Providing a results data set

Abstract

The invention relates to a computer-implemented method for providing a result data set, comprising: a) Capturing an image data set of an object under investigation, the image dataset spatially and temporally resolves the flow of contrast medium in a branched hollow organ of the subject of investigation, b) Identifying sections of the hollow organ in the image dataset based on the depicted contrast agent flow, c) Classifying subsections of the identified sections into singly fed and multiply fed subsections based on a depicted flow direction of the contrast medium flow and an identification of confluences of the identified sections of the hollow organ depicted in the image dataset, where sections of the hollow organ which are located downstream with respect to a confluence are classified as multiply fed sections, d) Providing the results dataset based on the image dataset and the classified subsections of the hollow organ, wherein the result data set contains a sub-data set for each of the classified sub-sections of the hollow organ, wherein the sub-datasets have a dedicated representation of the contrast agent flow in the respective classified sub-section of the hollow organ. The invention further relates to a provisioning unit, a medical imaging device and a computer program product.

Description

[0001] The present invention relates to a computer-implemented method for providing a result data set, a provisioning unit, a medical imaging device and a computer program product.

[0002] Arteriovenous malformations (AVMs) are vascular malformations in which arteries are directly connected to veins via a complex vascular network, particularly a nidus. A complete understanding of these complex vascular networks is crucial for treatment planning, for example, to ensure complete removal of the vascular malformation and prevent rupture during treatment. Three-dimensional (3D) or four-dimensional (4D) digital subtraction angiography (DSA) is frequently used to plan AVM treatment, such as surgical intervention and / or embolization and / or radiotherapy. In this procedure, 3D or 4D DSA is performed with a catheter positioned in one of the major blood vessels of the patient's brain, such as the common carotid artery and / or the vertebral artery, depending on the blood supply to the AVM.

[0003] However, AVMs often have a blood supply from more than one major blood vessel, for example, on multiple sides. As a result, a single vascular injection with contrast medium provides only incomplete opacity of the nidus, and the 3D dataset of the AVM is insufficient for treatment planning. Therefore, multiple 3D-DSA and / or 4D-DSA scans are frequently performed and fused after each reconstruction. This has the disadvantage of increasing the radiation dose for the patient. Furthermore, additional procedures are required.

[0004] Complete opacification of the nidus can be achieved, for example, by injecting the contrast agent into an ascending thoracic aorta. Alternatively, this can also be achieved via venous injection of the contrast agent, although the opacification is reduced. In both cases, the resulting datasets are often difficult to interpret because nidi are very complex structures. Furthermore, the mixing of the contrast agent in different feeding vessels of the nidus can complicate interpretation.

[0005] Patent JP 2011-254861A discloses a device for extracting blood vessel regions from volumetric data, wherein a blood inflow point, a blood outflow point, and an abnormal blood vessel segment can be displayed together with a 3D image generated from the volumetric data. Furthermore, Patent US 2012 / 0014577A1 discloses a method for identifying a feeding vessel of a malformation, wherein a real-time image is displayed overlaid with a medical image of the malformation and surrounding vessels, and the feeding vessel is identified in the displayed image. Furthermore, Patent US 2016 / 0328846A1 discloses a method for three-dimensionally displaying a limited area of ​​a patient's vascular structure with simulated contrast medium flow.Furthermore, the publication DE 10 2023 200 770 A1 discloses a method for providing a result data set comprising several subtraction image data sets, which are determined based on a first image data set of an examination object and each a partial image data set, wherein the partial image data sets are identified in a second medical image data set, which depicts a contrast agent flow in the examination object with time resolution.

[0006] It is therefore the object of the present invention to enable improved detection of the dynamics of contrast medium flows in branched hollow organs.

[0007] The object of the invention is achieved by the subject matter of the independent claims. Advantageous embodiments with expedient further developments are the subject matter of the dependent claims. Regardless of the grammatical gender of a particular term, persons of male, female, or other gender identities are included.

[0008] The invention relates, in a first aspect, to a computer-implemented method for providing a result data set. In a first step a), an image data set of an object under investigation is acquired. This image data set depicts the flow of contrast medium in a branched hollow organ of the object with spatial and temporal resolution. In step b), sections of the branched hollow organ are identified in the image data set based on the depicted contrast medium flow. In step c), subsections of the identified sections are classified into singly fed and multiply fed subsections based on the depicted flow direction of the contrast medium and the identification of confluences of the identified sections of the branched hollow organ depicted in the image data set.In this process, segments of the branched hollow organ located downstream of a confluence are classified as multiply fed segments. In step d), the result dataset is provided based on the image dataset in the classified segments of the branched hollow organ. The result dataset contains a subset for each classified segment of the branched hollow organ. The subsets provide a dedicated representation of the contrast agent flow in the respective classified segment of the branched hollow organ.

[0009] The steps described herein can advantageously be carried out at least partially simultaneously or sequentially.

[0010] The acquisition of the image data set can include, in particular, the acquisition and / or reading of a computer-readable data storage device and / or the receipt of data from a data storage unit, such as a database. Furthermore, the image data set can be provided by a provisioning unit of a medical imaging device for the purpose of receiving the image data set. The medical imaging device can include, for example, a magnetic resonance imaging (MRI) system and / or a computed tomography (CT) system and / or a medical X-ray device, in particular a medical C-arm X-ray device, and / or an ultrasound device and / or a positron emission tomography (PET) system.

[0011] The image dataset can comprise a spatially and temporally resolved representation of the examination area of ​​the object under investigation, in particular the contrast agent flow within the branched hollow organ of the object. The image dataset can be spatially resolved in two dimensions (2D) and / or three dimensions (3D). Advantageously, the image dataset can depict a flow, in particular a propagation and / or flow motion, of the contrast agent, especially radiopaque, located within the branched hollow organ. The image dataset can contain multiple image points, in particular pixels and / or voxels, each with a time-intensity curve. The time-intensity curves can depict the temporal evolution of image values, in particular intensity values ​​and / or attenuation values, of the respective image points.The branched hollow organ may, for example, include a vascular tree, particularly arterial and / or venous, and / or at least part of a vascular malformation, particularly a vascular malformation. The vascular malformation may, for example, include a nidus and / or an aneurysm.

[0012] In particular, the branched hollow organ may be located on or within a body part and / or organ of the subject under investigation. For example, the branched hollow organ may comprise a vascular tree of the subject's head, especially the brain.

[0013] Identifying the sections of a branched hollow organ, such as arterial and / or venous vessel segments, within the image dataset can involve segmenting image pixels that represent the respective sections. This segmentation can be based, for example, on comparing the image values ​​of the image pixels with a predefined threshold. In particular, the image pixels can be identified as representing a section of the branched hollow organ, specifically those depicting the contrast agent, particularly the contrast agent flow. Each section of the branched hollow organ can represent a spatially contiguous region of the organ. Identifying these sections can also involve identifying a contiguous set of image pixels within the dataset that represent each section of the branched hollow organ.

[0014] Subsections of the identified sections of the branched hollow organ are classified, and in particular distinguished, as singly and multiply fed subsections based on the depicted flow direction of the contrast medium and the identification of confluences of the identified sections of the branched hollow organ depicted in the image dataset. Advantageously, the identified sections of the branched hollow organ can each have at least one subsection, and in particular several subsections. For example, one of the identified sections can have several subsections, each of which borders a branching point, such as a bifurcation or vascular malformation, in pairs. In particular, the respective identified section can be formed by one or more, in particular contiguous, subsections.

[0015] Advantageously, based on the time-intensity curves of the image data points representing the sections of the branched hollow organ, the direction of contrast agent flow can be identified, for example, by comparing the intensity changes depicted in the time-intensity curves over time. Specifically, for each image point representing a section of the branched hollow organ, a time point of contrast agent inflow, particularly a bolus arrival time, can be determined based on the respective time-intensity curve. By comparing the contrast agent inflow times, especially the bolus arrival times, the direction of contrast agent flow can be identified for each section of the branched hollow organ.

[0016] Furthermore, confluences of the identified sections of the branched hollow organ can be identified at branching points. Confluences can denote branching points of the hollow organ, which have at least three subsections of the hollow organ, wherein the contrast medium flow in at least two of the at least three subsections is directed towards the branching point, in particular as a feeder. The at least one further section of the hollow organ, which includes the branching point and whose contrast medium flow is directed away from the branching point, in particular as an effluxer, can be classified as a multiply fed subsection. In particular, the at least one multiply fed subsection can be located downstream with respect to the confluence, in particular the branching point and the at least two feeder subsections.Sections of the branched hollow organ that are not located downstream of a confluence can be classified as simply fed sections.

[0017] Providing the results data set may involve storing it on a computer-readable storage medium, displaying a graphical representation of the results data set on a display unit, and / or transferring it to a delivery unit. In particular, a graphical representation of the results data set may be displayed using the display unit.

[0018] The result dataset is provided based on the image dataset and the classified segments of the branched hollow organ, in particular those segments classified as singly or multiply fed. The result dataset comprises a subset for each classified segment of the branched hollow organ. Each subset contains a dedicated representation, such as an image and / or a model, of the contrast agent flow in the respective classified segment of the branched hollow organ. The model may, for example, include a centerline model, a volume mesh model, a vector model, and / or a tensor model.The sub-datasets can each contain data points, in particular image points, with time-intensity curves that represent, or in particular depict, the contrast agent flow in the respective sub-section. The sub-datasets can each be spatially resolved in 2D or 3D. Furthermore, the sub-datasets can each be temporally resolved. In particular, each of the sub-datasets can represent only the contrast agent flow in the respective sub-section of the hollow organ.

[0019] The proposed method can advantageously enable improved capture of the dynamics of contrast agent flow in branched hollow organs. The proposed method can advantageously reduce the acquisition time and / or exposure, particularly the radiation dose, to the subject when imaging complex vascular malformations, especially AVMs. Furthermore, the fusion of multiple images, particularly 3D or 4D DSAs, can be advantageously avoided, which in turn leads to a shorter workflow. Separation into subsets can reduce the complexity of analyzing complex vascular malformations, especially when dealing with overlapping segments of the branched hollow organ.A medical operator can advantageously evaluate any number of sections of the branched hollow organ separately or in combination, for example by selectively switching on or off a graphical representation of the respective section.

[0020] In a further advantageous embodiment of the proposed method, the hollow organ can comprise a vascular tree. In this case, vascular segments, particularly arterial and / or venous, and / or at least part of a vascular malformation, especially the vascular malformation itself, can be identified as the segments of the vascular tree.

[0021] Advantageously, the branched hollow organ can comprise a vascular tree, in particular a branched structure comprising several vessel segments, which are at least partially interconnected at branching points. These multiple vessel segments can include arterial and / or venous vessel segments. Furthermore, the vascular tree can include a vascular malformation, such as a nidus and / or an aneurysm, as one segment.

[0022] The proposed embodiment can advantageously enable improved detection of the dynamics of contrast agent flows in branched vascular trees.

[0023] In a further advantageous embodiment of the proposed method, the partial data sets can have a spatially and temporally resolved representation of the contrast medium flow in the respective classified subsection of the hollow organ.

[0024] The sub-datasets can each contain a spatially and temporally resolved representation, in particular a 2D or 3D representation, such as an image and / or a model, of the contrast agent flow in the respective classified segment of the hollow organ. The image can, for example, depict the contrast agent concentration in the respective classified segment at different time points. The model can, for example, model or simulate the contrast agent kinetics in the respective classified segment, for example, based on a physical or empirical model.

[0025] The proposed embodiment can advantageously enable improved, in particular dedicated, detection of the dynamics of contrast medium flows in subsections of a branched hollow organ.

[0026] In a further advantageous embodiment of the proposed method, the dedicated representations of the contrast medium flow in the respective classified subsection of the hollow organ can have a visual distinguishing feature.

[0027] The visual distinguishing feature can, for example, include color coding and / or intensity coding, particularly grayscale coding, and / or annotation and / or at least partial transparency. Advantageously, the result dataset can be provided in such a way that the dedicated representations of contrast agent flow in the various subsections of the hollow organ are distinguishable based on the visual distinguishing feature. For example, the multiple subsections can each have a different color coding and / or intensity coding and / or annotation and / or at least partial transparency. The visual distinguishing feature can also be adapted to the classification of the respective subsection.

[0028] The proposed embodiment can enable improved differentiation of the respective dynamics of the contrast medium flows in the subsections of the branched hollow organ.

[0029] In a further advantageous embodiment of the proposed method, the result data set can comprise at least a partial superimposition and / or nesting and / or composition of the multiple sub-data sets.

[0030] Advantageously, the resulting dataset can comprise at least a partial, and in particular a complete, overlay of the multiple sub-datasets. The areas of the multiple sub-datasets that do not represent any of the multiple sub-sections can advantageously be overlaid at least partially, and in particular completely, transparently. Alternatively or additionally, the resulting dataset can comprise a composition, in particular a compilation and / or merging and / or reconstruction, of the multiple sub-datasets. The overlay and / or nesting and / or composition of the multiple sub-datasets can advantageously be carried out such that the sub-sections are arranged relative to each other according to their respective arrangement in the image dataset, in particular their respective anatomical arrangement.

[0031] The proposed embodiment can advantageously enable, in particular simultaneously, the detection of the dynamics of contrast medium flows in the several subsections of a branched hollow organ.

[0032] In a further advantageous embodiment of the proposed method, the image data set can contain multiple pixels, each with a time-intensity curve. Identifying the sections of the branched hollow organ can then involve identifying contrast agent influx based on the time-intensity curves of the respective pixels.

[0033] The time-intensity curves of the pixels in the image dataset can each depict a temporal progression of intensity values, particularly the image values ​​of the pixels in the image dataset. Advantageously, the contrast agent influx, and especially the time of contrast agent influx (e.g., bolus arrival time), can be determined for each pixel based on the respective time-intensity curve. The contrast agent influx, and especially the time of contrast agent influx, can be identified, for example, by the point in time of an increase, particularly a positive gradient, in the respective time-intensity curve and / or by the respective time-intensity curve exceeding a predefined threshold.Advantageously, pixels whose time-intensity curves show a contrast agent influx can be identified, and in particular segmented, as representing a section of the branched hollow organ. Furthermore, the sections, and especially a respective representation of the sections in the image dataset, can be identified based on the identified pixels.

[0034] The proposed embodiment can enable a particularly precise identification of the sections of the branched hollow organ.

[0035] In a further advantageous embodiment of the proposed method, step c) can include classifying at least some of the subsections into feeding and draining subsections based on the flow direction of the contrast agent depicted in edge pixels of the image dataset. The edge pixels can depict the branched hollow organ in an edge region of the image dataset. The classification of the subsections into singly fed and multiply fed subsections can additionally be based on classifying at least some of the subsections as feeding and draining.

[0036] Advantageously, sections of the branched hollow organ depicted by edge pixels of the image dataset can be identified. Edge pixels can comprise image pixels of the image dataset that are arranged within a, in particular spatial, edge region of the image dataset and each depict at least a section of the branched hollow organ. The edge region of the image dataset can be defined by a spatial boundary, in particular a boundary contour and / or a boundary surface, of a volume and / or a surface of the object under investigation depicted by the image dataset.The edge area image points can include the image points of the image data set that depict at least a partial section of the branched hollow organ, which are directly adjacent to the boundary contour and / or boundary surface and / or are arranged within a specified spatial distance with respect to the boundary contour and / or the boundary surface.

[0037] Advantageously, the segments of the branched hollow organ depicted by edge pixels of the image dataset can be identified as sectioned segments of the branched hollow organ in the image. Furthermore, the flow direction of the contrast agent can be identified for each sectioned segment of the hollow organ, particularly based on the time-intensity curves of the pixels representing the respective sectioned segment. If the depicted flow direction of the contrast agent in one of the sectioned segments is directed away from the edge region, the respective sectioned segment can be classified as an afferent segment. If the depicted flow direction of the contrast agent in one of the sectioned segments is directed towards the edge region, the respective sectioned segment can be classified as an efferent segment.

[0038] Advantageously, the classification of the identified sections into single and multiply fed subsections can additionally be based on the classification of at least some of the subsections as incoming and outgoing. In particular, incoming subsections can be excluded from the classification as multiply fed subsections.

[0039] The proposed embodiment can advantageously enable an improved classification of the identified sections into single and multiply fed subsections.

[0040] In a further advantageous embodiment of the proposed method, a time point of contrast agent inflow can be identified for each edge-area pixel based on the time-intensity curve. Step b) can include a comparison of the respective contrast agent inflow times. Furthermore, subsections with earlier contrast agent inflow times can be classified as inflowing subsections, and subsections with later contrast agent inflow times as outflowing subsections.

[0041] Advantageously, for each of the peripheral image points representing one of the segments of the branched hollow organ, a time point of contrast agent inflow, in particular a bolus arrival time, can be determined based on the respective time-intensity curve. By comparing the times of contrast agent inflow, especially the bolus arrival times, segments with comparatively earlier and comparatively later times of contrast agent inflow, especially bolus arrival times, can be identified. Segments of the branched hollow organ that exhibit a comparatively early time point of contrast agent inflow, especially an early bolus arrival time, can be classified as receiving segments.Furthermore, sections of the branched hollow organ which have a comparatively late time of contrast agent influx, in particular a late bolus arrival time, can be classified as defecation sections.

[0042] The proposed embodiment can advantageously enable an improved classification of the identified sections into single and multiply fed subsections.

[0043] In a further advantageous embodiment of the proposed method, step b) can include applying a related component analysis to the image dataset. This allows the identification of the image pixels that represent the sections of the branched hollow organ. Sub-datasets can then be provided based on these identified pixels.

[0044] Advantageously, connected component analysis can be based on graph theory. The analysis can be performed starting from a starting pixel of the image dataset. This starting pixel could, for example, be a boundary pixel representing the contrast agent flow. The boundary pixel could be a pixel of the image dataset that directly borders the boundary contour and / or boundary surface and / or is located within a predefined spatial distance of the boundary contour and / or boundary surface. The starting pixel can be identified, for example, by segmentation. This segmentation could be based on comparing the image values ​​of the pixels in the image dataset with a predefined threshold.Starting from the initial image point, several image points, particularly those arranged in pairs adjacent to each other, can be identified based on the analysis of related components. These image points represent one of the sections of the branched hollow organ. Advantageously, the identification of image points along a section of the branched hollow organ can be stopped when another peripheral image point is identified as representing that section, particularly the contrast agent flow.

[0045] Advantageously, by repeatedly and / or in parallel, especially simultaneously, applying the analysis of related components to the image data set, all sections of the branched hollow organ depicted therein, in particular the respective associated pixels of the image data set, can be identified.

[0046] The sub-datasets can be provided based on the identified pixels. In particular, the sub-datasets can include the pixels identified for the respective classified sub-section.

[0047] The proposed embodiment can advantageously enable improved identification of the sections of the branched hollow organ, particularly in the case of at least partial overlap of sections of the branched hollow organ.

[0048] In a further advantageous embodiment of the proposed method, common image points of the identified image points can be identified along the depicted flow direction of the contrast medium flow in the identified sections, each of which depicts at least two of the identified sections. Subsections of the identified sections depicted by common image points can then be classified as multiply supplied subsections of the branched hollow organ.

[0049] At least some of the identified sections of the branched hollow organ may exhibit at least one branch point, and in particular, multiple branch points. Advantageously, along the depicted direction of contrast medium flow in the identified sections, common image points can be identified, each of which depicts at least two of the identified sections. In other words, the identified image points that depict a common subsection of at least two partially distinct sections of the branched hollow organ can be identified as common image points. The common subsection depicted by the common image points can be classified as a multiply fed subsection of the branched hollow organ.

[0050] The proposed embodiment can advantageously enable an improved classification of the identified sections into single and multiply fed subsections.

[0051] In a further advantageous embodiment of the proposed method, step a) can include acquiring a mask image of the object under investigation, which depicts the object without the contrast medium flow in the branched hollow organ. Furthermore, the image acquisition can include acquiring a fill image of the object under investigation, which depicts the object with the contrast medium flow in the branched hollow organ. Finally, the image data set can be provided as a difference image set of the fill image and the mask image.

[0052] The acquisition of the mask data set and the fill data set can, in particular, include acquiring and / or reading a computer-readable data storage device and / or receiving data from a data storage unit, such as a database. Furthermore, the mask and image data sets can be provided by a provisioning unit of a medical imaging device for the purpose of receiving the mask and fill data sets. The medical imaging device can, for example, include a magnetic resonance imaging (MRI) system and / or a computed tomography (CT) system and / or a medical X-ray device, in particular a medical C-arm X-ray device, and / or an ultrasound device and / or a positron emission tomography (PET) system.

[0053] The mask dataset can image the object of investigation, in particular the hollow organ, within a first temporal phase, specifically a mask phase. Advantageously, no contrast medium is present in the object of investigation, particularly the branched hollow organ, during this first temporal phase. The filling dataset can image the object of investigation, in particular the branched hollow organ, in a second temporal phase, specifically a filling phase. During this second temporal phase, the contrast medium is present in the object of investigation, in particular the branched hollow organ. Specifically, during the second temporal phase, the contrast medium flows from an injection site on the object of investigation through the branched hollow organ.

[0054] The mask and fill data sets can spatially represent the object of study, in particular the branched hollow organ, in 2D or 3D within the respective temporal phase. Furthermore, the mask and / or fill data sets can represent the object of study, in particular the branched hollow organ, in a temporally resolved manner.

[0055] Providing the image data set can include a subtraction of the mask and fill data set, particularly pixel-by-pixel. The image data set can be provided as a difference image data set, specifically a subtraction image data set, of the fill and mask data set.

[0056] The proposed embodiment can advantageously enable improved identification of the sections of the branched hollow organ.

[0057] In a further advantageous embodiment of the proposed method, step a) can include capturing several projection images of the object under investigation from different projection directions. The image data set can then be reconstructed from the multiple projection images.

[0058] Advantageously, the imaging device for capturing the projection images can comprise a source, in particular an X-ray source, and a detector, in particular an X-ray detector. The source and the detector can be arranged in a defined configuration relative to each other, in particular opposite each other. Furthermore, the defined configuration of source and detector can be movably mounted, in particular rotatably and / or translatably, for example, with respect to the object under investigation. The source can be configured to emit radiation, in particular X-rays, for illuminating the object under investigation. In particular, the source can be configured to emit a cone beam or a fan beam for illuminating the object under investigation. A central ray and / or mid-ray of the radiation emitted by the source can define a projection direction.The detector can be configured to detect radiation, particularly after interaction with the area under investigation. Furthermore, the detector can be configured to generate projection images based on the detected radiation.

[0059] The imaging device may be designed in particular as an X-ray device, for example as a C-arm X-ray device, O-arm X-ray device or computed tomography system (CT system).

[0060] Advantageously, the image data set can be reconstructed from the multiple projection images, for example by means of a back projection, especially a filtered one.

[0061] The proposed embodiment can advantageously enable 3D detection of the dynamics of the contrast agent flows in the branched hollow organ.

[0062] In a second aspect, the invention relates to a provisioning unit which is configured to execute a proposed method for providing a result data set.

[0063] The advantages of the proposed provisioning unit essentially correspond to the advantages of the proposed computer-implemented method for providing a result data set. Features, advantages, or alternative embodiments mentioned here can likewise be transferred to the other claimed subject matter and vice versa.

[0064] The provisioning unit may advantageously comprise an interface, a processing unit, and / or a storage unit. The interface, processing unit, and storage unit may be configured to perform steps a) through d) of the proposed procedure for providing a result data set. In particular, the interface may be configured to perform steps a) and d). Furthermore, the processing unit and / or the storage unit may be configured to perform steps b) and c).

[0065] In a third aspect, the invention relates to a medical imaging device comprising a proposed delivery unit. The imaging device can be configured to acquire the image data set. The advantages of the proposed imaging device essentially correspond to the advantages of the proposed method for providing a result data set and / or the proposed delivery unit. Features, advantages, or alternative embodiments mentioned here can likewise be transferred to the other claimed items and vice versa.

[0066] The medical imaging device may include, for example, a magnetic resonance imaging (MRI) system and / or a computed tomography (CT) system and / or a medical X-ray machine, in particular a medical C-arm X-ray machine, and / or an ultrasound machine and / or a positron emission tomography (PET) system.

[0067] Advantageously, the imaging device can be designed to capture, in particular record, the image data set.

[0068] In a fourth aspect, the invention relates to a computer program product comprising a computer program that can be directly loaded into a memory of a provisioning unit, with program sections to execute all steps of a proposed method for providing a result data set when the program sections are executed by the provisioning unit.

[0069] The computer program product can comprise software with source code that still needs to be compiled and bound or only interpreted, or executable software code that only needs to be loaded into the deployment unit for execution. The computer program product enables the method for providing a result data set via a deployment unit to be executed quickly, identically, and robustly. The computer program product is configured to execute the process steps according to the invention via the deployment unit.

[0070] The computer program product is, for example, stored on a computer-readable storage medium or on a network or server, from where it can be loaded into the processor of a delivery unit, which may be directly connected to the delivery unit or be designed as part of the delivery unit. Furthermore, control information of the computer program product can be stored on an electronically readable data carrier. The control information of the electronically readable data carrier can be designed such that, when the data carrier is used in a delivery unit, it performs a method according to the invention. Examples of electronically readable data carriers are a DVD, a magnetic tape, or a USB flash drive on which electronically readable control information, in particular software, is stored.If this control information is read from the data carrier and stored in a provisioning unit, all embodiments of the methods described above can be carried out according to the invention.

[0071] A largely software-based implementation has the advantage that existing deployment units can be easily retrofitted via a software update to operate according to the invention. Such a computer program product may, in addition to the computer program itself, optionally include additional components such as documentation and / or additional components, as well as hardware components such as hardware keys (dongles, etc.) for using the software.

[0072] Exemplary embodiments of the invention are shown in the drawings and are described in more detail below. The same reference numerals are used for identical features in different figures. The figures show: Fig. 1 to 4 schematic representations of various advantageous embodiments of a proposed method for providing a result data set, Fig. 5 a schematic representation of an advantageous embodiment of a proposed provisioning unit, Fig. 6 a schematic representation of an exemplary identification of sections and classification of subsections of the identified sections of a branched hollow organ, Fig. 7 a schematic representation of an advantageous embodiment of a proposed medical imaging device as a medical C-arm X-ray device.

[0073] Fig. Figure 1 shows a schematic representation of an advantageous embodiment of a proposed method for providing a result data set PROV-ED. In a first step a), an image data set of an examination object can be acquired CAP-BD. This image data set can depict the contrast agent flow in a branched hollow organ of the examination object with spatial and temporal resolution. In a further step b), sections of the branched hollow organ can be identified in the image data set based on the depicted contrast agent flow ID-A. For example, the image data set can contain several pixels, each with a time-intensity curve. Identifying the sections ID-A of the branched hollow organ can involve identifying the contrast agent influx based on the time-intensity curves of the respective pixels.In a further step c), subsections of the identified segments can be classified as singly fed or multiply fed subsections based on the depicted flow direction of the contrast agent flow and the identification of confluences of the identified segments of the hollow organ (CL-TA) depicted in the image dataset. Subsections of the branched hollow organ located downstream of a confluence can be classified as multiply fed subsections (CL-TA). In a further step d), the result dataset can be generated based on the image dataset and the classified subsections of the branched hollow organ (PROV-ED). The result dataset can contain a subset for each of the classified subsections of the hollow organ.Furthermore, the sub-datasets can provide a dedicated representation of the contrast agent flow in the respective classified sub-section of the branched hollow organ.

[0074] Advantageously, the branched hollow organ can comprise a vascular tree. In this context, vascular segments, particularly arterial and / or venous, and / or at least part of a vascular malformation, can be identified as the segments of the vascular tree ID-A. Furthermore, the sub-datasets can provide a spatially and temporally resolved representation of the contrast agent flow in the respective classified sub-segment of the hollow organ. The dedicated representations of the contrast agent flow in the respective classified sub-segment of the branched hollow organ can include a visual distinguishing feature. Furthermore, the result dataset can comprise at least a partial superimposition, nesting, and / or composition of the multiple sub-datasets.

[0075] Advantageously, step b) can include applying a related component analysis to the image dataset. This allows the identification of the image pixels that represent the identified sections of the branched hollow organ. Furthermore, subsets can be provided based on these identified pixels. Additionally, along the depicted contrast agent flow direction within the identified sections, common pixels can be identified that represent at least two of the identified sections. Subsections of the identified sections represented by common pixels can then be classified as multiply fed subsections of the branched hollow organ (CL-TA).

[0076] Fig. Figure 2 shows a schematic representation of a further advantageous embodiment of a proposed method for providing a result data set PROV-ED. Step c) can include classifying CL2-TA at least some of the subsections into feeding and draining subsections based on the flow direction of the contrast agent depicted in edge pixels of the image data set. Furthermore, the edge pixels can depict the branched hollow organ in an edge region of the image data set. Additionally, the CL2-TA classification of the subsections into singly fed and multiply fed subsections can be based on the CL2-TA classification of at least some of the subsections as feeding and draining.

[0077] Furthermore, for each peripheral pixel, a time point of contrast agent influx can be identified based on the time-intensity curve. Step b) can include a comparison of the respective contrast agent influx times. Sub-segments with earlier contrast agent influx times can be classified as influxing segments, and sub-segments with later contrast agent influx times as outfluxing segments (CL2-TA).

[0078] The following is an example of a procedure: In this procedure, a 3D or 4D DSA dataset comprising multiple projection images can be acquired using a medical imaging device, such as a medical C-arm X-ray unit, with a contrast agent positioned within the subject during a filling phase. The contrast agent may have been injected into the branched hollow organ, particularly an aorta or vein, of the subject prior to the start of the filling phase. For example, the contrast agent injection may have been performed according to an injection protocol as described in the publication by Klostranec, Jesse M., et al., "Comparison of aortic arch and intravenous contrast injection techniques for C-arm cone beam CT: implications for cerebral perfusion imaging in the angiography suite." Academic Radiology 20.4 (2013): 509-518.In a further step, a 3D DSA reconstruction of the projection images can be performed to obtain a complete representation of the contrast-enhanced vascular system of the subject's head, particularly the brain. The vascular system, specifically a vascular tree, of the subject's head can represent the branched hollow organ of the subject. In a further step, a 4D DSA reconstruction can be performed using the projection images, for example, by means of a plausibility-based flow condition, with the image dataset being provided. The steps described above can be encompassed by the acquisition of the CAP-BD image dataset. Furthermore, the segments of the vascular system in the image dataset can be identified based on the depicted contrast agent flow (ID-A).Furthermore, subsections of the identified segments can be classified as singly fed and multiply fed subsections based on the depicted flow direction of the contrast agent flow and the identification of confluences of the identified vascular segments (CL-TA), which are depicted in the image dataset. Additionally, time information can be assigned to each pixel or defined subelement of the image dataset. In a further step, a time of contrast agent inflow, in particular a bolus arrival time, can be assigned to each pixel of the image dataset, especially by means of a plausibility-based flow condition, whereby the time of contrast agent inflow is determined for each pixel. In a further step, afferent and efferent vascular segments can be classified (CL2-TA), which represent an imaging window (English:Field-of-view (FOV) segments can be identified, for example, arteries entering the head, particularly the brain, of the subject and / or veins exiting the subject's head. Alternatively or additionally, feeding segments can be classified based on temporal information (CL2-TA), for example, by a comparatively low bolus arrival time or by identifying a flow direction, particularly an increase in the bolus arrival time of adjacent pixels. A 4D volume can be calculated for each feeding segment. An analysis of contiguous components can be applied to a boundary pixel of the feeding segment as the starting point. This can identify multiple volumes, in particular one volume for each of the feeding segments.The volumes may exhibit some overlap, for example, in an area of ​​a vascular malformation, particularly a nidus, and / or efferent veins. In a further step, the overlap of these volumes can be determined. Specifically, the common pixels that depict more than one of the connected segments can be identified. In a further step, the segments depicted in the volumes can be separated into subvolumes. Specifically, a sub-dataset can be provided for each of the classified segments of the branched vasculature. These sub-datasets can depict a respective sub-volume. The distinction can include afferent segments before a connection with other afferent segments and segments with multiple feeding sites.In a further step, a graphical representation of the partial volumes can be provided, individually or together, for example, color-coded. The respective image impression of the graphical representation of the partial volumes may have been processed differently in post-processing, for example, by windowing, to compensate for different mixtures of blood and contrast agent in the various feeding segments. For a future, especially planned, embolization treatment via an arterial access, the images of the feeding segments can, for example, be provided as a superimposition of a current image and / or representation of an examination area of ​​the subject during the procedure.

[0079] The proposed method advantageously reduces the number of images required for imaging complex AVMs, particularly reducing exposure time and radiation dose. Furthermore, it eliminates the need to fuse multiple 3D images, thus streamlining workflows. Separating data into subsets reduces the complexity of analyzing complex AVM structures, especially in cases of overlapping segments. A medical operator can advantageously evaluate any number of segments of the branched hollow organ separately or in combination, for example, by selectively enabling or disabling a graphical representation of the respective segment.

[0080] Fig. Figure 3 shows a schematic representation of a further advantageous embodiment of a proposed method for providing a result data set PROV-ED. Step a) can include acquiring a mask data set CAP-MD of the object under investigation, which depicts the object without the contrast medium flow in the hollow organ. Furthermore, step a) can include acquiring a fill data set CAP-FD of the object under investigation, which depicts the object with the contrast medium flow in the hollow organ. Additionally, step a) can include providing the image data set PROV-BD as a difference image data set between the fill data set and the mask data set.

[0081] Fig. Figure 4 shows a schematic representation of another advantageous embodiment of a proposed method for providing a result data set PROV-ED. Step a) can include capturing several projection images CAP-Pl of the object under investigation from different projection directions. Furthermore, the image data set can be reconstructed from the several projection images RECO-BD.

[0082] Fig. Figure 5 shows a schematic representation of a proposed provisioning unit PRVS. The provisioning unit PRVS can comprise a processing unit CU, a storage unit MU, and / or an interface IF. The provisioning unit PRVS can be configured to execute a proposed procedure for providing a result data set, in which the interface IF, the processing unit CU, and / or the storage unit MU are configured to execute the corresponding procedure steps. The interface IF, processing unit CU, and storage unit MU can be configured to execute steps a) through d) of the proposed procedure for providing a result data set. In particular, the interface IF can be configured to execute steps a) and d). Furthermore, the processing unit CU and / or the storage unit MU can be configured to execute steps b) and c).

[0083] Fig. 6. A schematic representation of an exemplary identification of sections and classification of subsections of the identified sections of a branched hollow organ HO. BD illustrates this with an image dataset containing a 3D representation of the branched hollow organ HO. Furthermore, in Fig. Figure 6 illustrates three subsets TD1, TD2, and TD3. The first subset, TD1, may contain an image of a first singly fed segment A1 of an identified segment of the branched hollow organ HO, for example, a left internal carotid artery. Furthermore, the second subset, TD2, may contain an image of a second singly fed segment A2 of an identified segment of the branched hollow organ HO, for example, a right internal carotid artery. In addition, the third subset, TD3, may contain an image of a vascular malformation GM, for example, a nidus, and a multiply fed segment A3 of an identified segment of the branched hollow organ HO, for example, a common vascular segment.

[0084] Fig.Figure 7 shows, as an example of a medical imaging device, a schematic representation of a medical C-arm X-ray device 37, comprising a PRVS delivery unit. The medical C-arm X-ray device 37 advantageously comprises a detector 34, in particular an X-ray detector, and a source 33, in particular an X-ray source, which are arranged in a defined configuration on a C-arm 38. The C-arm 38 of the C-arm X-ray device 37 can be movably mounted about one or more axes. To acquire, in particular record, the image data set CAP-BD of the examination object 31, which is positioned on a patient positioning device 32, the PRVS delivery unit can send a signal 24 to the X-ray source 33. The X-ray source 33 can then emit a beam of X-rays.When the X-ray beam strikes a surface of the detector 34 after interacting with the object under investigation 31, the detector 34 can send a signal 21 to the PRVS delivery unit. Based on the signal 21, the PRVS delivery unit can acquire projection images of the object under investigation (CAP-PI) and reconstruct the image data set (RECO-BD). The acquisition of the image data (CAP-BD) can include the acquisition of the projection images (CAP-PI) and the reconstruction of the image data set (RECO-BD).

[0085] The C-arm X-ray unit 37 can further comprise an input unit 42, for example a keyboard, and a display unit 41, for example a monitor and / or a display and / or a projector. The input unit 42 can preferably be integrated into the display unit 41, for example in the case of a capacitive and / or resistive input display. The input unit 42 can advantageously be configured to detect user input. For this purpose, the input unit 42 can, for example, send a signal 26 to the PRVS delivery unit. The PRVS delivery unit, in particular the C-arm X-ray unit 37, can be configured to be controlled depending on the user input, in particular the signal 26, especially for executing a method for providing a result data set PROV-ED.

[0086] The display unit 41 can advantageously be configured to display a graphical representation of the result data set. For this purpose, the provision unit PRVS can send a signal 25 to the display unit 41.

[0087] Using the proposed method, vascular, particularly neurovascular, areas can be separated in a 3D or 4D reconstruction of a single rotational scan using temporal information. It is advantageous to inject a contrast agent into the aorta or a vein before the rotational scan begins.

[0088] The schematic representations contained in the described figures do not depict any scale or size ratios.

[0089] Finally, it should be noted once again that the methods described in detail above and the devices shown are merely exemplary embodiments which can be modified in various ways by a person skilled in the art without departing from the scope of the invention. Furthermore, the use of the indefinite articles "a" or "an" does not preclude the possibility that the features in question may be present multiple times. Likewise, the terms "unit" and "element" do not preclude the possibility that the components in question consist of several interacting sub-components, which may also be spatially distributed.

[0090] The expression “based on” can, in the context of the present application, be interpreted in particular as “using”. Specifically, a formulation stating that a first feature is generated (alternatively: determined, ascertained, etc.) based on a second feature does not preclude the possibility that the first feature may be generated (alternatively: determined, ascertained, etc.) based on a third feature.

Claims

[1] Computer-implemented procedure for providing a result data set (PROV-ED), comprising: a) Acquiring an image data set (CAP-BD) of an investigation object (31), wherein the image data set spatially and temporally resolves a contrast medium flow in a branched hollow organ (HO) of the investigation object (31), b) Identifying sections (ID-A) of the branched hollow organ (HO) in the image dataset based on the depicted contrast agent flow, c) Classifying subsections (CL-TA) of the identified sections into singly fed and multiply fed subsections based on a depicted contrast medium flow direction and an identification of confluences of the identified sections of the branched hollow organ (HO) depicted in the image dataset, wherein sections of the branched hollow organ (HO) which are arranged downstream with respect to a confluence are classified as multiply fed sections, d) Providing the result data set (PROV-ED) based on the image data set and the classified subsections of the branched hollow organ (HO), wherein the result data set includes a subset of the classified subsections of the branched hollow organ (HO), wherein the subsets of data provide a dedicated representation of the contrast agent flow in the respective classified subsection of the branched hollow organ (HO). [2] Method according to claim 1, wherein the branched hollow organ (HO) comprises a vascular tree, wherein, in particular, arterial and / or venous, vessel segments and / or at least part of a vascular malformation are identified as the segments of the vascular tree. [3] Method according to claim 1 or 2, wherein the partial data sets have a spatially and temporally resolved representation of the contrast medium flow in the respective classified subsection of the branched hollow organ (HO). [4] Method according to any of the preceding claims, wherein the dedicated representations of the contrast medium flow in the respective classified subsection of the branched hollow organ (HO) have a visual distinguishing feature. [5] Method according to claim 4, wherein the result data set comprises at least a partial superimposition and / or nesting and / or composition of the multiple sub-data sets. [6] Method according to any of the preceding claims, wherein the image data set comprises several pixels, each with a time intensity curve, wherein the identification of the sections of the branched hollow organ (OH) comprises identifying a contrast agent influx based on the time intensity curves of the respective pixels. [7] Method according to any of the preceding claims, where step c) classifies (CL2-TA) at least some of the subsections into supplying and draining subsections based on a flow direction of the contrast agent mapped in edge pixels of the image dataset, where the edge pixels depict the hollow organ in an edge region of the image data set, where the classification of the subsections (CL-TA) into singly fed and multiply fed subsections is additionally based on the classification (CL2-TA) of at least one part of the subsections as feeding and draining. [8] Method according to claims 6 and 7, where, for each peripheral pixel, a time point of contrast agent influx is identified based on the time intensity curve, where step b) includes a comparison of the respective times of contrast agent inflow, whereby sections with earlier times of contrast agent inflow are classified as inflowing sections and sections with later times of contrast agent inflow are classified as outflowing sections. [9] Method according to any of the preceding claims, where step b) involves applying a related components analysis to the image dataset, wherein the image datasets are identified which represent the identified sections of the branched hollow organ (HO), wherein the subsets are provided based on the identified image datasets. [10] Method according to claim 9, wherein, along the depicted flow direction of the contrast agent flow in the identified sections, common image points of the identified image points are identified, each of which depicts at least two of the identified sections, where subsections of the identified sections, which are mapped by common pixels, are classified as multiply fed subsections of the branched hollow organ (HO). [11] Method according to any of the preceding claims, wherein step a) comprises: - Acquisition of a mask data set (CAP-MD) of the subject (31), which depicts the subject (31) without the contrast medium flow in the branched hollow organ (HO), - Acquisition of a filling data set (CAP-FD) of the object of investigation (31), which depicts the object of investigation (31) with the contrast medium flow in the branched hollow organ (HO), - Providing the image data set (PROV-BD) as a difference image data set of the fill data set and the mask data set. [12] Method according to any of the preceding claims, where step a) includes capturing several projection images (CAP-PI) of the object under investigation from different projection directions, the image data set is reconstructed from the multiple projection images (RECO-BD). [13] Provisioning unit (PRVS) which is configured to perform a method according to any of the preceding claims. [14] Medical imaging device comprising a provisioning unit (PRVS) according to claim 13, wherein the imaging device is configured to capture the image data set (CAP-BD). [15] Computer program product comprising a computer program which can be directly loaded into a memory of a provisioning unit (PRVS), comprising program sections to execute all steps of a method according to any one of claims 1 to 12 when the program sections are executed by the provisioning unit (PRVS).

Images