Providing synthetic contrast scenarios

Abstract

This invention relates to a computer-executed method for providing a synthetic contrast imaging scene, comprising: b.1) receiving preoperative image data of an examination region having a hollow organ, which images a contrast agent flow within the hollow organ; b.2) receiving intraoperative image data of an examination region of an examination subject, which images a medical object at least partially disposed within the hollow organ; b.3) generating a synthetic contrast imaging scene by applying a trained function to input data, said input data being based on the preoperative and intraoperative image data, wherein the synthetic contrast imaging scene images a virtual contrast agent flow within the hollow organ, taking into account the medical object disposed within the hollow organ, at least one parameter of the trained function being based on a comparison between a training scene and a comparison scene; and b.4) providing the synthetic contrast imaging scene. This invention relates to a computer-executed method for providing a trained function, a providing unit, a medical imaging device, a training unit, and a computer program product.

Description

Technical Field

[0001] The present invention relates to a computer-executed method for providing a synthetic imaging scene, a computer-executed method for providing a trained function, a providing unit, a medical imaging device, a training unit, and a computer program product. Background Technology

[0002] To detect changes over time in an area of ​​an examination subject, such as in human and / or animal patients, X-ray-based imaging methods are typically used. The changes to be detected over time may include, for example, the propagation and / or flow of contrast agents, particularly contrast agent streams and / or contrast agent blobs, within hollow organs of the subject, such as vascular segments.

[0003] X-ray-based imaging methods typically include digital subtraction angiography (DSA), in which at least two X-ray images acquired in chronological order are subtracted, the X-ray images at least partially imaging the entire examination area. Furthermore, DAS typically distinguishes between a mask phase for acquiring at least one mask image and a fill phase for acquiring at least one fill image. Here, the mask image typically images the examination area without contrast agent. The fill image images the examination area during the period when contrast agent is applied to the examination area. As a result of DSA, a difference image is typically provided, obtained by subtracting the mask image from the fill image. This difference image can usually reduce and / or remove components that are not important for treatment and / or diagnosis, and those components are particularly time-invariant.

[0004] To image the temporal dynamics of contrast agent flow in the examined area, such as propagation direction, multiple time-sequential filled images are typically captured and combined with a mask image to form a time-resolved difference image. Preoperative DSA is usually performed to plan treatment for vascular malformations and / or embolisms and / or stenosis. Because this temporal dynamics of contrast agent flow can be altered by the placement of medical objects, such as catheters and / or endoscopes, into hollow organs, additional DSA is often performed intraoperatively to monitor treatment. Placement of medical objects can particularly cause deformation of the hollow organ and / or surrounding tissues, thereby potentially altering the temporal dynamics of contrast agent flow. Disadvantageously, current DSA methods can only image this altered temporal dynamics by re-imagening X-rays while contrast agent is being administered. This additionally subjectes the examined subject to higher doses of X-rays and / or contrast agent. Summary of the Invention

[0005] Therefore, the technical problem to be solved by the present invention is to provide the temporal dynamics of the contrast agent flow in the hollow organ during surgery as a result of the medical subject.

[0006] The technical problem is solved according to the present invention by a computer-executed method for providing a synthetic imaging scene, a computer-executed method for providing a trained function, a providing unit, a medical imaging device, a training unit, and a computer program product.

[0007] The following describes solutions to the aforementioned technical problems according to the present invention, relating both to methods and apparatus for providing synthetic imaging scenes and to methods and apparatus for providing trained functions. Here, the features, advantages, and alternative implementations of data structures and / or functions in the methods and apparatus for providing synthetic imaging scenes can be similarly applied to data structures and / or functions in the methods and apparatus for providing trained functions. Similar data structures are indicated herein, in particular, by using the prefix "training". Furthermore, the trained functions used in the methods and apparatus for providing synthetic imaging scenes can be adjusted and / or provided, in particular, by means of the methods and apparatus for providing trained functions.

[0008] The invention relates in a first aspect to a computer-executed method for providing a synthetic contrast imaging scene. In this method, in a first step b.1), preoperative image data of an examination region of a subject is received. This examination region has a hollow organ. Furthermore, the medical image data images a contrast agent flow within the hollow organ. In a second step b.2), intraoperative image data of the examination region of the subject is received. This intraoperative image data images a medical object at least partially disposed within the hollow organ. In a third step b.3), a synthetic contrast imaging scene is generated by applying a trained function to the input data, wherein the input data is based on the preoperative and intraoperative image data. Furthermore, the synthetic contrast imaging scene images a virtual contrast agent flow within the hollow organ, taking into account the medical object at least partially disposed within the hollow organ. Advantageously, the synthetic contrast imaging scene represents a time-resolved imaging of the virtual contrast agent flow within the hollow organ, wherein the virtual contrast agent flow has a predefined contrast, particularly contrast values ​​and / or intensity curves, with respect to the anatomical structure and / or tissue of the subject. The predefined contrast of the virtual contrast agent flow can be particularly consistent with the contrast of the real contrast agent, especially the contrast of the contrast agent flow. Because the synthetic contrast scene images a virtual contrast agent flow in the hollow organ during surgery, this scene can be called synthetic contrast.

[0009] Furthermore, at least one parameter of the trained function is based on a comparison between the training scene and the comparison scene. In step four (b.4), a synthetic imaging scene is provided.

[0010] Receiving preoperative and / or intraoperative image data may in particular include detecting and / or reading computer-readable data storage and / or receiving from data storage units, such as databases. Furthermore, preoperative and / or intraoperative image data may be provided by a providing unit of the medical imaging equipment.

[0011] The subject of examination may be, for example, a human and / or animal patient. Furthermore, the examination area of ​​the subject may be a spatial segment, particularly a volumetric segment, of the subject having hollow organs. These hollow organs may, for example, include lung and / or vascular segments, such as arteries and / or veins, and / or the heart.

[0012] Preoperative image data can advantageously image the examination area, particularly hollow organs, at a time point earlier than intraoperative image data, prior to surgery. Here, preoperative image data can include two-dimensional (2D) and / or three-dimensional (3D) imaging of the examination area. Furthermore, the preoperative image data images the examination area temporally. Thus, the preoperative image data can advantageously image the contrast agent flow within the hollow organ temporally. The contrast agent flow can, for example, include the flow and / or propagation of the contrast agent within the hollow organ. Furthermore, the preoperative image data can have multiple image points, particularly pixels and / or volume pixels, wherein each image point can have a time-intensity curve. For example, the contrast agent flow can be imaged and / or detected by exceeding and / or falling below a preset intensity threshold in the time-intensity curve. Additionally, the preoperative image data can have metadata, which can include, for example, information about the imaging parameters and / or operating parameters of the medical imaging equipment and / or information about the contrast agent flow at the time point at which the preoperative image data was acquired.

[0013] Intraoperative imaging data can advantageously image the examination area, particularly hollow organs, at a later point in time than preoperative imaging data. Here, intraoperative imaging data can include 2D and / or 3D imaging of the examination area. Furthermore, intraoperative imaging data can advantageously image medical objects at least partially disposed within the hollow organ. These medical objects can particularly include surgical and / or diagnostic instruments, such as guide wires and / or catheters and / or laparoscopes and / or endoscopes.

[0014] Here, the medical imaging equipment used to capture intraoperative image data may be the same as or different from the medical imaging equipment used to capture preoperative image data.

[0015] Preoperative and / or intraoperative image data may, for example, include at least one projection image, particularly an X-ray projection image, and / or a slice image of the examination area. Here, the preoperative image data may image the examination area according to digital subtraction angiography (DSA), wherein the preoperative image data may include at least one mask image, at least one filled image, and / or at least one difference image. Furthermore, the preoperative and / or intraoperative image data may image the examination area along at least one, particularly multiple, and / or different spatial directions, particularly projection directions and / or angles. Additionally, the preoperative image data may have time-resolved 3D imaging of the examination area, for example, according to 4D-DSA, particularly time-resolved 3D-DSA.

[0016] In step b.3), a synthetic contrast scene can advantageously be generated by applying a trained function to the input data. The synthetic contrast scene can have temporally resolved 2D and / or 3D imaging of the examination area, particularly intraoperatively. The synthetic contrast scene can particularly include a sequence of individual imaging sequences, wherein a virtual contrast agent flow is imaged in the synthetic contrast scene, particularly in the individual imaging sequences. Here, the synthetic contrast scene can be at least partially consistent with the intraoperative image data, particularly in terms of imaging of the examination area. In particular, the spatial and / or temporal trajectories of the hollow organs and / or medical objects in the corresponding imaging of the synthetic contrast scene can be consistent. Furthermore, the synthetic contrast scene can image a virtual contrast agent flow within the hollow organs, taking into account the medical objects arranged within them. Medical objects arranged intraoperatively within hollow organs may cause changes in the hydrodynamics within the hollow organs. Synthetic imaging scenes can advantageously image hollow organs, particularly intraoperatively, where a virtual contrast agent flow can be imaged temporally, taking into account the altered hydrodynamics due to the medical object positioned within the hollow organ. For this purpose, the synthetic imaging scene can advantageously have multiple image points, particularly pixels and / or volume pixels, each image point potentially possessing a temporal intensity profile. For example, a virtual contrast agent flow can be imaged by changes in the temporal intensity profile, particularly exceeding and / or falling below a preset intensity threshold.

[0017] The trained function can be advantageously trained using machine learning methods. The trained function can be, in particular, a neural network, especially a convolutional neural network (CNN) or a network that includes convolutional layers.

[0018] A trained function images (or maps) input data to output data. Here, the output data may also depend, in particular, on one or more parameters of the trained function. These one or more parameters can be determined and / or adjusted through training. The determination and / or adjustment of one or more parameters of the trained function can be based, in particular, on a pair of training input data and corresponding training output data, wherein the trained function is used to generate training image data (or training mapping data) from the training input data. The determination and / or adjustment can be based, in particular, on a comparison between the training image data and the training output data. Generally, a function that can be trained, i.e., a function with one or more parameters that have not yet been adjusted, can also be referred to as a trained function.

[0019] Other terms for a trained function include trained imaging rule, imaging rule with trained parameters, function with trained parameters, artificial intelligence-based algorithm, and machine learning algorithm. An example of a trained function is an artificial neural network, where the edge weights of the artificial neural network correspond to the parameters of the trained function. The term "neural network" can also be used instead of "neural network." A trained function can also be, in particular, a deep artificial neural network. Another example of a trained function is a "support vector machine," and other machine learning algorithms can also be used as trained functions.

[0020] The trained function can be trained, in particular, using backpropagation. First, training imaging data is determined by applying the trained function to the training input data. Then, the deviation between the training imaging data and the training output data is determined by applying an error function to both the training imaging data and the training output data. Furthermore, the parameters of the trained function, especially the parameters of the neural network, particularly the weights, can be iteratively adjusted based on the gradient of the error function with respect to at least one parameter of the trained function. This advantageously minimizes the deviation between the training imaging data and the training output data during the training of the trained function.

[0021] The trained function, especially a neural network, advantageously has an input layer and an output layer. Here, the input layer can be designed to receive input data. Furthermore, the output layer can be designed to provide imaging data. Here, the input layer and / or output layer can each include multiple channels, especially neurons.

[0022] At least one parameter of the trained function may preferably be based on a comparison between a training scenario and a comparison scenario. Here, the training scenario and / or the comparison scenario may advantageously be determined as part of the proposed computer-executed method for providing the trained function, which is described below in the specification. The trained function can be provided, in particular, through one embodiment of the proposed computer-executed method for providing the trained function.

[0023] Furthermore, in step b.4), providing the composite imaging scene may, for example, include storing it on a computer-readable storage medium and / or displaying it on a display unit and / or transmitting it to the providing unit. In particular, a graphical view of the composite imaging scene may be displayed on the display unit.

[0024] The proposed method enables the provision of a virtual contrast agent flow during surgery within a hollow organ containing a medical object, exhibiting temporal dynamics, and particularly hydrodynamics. This allows for the advantageous elimination of intraoperative contrast agent administration. Furthermore, it can assist users, such as medical operators, when planning repositioning of the medical object, particularly changes in arrangement, and / or when pre-setting the operating parameters of the medical object.

[0025] In another advantageous embodiment of the proposed method for providing a synthetic imaging scene, the synthetic imaging scene may include time-resolved 3D imaging of a virtual contrast agent flow taking into account a medical object at least partially arranged in a hollow organ.

[0026] Advantageously, preoperative image data can be used to image the examination area in 2D with time resolution from at least two different spatial directions, particularly projection directions and / or angles. Similarly, intraoperative image data can be used to image the examination area in 2D with time resolution from at least two different spatial directions, particularly projection directions and / or angles. Alternatively or supplementarily, preoperative and / or intraoperative image data can have time-resolved 3D imaging of the examination area. This allows for the generation of a synthetic contrast scene by applying a trained function to the input data, the synthetic contrast scene including time-resolved 3D imaging of a virtual contrast agent flow. Therefore, the synthetic contrast scene can advantageously image the volumetric locality of the examination area in three dimensions with time resolution. This enables the detection of three-dimensional curvatures of hollow organs and / or medical objects. Furthermore, virtual contrast agent flows can be detected, particularly distinguished, in overlapping and / or branching vascular segments. The time-resolved 3D imaging of the virtual contrast agent flow can be particularly compatible with and / or feature 4D-DSA imaging.

[0027] Once the intraoperative image data is used to temporally and three-dimensionally image the examination area, a synthetic contrast scene, including a temporally resolved 3D image of a virtual contrast agent flow, can be overlaid with the intraoperative image data and / or, in particular, weighted and averaged. This allows for the advantageous integration of the virtual contrast agent flow into the intraoperative image data, for example, within a shared graphical view.

[0028] In another advantageous embodiment of the proposed method for providing a synthetic imaging scene, the synthetic imaging scene may include at least one synthetic time-resolved 2D image of a virtual contrast agent flow taking into account a medical object at least partially arranged in a hollow organ.

[0029] Here, the synthetic contrast scene can be temporally and 2D imaged along the projection direction of preoperative and / or intraoperative image data to represent a virtual contrast agent flow within the hollow organ. As long as the intraoperative image data 2D images the examination area, the synthetic contrast scene, including the temporally resolved 2D image of the virtual contrast agent flow, can be overlaid with the intraoperative image data and / or, in particular, weighted and averaged. This advantageously integrates the virtual contrast agent flow into the intraoperative image data, for example, in a common graphical view. In cases where the imaging geometry used to capture the intraoperative image data changes, the generation of the synthetic contrast scene can be advantageously repeated by applying a trained function to the input data. This ensures that the projection direction of the synthetic contrast scene is always consistent with the projection direction used for imaging the intraoperative image data.

[0030] In another advantageous embodiment of the proposed method for providing a synthetic imaging scene, the at least one synthetic time-resolved 2D image can be generated by virtual projection of the time-resolved 3D image.

[0031] The generation of time-resolved 2D imaging can advantageously involve virtually projecting 3D images onto a virtual detector along the projection direction of preoperative and / or intraoperative image data. The imaging geometry of the virtual projection can, in particular, be consistent with the imaging geometry used to capture the preoperative and / or intraoperative image data. Because time-resolved 2D imaging is generated through virtual projection of 3D imaging, the projection direction can be adjusted or adapted computationally, for example, to adapt to changes in the projection direction used to capture intraoperative image data. Furthermore, the user can, in particular, dynamically adjust the projection direction and / or imaging geometry used to generate time-resolved 2D imaging via input from the input unit. This enables the particularly intuitive generation and / or graphical display of synthetic scenes, especially time-resolved 2D imaging.

[0032] In another advantageous embodiment of the proposed method for providing synthetic imaging scenarios, the input data may also be based on the material parameters and / or operational parameters and / or shape information of the medical object and / or on the physiological parameters of the object being examined.

[0033] Material parameters may include, for example, material information about the medical object, particularly regarding its porosity and / or rigidity and / or flexibility and / or compressibility. Furthermore, operational parameters may include information about the medical object, particularly its current location, and / or orientation. Additionally, shape information may include information about the medical object's diameter and / or perimeter and / or cross-section and / or surface, particularly its volume model and / or volume mesh model, and / or curvature and / or spatial orientation. Moreover, shape information may advantageously include information about the surface characteristics of the medical object, particularly information about depressions and / or protrusions on the surface of the medical object. Material parameters and / or operational parameters and / or shape information can advantageously describe the state of the medical object at a given point in time during the acquisition of intraoperative image data.

[0034] In addition, physiological parameters may include, for example, respiratory and / or kinematic and / or pulse and / or blood pressure information of the subject, particularly at the time points when the preoperative and / or intraoperative image data were acquired. As long as the physiological parameters reflect the subject's condition during the acquisition of the preoperative image data, the contrast agent flow imaged in the preoperative image data can be characterized with particular precision.

[0035] Material parameters and / or operational parameters and / or shape information of the medical object and / or physiological parameters of the object being examined can be received, particularly in steps b.1) and / or b.2). Receiving material parameters and / or operational parameters and / or shape information of the medical object and / or physiological parameters of the object being examined can particularly include detecting and / or reading from a computer-readable data storage device and / or receiving from a data storage unit, such as a database. Furthermore, material parameters and / or operational parameters and / or shape information can be provided by the medical object, particularly by its processing unit. Additionally, the physiological parameters of the object being examined can be provided by sensors used to detect physiological parameters, such as respiratory sensors and / or motion sensors and / or pulse sensors and / or blood pressure sensors.

[0036] By increasing the material parameters and / or operational parameters and / or shape information of the medical object and / or the physiological parameters of the object being examined, it is possible to advantageously adapt the synthetic imaging scene, especially the virtual contrast agent flow, to the medical object and / or the object being examined, especially to its current state, with high accuracy and particularly good performance.

[0037] In another advantageous embodiment of the proposed method for providing a synthetic contrast imaging scenario, the input data may also be based on parameters of the contrast agent flow and / or a virtual contrast agent flow. Here, these parameters may be preset or specified as the dose and / or velocity and / or direction of the virtual contrast agent flow.

[0038] Parameters of the contrast agent flow can be provided, for example, by the device used to control the injection of the contrast agent at the time point when preoperative image data was acquired. Alternatively, parameters of the contrast agent flow can be determined from the preoperative image data, for example, by means of methods for bolus tracking and / or analysis of the time-intensity profiles of the preoperative image data. Parameters of the contrast agent flow may include, for example, information about the dose and / or injection rate and / or flow velocity and / or viscosity and / or injection location of the contrast agent flow at the time point when the preoperative image data was acquired.

[0039] The parameters of the virtual contrast agent flow can be advantageously input and / or adjusted, particularly dynamically, via user input, especially through input units. Furthermore, a synthetic imaging scene, particularly a virtual contrast agent flow, can be generated for a preset range of values ​​for the parameters of the virtual contrast agent flow, wherein the value range can include multiple different values ​​for the virtual contrast agent flow parameters. The parameters of the virtual contrast agent flow can, for example, be preset for the dose and / or injection rate of the virtual contrast agent flow. Additionally, the parameters of the virtual contrast agent flow can be preset for the movement velocity of the virtual contrast agent flow, particularly the movement velocity and / or propagation velocity and / or viscosity of the virtual contrast agent mass. The virtual contrast agent flow can be generated, for example, such that at least one virtual contrast agent mass, particularly multiple sequentially arranged virtual contrast agent masses, are imaged in the synthetic imaging scene.

[0040] The virtual contrast agent flow can be advantageously and dynamically adjusted in a composite imaging scene by means of user-preset parameters of the virtual contrast agent flow. Specifically, a graphical view of the composite scene can be displayed using a display unit, where the user can input and / or adjust the parameters of the virtual contrast agent flow using an input unit.

[0041] The invention relates in a second aspect to a computer-executed method for providing a trained function. In this method, in a first step t.1), preoperative training image data of a training examination region of a training examination subject is received. Furthermore, the training examination region has a hollow organ. Additionally, the preoperative training image data images a contrast agent flow within the hollow organ. In a second step t.2), intraoperative training image data of the training examination region of the training examination subject is received. Here, the intraoperative training image data images a medical object at least partially disposed within the hollow organ. In a third step t.3), a contrast comparison scene from a medical imaging device is received. Here, the contrast comparison scene images other contrast agent flows within the hollow organ, wherein the medical object is at least partially disposed within the hollow organ. Alternatively or supplementarily, in the third step t.3), a synthetic contrast comparison scene is generated by applying deformation corrections to the preoperative training image data. Here, the deformation corrections are based on the intraoperative training image data. Furthermore, the synthetic contrast imaging comparison scene images a virtual contrast agent flow in the hollow organ, taking into account the medical object at least partially arranged in the hollow organ, wherein the virtual contrast agent flow is simulated.

[0042] In step t.4), a synthetic angiography training scenario is generated by applying the trained function to input data based on preoperative and intraoperative training image data. In step t.5), at least one parameter of the trained function is adjusted based on a comparison between the training scenario and the comparison scenario. In step t.6), the trained function is provided.

[0043] Receiving preoperative training image data and / or intraoperative training image data and / or contrast imaging comparison scenarios may particularly include detecting and / or reading from computer-readable data storage and / or receiving from a data storage unit, such as a database. Furthermore, the preoperative training image data and / or intraoperative training image data and / or contrast imaging comparison scenarios may be provided by a unit providing other medical imaging equipment. Here, the other medical imaging equipment may be the same as or different from the medical imaging equipment used to acquire preoperative image data and / or intraoperative image data. Furthermore, the other medical imaging equipment used to acquire intraoperative training image data may be the same as or different from the other medical imaging equipment used to acquire preoperative training image data.

[0044] Preoperative training image data can, in particular, possess all the characteristics of preoperative image data described with reference to computer-executed methods for providing synthetic angiography scenarios, and vice versa. Furthermore, intraoperative training image data can possess all the characteristics of intraoperative image data described with reference to computer-executed methods for providing synthetic angiography scenarios, and vice versa. Additionally, preoperative training image data and / or intraoperative training image data can be simulated.

[0045] Preoperative training image data can advantageously image the training examination area, particularly hollow organs, before surgery, i.e., at a time point earlier than intraoperative training image data. The training examination area can, in particular, possess all the characteristics described by a computer-executed method for providing a synthetic imaging scene, and vice versa. The training examination area, particularly hollow organs, can advantageously include at least partially an anatomical region substantially similar to the examination area. The training subject can be a human and / or animal patient and / or a simulated blood vessel. Furthermore, the training subject can advantageously differ from the examination subject, especially by receiving preoperative training image data and / or intraoperative training image data and / or contrast imaging scenes for multiple different training subjects and / or training examination areas. The medical object described by a computer-executed method for providing a trained function can advantageously be at least partially consistent with the medical object described by a computer-executed method for providing a synthetic imaging scene, particularly in terms of type and / or material and / or material parameters and / or shape and / or operating parameters.

[0046] The contrast-enhanced imaging scenario can advantageously be captured using the same or different medical imaging equipment as the preoperative training image data and / or intraoperative training image data. Here, the contrast-enhanced imaging scenario can have time-resolved 2D and / or 3D imaging of the training examination area, particularly the hollow organ, especially intraoperatively, wherein the other contrast agent streams and medical objects are at least partially arranged in the training examination area, particularly the hollow organ. The contrast-enhanced imaging scenario can particularly be designed as a sequence of separate imaging, wherein the other contrast agent streams are imaged in the contrast-enhanced imaging scenario, particularly in separate imaging. Here, the contrast-enhanced imaging scenario can be at least partially consistent with the intraoperative training image data, particularly in terms of imaging of the examination area. Medical objects arranged intraoperatively in the hollow organ may cause changes in the hydrodynamics within the hollow organ. The contrast-enhanced imaging scenario can advantageously image the hollow organ, particularly intraoperatively, wherein the other contrast agent streams and medical objects arranged in the hollow organ can be imaged temporally and at a resolution within the contrast-enhanced imaging scenario. Therefore, the contrast-enhanced scene can advantageously have multiple image points, particularly pixels and / or volume pixels, wherein each image point can have a contrast-time intensity curve. For example, the other contrast agent flows can be imaged by changes in the contrast-time intensity curves, particularly exceeding and / or falling below a preset intensity threshold.

[0047] As an alternative or supplement, a synthetic contrast-enhanced scenario can be generated by applying deformation correction to preoperative training image data. Here, the deformation correction can be based on the orientation of the medical object in the intraoperative training image data. The orientation of the medical object in the hollow organ can be determined, in particular, by segmenting the medical object and / or identifying, especially locating, marker structures placed on the medical object in the intraoperative image data. Next, based on the intraoperative orientation of the medical object in the hollow organ, an algorithm for deformation correction can be applied to the preoperative training image data, which images the contrast agent flow in the hollow organ. This allows the preoperative training image data, especially the hollow organ imaged in the preoperative training image data, to be adapted to the spatial orientation of the hollow organ in the intraoperative training image data. It also allows the contrast agent flow, imaged temporally in the preoperative training image data, to follow the intraoperative orientation of the hollow organ and / or the medical object in the intraoperative training image data. However, it should be noted that after applying deformation correction to the preoperative training image data, the trajectory of the preoperative contrast agent is still imaged in the deformation-corrected preoperative training image data without considering the medical object placed in the hollow organ during surgery.

[0048] Simulations of virtual contrast agent flows can advantageously be based on computational fluid dynamics (CFD), where the fluid dynamics of the virtual contrast agent flow are considered due to the changes in the fluid dynamics caused by the medical object positioned within the hollow organ during surgery. For this purpose, the simulation can, for example, include a virtual representation of the medical object, particularly a 3D model, which can be generated based on intraoperative image data, for example by segmenting the medical object and / or identifying, particularly locating, marker structures positioned on the medical object. The simulation can advantageously be based on deformably corrected preoperative and intraoperative training image data, and the simulation can be particularly based on deformably corrected spatial orientation of the hollow organ and / or the spatial orientation of the medical object as imaged intraoperatively.

[0049] In step t.4), a synthetic angiography training scenario is generated by applying a trained function to the input data. Here, the input data may be based on preoperative training image data and intraoperative training image data. In step t.5), at least one parameter of the trained function can be adjusted by comparing the training scenario with a comparison scenario. The comparison between the training scenario and the comparison scenario can advantageously be performed in a pixel-by-pixel manner. The comparison may particularly include determining the deviation between the training time-intensity curve of the pixel in the training scenario and the comparison intensity curve of the pixel in the comparison scenario. Here, at least one parameter of the trained function can advantageously be adjusted such that the deviation between the training scenario and the comparison scenario, particularly the deviation between the training time-intensity curve of the pixel in the training scenario and the comparison intensity curve of the pixel in the comparison scenario, is minimized.

[0050] This can advantageously improve accuracy when applying the trained function to the input data to generate a synthetic imaging scene.

[0051] Providing trained functions may in particular include storing them on a computer-readable storage medium and / or transferring them to the providing unit.

[0052] The proposed method can advantageously provide a trained function that can be used in one implementation of a computer-executed method for providing a synthetic imaging scene.

[0053] In another advantageous embodiment of the proposed method for providing the trained function, the comparison scenario may include time-resolved 3D comparative imaging of the other contrast agent streams and / or virtual comparative contrast agent streams, taking into account a medical object at least partially arranged in a hollow organ.

[0054] The comparison scene can advantageously have at least one time-resolved 3D comparative imaging that images the training examination area. This time-resolved 3D comparative imaging can, in particular, image the volumetric localization of the training examination area in time-resolved 3D. This enables intraoperative imaging of the 3D curvature of hollow organs and / or at least partially located medical objects within hollow organs. Furthermore, other contrast agent flows and / or virtual comparative contrast agent flows in overlapping and / or branching vascular segments of hollow organs can be detected, and in particular distinguished, by means of 3D comparative imaging. The time-resolved 3D comparative imaging of these other contrast agent flows and / or virtual comparative contrast agent flows can be, for example, 4D-DSA imaging, especially time-resolved 3D-DSA, and / or have their characteristics. Moreover, the time-resolved 3D comparative imaging can have multiple image points, each of which can have a comparative time-intensity curve. For example, the other contrast agent flows and / or virtual comparative contrast agent flows can be imaged and / or detected by exceeding and / or falling below a preset intensity threshold in the comparative time-intensity curve.

[0055] Advantageously, the trained function generates a synthetic imaging training scene with at least one time-resolved 3D training image. Here, the comparison between the training scene and the comparison scene can advantageously include a comparison between the time-resolved 3D training image and the time-resolved 3D comparison image. This comparison can be performed at the image point level, specifically comparing corresponding image points between the training time-intensity curve of the 3D training image and the comparison time-intensity curve of the 3D comparison image.

[0056] In another advantageous embodiment of the proposed method for providing the trained function, the comparison scenario may include time-resolved 2D comparative imaging of at least one time-resolved contrast agent flow and / or virtual comparative contrast agent flow, taking into account a medical object at least partially arranged in a hollow organ.

[0057] The comparison scene can advantageously have at least one time-resolved 2D comparison image that images the training examination region along at least one spatial direction. The 2D comparison image can, for example, have a projection image of the object being examined, particularly an X-ray projection image. Alternatively or supplementarily, the 2D comparison image can have a layer image of the training examination region. Furthermore, the comparison scene can have multiple 2D comparison images that image the training examination region at different spatial directions at a time-resolved resolution, for example, projection images along multiple projection directions. Additionally, at least one time-resolved 2D comparison image can have multiple image points, wherein each image point can have a comparison time-intensity curve. For example, the other contrast agent flow and / or a virtual comparison contrast agent flow can be imaged and / or detected by exceeding and / or falling below a preset intensity threshold in the comparison time-intensity curve.

[0058] This allows for the advantageous realization that the trained function generates a synthetic imaging training scene with at least one time-resolved 2D training image. Here, the comparison between the training scene and the comparison scene can advantageously include a comparison between at least one time-resolved 2D training image and at least one time-resolved 2D comparison image. The comparison between at least one time-resolved 2D training image and at least one time-resolved 2D comparison image can be performed, particularly at the image point level, especially by comparing corresponding image points between the training time-intensity curve of the 3D training image and the comparison time-intensity curve of the 3D comparison image.

[0059] In another advantageous embodiment of the proposed method for providing a trained function, the at least one time-resolved 2D comparative imaging can be generated by virtual projection of the time-resolved 3D comparative imaging.

[0060] The generation of time-resolved 2D comparative imaging can advantageously include a virtual projection, particularly an intensity projection, onto a virtual detector by 3D comparative imaging originating from a virtual X-ray source along the projection direction, especially the projection direction of preoperative training image data and / or intraoperative training image data. The imaging geometry of the virtual projection can, in particular, be consistent with the imaging geometry used to capture the preoperative and / or intraoperative training image data. Because time-resolved 2D comparative imaging is generated through virtual projection of 3D comparative imaging, the projection direction can be adjusted or adapted computationally, for example, to adapt to changes in the projection direction of the intraoperative training image data.

[0061] The embodiments described herein are particularly advantageous when a synthetic training scene is generated by applying a trained function to produce at least one 2D training image with a training check region, wherein a comparison scene has a 3D comparison image with the training check region. The virtual projection used to generate the 3D comparison image of the at least one 2D comparison image enables a comparison between the at least one 2D training image and the at least one 2D comparison image, wherein at least one parameter of the trained function can be adjusted based on this comparison.

[0062] In another advantageous embodiment of the proposed method for providing the trained function, the virtual comparative contrast agent flow is simulated taking into account the deformation and / or narrowing of the cross-section of the hollow organ due to the medical object.

[0063] Advantageously, the orientation and / or volume model of the hollow organ can be generated and / or adjusted based on preoperative training image data, for example, by segmenting the hollow organ in the preoperative training image data. This is achieved particularly by placing contrast agent flow in the hollow organ at the time point of capturing the preoperative training image data. The preoperative training image data can thus have contrast-like (or comparative) imaging of the hollow organ. The orientation and / or volume model of the hollow organ can, for example, have a deformable, 2D or 3D centerline model and / or volume mesh model. This orientation and / or volume model can be advantageously adapted to the orientation of the medical object in the intraoperative training image data by applying deformation corrections to the preoperative training image data, particularly by being deformed. Furthermore, the simulation of virtual comparative contrast agent flow can be based on the deformation-corrected orientation and / or volume model of the hollow organ. Thus, the simulation of virtual comparative contrast agent flow can advantageously take into account the deformation of the hollow organ due to the medical object. Furthermore, it is advantageous to determine the virtual contrast agent flow imaged in the synthetic imaging training scenario, taking into account the deformation of the hollow organ due to the medical object arranged at least partially within the hollow organ.

[0064] Furthermore, the simulation of the virtual comparative contrast agent flow can take into account the narrowing of the cross-section of the hollow organ due to the medical object placed within it. To this end, the simulation of the virtual comparative contrast agent flow can be based on training shape information of the medical object, such as its diameter and / or perimeter and / or cross-section and / or surface, particularly a volumetric model and / or volumetric mesh model, and / or curvature and / or spatial orientation. Furthermore, the training shape information can advantageously include information regarding the surface characteristics of the medical object, particularly information regarding depressions and / or protrusions on the surface of the medical object. Additionally, the simulation of the virtual comparative contrast agent flow can be based on training material parameters and / or training operation parameters of the medical object, particularly at the time point during the imaging process when training image data. Here, the training material parameters can, for example, describe the porosity and / or material properties of the medical object. Furthermore, the training operation parameters can have information regarding the positioning of the medical object, particularly its orientation and / or spatial location.

[0065] Advantageously, the virtual comparative contrast agent flow can be accurately simulated, taking into account the narrowing of the hollow organ's cross-section due to the medical object at least partially arranged within the hollow organ. Here, the simulation of the virtual comparative contrast agent flow based on digital fluid dynamics can advantageously be adapted to the spatial orientation of the hollow organ and / or the medical object. This allows for the advantageous determination of the virtual contrast agent flow imaged in the synthetic contrast training scenario, also taking into account the narrowing of the hollow organ due to the medical object at least partially arranged within it.

[0066] In another advantageous embodiment of the proposed method for providing a trained function, the input data may also be based on training material parameters and / or training operation parameters and / or training shape information of the medical object and / or on physiological training parameters of the training examination object.

[0067] The training material parameters and / or training operation parameters and / or training shape information of the medical object may, in particular, possess all the characteristics of the material parameters and / or operation parameters and / or shape information of the medical object described above and / or with respect to the computer-executed method for providing the synthetic imaging scene, and vice versa. Furthermore, the training physiological parameters of the examination object may advantageously possess all the characteristics of the physiological parameters of the examination object described with respect to the computer-executed method for providing the synthetic imaging scene, and vice versa.

[0068] Specifically, in steps t.1) and / or t.2) and / or t.3, training material parameters and / or training operation parameters and / or training shape information and / or physiological training parameters of the medical subject can be received. Receiving the training material parameters and / or training operation parameters and / or training shape information and / or physiological training parameters of the medical subject can particularly include detecting and / or reading from a computer-readable data storage device and / or receiving from a data storage unit, such as a database. Furthermore, the training material parameters and / or training operation parameters and / or training shape information can be provided by the medical subject, especially by the processing unit of the medical subject. Additionally, the training physiological parameters of the training subject can be provided by sensors used to detect the training physiological parameters, such as respiratory sensors and / or motion sensors and / or pulse sensors and / or blood pressure sensors.

[0069] By increasing the training material parameters and / or training operation parameters and / or training shape information and / or training physiological parameters of the medical subject, the synthetic imaging scene, especially the virtual contrast agent flow, can be advantageously adapted to the state of the medical subject and / or the training examination subject with high accuracy and particularly well, especially at the time point of training image data during the imaging procedure.

[0070] In another advantageous embodiment of the proposed method for using a trained function, the input data may also be based on training parameters of the other contrast agent streams and / or virtual comparative contrast agent streams. Here, the training parameters may preset the dose and / or velocity and / or direction of the virtual contrast agent stream.

[0071] The training parameters for the other contrast agent streams and / or the virtual comparative contrast agent streams can, in particular, possess all the characteristics of the contrast agent stream parameters described with respect to the computer-executed method for providing a synthetic contrast scene, and vice versa. Advantageously, the virtual contrast agent streams in the training scene can be generated based on the training parameters of the other contrast agent streams and / or the virtual comparative contrast agent streams by applying a trained function to the input data. This is particularly advantageous when the other contrast agent streams and / or the virtual comparative contrast agent streams differ from the contrast agent streams at the time point in time of capturing the preoperative training image data in terms of dosage and / or motion speed and / or motion direction. This allows for a particularly flexible, especially dynamic, adaptation of the virtual contrast agent streams, for example, through user input via an input unit.

[0072] In a third aspect, the present invention relates to a providing unit designed to implement the aforementioned computer-executed method and aspects thereof for providing a synthetic imaging scene according to the invention. The providing unit may advantageously include a computing unit, a storage unit, and an interface. The providing unit may be designed to implement these methods and aspects in such a way that the interface and computing unit are designed to implement the corresponding method steps. The interface may in particular be designed to implement steps b.1), b.2), and / or b.4). Furthermore, the computing unit and / or storage unit may be designed to implement step b.3).

[0073] The advantages of the proposed providing unit substantially correspond to the advantages of the proposed computer-executed method for providing a synthetic imaging scene. The features, advantages, or alternative embodiments mentioned herein can also be applied to other claimed technical solutions and vice versa.

[0074] The invention relates, in a fourth aspect, to a training unit designed to implement the aforementioned computer-executed methods and aspects thereof for providing trained functions according to the invention. The training unit advantageously includes a training interface, a training storage unit, and a training computation unit. The training unit can be designed to implement these methods and aspects in such a way that the training interface, training storage unit, and training computation unit are designed to implement corresponding method steps. The training interface can be particularly designed to implement steps t.1), t.2), t.3), and / or t.6). Furthermore, the training computation unit and / or training storage unit can be designed to implement steps t.3) to t.5).

[0075] The advantages of the proposed training unit substantially correspond to the advantages of the proposed computer-executed method for providing the trained function. The features, advantages, or alternative implementations mentioned herein can also be applied to other claimed technical solutions and vice versa.

[0076] The invention relates, in a fifth aspect, to a medical imaging apparatus, which particularly includes the proposed providing unit. Here, the medical X-ray apparatus, especially the providing unit, is designed to implement the proposed computer-executed method for providing a synthetic imaging scene. Furthermore, the medical imaging apparatus is designed to capture and / or receive and / or provide preoperative and / or intraoperative image data. The medical imaging apparatus may, for example, be designed as a medical X-ray apparatus, particularly a C-arm X-ray apparatus, and / or as a computed tomography (CT) scanner and / or as a magnetic resonance imaging (MRT) scanner and / or as an ultrasound scanner.

[0077] The advantages of the proposed medical imaging device substantially correspond to the advantages of the proposed computer-executed method for providing synthetic imaging scenes. The features, advantages, or alternative embodiments mentioned herein can also be applied to other claimed technical solutions and vice versa.

[0078] In a sixth aspect, the present invention relates to a computer program product having a computer program that can be directly loaded into the memory of a providing unit, the computer program having program segments to implement all steps of a computer-executed method for providing a synthetic imaging scene when the providing unit executes the program segments; and / or the computer program can be directly loaded into the training memory of a training unit, the computer program having program segments to implement all steps of a proposed method for providing a trained function and / or aspects thereof when the training unit executes the program segments.

[0079] The invention relates in a seventh aspect to a computer-readable storage medium storing program segments that can be read and executed by a providing unit, so as to implement all steps of a computer-executed method for providing a synthetic imaging scene when the providing unit executes the program segments; and / or storing program segments that can be read and executed by a training unit, so as to implement all steps of a method for providing a trained function and / or aspects thereof when the training unit executes the program segments.

[0080] The invention relates in an eighth aspect to a computer program or computer-readable storage medium comprising a trained function provided by a proposed computer-executed method or one aspect thereof.

[0081] A significant advantage of the software implementation is that the provisioning and / or training units already in use can be easily added via software updates to operate in accordance with the present invention. Such a computer program product may, in addition to the computer program, include, if necessary, additional components such as document assemblies and / or additional parts, as well as hardware components, such as hardware locks (Dongles, etc.) for using the software. Attached Figure Description

[0082] Embodiments of the present invention are shown in the accompanying drawings and are described in detail below. The same reference numerals are used for the same features in different drawings. In the drawings:

[0083] Figure 1 A schematic diagram illustrating an advantageous implementation of the proposed method for providing a synthetic imaging scene;

[0084] Figure 2A schematic diagram showing preoperative image data;

[0085] Figure 3 A schematic diagram showing intraoperative image data;

[0086] Figure 4 A schematic diagram of a composite imaging scene is shown;

[0087] Figure 5 A schematic diagram illustrating another advantageous implementation of the proposed method for providing synthetic imaging scenes;

[0088] Figures 6 to 8 Schematic diagrams illustrating different implementations of the proposed method for providing a trained function;

[0089] Figure 9 A schematic diagram of the providing unit is shown;

[0090] Figure 10 A schematic diagram of the training unit is shown;

[0091] Figure 11 A schematic diagram of a medical C-arm X-ray device is shown as an exemplary embodiment of the proposed medical imaging device. Detailed Implementation

[0092] Figure 1 A schematic diagram illustrating an advantageous implementation of the proposed method for providing a synthetic contrast imaging scene PROV-S is shown. Here, in a first step b.1), preoperative image data pID of the examination region of the REC-pID examination subject is received. The examination region may have a hollow organ, wherein the preoperative image data pID can image the contrast agent flow within the hollow organ. In a second step b.2), intraoperative image data iID of the examination region of the REC-iID examination subject is received. Here, the intraoperative image data iID can image a medical object at least partially arranged within the hollow organ. In a third step b.3), the synthetic contrast imaging scene S is generated by applying a trained function TF to the input data, wherein the input data for the trained function may be based on the preoperative image data pID and the intraoperative image data iID. Furthermore, the synthetic contrast imaging scene S can image a virtual contrast agent flow within the hollow organ, taking into account the medical object at least partially arranged within the hollow organ. Additionally, at least one parameter of the trained function TF may be based on a comparison between a training scene and a comparison scene. In step 4 (b.4), the PROV-S composite imaging scene S is provided.

[0093] Furthermore, the proposed method for providing a PROV-S synthetic imaging scenario may include receiving REC-PARAM parameters PARAM, which may include, for example, material parameters and / or operational parameters and / or shape information and / or physiological parameters of the medical object MO. Advantageously, the input data of the trained function TF may be additionally based on the parameters PARAM. Here, the parameters PARAM may be provided by the medical object MO, especially the processing unit of the medical object. Furthermore, the parameters PARAM, especially the physiological parameters of the examined object, may be provided by sensors for detecting physiological parameters, such as respiratory sensors and / or motion sensors and / or pulse sensors and / or blood pressure sensors.

[0094] Furthermore, the PARAM parameter may include parameters for the contrast agent flow pFI and / or a virtual contrast agent flow vFI, which can preset the dose and / or velocity and / or direction of the virtual contrast agent flow. The parameters of the virtual contrast agent flow vFI can, for example, be preset and / or adjusted by the user using an input unit. Additionally, the parameters of the contrast agent flow pFI can be provided by the device controlling the contrast agent injection at the time point of acquiring preoperative image data pID.

[0095] exist Figure 2 The image schematically illustrates the preoperative image data pID and... Figure 3 The intraoperative image data iID is schematically shown. Preoperative image data pID can be used to preoperatively image hollow organs pHO, such as contrast agent flow pFI in vascular segments (illustrated schematically here by arrows). Preoperative image data pID can be acquired, for example, using DSA.

[0096] exist Figure 3 The intraoperative image data iID schematically shown can advantageously image the medical object MO at least partially arranged in the hollow organ iHO. The hollow organ iHO may be deformed during surgery due to the medical object MO at least partially arranged in the hollow organ iHO, relative to its preoperative orientation.

[0097] exist Figure 4The image schematically illustrates a synthetic contrast imaging scene S, and in particular, a separate image of the synthetic contrast imaging scene S. The synthetic contrast imaging scene S can advantageously image a virtual contrast agent flow vFI within the hollow organ iHO during surgery, taking into account the medical object MO arranged at least partially within the hollow organ. The synthetic contrast imaging scene S can particularly image the virtual contrast agent flow vFI (schematically shown here by arrows) taking into account the deformation and / or narrowing of the cross-section of the hollow organ iHO due to the medical object. The synthetic contrast imaging scene S can particularly have a graphical view of the virtual contrast agent flow vFI, such as a color-coded and / or grayscale-based and / or a generalized view that can be visually perceived by the user. Furthermore, the graphical view of the virtual contrast agent flow vFI can each include a time-intensity curve of at least one image point of the synthetic contrast imaging scene. When planning the repositioning of the medical object MO, especially a change in its arrangement, and / or when pre-setting the operating parameters of the medical object MO, the graphical view of the synthetic contrast imaging scene S, especially the virtual contrast agent flow vFI, can advantageously assist the user, such as a medical operator.

[0098] Furthermore, the graphical view of the synthetic contrast scene S, particularly the virtual contrast agent flow vFI, can include, in particular, color-coded imaging of the deviation between the virtual contrast agent flow vFI and the contrast agent flow pFI. For this purpose, deformation correction can be applied to the preoperative image data pID, wherein the graphical view can, for example, have the difference and / or quotient between the deformed preoperative image data pID and the synthetic contrast scene S. This allows for particularly intuitive detection of changes in hydrodynamics caused by the medical object MO, which is at least partially arranged within the hollow organ during surgery.

[0099] Figure 5 Another advantageous implementation of the proposed computer-executed method for providing a PROV-S synthetic imaging scene is shown. Here, the synthetic imaging scene S can include time-resolved 3D imaging 3D-SD of a virtual contrast agent flow vFI, taking into account a medical object at least partially arranged in a hollow organ iHO.

[0100] As an alternative or supplement, the synthetic imaging scene S may include at least one synthetic time-resolved 2D imaging 2D-SD of a virtual contrast agent flow vFI, taking into account a medical object MO arranged at least partially in a hollow organ iHO. Here, the at least one synthetic time-resolved 2D imaging 2D-SD may be generated by a virtual projection PROJ, particularly an intensity projection, of the time-resolved 3D imaging 3D-SD.

[0101] exist Figure 6The diagram schematically illustrates an advantageous implementation of a computer-executed method for providing a trained function PROV-TF. Here, in a first step t.1), preoperative training image data pTID of the training examination region of the REC-pTID training examination subject is received. This training examination region may have a hollow organ pHO. Furthermore, the preoperative training image data pTID can image the contrast agent flow pFI within the hollow organ pHO. In a second step t.2), intraoperative training image data iTID of the training examination region of the REC-iTID training examination subject is received. Here, the intraoperative training image data iTID can image a medical object MO at least partially disposed within the hollow organ iHO. In a third step t.3), a contrast comparison scene VS of a REC-VS medical imaging device is received. Here, the contrast comparison scene VS can, particularly intraoperatively, image other contrast agent flows within the hollow organ iHO, wherein the medical object MO is at least partially disposed within the hollow organ. In step t.4), a synthetic angiography training scenario TS can be generated by applying the trained function TF to the input data. Here, the input data can be based on preoperative training image data pTID and intraoperative training image data iTID. In step t.5), at least one parameter of the ADJ-TF trained function TF can be adjusted based on a comparison between the training scenario TS and the comparison scenario. In step t.6), the PROV-TF trained function TF can be provided.

[0102] Furthermore, the proposed method for providing the PROV-TF trained function may include receiving REC-TPARAM training parameters TPARAM, which may include, for example, training material parameters and / or training running parameters and / or training shape information and / or training physiological parameters of the medical object MO. Advantageously, the input data for the trained function TF may be additionally based on the training parameters TPARAM. Here, the training parameters TPARAM may be provided by the medical object MO, especially by the processing unit of the medical object MO. Furthermore, the training parameters TPARAM, especially the training physiological parameters of the training object, may be provided by sensors for detecting the training physiological parameters, such as respiratory sensors and / or motion sensors and / or pulse sensors and / or blood pressure sensors.

[0103] In addition, the training parameters TPARAM may include training parameters for the contrast agent flow pFI and / or the virtual contrast agent flow vFI, which may preset the dose and / or motion speed and / or motion direction of the virtual contrast agent flow vFI.

[0104] Figure 7Another advantageous implementation of the proposed computer-executed method for providing the trained function PROV-TF is shown. Here, in particular, as a way to… Figure 6 An alternative or supplement to the illustrated embodiment is to generate the GEN-VS synthetic contrast-enhanced scene VS in step t.3) by applying deformation correction to the preoperative training image data pTID. Here, the deformation correction can advantageously be based on the intraoperative training image data iTID, particularly on the spatial orientation of the medical object iHO imaged in the intraoperative training image data iTID. Furthermore, the synthetic contrast-enhanced scene VS can image a virtual contrast agent flow in the hollow organ iHO, taking into account the medical object MO arranged at least partially in the hollow organ. Moreover, the virtual contrast agent flow can be simulated based on the training parameters TPARAM, particularly the training material parameters and / or training running parameters and / or training shape information of the medical object MO.

[0105] Advantageously, the virtual comparative contrast agent flow can be simulated taking into account the deformation and / or narrowing of the cross-section of the hollow organ iHO due to the medical object MO.

[0106] Figure 8 Another advantageous implementation of the proposed computer-executed method for providing a trained function PROV-TF is shown. Here, the comparison scene VS can include time-resolved 3D comparative imaging 3D-VSD of the other contrast agent streams and / or virtual comparative contrast agent streams, taking into account the medical object MO arranged at least partially in the hollow organ iHO. Alternatively or supplementarily, the comparison scene VS can include at least one time-resolved 2D comparative imaging 2D-VSD of the other contrast agent streams and / or virtual comparative contrast agent streams, taking into account the medical object MO arranged at least partially in the hollow organ iHO. The at least one time-resolved 2D comparative imaging 2D-SD can be generated, in particular, by a virtual projection PROJ, especially an intensity projection, of the time-resolved 3D comparative imaging 3D-VSD.

[0107] Similarly, the training scenario TS may include a time-resolved 3D training image 3D-TSD of a virtual contrast agent flow taking into account a medical object MO at least partially arranged in the hollow organ iHO. Alternatively or supplementarily, the training scenario TS may include at least one time-resolved 2D training image 2D-TSD of a virtual contrast agent flow taking into account a medical object MO at least partially arranged in the hollow organ iHO. The comparison between the training scenario TS and the comparison scenario VS may advantageously include a comparison between the time-resolved 3D training image 3D-TSD and the time-resolved 3D comparison image 3D-VSD. Alternatively or supplementarily, the comparison between the training scenario TS and the comparison scenario VS may include a comparison between at least one time-resolved 2D training image 2D-TSD and at least one time-resolved 2D comparison image 2D-VSD. Advantageously, at least one parameter of the ADJ-TF trained function TF may be adjusted such that the corresponding bias is minimized.

[0108] The at least one time-resolved 2D training image (2D-TSD) can be generated, in particular, by virtual projection of the time-resolved 3D training image (3D-TSD). This is especially advantageous for comparisons between training and comparison scenes when the comparison scene has only at least one 2D comparison image (2D-VSD).

[0109] exist Figure 9 The proposed providing unit PRVS is schematically illustrated. Here, the providing unit PRVS may include an interface IF, a computing unit CU, and a storage unit MU. The providing unit PRVS can be designed to implement methods and aspects thereof for providing a synthetic imaging scene in such a way that the interface IF, computing unit CU, and storage unit MU are designed to implement the corresponding method steps. The interface IF can be particularly designed to implement steps b.1), b.2), and / or b.4). Furthermore, the computing unit CU and / or storage unit MU can be designed to implement step b.3).

[0110] Figure 10 A schematic diagram of the proposed training unit TRS is shown. The training unit TRS may advantageously include a training interface TIF, a training storage unit TMU, and a training computation unit TCU. The training unit TRS can be designed to implement the method and aspects thereof for providing the trained function PROV-TF in such a way that the training interface TIF, the training storage unit TMU, and the training computation unit TCU are designed to implement the corresponding method steps. The training interface TIF can be particularly designed to implement steps t.1), t.2), t.3), and / or t.6). Furthermore, the training computation unit TCU and / or the training storage unit TMU can be designed to implement steps t.3) to t.5).

[0111] The providing unit PRVS and / or training unit TRS can be, in particular, a computer, a microcontroller, or an integrated circuit. Alternatively, the providing unit PRVS and / or training unit TRS can be a real or virtual collection of computers (the technical term for a real collection is "Cluster," and the technical term for a virtual collection is "Cloud"). The providing unit PRVS and / or training unit TRS can also be designed as a virtual system implemented on a real computer or a real or virtual collection of computers (virtualization).

[0112] The interface IF and / or training interface TIF can be hardware or software interfaces (e.g., PCI bus, USB, or FireWire). The compute unit CU and / or training compute unit TCU can be hardware or software components, such as a microprocessor or a so-called FPGA (Field Programmable Gate Array). The storage unit MU and / or training storage unit TMU can be implemented as non-persistent working memory (Random Access Memory, or RAM) or designed as persistent mass storage (hard drive, USB flash drive, SD card, solid-state drive).

[0113] The interface IF and / or training interface TIF may in particular include multiple sub-interfaces, which implement different steps of the corresponding method. In other words, the interface IF and / or training interface TIF can also be understood as multiple interface IFs or multiple training interface TIFs. The computation unit CU and / or training computation unit TCU may in particular include multiple sub-computation units, which implement different steps of the corresponding method. In other words, the computation unit CU and / or training computation unit TCU can also be understood as multiple computation units CU or multiple training computation units TCU.

[0114] exist Figure 11 The diagram schematically illustrates a medical C-arm X-ray device 37 as an example of the proposed medical imaging device. Here, the medical C-arm X-ray device 37 may advantageously include a providing unit PRVS. Furthermore, the medical C-arm X-ray device 37, and in particular the providing unit PRVS, can be designed to implement the proposed method for providing a synthetic contrast scene PROV-S.

[0115] Here, the medical C-arm X-ray device 37 advantageously includes a detector 34, particularly an X-ray detector, and an X-ray source 33. For acquiring preoperative image data pID and intraoperative image data iID, the arm 38 of the C-arm X-ray device 37 is movably supported about one or more axes. Furthermore, the medical C-arm X-ray device 37 may include a motion device 39 that enables movement of the C-arm X-ray device 37 in space.

[0116] To acquire preoperative image data pID and intraoperative image data iID for the examination area UB of the examination subject 31 arranged on the patient support device 32, the providing unit PRVS can send signal 24 to the X-ray source 33. The X-ray source 33 can thereby emit an X-ray beam. When the X-ray beam reaches the surface of the detector 34 after interacting with the examination area UB, the detector 34 can send signal 21 to the providing unit PRVS. The providing unit PRVS can, for example, receive the preoperative image data pID and / or the intraoperative image data iID according to signal 21.

[0117] Furthermore, the medical C-arm X-ray device 37 may include an input unit 42, such as a keyboard, and / or a display unit 41, such as a monitor and / or a display screen. For example, for a capacitive input display screen, the input unit 42 may preferably be integrated into the display unit 41. Here, the medical C-arm X-ray device 37, especially the proposed method for providing a synthetic contrast imaging scene (PROV-S), can be controlled by user input on the input unit 42. Additionally, the input unit 42 can allow user input for preset parameters PARAM. For this purpose, the input unit 42 may, for example, send signal 26 to the providing unit PRVS.

[0118] Furthermore, the display unit 41 can be designed to display a graphical view of information and / or data from the medical C-arm X-ray device 37 and / or the providing unit PRVS and / or other components. For this purpose, the providing unit PRVS can, for example, send signal 25 to the display unit 41. The display unit 41 can be particularly designed to display a graphical view of preoperative image data pID and / or intraoperative image data iID and / or the composite contrast scene S.

[0119] The schematic views included in the accompanying drawings do not reflect scale or dimensional relationships.

[0120] Finally, it should be reiterated that the methods and apparatus described in the above description are merely embodiments, and those skilled in the art can modify the embodiments in different ways without departing from the scope of the invention. Furthermore, the use of the indefinite article "a" does not preclude the possibility that the relevant feature may be present multiple times. Similarly, the terms "unit" and "element" do not preclude the possibility that the relevant component consists of multiple working sub-components, which may, if necessary, be distributed in space.

Claims

1. A computer-executed method for providing a composite imaging scene (S) (PROV-S), the method comprising: b.1) Receive (REC-pID) preoperative image data (pID) of the examination area (UB) of the examination subject (31), the preoperative image data being captured by means of a medical imaging device for capturing preoperative image data (pID). The examination area (UB) has a hollow organ. Among them, preoperative image data (pID) imaged the contrast agent flow (pFI) in the hollow organ. b.2) Receive intraoperative image data (iID) of the examination area (UB) of the examination object (31), the intraoperative image data being captured by means of a medical imaging device for capturing intraoperative image data (iID). The medical imaging device used to capture intraoperative image data (iID) may be the same as or different from the medical imaging device used to capture preoperative image data (pID). The intraoperative image data (iID) images a medical object (MO) that is at least partially positioned within a hollow organ. b.3) A synthetic imaging scene (S) is generated by applying a trained function (TF) to the input data. The input data is based on preoperative image data (pID) and intraoperative image data (iID). The synthetic imaging scene (S) images a virtual contrast agent flow (vFI) within a hollow organ, taking into account a medical object (MO) at least partially disposed within the hollow organ. This virtual contrast agent flow has predefined contrast values ​​and / or intensity curves relative to the anatomical structures and / or tissues of the examined object. In this context, at least one parameter of the trained function (TF) is based on a comparison between the training scenario (TS) and the comparison scenario (VS) (ADJ-TF). b.4) Provide a composite imaging scene (S) via a display on the display unit (41). Specifically, a synthetic angiography training scenario is generated by applying a trained function to input data, the input data being based on preoperative training image data and intraoperative training image data, and a synthetic angiography comparison scenario is generated by applying deformation correction to the preoperative training image data, the deformation correction being based on the intraoperative training image data.

2. The method according to claim 1, characterized in that, The synthetic imaging scene (S) includes time-resolved 3D imaging (3D-SD) of a virtual contrast agent flow (vFI) taking into account a medical object (MO) arranged at least partially in a hollow organ.

3. The method according to claim 1, characterized in that, The synthetic imaging scene (S) includes at least one synthetic time-resolved 2D imaging (2D-SD) of a virtual contrast agent flow (vFI) taking into account a medical object (MO) at least partially arranged in a hollow organ.

4. The method according to claim 2 or 3, characterized in that, The at least one synthesized time-resolved 2D image (2D-SD) is generated by virtual projection (PROJ) of the time-resolved 3D image (3D-SD).

5. The method according to claim 1, characterized in that, The input data is also based on the material parameters and / or operating parameters and / or shape information of the medical object (MO) and / or the physiological parameters of the examination object (31).

6. The method according to claim 1, characterized in that, The input data is also based on parameters of a virtual contrast agent flow (vFI). The parameters preset the dose and / or velocity and / or direction of the virtual contrast agent flow (vFI).

7. A computer-executed method for providing (PROV-TF) trained functions (TF), the method comprising: t.1) Receive preoperative training image data (pTID) of the training examination area of ​​the training examination subject (REC-pTID), wherein the preoperative training image data is simulated or captured using other medical imaging equipment used to capture preoperative training image data (pTID). The examination area contains a hollow organ. Among them, preoperative training image data (pTID) was used to image the contrast agent flow in the hollow organ. t.2) Receive intraoperative training image data (iTID) of the training examination area of ​​the training examination subject (REC-iTID), wherein the intraoperative training image data is simulated or acquired using other medical imaging equipment used to acquire intraoperative training image data (iTID). The other medical imaging devices used for capturing intraoperative training image data (iTID) may be the same as or different from the other medical imaging devices used for capturing preoperative training image data (pTID). The intraoperative training image data (iTID) images at least partially depict medical objects (MOs) arranged within hollow organs. t.3) Receive contrast comparison scene (VS) from medical imaging equipment. The contrast-enhanced scene (VS) imaging reveals other contrast agent flows within the hollow organ. In this context, the medical object (MO) is at least partially positioned within a hollow organ, and / or a (GEN-VS) synthetic contrast-enhanced scene (VS) is generated by applying deformation corrections to preoperative training image data (pTID). The deformation correction is based on intraoperative training image data (iTID). Specifically, the synthetic contrast comparison scene (VS) images a virtual contrast agent flow within the hollow organ, taking into account the medical object (MO) at least partially disposed within it. Among them, the simulated virtual contrast agent flow, t.4) Synthetic imaging training scenes (TS) are generated by applying the trained function (TF) to the input data. The input data is based on preoperative training image data (pTID) and intraoperative training image data (iTID), and a synthetic contrast comparison scene is generated by applying deformation correction to the preoperative training image data. t.5) Adjust at least one parameter of the trained function (TF) based on the comparison between the training scenario (TS) and the comparison scenario (VS) (ADJ-TF). t.6) The trained function (TF) is provided (PROV-TF) by transmitting it to the Providing Unit (PRVS).

8. The method according to claim 7, characterized in that, The comparison scene (VS) includes time-resolved 3D comparison imaging (3D-VSD) of the other contrast agent flow and / or virtual comparison contrast agent flow, taking into account a medical object (MO) at least partially arranged in a hollow organ.

9. The method according to claim 7, characterized in that, The comparison scene (VS) includes at least one time-resolved 2D comparison imaging (2D-VSD) of the other contrast agent flow and / or virtual comparison contrast agent flow, taking into account a medical object (MO) at least partially arranged in a hollow organ.

10. The method according to claim 8, characterized in that, The at least one time-resolved 2D comparison imaging (2D-VSD) is generated by virtual projection (PROJ) of the time-resolved 3D comparison imaging (3D-VSD).

11. The method according to claim 7, characterized in that, The virtual comparative contrast agent flow is simulated taking into account the deformation and / or narrowing of the cross-section of the hollow organ due to the medical object (MO).

12. The method according to claim 7, characterized in that, The input data is also based on the material parameters and / or operating parameters and / or shape information of the medical object (MO) and / or the physiological training parameters of the training examination object.

13. The method according to any one of claims 7 to 12, characterized in that, The input data is also based on training parameters of the other contrast agent streams and / or virtual comparative contrast agent streams. The training parameters preset the dose and / or velocity and / or direction of the virtual contrast agent flow (vFI).

14. A providing unit (PRVS) designed to implement a computer-executed method as described in any one of claims 1 to 6.

15. A medical imaging apparatus (37) comprising a providing unit (PRVS) as described in claim 14, the medical imaging apparatus being designed to implement a computer-executed method as described in any one of claims 1 to 6. in, Medical imaging devices (37) are designed to capture and / or receive and / or provide preoperative image data (pID) and / or intraoperative image data (iID).

16. A training unit (TRS) designed to implement the computer-executed method as described in any one of claims 7 to 13.

17. A computer program product having a computer program, said computer program being directly loadable into the memory (MU) of a providing unit (PRVS), said computer program having program segments for implementing all steps of the computer-executed method according to any one of claims 1 to 6 when executed by said program segments by the providing unit (PRVS); and / or said computer program being directly loadable into the training memory (TMU) of a training unit (TRS), said computer program having program segments for implementing all steps of the computer-executed method according to any one of claims 7 to 13 when executed by said program segments by the training unit (TRS).

Images