Method and apparatus for processing a series of medical x-ray-based images

Abstract

The invention relates to a method for processing a series of medical x-ray-based images (B) which were taken at time intervals during the influx of the contrast medium after administration of a contrast medium from a patient (P), the method comprising the steps: - Provision of the series of image recordings (B), and selection of a series of temporally successive relevant images (B) from these image recordings (B), - Examining the image elements (V) of the relevant images (B) for a maximum image value, wherein when examining an image element (V), the value of this image element (V) is considered in all relevant images (B) at the same image coordinate, and the image in which the value of this image element (V) is maximal is identified as the maximum image for this image element (V). - Setting a start image (S) from the relevant images, - Generating a result image (E) in which those image elements (V) whose maximum image is the start image (S) are displayed in a different color than those image elements (V) whose maximum image is another relevant image that was taken after the start image (S) and / or than those image elements (V) whose image values ​​in all relevant images (B) lie within a specified range of variation, - Outputting the result image (E). Furthermore, the invention comprises a device, a control unit and a medical imaging system.

Description

[0001] The invention relates to a method and a device for processing a series of medical x-ray-based images, a control device for controlling a medical imaging system, in particular a CT system, and a medical imaging system.

[0002] The present invention is in the field of image data analysis of medical image data, which is used to create data-based representations of organs or tissues, in particular for visual representation and / or functional representation of organs or tissues.

[0003] In the case of an acute stroke, several imaging studies (scans) must be performed to determine the cause and severity of the stroke. Typically, a computed tomography (CT) scan without contrast is performed first to determine if the patient has a hemorrhage and to look for early signs of a stroke (e.g., signs of hyperdense media). If no hemorrhage is detected, a CT angiography (CTA) is performed, followed by a perfusion CT scan. CTA can locate the clot, while perfusion CT provides information about the severity of the stroke and answers the question of whether there are still areas of tissue that can be saved by treatment.

[0004] Contrast-enhanced imaging data, obtained through X-ray computed tomography (CT) using contrast agents, is frequently employed. Contrast agents for X-ray CT are often iodine-based. This is useful for highlighting structures such as blood vessels, which would otherwise be difficult to distinguish from their surroundings. Images are often acquired both with and without X-ray contrast. The use of contrast agents can also help obtain functional information about tissues, such as perfusion information, which describes the degree of blood flow to a tissue area.

[0005] Perfusion CT thus provides important data, especially when the stroke occurred recently. It is by far the most frequently used method in practice for determining the core infarct and the penumbra.

[0006] Unfortunately, perfusion CT has some disadvantages. These include the need for additional contrast agent administration (after an initial contrast agent administration for CTA), the inability to use bolus triggering (a baseline is always required for perfusion calculation), and the relatively high radiation exposure due to prolonged exposure.

[0007] In the first six hours after the onset of stroke symptoms, it is sufficient to obtain a plain radiograph and a CT angiogram (CTA) of the patient. Further examinations are only necessary after a longer period. In many cases, however, a perfusion scan is also performed within the first six hours, as this provides the best insight into the processes in the brain. Often, only a multiphase CTA is used (usually consisting of three phases), the first phase of which is scanned using bolus triggering precisely at the moment of arterial contrast enhancement. This first phase covers the aortic arch up to the skullcap; the subsequent phases, at intervals of 7-10 seconds, each image only of the brain itself. These images are normally evaluated by simply viewing them side by side. The time of maximum contrast enhancement provides information about the patient's condition.It is also possible to calculate “quasi-perfusion results” from these data, but due to the lack of a (contrast-free, but otherwise equivalently acquired) baseline dataset, the limited temporal sampling, and the possible absence of the time of maximum enhancement, these data can only represent an approximation.

[0008] Furthermore, it is possible to color-code the temporal increase of the contrast agent. Vessels with the highest contrast agent level in the first image are shown in red, those in the second image in green, and those in the last image in blue. However, only the vessels can be displayed (similar to CT flow visualization), and no conclusions can be drawn about areas that show no enhancement.

[0009] It is an object of the present invention to provide a method and a device for processing a series of medical X-ray-based images, a control device for controlling a medical imaging system, in particular a CT system, and a medical imaging system, with which the disadvantages described above are avoided. In particular, it is an object of the invention to process CTA images in such a way that they can be used more effectively by a physician for the examination of a stroke.

[0010] This problem is solved by a method according to claim 1, a device according to claim 10, a control device according to claim 12 and a medical imaging system according to claim 13.

[0011] A method according to the invention serves to process a series of medical X-ray-based images, in particular CTA, which were taken at time intervals during the influx of contrast medium from a patient after administration of contrast medium. The method comprises the following steps: - Provision of the series of image recordings, and selection of a series of chronologically successive relevant images from these image recordings, - Examining the image elements of the relevant images for a maximum image value, whereby when examining an image element, the value of this image element is considered in all relevant images at the same image coordinate, and the image in which the value of this image element is maximal is identified as the maximum image for this image element. - Setting a start image from the relevant images, - Generating a result image in which those image elements whose maximum value is the start image are displayed in a different color than those image elements whose maximum value is another relevant image taken after the start image and / or than those image elements whose image values ​​in all relevant images lie within a specified range of variation. - Outputting the result image.

[0012] The method according to the invention is preferably a computer-implemented method.

[0013] The provision of the image series can be achieved by providing the corresponding image data. This image data can be provided by an imaging device, an image data storage system, a network with access to stored image data, or any other data-based provision.

[0014] The resulting image can be output to an output unit (e.g., a screen) and / or in a data-based manner, e.g., by saving it to a data storage device or by making it available on a network.

[0015] This method is particularly advantageous for supplementing CTA images in stroke investigations. However, other beneficial applications are also possible, such as processing a series of medical X-ray-based images taken at time intervals during the contrast agent infusion of a patient (human or animal) after administration of a contrast agent. This processing includes at least color highlighting of areas within these images.

[0016] First, the series of images is provided. This can be done by taking the images or downloading them from a database. The images are preferably CT scans, especially from a CTA, but could also be MRI scans.

[0017] From these images, a series of temporally sequential relevant images is selected. These relevant images can be all images or only a subset. Thus, the images can certainly be 3D image stacks (i.e., a three-dimensional image of a region of interest, ROI), and the series of relevant images can comprise a cross-sectional view (at the same coordinate) of the respective images. Furthermore, images from a smaller time interval can be selected from the images. For the sake of simplicity, let's assume an example series of three to five CT scans, with the relevant images from each scan comprising a 2D cross-sectional view, all depicting the same area of ​​the brain.

[0018] Once the relevant images are available, their image elements (essentially pixels or voxels) are examined for a maximum image value. The same image coordinate is always considered for each image element, meaning essentially the same location within the imaged organ. For each examined image element, the image representing the maximum value for that pixel is determined as the maximum image. It should be noted that noise is generally irrelevant. Only a value that is clearly identifiable as a maximum should be considered, as will be explained in more detail below. There may be image elements that do not exhibit a maximum but, at most, lie within a (small) range of values. These can represent parenchyma and will be discussed in more detail below.

[0019] During this examination, irrelevant areas such as the background or bones should be excluded. This can be achieved by defining a mask before the examination that excludes areas that should not be examined. Alternatively, image segmentation can be performed, examining only image elements belonging to a predefined segment. Such techniques are well-established. Therefore, if relevant brain images are available and a stroke is to be investigated, it is advantageous to remove the skull bones and the background. The ventricles and atrium can also be excluded from the examination.

[0020] In practice, the grayscale values ​​of individual voxels (image elements) in layered images can be examined for maximum brightness. Preferably, all image elements to be examined are first rasterized, and ideally, a mask is created during this process. For each pixel, this mask contains the number of the maximum image in the series. An entry can also be made if no maximum was found, e.g., the value "0". This mask can then be used for subsequent processing steps. This has the advantage that the voxels only need to be rasterized once. It should be noted that the mask can also contain information about the values ​​of voxels in preceding and subsequent images, e.g., in the form of a vector (preceding value, maximum image number, subsequent value).

[0021] It is therefore now known for each examined pixel which of the images is the maximum image, or possibly that this pixel does not have a maximum.

[0022] Next, a starting image is selected from the relevant images. Generally, the starting image should show the first contrast agent peak, which is usually interpreted as an arterial blood supply. In the brain, for example, this characterizes healthy areas. In the liver, however, arterially perfused areas can be interpreted as diseased areas. The starting image can be selected in several ways. For example, the entire series of relevant images may have already been selected so that the first image shows the initial influx of contrast agent (i.e., usually arterial blood). In this case, the first relevant image in time can simply be selected as the starting image. Alternatively, it can be determined first in which image of the series of relevant images this peak is found.The earliest relevant image can be identified, which has been identified as the maximum image for more than a predefined number of image elements. It should be noted that, statistically, some image element will have its maximum in the earliest image. To exclude such outliers, a minimum number of image elements should be specified that must have their maximum in this first image.

[0023] It should be noted that different areas of an organ receive arterial blood at different times. Therefore, multiple starting images can be defined for different areas of the organ. However, these areas should be predefined or determined beforehand. Below is a method (comparing pre- and post-values) for differentiating arterial from collateral blood flow with respect to a maximum. This method can be considered as an additional check for each image point whose maximum value follows the starting image.

[0024] It is now known relative to which starting time value (the starting image represents a point in time) the maxima of the image values ​​should be considered.

[0025] The resulting image is now being generated. This resulting image can be a relevant image that is modified, or it can be created anew. Since the subject of the relevant images should be shown later, the resulting image should either depict this subject or be able to be overlaid on an additional image with the subject (possibly one of the relevant images), so that it complements the former with color markings.

[0026] In this result image, those image elements whose maximum image is the start image are displayed in a different color than those image elements whose maximum image is another relevant image that was taken after the start image and / or those image elements whose image values ​​in all relevant images lie within a specified range of variation (e.g. around a mean value).

[0027] Those image elements whose maximum enhancement is the starting image (i.e., those showing arterial blood flow) can be displayed in normal grayscale, i.e., unchanged, especially if the subject is the brain. It may be possible to use a different image as the starting image (e.g., the last image in the series), since the starting image shows maximum enhancement. However, if the subject is something other than the brain, such as the liver, these image elements could be color-coded and, in particular, displayed in a warning color (e.g., red), as they might indicate, for example, bleeding in the liver.

[0028] Image elements whose maximum image is another relevant image acquired after the initial image are preferentially marked with an initial warning color, e.g., yellow, at least when the subject is the brain. These typically indicate collaterally perfused tissue (in the brain, the penumbra). Here, it is particularly advantageous if the color coding correlates with the position of the maximum image in the series of relevant images (the image number), meaning different colors or shades are used to indicate the point of maximum. However, if the subject is something other than the brain, e.g., the liver, these image elements could be displayed in their normal gray tones, as these could also represent healthy tissue.

[0029] Image elements whose values ​​in all relevant images fall within a predefined range (e.g., around a mean value) do not receive contrast agent. Since the contrast agent essentially represents blood flow, it can be assumed that these areas are not perfused and therefore severely damaged (in the brain, this corresponds to parenchyma). These image elements are preferentially marked with a second warning color, e.g., red.

[0030] The resulting image, which may show (different) color markings and preferably also the subject in grayscale, is then output so that it can be reviewed for diagnostic purposes, e.g., by a physician or an examination algorithm. The examination itself is not part of the invention, the purpose of which is to generate the resulting image.

[0031] An apparatus according to the invention serves to process a series of medical X-ray-based images which have been taken at time intervals during the influx of contrast medium from a patient after administration of the contrast medium, in particular according to a method according to the invention. The apparatus comprises the following components: - a data interface designed to receive the series of image recordings, - a selection unit designed for selecting a series of temporally successive relevant images from these image recordings, - a unit of analysis designed to examine the image elements of the relevant images for a maximum image value, wherein when examining an image element the value of this image element is considered in all relevant images at the same image coordinate, and the image in which the value of this image element is maximal is identified as the maximum image for this image element, - a start image unit designed to define a start image from the relevant images, - a marking unit designed to generate a result image in which those image elements whose maximum image is the start image are displayed in a different color than those image elements whose maximum image is another relevant image that was taken after the start image and / or than those image elements whose image values ​​in all relevant images lie within a specified range of variation, - a data interface designed to output the result image.

[0032] The function of the device's components has already been described. The device is preferably designed for carrying out a method according to the invention.

[0033] Further embodiments of the device according to the invention follow directly from the various embodiments of the method according to the invention, and vice versa. In particular, individual features and corresponding explanations, as well as advantages relating to the various embodiments of the method according to the invention, can be transferred analogously to corresponding embodiments of the device according to the invention. In this case, the functional features of the method are embodied by corresponding units or modules of the system. In particular, the device according to the invention is designed or programmed to execute the computer-implemented method according to the invention.

[0034] The invention thus provides a false-color image as a result image, which, for example, can highlight the penumbra and / or parenchyma in a brain examination. For example, in three image acquisitions, the first of which is the starting image, voxels that appear brightest in the second image are displayed in yellow, and voxels that appear brightest in the third image are displayed in orange. Voxels that have the same gray value in all images (possibly within a predefined range) are displayed in red to identify them as stroke core areas. Voxels that exhibit the highest enhancement in the first image are further represented by their HU value on a grayscale.

[0035] Unlike other approaches that only consider the vessels, the approach described here considers both the vessels and the parenchyma. This allows parenchymal areas through which a contrasted vessel still runs, but which no longer show blood flow, to be marked with different colors. It also allows for the visualization of areas that no longer show blood flow, i.e., areas where no vessel is visible. This makes it possible to visualize the infarct core even if it is not yet visible in the native phase.

[0036] A control device according to the invention serves to control a medical imaging system, in particular a CT system or a diagnostic system. It comprises a device according to the invention and / or is designed to carry out a method according to the invention.

[0037] A medical imaging system according to the invention is preferably a CT system (but could also be an MRI system) or a diagnostic system, and comprises a control device according to the invention.

[0038] The invention can be implemented, in particular, in the form of a computer unit with suitable software. The computer unit can, for example, comprise one or more cooperating microprocessors or the like. In particular, it can be implemented in the form of suitable software program components within the computer unit. A largely software-based implementation has the advantage that even previously used computer units can be easily retrofitted by a software or firmware update to operate according to the invention. In this respect, the problem is also solved by a corresponding computer program product with a computer program that can be directly loaded into a memory device of a computer unit, containing program sections to execute all steps of the method according to the invention when the program is run in the computer unit.In addition to the computer program itself, such a computer program product may include additional components such as documentation and / or additional components, including hardware components such as hardware keys (dongles, etc.) for using the software.

[0039] For transport to the computer unit and / or for storage on or in the computer unit, a computer-readable medium, such as a memory stick, a hard drive or other portable or permanently installed data carrier, can be used, on which the program sections of the computer program that can be read and executed by a computer unit are stored.

[0040] Further, particularly advantageous embodiments and developments of the invention result from the dependent claims and the following description, wherein the claims of one claim category may also be further developed analogously to the claims and description parts of another claim category and, in particular, individual features of different embodiments or variants may be combined to form new embodiments or variants.

[0041] According to a preferred embodiment of the method, in the resulting image, image elements with different maximum images are displayed in different colors with respect to those image elements whose maximum image was captured after the start image. Voxels that appear brightest in a later phase (an image captured after the start image) are preferably displayed in different colors or color shades (e.g., yellow / orange or different shades of yellow or a transition from yellow to red) according to their capture time (the image number in the series).

[0042] According to a preferred embodiment of the method, in the resulting image, image elements for which a maximum image has been identified are displayed in a different color than those whose image values ​​lie within a predefined range across all relevant images, with regard to those image elements whose maximum image was captured after the start image. Image values ​​that do not have a maximum in the series are considered unfilled. They can, for example, be marked in red, while image points whose maximum lies after the start image are displayed in yellow or orange.

[0043] According to a preferred embodiment of the method, for image coordinates of the relevant images relative to the respective maximum image, a prior value of the corresponding image element in a temporally preceding image and a subsequent value of a corresponding image element in a temporally subsequent image are compared. For the image points preferred here, the maximum is not located in the first image. Thus, for an image point, the image that precedes the maximum image in the series can be considered, and the value of the corresponding image point can be taken as the "prior value." For image points whose maximum is not located in the last image, a corresponding subsequent value can also be determined (in the image that follows the maximum image in the series).

[0044] In tissue with collateral blood flow, no contrast agent is initially visible, then the maximum contrast level is reached, followed by a decrease. The pre-image will therefore show less contrast agent (and be darker) than the post-image. The situation is different in tissue with arterial blood flow. Here, a pixel before the maximum will appear brighter than after. In this arterially perfused tissue, the pre-image value will therefore be higher than the post-image value.

[0045] Therefore, according to the procedure, if the initial value is greater than the subsequent value, an image element is preferably assigned to arterially perfused tissue. Alternatively or additionally, if the initial value is less than the subsequent value, this image element is preferably assigned to non-arterially perfused, particularly collaterally perfused, tissue. In the resulting image, arterially perfused tissue is then preferably marked with a different color than non-arterially perfused tissue. Here, it can preferably be assumed that voxels that have their (significant) maximum in the initial image are always arterially perfused, and voxels that have their (significant) maximum in subsequent images are generally collaterally perfused, unless this voxel is darker in the subsequent image than in the initial image. In that case, the voxel is also arterially perfused.In this way, areas can be defined where the contrast agent only begins to be delivered after the initial image, and a new, individual initial image can be defined for each area. Different areas can then be marked differently.

[0046] As mentioned above, simply considering the highest value of a pixel can be disadvantageous in practice, as it may be subject to noise. A maximum should be clearly identifiable. Otherwise, this pixel should be assigned to a parenchyma. According to a preferred embodiment of the method, therefore, when examining the image elements for a maximum image value, only those image elements whose maximum - exceeds a predetermined absolute limit, and / or - exceeds a predetermined distance from the mean value of the pixel, and / or - exceeds a specified distance from the average value of the neighboring pixels and / or - lies outside the specified range of variation.

[0047] It can also be advantageous to perform image smoothing or morphological operations before evaluation. This is a known technique and reduces unwanted fluctuations in image values.

[0048] According to a preferred embodiment of the method, the images are brain images, preferably CT scans of the brain. It is preferred that the relevant images are cross-sectional images from successive CT scans. It is particularly preferred that image areas for which a maximum image has been identified, acquired after the initial image, and / or which are assigned to non-arterially, particularly collaterally, perfused tissue, are considered the penumbra, and / or that image elements whose image values ​​lie within a predetermined range of variation in all relevant images are considered the parenchyma.

[0049] The series of images preferably comprises fewer than ten consecutive shots, ideally fewer than five. In principle, three shots are sufficient, the first of which should be the starting image.

[0050] It is preferred that, in one embodiment of the method, image values ​​of image elements whose maximum image was acquired after the start value are compared with image values ​​of image elements whose maximum image is the start image, and from this, blood flow or the number of blood vessels is estimated. Thus, the gray tones of collaterally perfused tissue are compared with the gray tones of arterially perfused tissue.

[0051] Sometimes it is very helpful to know, in addition to the previously determined information about when blood arrives at collaterally perfused tissue, how much blood arrives compared to healthy tissue. If the blood arrives late (in the case of collateral supply), it makes a significant difference to the physician whether this collateral supply is still able (at least at the time of the examination) to transport a similar amount of blood, and thus oxygen, to the affected region, or whether almost none arrives. There are various ways to quantify this amount or the difference. One example would be to create an image by averaging all time points, which would then show the average amount of blood arriving. Another possibility would be to create an image that represents the maximum across all time points, thus showing the maximum amount of blood arriving.

[0052] For this purpose, individual pixels should not be considered, but rather (larger) collaterally supplied areas. Averaged gray values ​​(over a given area) would be compared between different such areas to determine the relative blood volume. For example, an area of ​​collaterally perfused tissue could be compared with a corresponding area of ​​healthy, i.e., arterially perfused, tissue. A comparison of areas on the left hemisphere with their counterparts on the right hemisphere, and vice versa, is particularly suitable for this purpose, since most strokes are unilateral and the brain is otherwise largely symmetrical.

[0053] Unfortunately, "reference" time phases without any contrast agent are rarely available in standard mCTA scans. However, there is the (optional) possibility of including a prior non-contrast scan, which is usually acquired. While the grayscale values ​​would differ somewhat from those in the CTA scan due to the different acquisition method, it is quite possible to better identify certain areas (e.g., gray vs. white matter) in the non-contrast image. During the comparison, gray and white areas could then be considered separately to better distinguish grayscale differences caused by contrast agent or hemodynamics from those caused by the inherently different tissue types. Alternatively, an atlas registered to the image could be used to determine gray and white matter (even without a non-contrast image).

[0054] It is therefore preferred that the relevant image values ​​are read from other relevant images than the maximum image, preferably in a previous image.

[0055] Furthermore, it is preferred that in one embodiment of the method the comparison - with corresponding areas of the opposite hemispheres of the brain or - with areas with corresponding image values ​​or - is performed on areas with corresponding brain matter (gray matter, white matter). It is preferred that the mean values ​​of the image data in the respective areas are calculated and compared.

[0056] Before determining the phase of highest enhancement per voxel, relatively strong smoothing should be performed in the image space to reduce misattributions due to noise and image artifacts. A suitable method for this is one that avoids mixing the signals from (larger) vessels, bone, and parenchyma. This can be achieved, for example, by prior segmentation of bones and vessels and corresponding masking in the filter core, or alternatively / additionally by using bilateral filters that introduce an additional weighting of the contribution based on voxel intensity.

[0057] According to a preferred embodiment of the method, the image acquisitions or the relevant images in the image space are smoothed before the image elements are examined. This has the advantage that incorrect assignments due to noise and image artifacts can be reduced. It is preferred that bones and vessels are first segmented and then masked out in the filter core, and / or that a bilateral filter is used that weights the contribution of voxels based on their voxel intensity. This embodiment avoids the mixing of signals from (larger) vessels, bones, and parenchyma.

[0058] A preferred device comprises a filter unit designed for smoothing images in the image space. It is preferred that the filter unit is designed for segmenting bones and vessels in images, subsequently masking out the segmented areas, and filtering the other areas, and / or that the filter unit comprises a bilateral filter.

[0059] The use of AI-based methods (AI: "Artificial Intelligence") is preferred for the method according to the invention. Artificial intelligence is based on the principle of machine learning and is generally implemented with a learning algorithm that has been trained accordingly. The English term "machine learning" is frequently used for machine learning, and this also includes the principle of "deep learning".

[0060] Preferably, components of the invention are provided as a "cloud service." Such a cloud service serves to process data, particularly using artificial intelligence, but can also be a service based on conventional algorithms or a service where human evaluation takes place in the background. Generally, a cloud service (hereinafter also referred to simply as "cloud") is an IT infrastructure in which, for example, storage space or computing power and / or application software is provided via a network. Communication between the user and the cloud takes place via data interfaces and / or data transmission protocols. In the present case, it is particularly preferred that the cloud service provides both computing power and application software.

[0061] In a preferred method, data obtained within the scope of the invention is provided to the cloud service via the network. This cloud service comprises a computing system that typically does not include the user's local computer. The method can be implemented using a command structure within a network. The data processed in the cloud is subsequently sent back to the user's local computer via the network.

[0062] The invention is explained in more detail below with reference to the accompanying figures and exemplary embodiments. The same components are designated with identical reference numerals in the various figures. The figures are generally not to scale. They show: Fig. Figure 1 shows a conventional CT system with a device according to the invention, Fig. 2 a block diagram of a method according to the invention, Fig. 3 a series of relevant recordings.

[0063] Fig. Figure 1 shows a computed tomography (CT) system 1 with a radiation detector 4 and an X-ray source 5. The X-ray source 5 is configured to expose the radiation detector 4 with X-rays. The CT system 1 shown comprises a gantry 2 with a rotor R. The rotor R includes the X-ray source 5 and the radiation detector 4.

[0064] The rotor R is rotatable about the axis of rotation 8. The patient P is positioned on the patient table L and can be moved along the axis of rotation 8 through the gantry 2. The processing unit 9 is provided for controlling the CT system 1 and / or for generating an image data set based on signals detected by the radiation detector 4.

[0065] Typically, a (raw) X-ray image dataset of the patient P is acquired from a variety of angular directions using the radiation detector 4. Subsequently, a (final) image dataset can be reconstructed from the (raw) X-ray image dataset using a mathematical procedure, for example, including a filtered backprojection or an iterative reconstruction method.

[0066] The processing unit 9 serves here as a control unit 9 for controlling the CT system 1. An input device 10 and an output device 11 are connected to this processing unit 9. The input device 10 and the output device 11 can, for example, enable interaction by a user or the display of a generated image data set B.

[0067] The control unit comprises a device 12 for processing a series of medical X-ray-based images according to the invention. Due to the large number of components, the device 12 is shown above the control unit 9. The device 12 comprises a data interface 13, a selection unit 14, an examination unit 15, a start image unit 16, and a marking unit 17.

[0068] The data interface 13 is used to receive the series of image recordings B and to output the resulting image E.

[0069] Selection unit 14 is used to select a series of temporally successive relevant images B from these image recordings B.

[0070] The investigation unit 15 serves to examine the image elements V of the relevant images B for a maximum image value, whereby when examining an image element V the value of this image element V is considered in all relevant images B at the same image coordinate, and the image in which the value of this image element V is maximal is identified as the maximum image for this image element V.

[0071] The start image unit 16 serves to define the first relevant image in time as the start image S, which has been identified as the maximum image for more than a given number of image elements V.

[0072] The marking unit 17 is used to generate a result image E in which those image elements V whose maximum image is the start image S are displayed in a different color than those image elements V whose maximum image is another relevant image that was taken after the start image S and / or than those image elements V whose image values ​​in all relevant images B lie within a specified range of variation.

[0073] Fig. Figure 2 shows a method according to the invention for processing a series of medical x-ray-based images B.

[0074] In step I, the series of image acquisitions B is provided. In this example, these are current images from the CT system. Fig. 1. From the image recordings B, a series of temporally successive relevant images B is selected, e.g. cross-sectional images of the same plane.

[0075] In step II, the image elements V of the relevant images B are examined for a maximum image value. When examining an image element V, its value is considered at the same image coordinate in all relevant images B, and the image in which the value of this image element V is maximal is identified as the maximum image for that image element V. The grayscale values ​​of the individual voxels are examined, where a maximum brightness corresponds to a maximum value.

[0076] In step III, a starting image S is defined. In this example, it is the first relevant image B in time, but it could also be the first image B that has been identified as the maximum image for more than a predefined number of image elements V.

[0077] In step IV, a result image E is generated in which those image elements V whose maximum value is the start image S are displayed in a different color than those image elements V whose maximum value is another relevant image taken after the start image S, and / or those image elements V whose image values ​​in all relevant images B lie within a predefined range. This result image is then output.

[0078] Fig. Figure 3 shows a series of relevant images B. A series of four images B is shown here, with the recording time running from top to bottom. These are then evaluated by the inventive method, and a result image E (bottom left) is generated. These images B are intended only to illustrate the invention by way of example.

[0079] In this case, the contrast agent had not yet reached its maximum level in the first image (B), and therefore the second image (B) is used as the starting image (S). Here it can be seen that an area in the left hemisphere is not yet permeated by the contrast agent. It is still unclear which part is the penumbra (X) and which is the parenchyma (Y).

[0080] In the second image, B, it can be seen that the contrast agent has disappeared from most of the brain. However, part of the area that was previously dark is now clearly visible, which suggests collateral blood flow (penumbra X).

[0081] In the last image B, the contrast agent has completely disappeared. It is now clear that the dark area in the third image B is parenchyma Y, since no contrast agent ever reached it. In the resulting image, parenchyma Y and penumbra X are marked with different colors.

[0082] Finally, it should be noted once again that the invention described in detail above merely represents exemplary embodiments, which can be modified in various ways by a person skilled in the art without departing from the scope of the invention. Furthermore, the use of the indefinite articles "a" or "an" does not preclude the possibility that the features in question may be present multiple times. Likewise, terms such as "unit" do not preclude the possibility that the components in question consist of several interacting sub-components, which may also be spatially distributed. The term "a number" should be read as "at least one." Regardless of the grammatical gender of a particular term, persons of male, female, or other gender identities are included.

Images