Radiography image generation

Abstract

The present invention relates to the generation of a radiological image of an examination area of ​​an object under examination. Based on measured radiological images of the examination area showing blood vessels in the examination area with a contrast intensity that decreases over time, the invention generates an artificial radiological image of the examination area showing blood vessels with a constant contrast intensity.

Description

[Technical Field]

[0001] The present invention relates to the generation of a radiological image of an examination area of ​​an examination subject. Based on such measured radiological images of the examination area showing blood vessels in the examination area with contrast enhancement that decreases over time, the invention generates an artificial radiological image of the examination area showing blood vessels with constant contrast enhancement. [Background technology]

[0002] Radiology is the medical branch that deals with imaging for diagnostic and therapeutic purposes.

[0003] Previously, X-ray radiation and films sensitive to X-ray radiation were primarily used in medical imaging, but nowadays, radiology includes a variety of different imaging modalities, such as computed tomography (CT), magnetic resonance imaging (MRI), or ultrasound.

[0004] All of these methods may use substances that facilitate the delineation or demarcation of particular structures within the subject, said substances being called contrast agents.

[0005] In computed tomography, iodine-containing solutions are usually used as contrast agents.In magnetic resonance imaging (MRI), superparamagnetic substances (e.g., iron oxide nanoparticles, superparamagnetic iron-platinum particles (SIPP)) or paramagnetic substances (e.g., gadolinium chelate, manganese chelate) are usually used as contrast agents.

[0006] Examples of contrast agents can be found in the literature (e.g., ASL Jascinth et al.: Contrast Agents in Computed Tomography: A Review, Journal of Applied Dental and Medical Sciences, 2016, Vol. 2, Issue 2, 143-149; H. Lusic et al.: X-ray-Computed Tomography Contrast Agents, Chem. Rev. 2013, 113, 3, 1641-1666, https: / / www.radiology.wisc.edu / wp-content / uploads / 2017 / 10 / contrast-agents-tutorial.pdf; MR Nough et al.: Radiographic and magnetic resonance contrast agents: Essentials and tips for safe practices, World J Radiol. 2017 Sep 28; 9(9): 339-349; LC Abonyi et al.: Intravascular Contrast Media in Radiography: Historical Development & Review of Risk Factors for Adverse Reactions, South American Journal of Clinical Research, 2016, Vol. 3, Issue 1, 1-10; ACR Manual on Contrast Media, 2020, ISBN: 978-1-55903-012-0, A. Ignee et al.:Ultrasound contrast agents, Endosc Ultrasound. 2016 Nov-Dec; 5(6): 355-362).

[0007] Based on their pattern of spread within tissue, contrast agents can be broadly divided into the following categories: extracellular contrast agents, mixed extracellular / intracellular contrast agents (often simply referred to as intracellular contrast agents), and blood pool contrast agents.

[0008] Extracellular MRI contrast agents include, for example, the gadolinium chelates gadobutrol (Gadovist®), gadoteridol (Prohance®), gadoteric acid (Dotarem®), gadopentetic acid (Magnevist®), and gadodiamide (Omnican®). The highly hydrophilic properties of the gadolinium chelates and their low molecular weight result in rapid diffusion into the interstitial space after intravenous administration. After a certain relatively short circulation period in the blood circulation system, they are excreted via the kidneys.

[0009] Mixed extracellular / intracellular contrast agents are taken up to a certain extent into the cells of tissues and then excreted. Mixed extracellular / intracellular MRI contrast agents based on gadoxetic acid are distinguished, for example, by their proportionally distinct uptake by liver cells (hepatocytes), their accumulation in functional tissue (parenchymal tissue), and their subsequent enhancement of contrast in healthy liver tissue before being excreted in the feces via the gallbladder. Examples of such contrast agents based on gadoxetic acid are described in U.S. Pat. No. 6,039,931 A and are commercially available, for example, under the trade names Primovist® and Eovist®. Another MRI contrast agent with lower uptake into hepatocytes is gadobenate meglumine (Multihance®).

[0010] Blood pool contrast agents, also known as intravascular contrast agents, are distinguished by significantly longer residence times in the blood circulation compared to extracellular contrast agents. Gadofosveset, for example, is a gadolinium-based intravascular MRI contrast agent. It has been used in the trisodium salt monohydrate form (Ablavar®). It binds to serum albumin, thereby achieving a long residence time of the contrast agent in the blood circulation (a half-life in blood of approximately 17 hours). However, Ablavar® was removed from the market in 2017. No other contrast agents approved as blood pool contrast agents for magnetic resonance imaging are commercially available. Similarly, no contrast agents approved as blood pool contrast agents for computed tomography are available on the market.

[0011] Therefore, no commercially available products are approved as blood pool contrast agents for radiological examinations. When generating radiological images with relatively long acquisition / scan times, for example, to depict the vasculature, for example, in free-breathing image acquisition of the chest and abdomen (e.g., diagnosing pulmonary embolism in free-breathing MRI), extracellular contrast agents are relatively quickly removed from the vasculature, meaning that the contrast quickly decreases. However, it would be advantageous to be able to maintain the contrast over a longer period of time. Summary of the Invention [Problem to be solved by the invention]

[0012] The present invention addresses this problem: it provides a means by which radiological images can be simulated based on blood pool contrast agents. [Means for solving the problem]

[0013] In a first aspect, the present invention provides a method for producing a pharmaceutical composition comprising: receiving a series of measured radiological images, the measured radiological images showing an examination region of the examination subject at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels within the examination region, the contrast enhancement of blood vessels in the measured radiological images decreasing with increasing time; calculating a series of artificial radiation images based on the received radiation images, wherein contrast enhancement of blood vessels in the artificial radiation images remains unchanged over time; outputting an artificial radiation image; The present invention provides a computer-implemented method, comprising:

[0014] The present invention provides A receiving unit; a control and calculation unit; an output unit, the control and computing unit is configured to cause the receiving unit to receive a series of measured radiological images, the measured radiological images showing an examination region of the examination subject at different successive time points after administration of a contrast agent, the contrast agent causing contrast enhancement of blood vessels in the examination region, the contrast enhancement of blood vessels in the measured radiological images decreasing with increasing time; a control and calculation unit configured to calculate a series of artificial radiation images based on the measured radiation images, wherein contrast enhancement of said blood vessels in the artificial radiation images remains unchanged over time; There is further provided a computer system, wherein the control and calculation unit is configured to prompt the output unit to output the artificial radiation image.

[0015] The present invention relates to a computer program product comprising a computer program that can be loaded into the memory of a computer, the computer performing the following steps: receiving a series of measured radiological images, the measured radiological images showing an examination region of the examination subject at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels within the examination region, the contrast enhancement of blood vessels in the measured radiological images decreasing with increasing time; calculating a series of artificial radiation images based on the received radiation images, wherein contrast enhancement of blood vessels in the artificial radiation images remains unchanged over time; outputting an artificial radiation image; There is further provided a computer program product for prompting a user to execute the method.

[0016] The present invention relates to the use of a contrast agent in a radiological examination method, the radiological examination method comprising the following steps: administering a contrast agent into a blood vessel of the vascular system to be examined; capturing a series of radiological images of an examination region of an examination object, the radiological images showing the examination region at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels within the examination region, the contrast enhancement of blood vessels in the radiological images decreasing with increasing time; calculating a series of artificial radiation images based on the acquired radiation images, wherein contrast enhancement of blood vessels in the artificial radiation images remains unchanged over time; outputting an artificial radiation image; Further provided is the use of a contrast agent in a radiological examination, comprising:

[0017] The present invention relates to a contrast agent for use in a radiological examination, the radiological examination comprising the steps of: administering a contrast agent into a blood vessel of the vascular system to be examined; capturing a series of radiological images of an examination region of an examination object, the radiological images showing the examination region at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels within the examination region, the contrast enhancement of blood vessels in the radiological images decreasing with increasing time; calculating a series of artificial radiation images based on the acquired radiation images, wherein contrast enhancement of blood vessels in the artificial radiation images remains unchanged over time; outputting an artificial radiation image; Further provided is an imaging agent for use in radiological examinations, comprising:

[0018] The present invention further provides a kit comprising an imaging agent and a computer program product according to the present invention.

[0019] Further subject matters of the invention and preferred embodiments of the invention can be found in the dependent claims, in the description and in the drawings.

[0020] The present invention will be elucidated in more detail below without distinguishing between the subject matter of the invention (method, computer system, computer program product, use, imaging agent for use, kit). In contrast, the following elucidations are intended to apply analogously to all subject matter of the invention, regardless of the context (method, computer system, computer program product, use, imaging agent for use, kit) in which they occur.

[0021] The present invention generates a series of artificial radiological images of an examination region of an examination object, the artificial radiological images showing the examination region after administration of a blood pool contrast agent, even though no blood pool contrast agent was administered. In other words, the present invention simulates a series of artificial radiological images of the examination region after administration of an intravascular contrast agent based on a series of measured radiological images of the examination region. In other words, the present invention generates a series of artificial radiological images based on a series of radiological images showing the examination region of an examination object, showing how the examination region would look if a blood pool contrast agent had been administered. Thus, a radiologist can generate a series of radiological images of an examination region of an examination object that look as if a blood pool contrast agent had been administered to the examination object, without the radiologist administering such an intravascular contrast agent.

[0022] Therefore, the term "artificial radiological image after administration of an intravascular contrast agent" is synonymous with the term "artificial radiological image showing how the examination area looks / will look after administration of an intravascular contrast agent."

[0023] The term "image" is used in this description to refer to both measured radiographic depictions of the examination area and artificially generated (calculated) radiographic depictions.

[0024] A "test subject" is typically an organism, preferably a mammal, and most preferably a human.

[0025] A part of the examination object - the examination area - is the subject of the radiological examination. The "examination area", also called image volume or field of view (FOV), is in particular the volume that is imaged in the radiological image. The examination area is typically defined by a radiologist, for example on an overview image (localizer). Of course, the examination area can alternatively or additionally be defined automatically, for example based on a selected protocol. The examination area can be or include, for example, the liver or part of the liver, the lungs or part of the lungs, the heart or part of the heart, the aorta or part of the aorta, abdominal vessels, leg / pelvic vessels, the esophagus or part of the esophagus, the stomach or part of the stomach, the small intestine or part of the small intestine, the large intestine or part of the large intestine, the abdomen or part of the abdomen, the pancreas or part of the pancreas, and / or other parts of the examination object.

[0026] The radiological examination is preferably an MRI examination, and therefore the at least one (measured) radiological image captured of the examination region is preferably an MRI image, as is the at least one artificially generated radiological image.

[0027] In a further preferred embodiment, the radiological examination is a CT examination, and therefore the at least one (measured) radiological image captured of the examination area is in this embodiment a CT image, and the at least one artificially generated radiological image is likewise a CT image.

[0028] Measured / measurement-generated radiographic images and artificially generated radiographic images can be presented as two-dimensional images showing cross-sectional planes through the object under examination. Radiographic images can be presented as stacks of two-dimensional images, with each image of the stack showing a different cross-sectional plane. Radiographic images can be presented as three-dimensional images (3D images). For simpler illustration, the present invention is explained in some points in the present description based on the existence of two-dimensional radiographic images, but it is not desired in any way to limit the present invention to two-dimensional radiographic images. It will be clear to those skilled in the art that what is respectively described can be applied to stacks of two-dimensional images and 3D images (see, in this connection, for example, M. Reisler, W. Semmler: Magnetresonanztomographie [Magnetic resonance imaging], Springer Verlag, 3rd edition, 2002, ISBN: 978-3-642-63076-7).

[0029] Usually, the measured radiographic images are presented as digital image files. The term "digital" means that the radiographic images can be processed by a machine, generally a computer system. "Processing" is understood to mean known methods for electronic data processing (EDP).

[0030] Digital image files can be represented in various formats. For example, they can be coded as raster graphics. Raster graphics consist of a grid arrangement of so-called picture elements (pixels) or voxels, to which a color or gray value is assigned in each case. Therefore, the main characteristics of 2D raster graphics are the image size (width and height measured in pixels, more informally called image resolution) and color depth. Colors are usually assigned to the pixels of a digital image file. The color coding used for the pixels is specified, among other things, in terms of a color space and a color depth. The simplest case is a binary image, in which pixels store black and white values. In the case of an image, its color is specified in terms of the so-called RGB color space (RGB stands for the primary colors red, green, and blue), where each pixel consists of three subpixels: one for red, one for green, and one for blue. The color of a pixel results from the superposition (additive mixing) of the color values ​​of the subpixels. The color values ​​of the subpixels can be divided into, for example, 256 color nuances, called tone values, which typically range from 0 to 255. The color nuance "0" of each color channel is the darkest. If all three channels have a tone value of 0, the pixel appears black, and if all three channels have a tone value of 255, the corresponding pixel appears white. When implementing the present invention, digital image files (radiographic images) are subject to specific operations. In this regard, the operations primarily affect the pixels or the tone values ​​of individual pixels. There are many possible digital image formats and color codings. For simplicity, in this description, it is assumed that the current image is a grayscale raster graphic with a specific number of pixels, each pixel being assigned a tone value that represents the gray value of the image. However, such assumption should not be understood as limiting in any way. It will be clear to those skilled in the art of image processing how the teachings of the above description can be applied to image files present in other image formats and / or in which color values ​​are coded differently.

[0031] In a first step, a series of measured radiological images are received, which may be T1-weighted, T2-weighted, and / or diffusion-weighted depictions and / or images generated with the aid of different image acquisition sequences.

[0032] The series of measured radiographic images includes at least two radiographic images.

[0033] The term "sequence" means a temporal sequence, i.e., a plurality of (at least two) radiological images showing the examination area at successive time points are generated by the measurement. A time point is assigned to each image, or each image can be assigned a time point. Typically, the time point is the time point at which the image was generated (absolute time). However, it is also conceivable that the radiological images are assigned arbitrary time points (e.g. relative time points).

[0034] Those skilled in the art are aware that the generation of a radiological image takes a certain amount of time. Images may be assigned, for example, a time point of the start of image acquisition or a time point of the completion of image acquisition. The time points allow one radiological image to be placed chronologically in relation to another radiological image, and the time point of a radiological image allows establishing whether a moment shown in a radiological image occurred before or after a moment shown in another radiological image. Preferably, the radiological images are placed in chronological order in the sequence such that an image showing an earlier state of the examination region is placed before an image showing a later state of the examination region in the sequence.

[0035] The time span between two immediately successive images in a sequence is preferably the same for all pairs of immediately successive images in the sequence, i.e. the images are preferably produced at a constant image acquisition rate.

[0036] Preferably, the measured radiographic images of the sequence show the examination region of the subject at different successive time points after administration of a contrast agent, the contrast agent causing contrast enhancement of blood vessels in the measured radiographic images within the examination region, the contrast enhancement of blood vessels in the measured radiographic images decreasing with increasing time.

[0037] The sequence may also include native radiological images (native images), such native images showing the examination region without the administration of a contrast agent.

[0038] The administered contrast agent may be an extracellular and / or mixed extracellular / intracellular contrast agent. In a preferred embodiment, the contrast agent is an extracellular contrast agent. In a further preferred embodiment, the contrast agent is a mixed extracellular / intracellular contrast agent. In a first step, at least one first radiological image of the examination region may be captured without administering a contrast agent (native image). The examination subject is (in a further step) administered with a contrast agent. The contrast agent may be an MRI contrast agent or a CT contrast agent. Preferably, the contrast agent is an extracellular MRI contrast agent such as gadobutrol, gadoteridol, gadoteric acid, gadopentetic acid, and / or gadodiamide. Further extracellular MRI contrast agents are described in the literature (see, for example, Yu. Dong Xiao et al.: MRI contrast agents: Classification and application (Review), International Journal of Molecular Medicine 38: 1326 (2016)).

[0039] In alternative embodiments, the contrast agent is a mixed extracellular / intracellular MRI contrast agent, such as Gd-EOB-DTPA (Primovist®), Mn-DPDP (mangafodipir), Gd-BOPTA (gadobenate meglumine), and / or Gd-DTPA mesoporphyrin (gadophyllin). Additional mixed extracellular / intracellular MRI contrast agents are described in the literature (see, e.g., Yu, Dong Xiao et al.: MRI contrast agents: Classification and application (Review), International Journal of Molecular Medicine 38: 1326 (2016)).

[0040] The contrast agent is preferably introduced into the blood vessels of the subject, for example into a vein in the arm, from where it travels with the blood along the blood circulation system.

[0041] The "circulatory system" is the blood-filled pathways in the body of humans and most animals. It is the system of blood flow formed by the heart and by the network of blood vessels (cardiovascular system, vasculature).

[0042] Blood vessels can be divided into several types based on their structure and function. Arteries transport blood under high pressure and at high flow rates. This allows blood to pass from the heart to various tissues. Arteries branch into arterioles, which have strong muscular walls that act as valve regulators and can constrict (vasoconstriction) or dilate (vasodilation) blood vessels. These further branch into capillaries, which exchange fluids, nutrients, electrolytes, hormones, and other substances between blood and tissues and have thin walls that are permeable to low-molecular-weight substances. In some organs (liver, spleen), capillaries widen and the endothelium becomes discontinuous. This refers to sinusoids. Venules have thin walls and collect blood from capillaries and supply it to veins, which transport blood from the periphery back to the heart. The extracellular contrast agent circulates within the blood circulation for a period of time that depends on the subject, the agent, and the dose, but it is continuously cleared from the blood circulation via the kidneys.

[0043] While the contrast agent spreads and / or circulates within the vasculature of the subject, at least one radiographic image of the vasculature, or a portion thereof, is captured. Preferably, at least one radiographic image is captured of a portion of the vasculature located within the examination region. Multiple radiographic images showing different phases of the spread of the contrast agent in the vasculature, or a portion thereof, (e.g., a distribution phase, an arterial phase, a venous phase, and / or the like) may be captured. The capture of multiple images allows for subsequent differentiation of vessel types.

[0044] The measured radiographic images show the vasculature or parts thereof, in particular parts located within the examination region, with contrast enhancement compared to the surrounding tissue. Preferably, at least one first radiographic image shows arteries with contrast enhancement (arterial phase), while at least one second radiographic image shows veins with contrast enhancement (venous phase).

[0045] The measured radiographic image is used as a basis for generating an artificial radiographic image, which preferably shows the same examination area as the measured radiographic image. If multiple measured radiographic images of the examination area are captured at different time points after administration of the contrast agent, the subsequent radiographic images will show the blood vessels, in particular, with increasingly reduced contrast compared to the surrounding tissue, since the contrast agent is gradually removed from the blood vessels. In contrast, the artificial radiographic image shows the blood vessels with a constant high contrast compared to the surrounding tissue.

[0046] The measured radiographic images are used as a basis for generating artificial radiographic images with the aid of a computer system. It is conceivable that exactly one artificial radiographic image is generated from each measured radiographic image, this artificial radiographic image showing the same examination area as the measured radiographic image, and this artificial radiographic image showing the examination area at the same time point as the measured radiographic image, with the difference being that the contrast enhancement in the measured radiographic image decreases over time, while the contrast enhancement remains unchanged (does not decrease) in the artificially generated radiographic image.

[0047] This can be achieved in different ways.

[0048] In a preferred embodiment, a predictive model is used. The predictive model is trained based on reference data to compensate for the decreasing contrast enhancement of blood vessels over time. The predictive model is trained based on a series of measured radiographic images showing an examination region of an examination subject after administration of an extracellular or mixed extracellular / intracellular contrast agent to generate a series of artificial radiographic images showing the examination region after administration of a blood pool contrast agent. The predictive model is trained based on the reference data for a series of measured radiographic images showing blood vessels in the examination region at different time points after administration of the contrast agent to generate a series of artificial radiographic images showing blood vessels in the examination region with contrast enhancement and with a contrast that remains unchanged over time compared to surrounding tissue.

[0049] The reference data used for training and validating such predictive models typically include radiographic images of the examined region after administration of an extracellular or mixed extracellular / intracellular contrast agent. The reference data may further include radiographic images of the examined region after administration of a blood pool contrast agent. Such reference data may be verified, for example, in clinical studies. An example of an intravascular contrast agent that may be used in such clinical studies is ferumoxytol. Ferumoxytol is a colloidal iron-carbohydrate complex approved for the parenteral treatment of iron deficiency in chronic kidney disease when oral therapy is not feasible. Ferumoxytol is administered intravenously. Ferumoxytol is commercially available as a solution for intravenous injection under the trade names Rienso® or Ferahme®. Iron-carbohydrate complexes exhibit superparamagnetic properties and can therefore be used (off-label) for contrast enhancement in MRI examinations (see, e.g., L.P. Smits et al.: Evaluation of ultrasmall superparamagnetic iron-oxide (USPIO) enhanced MRI with ferumoxytol to quantify arterial wall inflammation, Atherosclerosis 2017, 263: 211-218). Therefore, the present invention further provides for the use of ferumoxytol, or another comparable blood pool contrast agent approved for intravenous injection, as a blood pool contrast agent for the generation of a training data set for the prediction of artificial radiographic images after administration of a blood pool contrast agent based on measured radiographic images after administration of an extracellular or mixed extracellular / intracellular contrast agent. It is also conceivable to use already existing radiographic images after administration of an intravascular contrast agent, e.g., from the time when Ablavar® was still on the market, as training data.

[0050] However, the reference data may also include artificially generated radiographic images in which the decreasing contrast enhancement of blood vessels over time in the measured radiographic images is subsequently compensated for by image processing methods, which are known to those skilled in the art (see, for example, MA Joshi: Digital Image Processing - An Algorithmic Approach, PHI Learning Private Limited, 2nd Edition 2018, ISBN: 978-93-81472-58-7).

[0051] The predictive model can be trained in a supervised learning process to learn the relationship between measured radiographic images and radiographic images after administration of a blood pool contrast agent or images processed using an image processing method. This learned relationship can then be used to calculate an artificial radiographic image for a new measured radiographic image, which shows how the examination region will appear after administration of a blood pool contrast agent, even though an extracellular contrast agent or a mixed extracellular / intracellular contrast agent was administered for the measured radiographic image, and in which blood vessels within the examination region exhibit enhanced contrast that remains unchanged over time compared to surrounding tissue. The predictive model is thus trained to compensate for the decreasing contrast enhancement of blood vessels over time in the measured radiographic images.

[0052] The predictive model may be, for example, an artificial neural network or may include such a network.

[0053] Such an artificial neural network includes at least three layers of processing elements: a first layer having input neurons (nodes), an Nth layer having at least one output neuron (node), and an N-2th inner layer, where N is a natural number greater than 2.

[0054] The input neurons serve to receive the measured (digital) radiographic image as input values. Usually, there is one input neuron for each pixel or voxel of the digital radiographic image. There may be additional input neurons for additional input values ​​(e.g., information about the examination area, information about the examination subject, and / or information about the conditions that are enabled when generating the radiographic image).

[0055] In such a network, the output neurons are responsible for outputting (providing) the artificial radiological image.

[0056] The processing elements of the layer between the input and output neurons are connected to each other in a predetermined pattern with predetermined relevance weights.

[0057] Preferably, the artificial neural network is a so-called convolutional neural network (abbreviated as CNN).

[0058] Convolutional neural networks can process input data in the form of matrices. This allows them to use digital radiographic images, represented as matrices (e.g., width x height x color channels), as input data. In contrast, conventional neural networks, e.g., in the form of multilayer perceptrons (MLPs), require vectors as input, i.e., radiographic images, whose pixels or voxels must be rolled out consecutively in long chains. As a result, conventional neural networks cannot, for example, recognize objects in radiographic images independently of their location in the image. The same object at different locations in the image would have completely different input vectors.

[0059] A CNN essentially consists of alternating filter (convolutional layers) and aggregation (pooling) layers, and finally one or more layers of "normal" fully connected neurons (dense / fully connected layers).

[0060] When analyzing sequences (time sequences of multiple radiological images), space and time can be treated as equivalent dimensions and processed, for example, by 3D convolution, as shown, for example, in the papers by Baccouche et al. (see, for example, Sequential Deep Learning for Human Action Recognition; International Workshop on Human Behavior Understanding, Springer 2011, pages 29-39) and Ji et al. (3D Convolutional Neural Networks for Human Action Recognition, IEEE Transactions on Pattern Analysis and Machine Intelligence, 35(1), 221-231). Furthermore, it is possible to train different networks responsible for time and space and finally fuse the features as described, for example, in publications by Karpathy et al. (see, e.g., Large-scale Video Classification with Convolutional Neural Networks; Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, 2014, pages 1725-1732) and Simonyan & Zisserman (Two-stream Convolutional Networks for Action Recognition in Videos; Advances in Neural Information Processing Systems, 2014, pages 568-576).

[0061] Recurrent neural networks (RNNs) are a family of artificial neural networks that contain feedback connections between layers. RNNs allow for the modeling of time series data through the shared use of parameter data across different parts of the neural network. The architecture of an RNN contains cycles. At least a portion of the output data from the RNN is used as feedback to process subsequent inputs in the sequence, so that the cycle represents the effect of the current value of a variable on its own value at a future time point.

[0062] Further details can be found in the prior art (see, for example, S. Khan et al.: A Guide to Convolutional Neural Networks for Computer Vision, Morgan & Claypool Publishers 2018, ISBN 1681730227, 9781681730226).

[0063] The training of a neural network can be carried out, for example, using the backpropagation method. The aim here for the network is the maximum reliability of the mapping of a given input vector to a given output vector. The mapping quality is described by an error function. The goal is to minimize the error function. In the backpropagation method, the artificial neural network is taught by modifying the relevance weights.

[0064] In the trained state, the association weights between the processing elements contain information about the relationship between measured radiographic images and artificially generated radiographic images that simulate radiographic images after administration of a blood pool contrast agent, which information can be used to predict at least one artificial radiographic image for at least one new measured radiographic image.

[0065] Cross-validation can be used to split the data into a training data set and a validation data set. The training data set is used in backpropagation training of the network weights. The validation data set is used to check the accuracy of predictions that the trained network can make when applied to unknown (novel) radiological images.

[0066] However, a blood pool contrast agent does not necessarily need to be / is not administered to generate the training and validation datasets. It is also conceivable that a different contrast agent, preferably an extracellular contrast agent, may be used to generate the training and validation datasets. Even if the contrast agent is not a blood pool contrast agent, it remains within the vasculature of the subject for a certain period of time. This period may be sufficient for (measurement-based) capture of radiological images showing the region of examination in which blood vessels have high contrast compared to the surrounding tissue. These captured images, after optional processing to compensate for the decrease in contrast enhancement over time, may then be used to train and validate the predictive model.

[0067] As already indicated, further information about the inspection object, inspection area, and / or inspection conditions can be used for training and validating the predictive model and for generating predictions using the predictive model.

[0068] Examples of information about the subject are sex, age, weight, height, medical history, nature and duration and amount of medications already taken, blood pressure, central venous pressure, respiratory rate, serum albumin, total bilirubin, blood glucose level, iron content, vital capacity, and the like, which can be obtained, for example, from a database or electronic patient file.

[0069] Examples of information regarding the area of ​​examination are medical history, surgery, partial resection, liver transplant, liver iron, fatty liver, and the like.

[0070] Preferably, the predictive model is taught to distinguish different blood vessels from one another, for example, to distinguish arteries from veins. This can be done, for example, by radiologists marking each blood vessel differently in the radiological images used for training. It is also conceivable that the predictive model learns to distinguish different blood vessels from one another based on the dynamics of the contrast agent in a series of radiological images after administration. After administration in the form of a bolus, the contrast agent is not immediately present in all blood vessels at the same concentration, but spreads from the site of administration into the vascular system with the blood flow. Depending on the site of administration, the first to pass through is either an artery or a vein. The predictive model can thus learn to distinguish different blood vessels from one another based on the dynamic behavior of the administered contrast agent.

[0071] It is also possible that multiple radiographic images are captured at different time points after administration of the contrast agent and then combined to form a single image in which the blood vessels exhibit uniform, high contrast compared to the surrounding tissue.

[0072] Thus, an artificial radiographic image can also be generated by summing multiple measured radiographic images showing the examination area at different time points after administration of a contrast agent. For example, a first measured radiographic image could show the arterial phase, while a second measured radiographic image could show the venous phase. These two measured radiographic images (and possibly further radiographic images) can be summed. The summation can be performed pixel by pixel or voxel by voxel. For example, the gray values ​​of the pixels can be summed (pair by pair). Subsequent normalization can ensure that the gray values ​​return to a normal range (e.g., 0 to 255).

[0073] If the subject under examination does not move during the acquisition of time-sequential radiographic images, the pixels or voxels of one image correspond exactly to the pixels or voxels of the next and / or previous image, and the corresponding pixels or voxels represent the same examination region at different time points. In such a case, an artificial radiographic image can be calculated by performing the mathematical operations described in this description using pairwise corresponding pixels or voxels. If the subject under examination moves between time-sequential radiographic images, motion correction must be performed before the described calculations are performed. Motion correction methods are described in the prior art (see, for example, EP 3118644, EP 3322997, US 20080317315, US 20170269182, US 20140062481, and EP 2626718).

[0074] Artificial radiation images generated in accordance with the present invention may be displayed on a monitor, output to a printer, and / or stored on a data storage medium.

[0075] Preferably, the artificial radiographic images are automatically generated and output (preferably displayed) in quasi-real time in addition to or instead of the corresponding measured radiographic images.

[0076] It is also conceivable that a vascular model is generated based on the measured radiographic images. The vascular model is a digital representation of the examination object or a part thereof (preferably the examination region) with structures that may be due to blood vessels marked in this representation or due to blood vessels that are only present in this representation. Preferably, the vascular model is a three-dimensional representation in which the spatial course of blood vessels is marked / recorded. Preferably, different types of blood vessels (e.g., arteries and veins) are marked differently.

[0077] In a preferred embodiment, the vascular model is generated based on at least one measured native image and at least one measured radiological image after administration of a contrast agent. The at least one native image shows the examination region of the examination subject without the contrast agent. The at least one radiological image after administration of the contrast agent preferably shows the same region, with some or all of the blood vessels in the region exhibiting contrast enhancement. By comparing the two images, it is possible to identify structures in the radiological image that may be due to blood vessels.

[0078] The vascular model can be generated by subtracting the native image from the measured radiographic image after administration of the contrast agent, followed by subsequent normalization. The subtraction can be performed pixel-by-pixel or voxel-by-voxel. For example, the gray values ​​of pixels can be subtracted from each other. Subsequent normalization ensures that the gray values ​​return to a normal range (e.g., 0 to 255) and that negative gray values ​​are not present.

[0079] If multiple radiographic images after administration of the contrast agent are available, showing the spread of the contrast agent within the blood vessels at different time points, it is possible to identify different blood vessel types (e.g., arteries and veins). This allows for differentiation and different marking of blood vessel types in the vascular model. In such cases, the vascular model can also be generated by summing multiple measured radiographic images after administration of the spreading contrast agent and subsequent normalization. The summation is preferably performed pixel-by-pixel or voxel-by-voxel. For example, pixel gray values ​​can be summed pairwise. Subsequent normalization ensures that the gray values ​​return to a normal range (e.g., 0 to 255).

[0080] Preferably, structures in the vascular model that may not be due to blood vessels are removed, for example, if blood vessels are displayed brightly, all pixels (or voxels) with a gray value below a threshold may be set to a gray value of zero, and in contrast, if blood vessels are displayed darkly, all pixels (or voxels) with a gray value above the threshold may be set to the highest gray value (e.g., 255). Using this procedure, structures that do not originate from blood vessels are reduced (in contrast) or removed entirely.

[0081] Vessel models can also be obtained from measured radiographic images by other segmentation methods, which are widely described in the literature. The following publications may be mentioned as examples: F. Conversano et al.: Hepatic Vessel Segmentation for 3D Planning of Liver Surgery, Acad Radiol 2011, 18: 461-470, S. Moccia et al.: Blood vessel segmentation algorithms - Review of methods, datasets and evaluation metrics, Computer Methods and Programs in Biomedicine 158(2018) 71-91, M. Marcan et al.: Segmentation of hepatic vessels from MRI images for planning of electroporation-based treatments in the liver, Radiol Oncol 2014; 48(3): 267-281, TA Hope et al.: Improvement of Gadoxetate Arterial Phase Capture With a High Spatio-Temporal Resolution Multiphase Three-Dimensional SPGR-Dixon Sequence, Journal of Magnetic Resonance Imaging 38: 938-945 (2013), WO 2009 / 135923, U.S. Pat. No. 6,754,376 B1, WO 2014 / 162273, WO 2017 / 139110, WO 2007 / 053676, EP 2750102 A1).

[0082] Preferably, the vascular model is present in the same digital (data) format as at least one measured radiographic image after administration of a contrast agent and / or at least one native radiographic image. If the same digital format is present, calculations can be more easily performed using associated files, and in particular, the vascular model can be more easily generated from the measured radiographic images.

[0083] The vascular model can be directly used and output as an artificial radiographic image. However, it is also conceivable that one or more measured radiographic images can be superimposed on the vascular model to generate one or more artificial radiographic images. For example, a native image can be superimposed to show blood vessels in the native image. Preferably, different blood vessels can be faded in and out independently of each other. Similarly, at least one measured radiographic image after administration of a contrast agent can also be superimposed on the vascular model. Superimposing a radiographic image after administration of a contrast agent is advantageous, for example, when small focal liver lesions are to be identified in an MRI examination of the liver (see, for example, P. Bannas: Combined Gadoxetic Acid and Gadofosveset Enhanced Liver MRI: A Feasibility and Parameter Optimization Study, Magnetic Resonance in Medicine 75:318-328 (2016)). It can be difficult to distinguish liver lesions from blood vessels in an MRI image. This can be improved by simulating a blood pool contrast agent according to the present invention.

[0084] When superimposing the vessel model onto at least one measured radiological image, it is preferable to select different color values ​​for different types of vessels (e.g., arteries and veins) (pseudocolor display). For example, in the artificial radiological image, it is possible to mark arteries with a first color value (e.g., a color value for red) and veins with a second color value (e.g., a color value for blue).

[0085] Preferably, pixels or voxels depicting blood vessels in the vascular model can be successively faded in into at least one measured radiographic image, for example using a (virtual) slider, and corresponding pixels or voxels in the artificial radiographic image thus generated increasingly assume the color value of the pixels or voxels of the vascular model during fading in. It is also conceivable that vessel types can be faded in independently of each other (e.g., arteries independently of veins and / or veins independently of arteries). The option of switching on and off structures or vessel types arising from vessels is also conceivable instead of or in addition to fading them in.

[0086] In this manner, the radiologist can visualize the vessels or vessel types in the measured radiographic images so that structures within the radiographic images can be assigned.

[0087] Further subjects and embodiments of the present invention: Embodiment 1 receiving at least one radiological image, the at least one radiological image showing an examination area of ​​an examination object; calculating at least one artificial radiation image based on the at least one radiation image, wherein blood vessels in the at least one artificial radiation image are depicted with contrast enhancement compared to surrounding tissue; outputting at least one artificial radiation image; 20. A computer-implemented method, comprising:

[0088] Embodiment 2 receiving at least one measured radiographic image, the at least one measured radiographic image showing an examination area of ​​an examination object; - feeding at least one measured radiographic image to a predictive model, the predictive model being trained on reference data in a supervised learning process to generate at least one artificial radiographic image for the at least one measured radiographic image showing an examination region of the examination object, the at least one artificial radiographic image showing the examination region after administration of a blood pool contrast agent; receiving at least one artificial radiographic image from the predictive model, the at least one artificial radiographic image showing the examination region after administration of a blood pool contrast agent; outputting at least one artificial radiation image; 2. The method of embodiment 1, further comprising:

[0089] Embodiment 3 receiving at least one measured radiological image, wherein the at least one measured radiological image is a radiological image or comprises an image showing an examination area of ​​an examination subject after administration of a contrast agent; feeding at least one measured radiographic image to a predictive model, the predictive model being trained on reference data in a supervised learning process to generate at least one artificial radiographic image for the at least one measured radiographic image showing an examination region of the subject after administration of a contrast agent, the at least one artificial radiographic image showing the examination region after administration of a blood pool contrast agent; receiving at least one artificial radiographic image from the predictive model, the at least one artificial radiographic image showing the examination region after administration of a blood pool contrast agent; outputting at least one artificial radiation image; 3. The method of embodiment 1 or 2 above, further comprising:

[0090] Embodiment 4 receiving a plurality of measured radiological images, the radiological images showing the examination area at different time points after administration of a contrast agent; feeding a plurality of measured radiographic images to a predictive model, the predictive model being trained on reference data in a supervised learning process to generate at least one artificial radiographic image for the plurality of measured radiographic images showing the examination region at different time points after administration of a contrast agent, the at least one artificial radiographic image showing blood vessels in the examination region with contrast enhancement and with contrast that does not change over time compared to surrounding tissue; receiving at least one artificial radiation image from the predictive model, the at least one artificial radiation image showing blood vessels in the examination region with contrast enhancement and with contrast that does not change over time compared to surrounding tissue; outputting at least one artificial radiation image; 4. The method of any one of the above embodiments 1-3, further comprising:

[0091] Embodiment 5 receiving a plurality of measured radiological images, the radiological images showing the examination region at different time points after administration of a contrast agent, wherein blood vessels within the examination region are depicted with contrast enhancement compared to surrounding tissue, the contrast enhancement decreasing over time; generating at least one artificial radiation image showing the same examination area, wherein blood vessels within the examination area are depicted with contrast enhancement compared to surrounding tissue, and the contrast enhancement does not decrease over time; outputting at least one artificial radiation image; 5. The method of any one of embodiments 1-4 above, further comprising:

[0092] Embodiment 6 receiving a plurality of measured radiological images, the radiological images showing the examination area at different time points after administration of a contrast agent; generating an artificial radiation image by summing the received radiation images; outputting at least one artificial radiation image; 6. The method of any one of embodiments 1-5 above, further comprising:

[0093] Embodiment 7 receiving a plurality of measured radiological images, the radiological images showing the examination area at different points in time before and / or after administration of a contrast agent; - generating a vascular model from the received radiological image, the vascular model being a representation of the examination area, and structures that can be attributed to blood vessels in the examination area being marked in the vascular model; generating at least one artificial radiographic image by superimposition of at least one measured radiographic image on a blood vessel model; outputting at least one artificial radiation image; 7. The method of any one of embodiments 1-6 above, further comprising:

[0094] Embodiment 8 8. The method of embodiment 7 above, wherein a native radiographic image of the examination area is superimposed on the vascular model.

[0095] Embodiment 9 8. The method of embodiment 7 above, wherein at least one measured radiographic image of the examination region is superimposed on the vascular model, and the at least one measured radiographic image shows the examination region after administration of a mixed extracellular / intracellular contrast agent, preferably a hepatobiliary contrast agent.

[0096] Embodiment 10 10. The method according to any one of the above embodiments 1-9, wherein different blood vessels in at least one artificial radiological image are marked differently.

[0097] Embodiment 11 A receiving unit; a control and calculation unit; an output unit, the control and computing unit is configured to prompt the receiving unit to receive at least one radiological image, the at least one radiological image showing an examination region of the examination object; The control and calculation unit is configured to calculate at least one artificial radiation image based on the at least one radiation image, wherein blood vessels in the at least one artificial radiation image are depicted with contrast enhancement compared to surrounding tissue; The computer system, wherein the control and calculation unit is configured to prompt the output unit to output at least one artificial radiation image.

[0098] Embodiment 12 the control and calculation unit is configured to prompt the receiving unit to receive at least one first measured radiological image of the vasculature or a portion of the vasculature under examination; the control and computing unit is configured to generate a model of the vasculature or a portion thereof based on the at least one first measured radiological image; the control and calculation unit is configured to prompt the receiving unit to receive at least one second measured radiation image of the examination area of ​​the examination object; the control and computing unit is configured to generate at least one third radiological image by superimposing a model of the vasculature or a portion thereof onto the at least one second radiological image; 12. The computer system of embodiment 11, wherein the control and calculation unit is configured to prompt the output unit to output at least one third radiation image.

[0099] Embodiment 13 A computer program product comprising a computer program that can be loaded into the memory of a computer, the computer program causing the computer to perform the following steps: receiving at least one radiological image, the at least one radiological image showing an examination area of ​​an examination object; calculating at least one artificial radiation image based on the at least one radiation image, wherein blood vessels in the at least one artificial radiation image are depicted with contrast enhancement compared to surrounding tissue; outputting at least one artificial radiation image; A computer program product that prompts the user to execute the program.

[0100] Embodiment 14 1. Use of a contrast agent in a radiological examination, the radiological examination comprising the steps of: administering a contrast agent into a blood vessel of the vascular system to be examined; capturing at least one radiological image of the vasculature or a portion thereof after administration of the contrast agent; generating a model of the vasculature or a portion thereof based on at least one radiological image; generating at least one artificial radiological image by superimposing a model of the vasculature or a portion thereof onto at least one radiological image; outputting at least one artificial radiation image; Use of contrast agents, including

[0101] Embodiment 15 1. A contrast agent for use in a radiological examination, the radiological examination comprising the steps of: administering a contrast agent into a blood vessel of the vascular system to be examined; capturing at least one radiological image of the vasculature or a portion thereof after administration of the contrast agent; generating a model of the vasculature or a portion thereof based on at least one radiological image; generating at least one artificial radiological image by superimposing a model of the vasculature or a portion thereof onto at least one radiological image; outputting at least one artificial radiation image; A contrast agent comprising:

[0102] Embodiment 16 A kit comprising an imaging agent and a computer program product according to the invention as described in embodiment 13 above.

[0103] The present invention will be explained in detail hereinafter with reference to the drawings, without any intention of limiting the invention to the features or combinations of features shown in the drawings.

[0104] 1 shows, in schematic form and by way of example, one embodiment of a computer system according to the invention. The computer system (10) comprises a receiving unit (11), a control and calculation unit (12), and an output unit (13).

[0105] A "computer system" is an electronic data processing system that processes data using programmable computational rules. Such a system typically comprises a control and computation unit, often also referred to as a "computer", which comprises a processor for performing logical operations and a memory for loading computer programs, as well as peripheral devices.

[0106] In computer technology, a "peripheral" refers to any device that is connected to a computer and used to control the computer and / or as an input and output device. Examples include monitors (screens), printers, scanners, mice, keyboards, joysticks, drives, cameras, microphones, speakers, etc. Internal ports and expansion cards are also considered peripherals in computer technology.

[0107] Modern computer systems are often divided into desktop PCs, portable PCs, laptops, notebooks, notebook and tablet PCs, and what are called handhelds (e.g., smartphones), all of which may be used to practice the present invention.

[0108] Input to the computer system (e.g., for user control) is achieved via input means, such as, for example, a keyboard, a mouse, a microphone, a touch-sensitive display, and / or the like. Output is achieved via an output unit (13), which may in particular be a monitor (screen), a printer, and / or a data storage medium.

[0109] The computer system (10) according to the invention is configured to receive measured radiation images and to generate (calculate) artificial radiation images based on the received radiation images.

[0110] The control and calculation unit (12) is responsible for controlling the receiving unit (11) and the output unit (13), for coordinating the flow of data and signals between the various units, for processing the radiological images and for generating the artificial radiological images. It is conceivable that there may be multiple control and calculation units.

[0111] The receiving unit (11) serves to receive the radiation image. The radiation image can be transmitted, for example, from a magnetic resonance imaging system, a computed tomography system, or can be read from a data storage medium. The magnetic resonance imaging system or the computed tomography system can be a component of the computer system according to the invention. However, it is also conceivable that the computer system according to the invention is a component of the magnetic resonance imaging system or the computed tomography system. The radiation image can be transmitted via a network connection or a direct connection. The radiation image can be transmitted via wireless communication (WLAN, Bluetooth, mobile communication, and / or the like) and / or via a cable. It is conceivable that there are multiple receiving units. The data storage medium can also be a component of the computer system according to the invention or can be connected to the computer system according to the invention, for example, via a network. It is conceivable that there are multiple data storage media.

[0112] The radiological image and possibly further data (such as, for example, information about the examination subject, image acquisition parameters, and / or the like) are received by the receiving unit and transmitted to the control and calculation unit.

[0113] The control and computing unit is configured to generate an artificial radiation image based on the received data.

[0114] Via the output unit (13), the artificial radiation image can be displayed (e.g. on a monitor), output (e.g. by a printer) or stored on a data storage medium. It is conceivable that there are multiple output units.

[0115] FIG. 2 shows, by way of example and in schematic form, in the form of a flow chart, the steps performed by one embodiment of a method (100) according to the invention or a computer program product according to the invention.

[0116] The steps are: (110) receiving a series of measured radiological images, the measured radiological images showing an examination region of the examination subject at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels in the examination region, the contrast enhancement of blood vessels in the measured radiological images decreasing with increasing time; (120) calculating a series of artificial radiation images based on the received radiation images, wherein contrast enhancement of blood vessels in the artificial radiation images remains unchanged over time; (130) Step of outputting an artificial radiation image is.

[0117] FIG. 3 shows, by way of example and in schematic form, in the form of a flow chart the steps carried out by a preferred embodiment of the method (200) according to the invention or by a computer program product according to the invention.

[0118] The steps are: (210) receiving a series of measured radiological images, the measured radiological images showing an examination region of the examination subject at different successive time points after administration of a contrast agent, the contrast agent producing vascular contrast enhancement in the measured radiological images of the examination region, the vascular contrast enhancement in the measured radiological images decreasing with increasing time; (220) feeding the radiological image to an artificial neural network, the artificial neural network being trained based on reference data in a supervised learning process to compensate for decreasing contrast enhancement of blood vessels over time in the radiological image; (230) receiving a series of calculated radiographic images from the artificial neural network, the calculated radiographic images showing the examination area at different successive time points, blood vessels in the calculated radiographic images being depicted with contrast enhancement, and the contrast enhancement of the blood vessels remaining constant as time increases; (240) Step of outputting the calculated radiographic image is.

[0119] Figures 4(a), (b), and (c) show, by way of example and in schematic form, radiographic images of the liver after intravenous administration of a contrast agent into a vein in the arm of the subject. In Figures 4(a), 4(b), and 4(c), the same cross-section through the liver (L) is always depicted at different successive time points. The reference numerals entered in Figures 4(a), 4(b), and 4(c) apply to all of Figures 4(a), 4(b), and 4(c), and they are entered only once each for the sake of clarity only.

[0120] In Figures 4(a), 4(b), and 4(c), arteries (A) and veins (V) are depicted with contrast enhancement relative to the surrounding tissue (liver cells). However, the contrast enhancement decreases over time from Figure 4(a) through Figure 4(b) to Figure 4(c).

[0121] FIG. 5 shows, in schematic form, by way of example, the generation of artificial radiographic images based on measured radiographic images with the aid of a prediction model (PM). Liver radiographic images (a), (b), and (c) depicted in FIG. 5 correspond to the liver images depicted in FIGS. 4(a), 4(b), and 4(c). The measured radiographic images (a), (b), and (c) are fed to the prediction model (PM). The prediction model generates three artificial radiographic images (a'), (b'), and (c') from the three measured radiographic images (a), (b), and (c). While the contrast enhancement of blood vessels (arteries A and veins V) decreases over time in the measured radiographic images, it remains unchanged over time in the artificially generated radiographic images.

[0122] Figures 6(a), 6(b), and 6(c) show, by way of example and in schematic form, radiological images of the liver before (6(a)) and after (6(b), 6(c)) intravenous administration of a contrast agent into a vein in the arm of the subject. In Figures 6(a), 6(b), and 6(c), the same cross-section through the liver (L) is always depicted at different successive time points. The reference numerals entered in Figures 6(a), 6(b), and 6(c) apply to all of Figures 6(a), 6(b), and 6(c), and they are entered only once each for purposes of clarity only.

[0123] Figure 6(a) shows a cross section through the liver (L) before intravenous administration of contrast agent. At a time point between the time points depicted by Figures 6(a) and 6(b), contrast agent was administered intravenously as a bolus. This reaches the liver via the hepatic artery (A) in Figure 6(b). The hepatic artery is therefore depicted with signal enhancement (arterial phase). At the time point depicted in Figure 6(c), contrast agent reaches the liver via the veins (venous phase).

[0124] Figure 6(a) is therefore the native radiographic image, Figure 6(b) is the first radiographic image after administration of contrast agent, and Figure 6(c) is the second radiographic image after administration of contrast agent. In Figure 6(b), the arteries are particularly visible, while in Figure 6(c), the veins are particularly visible.

[0125] FIG. 7 shows, by way of example and in schematic form, the generation of an artificial radiographic image (AI) based on measured radiographic images with the aid of a predictive model (PM). The predictive model (PM) is trained to generate, for at least one measured radiographic image showing an examination region of an object, at least one artificial radiographic image showing the examination region after administration of an intravascular contrast agent. In this example, the radiographic images from FIGS. 6(b) and 6(c) are fed into the predictive model (PM). The predictive model then automatically generates an artificial radiographic image (AI), which shows all blood vessels (arteries A, veins V) with contrast enhancement and with contrast that does not change over time compared to the surrounding tissue.

[0126] Figure 8 shows, by way of example and in schematic form, the generation of a vascular model from measured radiographic images. Figures 8(a), 8(b), and 8(c) are identical to Figures 6(a), 6(b), and 6(c).

[0127] The native radiographic image of FIG. 8(a) is combined with the radiographic image of FIG. 8(b) and the radiographic image of FIG. 8(c) to form a vascular model (FIG. 8(d)). This can be done, for example, by generating a difference image of FIG. 8(a) and FIG. 8(b) in a first step (FIG. 8(b)-FIG. 8(a)). In such a difference image, the artery (A) stands out particularly strongly, while all other structures disappear into the background. In a further step, a difference image of FIG. 8(a) and FIG. 8(c) can be generated (FIG. 8(c)-FIG. 8(a)). In such a difference image, the vein (V) stands out particularly strongly, while all other structures disappear into the background. In a further step, the two generated difference images can be combined, for example by summation, to form the vascular model (FIG. 8(d)). Preferably, the arteries and veins are marked differently in the vascular model (FIG. 8(d)) (in this case, the veins are marked with horizontal shading and the arteries with vertical shading).

[0128] FIG. 9 shows, by way of example and in schematic form, a measured radiological image of the liver (L) after intravenous administration of a hepatobiliary contrast medium into a vein in the arm of the examined subject. The hepatobiliary contrast medium is taken up by healthy liver cells. The radiological image shown in FIG. 9 shows a cross-section of the liver in the hepatobiliary phase, when the liver cells have already taken up the contrast medium. Structures T are visible, and it is unclear whether said structures are blood vessels or tumors.

[0129] FIG. 10 shows, by way of example and in schematic form, the superposition of a vessel model onto a measured radiographic image to form an artificial radiographic image.

[0130] FIG. 10(a) shows a measured radiographic image of the liver (L) in cross section. FIG. 10(a) is identical to FIG. 9. FIG. 10(b) shows a vascular model. FIG. 10(b) is identical to FIG. 8(d). FIG. 10(c) shows an artificial radiographic image. In the artificial radiographic image, pixels of those structures in the vascular model that can be attributed to blood vessels are replaced with the corresponding pixels in the measured radiographic image. In the artificial radiographic image, it is easily visible which structures can be attributed to healthy liver cells, which structures can be attributed to arteries (A), and which structures can be attributed to veins (V). Furthermore, it is visible in the artificial radiographic image that structure T is not a blood vessel. It is believed that a tumor is present.

Claims

1. receiving a series of measured radiological images, the measured radiological images showing an examination region of an examination subject at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels in the examination region, the contrast enhancement of the blood vessels in the measured radiological images decreasing with increasing time; calculating a series of artificial radiation images based on the received radiation images, wherein the contrast enhancement of the blood vessels in the artificial radiation images remains unchanged over time; outputting the artificial radiation image; Including, The calculation of the series of artificial radiation images comprises the steps of: feeding the measured radiological images to a predictive model, the predictive model being trained on reference data in a supervised learning process to generate a series of artificial radiological images for a series of measured radiological images showing an examination region of an examination object after administration of a contrast agent, the artificial radiological images showing the examination region after administration of a blood pool contrast agent; receiving a series of artificial radiological images from the predictive model, the artificial radiological images showing the examination region after administration of a blood pool contrast agent; 20. A computer-implemented method, comprising:

2. The calculation of the series of artificial radiation images comprises the steps of: feeding the measured radiographic images to a predictive model, the predictive model having been trained on reference data in a supervised learning process to compensate for decreasing vascular contrast enhancement over time; receiving a series of artificial radiation images from the predictive model, the artificial radiation images showing the examination area at different successive time points, blood vessels in the artificial radiation images being depicted with contrast enhancement, the contrast enhancement of the blood vessels in the artificial radiation images remaining constant as time increases; The method of claim 1 , comprising:

3. The calculation of the series of artificial radiation images comprises the steps of: feeding the measured radiological images to a predictive model, the predictive model being trained on reference data in a supervised learning process to generate a series of artificial radiological images for a series of measured radiological images showing an examination area at different time points after administration of a contrast agent, the artificial radiological images showing blood vessels in the examination area with contrast enhancement and with contrast that does not change over time compared to surrounding tissue; receiving a series of artificial radiation images from the predictive model, the artificial radiation images showing blood vessels in the examination region with contrast enhancement and with contrast that does not change over time compared to surrounding tissue; The method of claim 1 , comprising:

4. The calculation of the series of artificial radiation images comprises the steps of: generating a vascular model from the received radiological image, the vascular model being a representation of the examination area, and structures within the examination area that may be due to blood vessels being marked in the vascular model; generating a series of artificial radiation images by superimposing the blood vessel model on the received radiation images; The method of claim 1 , comprising:

5. The method of any one of claims 1 to 4, wherein the administered imaging agent is an extracellular imaging agent or a mixed extracellular / intracellular imaging agent.

6. The method according to any one of claims 1 to 5, wherein the measured radiological image is an MRI image.

7. The method according to any one of claims 1 to 6, wherein the predictive model is an artificial neural network.

8. 8. The method according to claim 1, wherein the reference data comprises radiographic images measured after administration of an extracellular contrast agent and radiographic images measured after administration of an intravascular contrast agent, and / or artificially generated radiographic images.

9. A receiving unit; a control and calculation unit; An output unit; A computer system comprising: the control and computing unit is configured to cause the receiving unit to receive a series of measured radiological images, the measured radiological images showing an examination region of the examination object at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels in the examination region, the contrast enhancement of the blood vessels in the measured radiological images decreasing with increasing time; the control and calculation unit is configured to calculate a series of artificial radiation images based on the measured radiation images, and the contrast enhancement of the blood vessels in the artificial radiation images remains unchanged over time; the control and calculation unit is configured to prompt the output unit to output an artificial radiation image; The calculation of the series of artificial radiation images comprises the steps of: feeding the measured radiological images to a predictive model, the predictive model being trained on reference data in a supervised learning process to generate a series of artificial radiological images for a series of measured radiological images showing an examination region of an examination object after administration of a contrast agent, the artificial radiological images showing the examination region after administration of a blood pool contrast agent; receiving a series of artificial radiological images from the predictive model, the artificial radiological images showing the examination region after administration of a blood pool contrast agent; 2. A computer system comprising:

10. A computer program that can be loaded into the memory of a computer, the computer being configured to perform the following steps: receiving a series of measured radiological images, the measured radiological images showing an examination region of an examination subject at different successive time points after administration of a contrast agent, the contrast agent providing contrast enhancement of blood vessels in the examination region, the contrast enhancement of the blood vessels in the measured radiological images decreasing with increasing time; calculating a series of artificial radiation images based on the received radiation images, wherein the contrast enhancement of the blood vessels in the artificial radiation images remains unchanged over time; outputting the artificial radiation image; Encourage them to carry out the following: The calculation of the series of artificial radiation images comprises the steps of: feeding the measured radiological images to a predictive model, the predictive model being trained on reference data in a supervised learning process to generate a series of artificial radiological images for a series of measured radiological images showing an examination region of an examination object after administration of a contrast agent, the artificial radiological images showing the examination region after administration of a blood pool contrast agent; receiving a series of artificial radiological images from the predictive model, the artificial radiological images showing the examination region after administration of a blood pool contrast agent; a computer program comprising:

11. A kit comprising a contrast agent and a computer program according to the invention as claimed in claim 10.

Images