Method and apparatus for providing a perfusion image data set of a patient

Abstract

The invention relates to a method of providing a perfusion image data set, comprising: detecting a first image data set representing a first x-ray attenuation distribution of a patient corresponding to a first x-ray quantum energy distribution and a second image data set representing a second x-ray attenuation distribution corresponding to a second x-ray quantum energy distribution, the first and second image data sets being recorded under administration of a contrast agent, or detecting a first and a second image data set of a patient, the first image data set being recorded under administration of a contrast agent and representing a first x-ray attenuation distribution with contrast agent, the second image data set being recorded under non-administration of a contrast agent and representing a first x-ray attenuation distribution without contrast agent; determining a contrast agent image data set and a non-contrast image data set based on a basis material decomposition of the first and second image data sets; calculating a perfusion image data set based on a ratio of image values of the contrast agent image data set and the non-contrast image data set corresponding in location; providing the perfusion image data set.

Description

Technical Field

[0001] This invention relates to a method and related apparatus for providing a patient's perfusion image dataset based on a first detected image dataset and a second detected image dataset. Furthermore, this invention relates to a medical imaging device, a computer program product, and a computer-readable storage medium. Background Technology

[0002] With the aid of modern imaging methods, two-dimensional or three-dimensional image data is typically generated, which can be used for visualization of the imaged object and other applications. Imaging methods are generally based on the acquisition of X-rays, in which so-called projective measurement data is generated. For example, projective measurement data can be acquired using a computed tomography (CT) scanner. In a CT system, a combination of an X-ray source arranged at a gantry and opposing X-ray detectors typically rotates around a measurement chamber containing the object being examined (hereinafter referred to as the patient, but not limited to this). The center of rotation (also called the "isocenter") coincides with the so-called system axis (also called the z-axis) extending in the z-direction. During one or more rotations, the patient is irradiated with X-ray radiation from the X-ray source, in which image datasets are acquired in the form of projective measurement data or X-ray projection data using opposing X-ray detectors. Based on the projective measurement data, an image dataset for a spatial representation of the patient in image space can be generated using appropriate reconstruction algorithms. The reconstruction or reconstruction algorithm can be any image reconstruction algorithm known in the field of expertise, such as weighted filtered back projection (WFBP), as it is commonly used in computed tomography. Alternatively, reconstruction algorithms are also feasible, applied within the scope of professional skill.

[0003] In spectral computed tomography and other radiographic methods, such as by using different tube voltages or energy-resolved detectors, multiple images of the same object volume can be reconstructed, differing in X-ray attenuation due to the different X-ray spectra (i.e., X-ray quantum energy distributions) recorded by the detectors due to the varying X-ray attenuation caused by the existing materials. Based on this, at least two materials can be identified, for example, through material decomposition. With spectral computed tomography, functional information can be obtained in addition to vascular morphology information. An example of this is perfusion imaging, which allows measurement of blood flow, for example, in a tissue region. A contrast agent (typically iodine) is injected into the patient. Then, partial information, such as the distribution of the contrast agent within the patient's body, can be calculated through material decomposition. Such images provide information about the local iodine concentration in the tissue (iodine per unit volume, e.g., mg / ml or mg / cm³). 3 From this information, the bleeding flow can be inferred.

[0004] One application of perfusion imaging is, for example, imaging of lung parenchymal perfusion. With each inhalation and exhalation, fresh air is carried across the blood-gas barrier in the lungs, reaching the alveoli. Gas exchange then occurs at the alveolar plane, resulting in the oxidation of the blood and the release of CO2. Two crucial factors here are adequate ventilation (the supply of fresh air) and adequate perfusion (the blood supply to the organ). Disorders of lung parenchymal perfusion can lead to life-threatening consequences. For example, pulmonary embolism can cause perfusion defects in the lung parenchyma, resulting from the closure of blood vessels supplying the lungs.

[0005] In diagnosis, contrast-assisted CT angiography (CTA) is often performed to detect thromboembolism in high-risk cases where computed tomography (CT) is available. However, CTA only provides a morphological representation indicating the location of the thrombus (the contrast agent abruptly ends in the vessel), not a functional representation such as a perfusion map. Therefore, small emboli may not be identified in CTA due to dissolution, but this could lead to larger perfusion defects.

[0006] As mentioned above, spectral computed tomography (CT) can obtain functional information in addition to vascular morphology, thus providing direct information about perfusion defects. Contrast maps generated from material breakdown provide information about local contrast agent concentrations in lung tissue (e.g., iodine per unit volume, such as mg / ml or mg / cm³). 3 This information allows for the inference of perfusion volume relationships within the lung parenchyma, hence it is also known as pulmonary perfusion volume (PBV). Compared to CTA, it provides not only an indication of the potential causes of perfusion defects but also an indication of the perfusion defects themselves. Alternatively, similar images can be obtained during / after contrast agent administration and the subtraction of the two CT scans, using an additional CT scan prior to contrast agent administration and the CT scan itself.

[0007] However, classic imaging using contrast agent maps to show local contrast agent concentrations neglects the morphological characteristics of the observed tissue. These morphological characteristics can vary across a range of physical and physiological influences and cannot always provide information as to whether a higher contrast agent concentration, as shown in the figure, is actually attributable to higher or lower tissue perfusion or to changes in tissue morphology relative to its normal state. This is particularly difficult in cases where tissue morphology may change over time, even under normal conditions, such as in lung parenchyma where only inhalation or exhalation occurs. Summary of the Invention

[0008] Therefore, the object of the present invention is to provide an improved perfusion image dataset as follows.

[0009] This objective is achieved by the method and apparatus according to the invention. Advantageous and inventive design solutions are also the subject of the following description.

[0010] The present invention relates to a method for providing a patient perfusion image dataset, comprising a detection step, a determination step, a calculation step, and a provision step.

[0011] The detection steps include detecting a first image dataset and at least one second image dataset by means of a first interface, the first image dataset representing a first X-ray attenuation distribution of the patient corresponding to a first X-ray quantum energy distribution, and the second image dataset representing at least one patient second X-ray attenuation distribution of the patient corresponding to at least one second X-ray quantum energy distribution, wherein the first image dataset and the second image dataset are recorded in the presence of a contrast agent.

[0012] Alternatively, the detection step includes detecting a first image dataset of the patient and a second image dataset of the patient, wherein the first image dataset is recorded with contrast agent applied and represents the patient's first X-ray attenuation distribution with contrast agent applied, and wherein the second image dataset is recorded without contrast agent applied and represents the patient's first X-ray attenuation distribution without contrast agent applied.

[0013] The image dataset can be generated using the medical imaging device according to the invention, as will be further described below, and is detected in the detection step by means of the first interface for use in other method steps according to the invention. The image dataset can also be stored, for example, on a computer-readable storage medium, or on a network or server, and is read in the detection step by means of the first interface and detected for use in other method steps according to the invention.

[0014] The first and second image datasets can, in particular, be computed tomography (CT) image datasets, determined, i.e., reconstructed, based on multiple projection datasets recorded from different projection angles. The first and second image datasets can be computed tomography (CT) image datasets determined by a computed tomography (CT) device. However, the first and second image datasets can also be determined by other radiographic methods, such as by a C-arm X-ray device. The result of such radiographic methods (e.g., computed tomography) is the attenuation of the X-ray beam along its path from the X-ray source to the X-ray detector. This X-ray attenuation is caused by the irradiating medium or material along the ray path. Attenuation is generally defined as the logarithm of the ratio of the attenuated radiation intensity to the main radiation intensity, and is called the material attenuation coefficient relative to the path normal. In some radiographic imaging applications, particularly in computed tomography applications, a value of the attenuation coefficient normalized to water—that is, a CT value expressed in Hornsfield units (HU)—is used instead of the attenuation coefficient itself to represent the attenuation distribution of X-rays in the examined object. This is calculated in a well-known manner based on the attenuation coefficient currently determined by measurement and a reference attenuation coefficient for water. The first or second image dataset can in particular reflect the spatial distribution of attenuation coefficients or CT values, i.e., the attenuation distribution of the patient's X-rays.

[0015] A first image dataset and at least one second image dataset, or in alternative embodiments only the first image dataset, are generated, particularly in the case of contrast agent administration, thus equivalent to a contrast agent-enhanced image dataset. The contrast agent-enhanced image dataset is particularly based on the fact that a contrast agent, e.g., intravenous injection, is administered to the patient prior to the measurement data on which the image dataset is based, so as to produce contrast enhancement in the acquired image dataset. In conjunction with the invention, a contrast agent is understood as any medium that, upon addition to the subject of examination, promotes an improvement or enhancement of contrast in absorption (i.e., in X-ray images). Contrast agents with atomic numbers greater than 20 or greater than 40 are preferably used. Contrast agents, in particular, have atomic numbers less than 83 or less than 70. Particularly advantageous contrast agents, for example, contain iodine. However, contrast agents can also be, for example, gadolinium-based or other material-based contrast agents.

[0016] The patient can be a human patient and / or an animal patient. The determined image dataset may involve a partial area of ​​the patient's body, such as a specific body region of the patient that should be imaged in the resulting image dataset. In this case, the image dataset contains only information about the X-ray attenuation distribution of the partial area of ​​the patient to be imaged (e.g., the chest or pelvis). Alternatively, the image dataset may involve the entire patient's body. In this case, the image dataset contains information about the X-ray attenuation distribution of the entire patient's body.

[0017] X-ray quantum energy distribution refers to the energy spectrum of X-ray radiation used to record one of multiple image datasets. According to an alternative approach, at least two image datasets can be generated using X-ray radiation with different X-ray quantum energy distributions. These distributions may differ, for example, in their average X-ray quantum energy or their peak energy, and may partially overlap or completely separate in their spectra, i.e., have no intersection.

[0018] At least two image datasets generated with different X-ray quantum energy distributions can be generated using a dual-energy or multi-energy imaging device. Here, the different X-ray quantum energy distributions are generated by different accelerating voltages at one or more X-ray sources used in the corresponding X-ray image recording apparatus. Alternatively, different X-ray quantum energy distributions can be generated using different spectral filters behind the X-ray sources, and / or an energy-selective detector can be used. Here, energy selectivity can be understood as spectral resolution or spectral separation. The energy-selective detector is configured to classify the incident X-ray quanta according to their quantum energy. The energy-selective X-ray detector can be specifically designed as a direct-conversion X-ray detector with photon counting.

[0019] In particular, the X-ray quantum energy distribution for recording image datasets can be preset or fixed in advance by the imaging equipment used and one or more of its accelerating voltages or other system parameters (such as parameters of energy-selective X-ray detectors). This includes adjusting these parameters directly before recording based on patient information of the patient to be imaged (e.g., positioning films, queries, etc. based on patient information stored in a database).

[0020] It is known that different materials or tissue types, such as water or bone, or added contrast agents, interact with X-rays to varying degrees. Furthermore, the energy correlation of X-ray attenuation as it passes through matter is also known. This means that low-energy X-rays are more readily absorbed by matter than high-energy X-rays. If a first image dataset and a second image dataset are generated with different X-ray quantum energy distributions, the first image dataset represents the spatial X-ray attenuation distribution of the patient with respect to a first X-ray energy spectrum, i.e., the X-ray quantum energy distribution, and at least one second image dataset represents the X-ray attenuation distribution of the patient with respect to at least one second X-ray energy spectrum.

[0021] If the detection step includes detecting a first image dataset and a second image dataset of the patient, wherein the first image dataset is recorded with contrast agent applied and represents the patient's first X-ray attenuation distribution with contrast agent, and the second image dataset is recorded without contrast agent applied and represents the patient's first X-ray attenuation distribution without contrast agent, then two time-delay imaging applications are performed on the patient, wherein the contrast agent is applied in the first application and the relevant image dataset is determined, and wherein the contrast agent is not applied in the second application and the relevant image dataset is determined again accordingly. However, the imaging application parameters, except for the contrast agent application, remain unchanged. In particular, the same X-ray quantum energy distribution is used not only to generate the first image dataset but also to generate the second image dataset. Thus, the first image dataset is contrast-enhanced, while the second image dataset is not. In an advantageous design, image datasets of as similar type as possible are generated, except for the contrast agent, and are as similar as possible in terms of imaging device parameters, the imaged patient region, and the patient's condition. Here, the first image dataset can be determined before or after the second image dataset. Typically, the image dataset without contrast agent is generated before the image dataset with contrast agent.

[0022] Prior to the determination steps described below, registration of the first image dataset to the second image dataset can be performed. This particularly facilitates advantageously improved results when the first and second image datasets do not represent exactly the same time and location, and movement of the recorded object cannot be ruled out during the recording of the first and second image datasets. For example, this can particularly be the case where two time-lapse imaging applications are performed on a patient, wherein a contrast agent is applied and the relevant image dataset is determined in the first application, and wherein no contrast agent is applied and the relevant image dataset is determined again in the second application, which is time-lapsed. However, it is also advantageous to record the first and second image datasets using a so-called dual-energy method by using two X-ray spectra. This can be particularly advantageously excluded by using an energy-selective, i.e., energy-resolved X-ray detector designed to simultaneously record data from the first and second image datasets in a first energy range, i.e., according to a first X-ray quantum energy distribution, and in a second energy range, i.e., according to a second X-ray quantum energy distribution.

[0023] Here, registration can be understood as a method that establishes explicit morphological relationships or correlations between image elements (pixels or voxels) in two images reflecting the same or similar objects. The determination of correlation is typically done through unique features, so-called "landmarks," which can be determined interactively by the user or automatically by the system. Landmarks can be one-dimensional structures, such as specific anatomical points, or multi-dimensional structures, such as the surfaces of certain organs previously segmented in the image. However, registration can also be performed through the distribution of intensity values ​​stored in the image. Image registration is a common task in medical image processing, and many solutions exist. Examples of available optimization methods for registration include gradient descent, downhill simplex, hill climbing, and simulated annealing. For registration, in particular, non-rigid, flexible, or elastic registration can be used. Elastic registration methods can be understood as image registration methods in which elastic transformations (also called "non-rigid transformations"), such as spline- or polynomial-based transformations, can be applied.

[0024] The determination step includes using a computing unit to determine a contrast agent image dataset and a non-contrast image dataset based on a first image dataset and at least one second image dataset.

[0025] An alternative approach involves detecting a first image dataset and a second image dataset of the patient, wherein the first image dataset is recorded with contrast agent applied and represents the patient's first X-ray attenuation distribution with contrast agent, and the second image dataset is recorded without contrast agent applied and represents the patient's first X-ray attenuation distribution without contrast agent. Based on this alternative approach, the contrast agent dataset can be determined by selectively subtracting the second image dataset from the first image dataset after registration, particularly in image space. The result of the subtraction corresponds to the contrast agent dataset. The non-contrast image dataset can be directly determined based on the second image dataset without contrast agent application, i.e., specifically directly corresponding to the second image dataset. While there are limitations due to the necessary time interval between the two scans, and the perfusion image dataset generated using this alternative approach may only allow for qualitative rather than quantitative statements, this presents a viable and advantageous possibility, for example, when no suitable imaging system is available for spectral generation of the image dataset using different X-ray quantum energy distributions.

[0026] An alternative detection step includes detecting, by means of a first interface, a first image dataset representing a patient corresponding to a first X-ray attenuation distribution corresponding to a first X-ray quantum energy distribution and at least one second image dataset representing a patient corresponding to at least one second X-ray quantum energy distribution, wherein the first image dataset and the second image dataset are recorded in the presence of a contrast agent. Based on this alternative, the determination of the contrast agent image dataset and the non-contrast image dataset can be performed as follows.

[0027] Then, in the determination step, material or base material decomposition, which is known per se, can be performed based on at least two image datasets of at least two (base) materials. Material decomposition assumes that X-ray attenuation values ​​measured by means of an X-ray image recording device can be described as a linear combination of the X-ray attenuation values ​​of the so-called base materials relative to the X-ray quantum energy distribution. The measured X-ray attenuation values ​​can be obtained from at least two image datasets with different X-ray quantum energy distributions. The material or base material can be any substance or arbitrary tissue, particularly water, contrast agents (such as iodine), soft tissue, bone, etc. The X-ray attenuation of the base material relative to X-ray radiation energy is substantially known, or can be determined by prior measurement of the phantom and stored in tabular form for retrieval within the scope of material decomposition. The result of material decomposition can be the spatial density distribution of at least two materials in the patient, from which the proportion or combination of base materials can be determined for each image element in the patient's body region to be imaged. For example, in W. Kalender et al., “Materials Selective Imaging and Density Measurement Using a Dual-Spectrum Method, I. Foundations and Methods, W. Kalende, W. Bautz, D. Felsenberg, C. Süß and E. Klotz, Digital Image Diagnostics 7, 1987, 66-77, Georg Thieme Publishing House,” a method for basic material decomposition in X-ray radiography is described. For those skilled in the art, basic material decomposition can be performed in an obvious manner in both image space and projection space. These two methods are equivalent to the method according to the invention; however, the computational steps can be advantageously and simply performed in image space, since calculations can be performed element-by-element of the image. Decomposition into two or more materials can be based, in particular, on two or more image datasets detected, wherein X-ray quantum energy distributions can be assigned to each image dataset, wherein these X-ray quantum energy spectra are different from each other. For example, example methods for multi-material decomposition based solely on two image datasets in more than two materials can be found in US 2010 / 0 135 453 A1 or US 2007 / 0 217 570 A1.

[0028] Within the scope of the method according to the invention, in addition to the contrast agent as the base material, the breakdown of the base material can be based on water or tissue material as a second base material, particularly soft tissue material, such as fat. For example, it may also include breaking down the material into two or more base materials, such as the contrast agent used, water, and fat. The second or additional base material may depend particularly on which examination area is being imaged and which structure should be imaged using a perfusion image dataset.

[0029] Contrast agent image datasets can essentially correspond to the proportion of X-ray attenuation caused by the contrast agent within the patient. In particular, contrast agent image datasets can reflect the local contrast agent concentration (contrast dose per volume) within the patient. Contrast agent image datasets can also be generated directly from the breakdown of a base material, where the base material represents the contrast agent used. Contrast agent datasets can correspond to a classical representation of tissue perfusion, such as in the sense of the PBV method for the lungs.

[0030] Based on the fundamental material decomposition, a (virtual) noncontrast image dataset (also known as VNC images) is further determined. The (virtual) noncontrast image dataset can particularly reflect the morphological characteristics of the tissue of interest.

[0031] In its simplest approximation, the (virtual) noncontrast image dataset, within the scope of the method according to the invention, can correspond to an image of the base material, representing a second base material, provided as a result of decomposing the base material into two base materials, such as contrast agent and water, or contrast agent and tissue material, such as fat. This provides particularly good results when essentially only the two materials in the display area of ​​interest are important and / or, for example, water corresponds to a sufficiently accurate approximation of the tissue to be displayed. If the perfusion image dataset is used for perfusion imaging of lung parenchyma, a (virtual) noncontrast image dataset, such as a water image dataset resulting from decomposing the base material into contrast agent and water, may be a sufficiently good choice and advantageously simple implementation.

[0032] However, based on the results of basic material decomposition, a (virtual) noncontrast image dataset can also be determined, which is different from the aforementioned material image dataset and considers the combination of materials present in the patient-related region. For this purpose, well-known methods for calculating (virtual) noncontrast images can be used. For example, see here Uhrig M et al.: Monitoring targeted therapy using dual-energy CT: Semi-automatic RECIST plus supplemental functional information by quantifying iodine uptake in melanoma metastasis (Cancer Imaging, 22 July 2013; 13(3): 306-13. doi:10.1102 / 1470-7330.2013.0031.PMID:23876444; PMCID: PMC3719051) and Martin Petersilka et al.: Technical principles of dual-source CT (European Journal of Radiology, Vol. 68, No. 3, 2008, pp. 362-368). https: / / doi.org / 10.1016 / j.ejrad.2008.08.013 ).

[0033] The calculation steps include using a computing unit to calculate the perfusion image dataset based on the ratio of image values ​​in the contrast agent image dataset to image values ​​corresponding to locations in the non-contrast image dataset.

[0034] In particularly simple and timely designs, the final perfusion image dataset directly corresponds to the quotient of the contrast-enhanced image dataset and the non-contrast-enhanced image dataset, or the quotient of the image values ​​at each location. However, this information may depend on the acquisition and reconstruction parameters. Nevertheless, this approach can be advantageously simple and timely as long as quantitative comparisons between different generated image datasets are not required. In other designs, the calculation may include additional computational steps or factors applied to the ratio, for example, to improve comparability and quantifiability across records of different generated image datasets.

[0035] The steps include providing the perfusion image dataset via a second interface.

[0036] The perfusion image dataset can be provided for display on a display unit (e.g., a monitor) or for further processing. In addition to the perfusion image dataset, a first or second image dataset, a contrast agent image dataset, and / or a non-contrast image dataset can also be provided via an interface. Thus, for example, the display of different datasets can be side-by-side or selected sequentially to comprehensively provide the different information contained therein.

[0037] The method according to the invention advantageously combines morphological and functional information, thereby allowing the correlation between perfusion image data and morphological conditions (particularly, for example, disease-related changes) to be taken into account. The invention described herein and the combination of these two types of information make it possible to potentially compensate for this effect.

[0038] A particularly advantageous application of the method according to the invention is to represent lung parenchyma perfusion by means of a perfusion image dataset according to the invention.

[0039] Unlike other human tissues such as muscle or fat, the density of lung parenchyma changes every second. With inhalation, the air-to-tissue ratio changes, and the density decreases. During exhalation, the opposite occurs, and the density increases. This physiological effect can be directly read from the measured CT values ​​(different HU values) in inspiratory and expiratory CT images, for example. Furthermore, even within a single time point, the density distribution is not uniform. Due to the typical supine position during recording and the physical force of gravity acting on the lung parenchyma, there is a non-uniform distribution within the lungs: the density of the lung parenchyma is highest in the back and steadily decreases towards the sternum.

[0040] If lung parenchymal blood flow is traditionally represented solely by contrast agent datasets, perfusion is not measured directly but instead replaced by the amount of iodine injected. In previous methods of conventional lung PBV using only contrast agent datasets, iodine maps showed how much "iodine per volume" (e.g., mg / ml) was enriched in the tissue, ignoring the density or volume of the lung parenchyma. Thus, while it is known where iodine is more and less concentrated, it is not known, except in cases of significant differences, whether lung parenchymal perfusion is locally lower or higher. This may be sufficient to represent large perfusion differences but insufficient to quantitatively compare small differences. Therefore, although perfusion in healthy lung tissue is the same everywhere, conventional lung PBV values, for example, show contrast agent concentrations that differ within a single layer solely due to the force of gravity. Consequently, previous visualizations of perfusion using contrast agent maps were not always able to indicate whether higher contrast agent concentrations were attributable to more parenchyma (higher density) or lower perfusion. Quantification also becomes problematic. Identifying hyperperfusion differences in pulmonary embolism is generally irrelevant, but other applications remain limited (e.g., assessment of perfusion outcomes in the presence of other disease-related changes in the lungs; assessment of incremental or local effects; quantification). The method and the resulting perfusion dataset according to the invention enable a particularly accurate display of considered morphology, which can compensate for potential influencing events, and thereby expands the possibilities for assessment and diagnosis.

[0041] The following design variations of the method of the present invention are particularly based on the following alternative according to the invention, namely, detecting in the detection step a first image dataset representing a patient corresponding to a first X-ray attenuation distribution of a first X-ray quantum energy distribution and at least one second image dataset representing a patient corresponding to at least one second X-ray quantum energy distribution, wherein the first image dataset and the second image dataset are recorded with the application of a contrast agent, and furthermore, the determination includes a basic material decomposition based on the first image dataset and at least one second image dataset.

[0042] Therefore, in the corresponding design variations, the ratio of image values ​​in the contrast agent image dataset and the non-contrast image dataset can be scaled using a scaling factor, which can specifically have a correlation with X-ray photon energy. Appropriate scaling can be advantageously used to generate perfusion image datasets, where the correlation with recording parameters is reduced, or to better display additional information relevant to the evaluation. In particular, appropriate scaling can improve the standardization and quantifiability of the dataset, eliminating differences in acquisition parameters.

[0043] Here, the scaling factor may specifically depend on at least one specific attenuation coefficient µ / ρ of the base material on which the base material decomposition is based, where µ is the energy-dependent absorption coefficient of the corresponding base material to X-rays, and ρ is the corresponding density. For example, if the base material is the contrast agent used (especially iodine) and water or tissue (especially soft tissue), the scaling factor may depend on the specific attenuation coefficient (µ / ρ) of the contrast agent used. I Specific attenuation coefficient (µ / ρ) of water or tissue W .

[0044] The scaling factor can in particular include the quotient of the specific attenuation coefficients µ / ρ of at least two base materials. Such a scaling factor can, for example, be used to generate a virtual single-energy perfusion image dataset for a specified X-ray photon energy.

[0045] Scaling factors at different X-ray photon energies can be stored, for example, in tabular form or provided using functional relationships. For instance, functional relationships can be fitted based on known reference data for the corresponding base material.

[0046] Furthermore, the present invention also relates to an apparatus for providing a patient perfusion image dataset, comprising:

[0047] – The first interface was designed as

[0048] ◦ Detect a first image dataset and at least one second image dataset, the first image dataset representing a first X-ray attenuation distribution of the patient corresponding to a first X-ray quantum energy distribution, and the second image dataset representing at least one second X-ray attenuation distribution of the patient corresponding to at least one second X-ray quantum energy distribution, wherein the first and second image datasets are recorded with contrast agent applied, or

[0049] ◦ Detect a first image dataset and a second image dataset of the patient, wherein the first image dataset is recorded with contrast agent applied and represents the patient's first X-ray attenuation distribution with contrast agent, wherein the second image dataset is recorded without contrast agent applied and represents the patient's first X-ray attenuation distribution without contrast agent.

[0050] – The computational unit is designed to determine a contrast agent image dataset and a non-contrast image dataset based on a first image dataset and at least one second image dataset, and to calculate the perfusion image dataset based on the ratio of image values ​​in the contrast agent image dataset to image values ​​corresponding to locations in the non-contrast image dataset.

[0051] – The second interface is designed to provide perfusion image datasets.

[0052] Such an apparatus for providing a perfusion image dataset can be specifically designed to perform the aforementioned method and aspects thereof for providing a perfusion image dataset according to the present invention. The apparatus can be designed to perform the method and aspects thereof by designing the interface and processing unit to perform the corresponding method steps.

[0053] The device or computing unit can be, in particular, a computer, a microcontroller, or an integrated circuit. Alternatively, it can be a real or virtual network of computers (the English term for a real network is "cluster," and the English term for a virtual network is "cloud"). The device can also be designed as a virtual system that runs on a real computer or a real or virtual computer network.

[0054] The interface can be a hardware or software interface (such as a PCI bus, USB, or FireWire). The processing unit can have hardware or software components, such as a microprocessor or a so-called FPGA (Field Programmable Gate Array).

[0055] An interface may specifically include several sub-interfaces. In other words, an interface may also include multiple interfaces. A computation unit may specifically include several sub-computation units that execute different steps of a corresponding method. In other words, a computation unit can also be understood as multiple computation units.

[0056] In addition, the device may include a storage unit. The storage unit can be implemented as non-persistent working memory (random access memory, or RAM) or permanent mass storage (hard disk, USB stick, SD card, solid-state drive).

[0057] The advantages of the proposed apparatus essentially correspond to the advantages of the proposed method for generating the resulting image dataset. The features, advantages, or alternative embodiments mentioned herein can also be applied to the apparatus, and vice versa.

[0058] Furthermore, the present invention relates to a medical imaging apparatus including means for providing a perfusion image dataset as described above, and including at least one X-ray source opposite at least one X-ray detector, wherein a patient may be located between the X-ray source and the X-ray detector. The medical imaging apparatus may be accordingly designed to generate first and second image datasets.

[0059] The advantages of the proposed medical imaging device essentially correspond to the advantages of the proposed method for providing perfusion image datasets. The features, advantages, or alternative implementations mentioned herein can also be applied to the imaging device, and vice versa.

[0060] Medical imaging devices can be designed to provide a first image dataset representing a patient’s first X-ray attenuation distribution corresponding to a first X-ray quantum energy distribution and at least one second image dataset representing a patient’s second X-ray attenuation distribution corresponding to at least one second X-ray quantum energy distribution, wherein the first image dataset and the second image dataset are recorded in the presence of a contrast agent.

[0061] Imaging equipment can be, in particular, X-ray equipment designed to record multiple X-ray projections from different projection angles, such as a computed tomography (CT) scanner with a rotating annular frame, or an imaging device can be a C-arm X-ray device. Images can be generated during, in particular, continuous rotational motion of the recording unit, which includes an X-ray source and an X-ray detector interacting with the X-ray source. The X-ray source can, in particular, be an X-ray tube with a rotating anode. X-ray detectors for CT scanners are, for example, line detectors with multiple lines. X-ray detectors for C-arm X-ray devices are, for example, planar detectors.

[0062] The X-ray detector can correspond to a spectrum-separated X-ray detector. The spectrum-separated X-ray detector is configured to classify incident X-ray quanta according to their quantum energy and assign them to one of the image datasets. Therefore, for the method according to the invention, only an X-ray source with a predetermined or fixed emission spectrum is required. According to this aspect of the invention, recording of the image dataset is particularly rapid and requires no additional dose to the patient. The X-ray detector can be a quantum counting detector or a two-layer detector. A quantum counting detector can generally be understood as a direct conversion detector that directly converts incident X-ray quanta into electrical signals using suitable detector materials. The quantum counting detector can operate with energy resolution, where the energy resolution can be adjusted by so-called compartmentalization. In other words, any energy range that can classify incident X-ray quanta can be defined. A first image dataset and at least one second image dataset are each formed from signals within one or more energy ranges. The image datasets can be assigned energy ranges according to the first X-ray quantum energy distribution and / or at least one second X-ray quantum energy distribution. Semiconductors such as cadmium telluride, zinc cadmium telluride, or gallium arsenide are particularly suitable as detector materials for quantum counting detectors, or, in the case of planar detectors, amorphous selenium is particularly suitable. Two-layer or bilayer detectors are designed to decompose the spectrum of an incident X-ray tube into low-energy and high-energy components. For this purpose, a two-layer detector consists of two layers. The detector layer facing the X-ray source measures photons of the incident X-rays at low energy and assigns the measured signal to a first image dataset. It is penetrated by high-energy X-rays. Photons with higher quantum energies are measured in the detector layer below or behind it, i.e., positioned away from the X-ray source, and assigned to a second image dataset. Typically, both detector layers include scintillators, thus the two-layer detector is an indirect conversion detector. Crystals such as cesium iodide and cadmium tungstate, or ceramic materials such as gadolinium oxysulfide, can be used as scintillator materials.

[0063] Imaging apparatuses can also include two-source detector systems operating with different emission spectra. In this case, the imaging apparatus includes two X-ray sources and two X-ray detectors, with each detector configured to receive X-ray radiation emitted from one of the X-ray sources. This is also known as a dual-source X-ray imaging apparatus. Furthermore, at least one of the two X-ray sources may include a filter, particularly a tin filter, to improve the spectral separation of the emitted X-rays.

[0064] Imaging equipment can also be used for so-called "kV swing," in which the X-ray source rapidly and continuously emits different emission spectra in the direction of the X-ray detector.

[0065] Imaging devices can also be designed as so-called "dual-beam" devices, which provide a source detector system and a filter composed of two materials arranged in front of the X-ray source, so that after passing through the filter there are two different X-ray quantum energy distributions that irradiate a portion of the X-ray detector respectively.

[0066] Furthermore, the present invention also relates to a computer program product having a computer program that can be directly loaded into the memory of the apparatus for providing perfusion image datasets as described above, the computer program having a program segment that, when run by the apparatus, performs all or aspects of one of the aforementioned methods for providing perfusion image datasets.

[0067] Furthermore, the present invention relates to a computer-readable storage medium having stored thereon a program segment that can be read and executed by an apparatus for providing an perfusion image dataset as described above, wherein when the program segment is executed by the apparatus, it performs all or aspects of one of the aforementioned methods for providing the perfusion image dataset.

[0068] Examples of computer-readable storage media include DVDs, magnetic tapes, hard drives, or USB sticks, on which electronically readable control information, particularly software, is stored.

[0069] The largely software-based implementation has the advantage that previously used devices and computing units can be easily modified through software updates to operate in accordance with the invention. In addition to the computer program, the computer program product may, if necessary, include additional components such as documentation and / or add-ons, as well as hardware components such as hardware keys (dongles, etc.) to facilitate the use of the software.

[0070] Furthermore, within the scope of this invention, features described regarding different embodiments and / or different categories of invention (methods, uses, apparatuses, systems, components, etc.) can also be combined to form other embodiments of the invention. For example, an invention related to an apparatus can also be extended to have features described or claimed in conjunction with the method, and vice versa. Here, the functional features of the method can be implemented through specific components of a corresponding design.

[0071] The use of the indefinite article "a" or "a kind" does not preclude the possibility that the related feature may appear multiple times. The use of the word "having" does not preclude that the terms connected by the word "having" may be the same. For example, a medical imaging device has a medical imaging device. The use of the word "unit" does not preclude that the object referred to by the word "unit" may have multiple components that are spatially separated from each other.

[0072] In the context of this application, the word "based on" can be understood in particular as the meaning of "by use". The statement that the first feature is generated (or determined, decided, etc.) based on the second feature does not exclude the possibility that the first feature can be generated (or determined, decided, etc.) based on the third feature. Attached Figure Description

[0073] The invention will be explained below with reference to the accompanying drawings and exemplary embodiments. The illustrations in the drawings are schematic, greatly simplified, and not necessarily drawn to scale. Wherein:

[0074] Figure 1 The illustrations depict different scenarios of lung tissue perfusion.

[0075] Figure 2 A schematic method flow for providing an infused image dataset is shown.

[0076] Figure 3 A schematic diagram of an apparatus for providing a perfusion image dataset is shown, and

[0077] Figure 4 A schematic diagram of an exemplary medical imaging device is shown. Detailed Implementation

[0078] Figure 1 This illustration schematically depicts different scenarios of lung tissue perfusion to illustrate the limitations of conventional assessments of lung parenchymal perfusion using traditional lung PBV values. In previous conventional lung PBV methods, contrast agent maps generated from the breakdown of the base material showed how much "contrast agent per volume" (e.g., mg / ml) was enriched in the tissue, neglecting the density or volume of the lung parenchyma. Thus, while it was known where iodine was more or less concentrated, it was not known, except in cases of significant variation, whether lung parenchymal perfusion was locally lower or higher.

[0079] The first row shows the lung volume (LV) with normally perfused vessels a at the microscopic level. Imaging using conventional PBV values ​​results in a first lung PBV value (PBV1) representing the first perfusion level at the macroscopic level.

[0080] The second row shows the lung volume LV with normally perfused vessels a at the microscopic level, while the first row reflects the inspiratory state, for example, during expiration. Thus, when imaging with conventional lung PBV values, a higher number of normally perfused vessels per unit volume results in a second lung PVB value PBV2, which, despite the presence of normal perfusion of vascular structures, exhibits a higher level of perfusion at the macroscopic level compared to the first perfusion level. Therefore, previous lung perfusion measurements using only contrast agent concentration per volume have not always been able to reveal whether a higher contrast agent concentration is attributable to more parenchyma (higher density) or lower perfusion. Quantification is also problematic.

[0081] The third row shows, for example, a lung volume LV during inspiration, which has vascular structures b with increased perfusion due to inflammation (e.g., due to viral infection). Here, imaging using conventional lung PBV values ​​can result in lung PBV2 values ​​that are the same or at least similar to those of normally perfused tissue during expiration (second row). For example, two different cases with different perfusion levels at the microscopic level can result in the same perfusion level at the macroscopic level based on conventional lung PBV values. On this basis, the states cannot be distinguished.

[0082] Furthermore, as shown in the fourth row, fluid deposits, such as c, may also occur, which could additionally alter the composition of lung tissue. In such cases, imaging using conventional lung PBV values ​​may result in PBV2 values ​​that are the same as or at least similar to those of normally perfused tissue during exhalation (second row). Distinction cannot be made in this situation either.

[0083] In this context, the current method of dividing lung parenchymal perfusion examination results into a purely functional portion (contrast-enhanced images) and a purely morphological portion (non-contrast-enhanced images) has its limitations. The present invention, and the combination of these two types of information, can potentially compensate for this effect and provide distinguishable or potentially quantifiable results.

[0084] Figure 2 A schematic method flow of the present invention for providing an irrigation image dataset is shown.

[0085] In a first alternative to the illustrative method flow, step S1 includes detecting a first image dataset of patient 39 and a second image dataset of patient 39, wherein the first image dataset is recorded with contrast agent applied and represents a first X-ray attenuation distribution of patient 39 with contrast agent, and wherein the second image dataset is recorded without contrast agent applied and represents a first X-ray attenuation distribution of patient 39 without contrast agent.

[0086] In a second alternative to the method flow, step S1 includes detecting a first image dataset and at least one second image dataset by means of a first interface IF1, the first image dataset representing a first X-ray attenuation distribution of the patient corresponding to a first X-ray quantum energy distribution, and the second image dataset representing at least one second X-ray attenuation distribution of the patient corresponding to at least one second X-ray quantum energy distribution, wherein the first image dataset and the second image dataset are recorded in the presence of a contrast agent.

[0087] Step S2 includes determining the contrast agent image dataset and the non-contrast image dataset using the computing unit CU.

[0088] Based on the first alternative, the contrast agent dataset can be determined by subtracting the second image dataset from the first image dataset in image space after the image datasets have been cross-registered. The result of the subtraction corresponds to the contrast agent dataset. The non-contrast image dataset can be determined directly based on the first image dataset without the application of contrast agent, i.e., specifically, it directly corresponds to the first image dataset.

[0089] Based on the second alternative, the contrast agent image dataset and the non-contrast image dataset are determined by means of a computing unit (CU), comprising a base material decomposition based on a first image dataset and at least one second image dataset. In one embodiment, the base material decomposition is based at least on a base material comprising the contrast agent and water used, or the contrast agent and tissue material (particularly soft tissue material) used. In other embodiments, more than two base materials may also be distinguished, such as contrast agent, water, and fat. The contrast agent dataset can be generated essentially directly from the base material decomposition and reflects the local contrast agent concentration (contrast dose per volume) in the patient's body. In a simple embodiment, the non-contrast images may directly correspond to the base material image dataset generated by the base material decomposition, which is assigned to the second material when the base material is decomposed into contrast agent and second material. For example, the non-contrast images may correspond to a water image dataset. This is particularly advantageous when the region of interest relating to the material composition can be adequately described in terms of morphological properties based on the second material. In other implementation variations, the non-contrast image dataset may also be provided in other aspects, thus also taking into account the different materials in the relevant image regions. In particular, known methods for computing virtual non-contrast images (so-called VNC images, VNC: virtual non-contrast) can be used. For example, if the perfusion image dataset is used for imaging lung parenchyma perfusion, however, within the scope of advantageous and simple assumptions and implementations as a non-contrast image dataset, a suitable perfusion image dataset can be obtained, for example, by selecting a water image dataset generated when the base material is decomposed into contrast agent and water.

[0090] Step S3 includes calculating the S3 perfusion image dataset using the computing unit CU based on the ratio of image values ​​in the contrast agent image dataset to image values ​​corresponding to locations in the non-contrast image dataset.

[0091] Step S4 includes providing the S4 perfusion image dataset via the second interface IF2.

[0092] In a particularly simple and time-sensitive design, the final perfusion image dataset directly corresponds to the quotient of the contrast agent image dataset and the non-contrast image dataset, or the quotient of the image values ​​corresponding to each location.

[0093] However, particularly in conjunction with the second alternative described above, the computation S3 of the perfusion image dataset can further include scaling the ratio of image values ​​of the contrast agent image dataset to those of the non-contrast image dataset using a scaling factor. The scaling factor can, in particular, depend on the X-ray photon energy L. For example, the scaling factor can depend on at least one specific attenuation coefficient µ / ρ of the base material on which the base material decomposition is based, where µ is the corresponding absorption coefficient of the base material for X-rays at the energy L of the specified X-ray photon energy, and ρ is the corresponding density. The scaling factor can, in particular, comprise the quotient of the specific attenuation coefficients µ / ρ of at least two base materials at a specific X-ray photon energy and is used to generate a virtual single-energy perfusion image dataset for that X-ray photon energy. Possible implementations of the scaling factor will be explained in more detail below.

[0094] For example, according to one implementation variant, the perfusion image dataset PD used to specify the X-ray photon energy L can be calculated as follows:

[0095]

[0096] Here, b w [Unit HU] and b I [Unit: g / cm³] 3 An exemplary image corresponding to the basic material of water or tissue. b w and the base material images of contrast agents (such as iodine) b I This originates from the decomposition of the base material into these two materials. If the base material decomposition is based on other base materials, then the corresponding base material image must be used. Choosing water as the base material, particularly in the lung region, is a suitable approximation, as mentioned above. This is a particularly advantageous application of the method of the present invention, and it allows for simple provision because the density of the relevant areas of the lung, especially the lung tissue, is thus adequately and well described, and few other materials (such as fat or calcifications) play a role.

[0097] hum(L) Corresponding to the X-ray photon energy L The scaling factor is based on the specific attenuation coefficient µ / ρ of the base materials upon which the base material decomposition is based (e.g., water and contrast agents) at X-ray photon energy L, where µ is the energy-dependent absorption coefficient of the base material to X-rays, and ρ is the corresponding density. Here, the scaling factor specifically includes the quotient of the specific attenuation coefficients of the two base materials at energy L:

[0098]

[0099] Specifically, the scaling factor can be defined as follows.

[0100]

[0101] Scaling factors at different energies L can be stored, for example, in tabular form or provided using functional relationships. For instance, functional relationships can be determined based on fitting known reference data of the corresponding underlying material (e.g., using the NIST database of X-ray attenuation coefficients, https: / / physics.nist.gov / ). Thus, by applying the scaling factor at the selected X-ray photon energy L, a perfusion image dataset for that X-ray photon energy can be calculated.

[0102] Energy-dependent scaling factors enable the perfusion image dataset to be displayed, to a certain extent, in a manner similar to the known display of a virtual single-energy image (VMI) at energy level L, in the sense of a virtual single-energy perfusion image dataset at a given energy level L. In the virtual single-energy image (VMI), the display is normalized by setting the energy level and the fixed attenuation characteristics of all associated materials to eliminate differences in CT acquisition parameters. This is particularly significant given the possibility of normalization across all recordings using new technologies, such as photon-counting detectors. For the generation of virtual single-energy images, see, for example, DE 10 2015 204 450 A1.

[0103] Virtual single-energy images can be simplified and calculated from a linear combination of defined base material images. For example, a virtual single-energy image can be simplified based on a two-material decomposition of water or tissue and a contrast agent (e.g., iodine) as follows:

[0104]

[0105] Based on the above calculations using the perfusion image dataset, b w [Unit HU] and b I [Unit: g / cm³] 3 This exemplifies an image corresponding to the base material of water or tissue and contrast agents, derived from the breakdown of the base material. This is achieved through a function... hum(L) It can calculate the CT value per concentration at a specified energy L. If base material decomposition is performed based on other base materials, the corresponding base material images must be used. If the calculation of a virtual single-energy image dataset must be performed regardless of the scope of X-ray imaging applications under contrast agent administration based on the first and second image datasets, intermediate results in monochrome can be used. b w and b I To calculate the perfusion image dataset.

[0106] Similar to the case of providing a virtual single-energy image dataset, the X-ray photon energy used for calculation can be specified to calculate the perfusion image dataset, or different energies L can be repeatedly selected for calculating the perfusion image dataset. Furthermore, a method may also include, in particular, the step of selecting the X-ray photon energy. Here, for example, a user can select the X-ray photon energy to be used to calculate the perfusion image dataset via an input unit such as a keyboard. This selection can also be automatically determined by means of other parameters (e.g., patient-specific parameters, such as patient weight or height).

[0107] In addition to the above, by means of hum(L) Beyond the scaling described above, other scaling methods may be necessary within the scope of this approach. For example, scaling using a background without a background (HU) might be meaningful.

[0108]

[0109] in b w [Unit HU] and b I [Unit: g / cm³] 3 Image of the base materials corresponding to water and contrast agents.

[0110] Furthermore, based on the already created system with two different energy levels ( L 1 and L 2) The virtual single-energy image is also feasible for subsequent calculations of the perfusion image dataset of this invention. This is possible because each VMI essentially represents only the result of... b I Imaging contrast agents and by b w The imaging is a linear combination of water or tissue, and the underlying material can be extracted through a clear understanding of the linearity factor.

[0111]

[0112] Wherein is in energy L 1 or L A virtual single-energy image dataset of 2 (in keV) hum ( L For the perfusion image dataset PD ( L Selectable energy L or at the corresponding energy L 1 or L Scaling factor at 2.

[0113] What is not explicitly stated in the above formula is that a pseudo density [in HU] is used instead of the actual HU value in the image dataset. This is equivalent to adding 1000 HU, in order to avoid division by zero, especially when water is used as the base material and the calculation is performed in HU units.

[0114] Figure 3 A schematic diagram of a device 45 for providing a perfusion image dataset of patient 39 is shown.

[0115] The device 45 includes a first interface IF1 designed to detect a first image dataset and at least one second image dataset, the first image dataset representing a first X-ray attenuation distribution of patient 39 corresponding to a first X-ray quantum energy distribution, and the second image dataset representing at least one second X-ray attenuation distribution of patient 39 corresponding to at least one second X-ray quantum energy distribution. Alternatively, the first interface IF1 may be designed to detect the first image dataset and the second image dataset of patient 39, wherein the first image dataset is recorded with contrast agent applied and represents the first X-ray attenuation distribution of patient 39 with contrast agent, and wherein the second image dataset is recorded without contrast agent applied and represents the first X-ray attenuation distribution of patient 39 without contrast agent.

[0116] In addition, the device 45 also includes a computing unit CU, which is designed to determine a contrast agent image dataset and a non-contrast image dataset based on a first image dataset and at least one second image dataset, and to calculate a perfusion image dataset based on the ratio of image values ​​in the contrast agent image dataset to image values ​​corresponding to the positions in the non-contrast image dataset.

[0117] In addition, device 45 also includes a second interface IF2, which is designed to output an infusion image dataset.

[0118] In addition, the device 45 may also include a storage unit MU. The detected image dataset or the calculated perfusion image dataset can be stored in a recallable manner on the storage unit MU. The storage unit can be implemented as non-persistent working memory (random access memory, or RAM) or permanent mass storage (hard disk, USB stick, SD card, solid-state drive).

[0119] Such an apparatus 45 for providing a perfusion image dataset can be specifically designed to perform the aforementioned method and aspects thereof for providing a perfusion image dataset according to the present invention. The apparatus can be designed to perform the method and aspects thereof by designing the interfaces IF1, IF2 and the arithmetic unit CU to perform the corresponding method steps.

[0120] In the illustrated embodiment, device 45 is connected to medical imaging equipment 32. Device 45 can be connected to imaging equipment 32, for example, via a network. The device may also be included within imaging equipment 32. Imaging equipment 32 may be, for example, a computed tomography (CT) scanner.

[0121] A network can be a local area network (LAN) or a wide area network (WAN). An example of a local area network is an intranet, and an example of a wide area network is the Internet. Networks can also be implemented wirelessly, particularly as a WLAN (Wireless LAN, often abbreviated as WiFi) or a Bluetooth connection. Networks can also be implemented as a combination of the above examples.

[0122] Furthermore, communication between the device 45 and the imaging device 32 can also be conducted offline, for example, through the exchange of data carriers.

[0123] Figure 4 A medical imaging device 32 in the form of a computed tomography (CT) scanner is shown.

[0124] The CT scanner has a gantry 33 with a rotor 35. The rotor 35 includes at least one X-ray source 37, particularly an X-ray tube, and at least one X-ray detector 36 opposite to the X-ray source 37. The X-ray detector 36 and the radiation source 37 can rotate about a common axis 43 (also referred to as the axis of rotation). A patient 39 is supported on a patient bed 41 and can be moved along the axis of rotation 43 via the gantry 33. Typically, the patient 39 may include, for example, an animal patient and / or a human patient.

[0125] The CT device 32 includes a computer system 20, which includes means 45 for providing a perfusion image dataset. Furthermore, the computer system 20 may also have a reconstruction unit 42 for reconstructing the image dataset based on data determined by the imaging device 32. Additionally, the computer system 20 may also have a control unit 48 for controlling the imaging device.

[0126] Furthermore, input device 47 and output device 49 are connected to computer system 20. Input device 47 and output device 49 can, for example, enable interaction, such as manual configuration, confirmation, or triggering of method steps by the user. For instance, computed tomography projection datasets and / or two-dimensional or three-dimensional image datasets can be displayed to the user on output device 49, which includes a monitor.

[0127] Typically, during the relative rotational motion between the radiation source and the patient, measurement data are recorded from multiple projection angles as multiple (raw) projection datasets of the patient 39, while the patient 39 moves continuously or sequentially through the rack 33 by means of the patient bed 41. Then, based on the projection datasets, a dataset of layer images at the corresponding z-positions along the rotation axis within the examination area can be reconstructed by means of mathematical methods, such as filtered backprojection or iterative reconstruction methods.

[0128] Imaging device 32 is specifically designed to generate at least a first image dataset and a second image dataset according to the invention. Thus, the means for providing perfusion image datasets, included within computer system 45, is specifically designed to perform the method for providing perfusion image datasets according to the invention based on these image datasets.

[0129] X-ray detector 36 can be a spectrum-separated X-ray detector, such as a quantum counting detector or a two-layer detector. Imaging device 32 can also be designed as a so-called "kV swing" device, in which the X-ray source rapidly and continuously emits different emission spectra in the direction of X-ray detector 36. X-ray device can also be designed as a so-called "dual-beam" device.

[0130] In other embodiments, the imaging apparatus may also include two X-ray source detector systems operating at different emission spectra. In this case, the imaging apparatus includes two X-ray sources and two X-ray detectors, wherein each detector is configured to receive X-ray radiation emitted from one of the X-ray sources. This is also known as a dual-source X-ray imaging apparatus. Furthermore, at least one of the two X-ray sources may also include a filter, particularly a tin filter, for improving the spectral separation of the emitted X-rays.

Claims

1. A method for providing a dataset of perfusion images of a patient, comprising the following steps: – Using a first interface (IF1), a first image dataset and at least one second image dataset are detected, the first image dataset representing a first X-ray attenuation distribution of the patient (39) corresponding to a first X-ray quantum energy distribution, and the second image dataset representing at least one second X-ray attenuation distribution of the patient (39) corresponding to at least one second X-ray quantum energy distribution, wherein the first image dataset and the second image dataset are recorded in the presence of a contrast agent, or Using a first interface (IF1), a first image dataset and a second image dataset of the patient (39) are detected, wherein the first image dataset is recorded with contrast agent applied and represents the first X-ray attenuation distribution of the patient (39) with contrast agent, and wherein the second image dataset is recorded without contrast agent applied and represents the first X-ray attenuation distribution of the patient (39) without contrast agent. – Using a computing unit (CU), a contrast agent image dataset and a non-contrast image dataset are determined based on the first image dataset and at least one second image dataset; – The perfusion image dataset is calculated using the computing unit (CU) based on the ratio of image values ​​in the contrast agent image dataset to image values ​​corresponding to the locations in the non-contrast image dataset; – The perfusion image dataset is provided using the second interface (IF2).

2. The method according to claim 1, wherein in the detection step, a first image dataset representing the patient (39) corresponding to a first X-ray attenuation distribution of a first X-ray quantum energy distribution and at least one second image dataset representing the patient (39) corresponding to at least one second X-ray attenuation distribution of at least one second X-ray quantum energy distribution are detected, wherein the first image dataset and the second image dataset are recorded with the application of a contrast agent, and wherein the determination includes a basic material decomposition based on the first image dataset and at least one second image dataset.

3. The method of claim 2, wherein the base material decomposition is based at least on a base material comprising the contrast agent used and water or tissue material, and the non-contrast image corresponds to a water image dataset or tissue image dataset generated by the base material decomposition.

4. The method according to claim 2 or 3, wherein in the calculation, the ratio of the image values ​​of the contrast agent image dataset and the non-contrast image dataset is scaled using a scaling factor.

5. The method of claim 4, wherein the scaling factor has a correlation with the X-ray photon energy.

6. The method of claim 4, wherein the scaling factor depends on at least one specific attenuation coefficient µ / ρ of the base material on which the base material decomposition is based, where µ is the energy-dependent absorption coefficient of the corresponding base material for X-rays, and ρ is the corresponding density.

7. The method of claim 6, wherein the scaling factor comprises the quotient of the specific attenuation coefficient µ / ρ of the base material on which the base material decomposition is based.

8. The method according to any one of claims 1 to 3, wherein the perfusion image dataset reflects the perfusion of the lung parenchyma.

9. An apparatus (45) for providing a perfusion image dataset of a patient (39), comprising: – A first interface (IF1) is designed as Detect a first image dataset and at least one second image dataset, wherein the first image dataset represents a first X-ray attenuation distribution of the patient (39) corresponding to a first X-ray quantum energy distribution, and the second image dataset represents at least one second X-ray attenuation distribution of the patient (39) corresponding to at least one second X-ray quantum energy distribution, wherein the first image dataset and the second image dataset are recorded with contrast agent applied, or A first image dataset and a second image dataset of the patient (39) are detected, wherein the first image dataset is recorded with the application of a contrast agent and represents the first X-ray attenuation distribution of the patient (39) with the contrast agent, and wherein the second image dataset is recorded without the application of a contrast agent and represents the first X-ray attenuation distribution of the patient (39) without the contrast agent. – A computing unit (CU) is designed to determine a contrast agent image dataset and a non-contrast image dataset based on the first image dataset and at least one second image dataset, and to calculate the perfusion image dataset based on the ratio of image values ​​in the contrast agent image dataset to image values ​​corresponding to the locations in the non-contrast image dataset; and – A second interface (IF2) is designed to provide the perfusion image dataset.

10. A medical imaging device (32) comprising the apparatus of claim 9, and comprising at least one X-ray source (37) opposite to at least one X-ray detector (2), wherein a patient (39) may be positioned between the X-ray source (37) and the X-ray detector (2).

11. A computer program product having a computer program capable of being directly loaded into the memory (25) of an apparatus (45) for providing a patient perfusion image dataset according to claim 9, the computer program having program segments that, when the program segments are run by the apparatus (45), perform all steps of the method according to any one of claims 1 to 8.

12. A computer-readable storage medium storing a program segment that can be read and executed by an apparatus (45) for providing a patient perfusion image dataset according to claim 9, such that when the program segment is run by the apparatus (45), all steps of the method according to any one of claims 1 to 8 are performed.

Images