenergetic" algorithm on the data simulated for 140 kV. The errors caused by simulating the data for 140 kV are visible in the difference image. The state of the art here was a simple method that scales the CT values ​​above a certain threshold (e.g., the HU value of bone).

[0076] Fig.Figure 5 shows a result according to the invention. On the left is the result image of the "Monoenergetic" algorithm for 70 keV with a data set measured at 120 kV. In the middle is the result image of the "Monoenergetic" algorithm for 70 keV with a data set simulated for 140 kV (simulated from the data measured at 120 kV). On the right is the difference image, which shows significantly fewer differences than the previous method.

[0077] Finally, it should be noted once again that the invention described in detail above merely represents exemplary embodiments, which can be modified in various ways by a person skilled in the art without departing from the scope of the invention. Furthermore, 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, terms such as "unit" do not preclude the possibility that the components in question consist of several interacting sub-components, which may also be spatially distributed. The term "a number" should be read as "at least one." Regardless of the grammatical gender of a particular term, persons of male, female, or other gender identities are included.

Images