Providing customized recording parameters

Abstract

Computer-implemented procedure for providing a trained function (PROV-TF), comprising: - Acquisition (CAP-TDP) of spatially resolved training dose profiles (TDP) for each real or simulated X-ray fluoroscopy of at least one training subject according to different initial training acquisition parameters (iTAP) using a photon-counting X-ray detector (36), wherein the X-ray detector (36) has several photon-counting detector elements, each of which provides a training dose value depending on a number of detected X-ray photons after an interaction of the X-ray radiation with the at least one training object, the training dose profiles (TDP) are formed by the training dose values ​​of the detector elements of each of the several fluoroscopic examinations, - first classification (CL1) of the training dose profiles (TDP) based on the respective initial training intake parameters (iTAP) to classified training dose profiles (CL1-TDP), - Second classification (CL2) of the classified training dose profiles (CL1-TDP) into compliant and deviating training dose profiles (CL2-TDP) by applying a dose-aware signal quality metric (QM), which assesses the signal quality taking into account a respective total dose, to the classified training dose profiles (CL1-TDP) within each first class, - Providing adapted comparison uptake parameters (aVAP) based on the initial training uptake parameters (iTAP) of the training dose profiles classified as compliant, - Providing adapted training input parameters (aTAP) by applying the trained function (TF) to training input data, where the training input data is based on the initial training input parameters (iTAP) and the training dose profiles (TDP), - Adjusting (ADJ-TF) at least one parameter of the trained function (TF) based on a comparison of the adjusted training input parameters (aTAP) with the adjusted comparison input parameters (aVAP), - Deploying (PROV-TF) the trained function (TF).

Description

[0001] The present invention relates to a computer-implemented method for providing a trained function, a computer-implemented method for providing adapted recording parameters, a provisioning unit, a medical X-ray device, a training unit and a computer program product.

[0002] Computed tomography (CT) systems typically include at least one rotating imaging assembly around the object being examined, comprising an X-ray source and an X-ray detector. This rotating assembly, containing the X-ray source and detector, is part of the CT system's rotating component, which may be guided, for example, within a gantry that forms part of the stationary component. To generate CT image datasets of the object being examined, measurement data detected by the X-ray detector are acquired after X-ray fluoroscopy of the object. A CT image dataset can then be reconstructed from this measurement data, such as a set of cross-sectional images and / or a three-dimensional image volume.

[0003] With CT systems, and X-ray equipment in general, the requirement is that the applied X-ray radiation must always be converted into a diagnostic X-ray image. The image quality of the X-ray images is proportional to the applied X-ray dose. A disadvantage is that unnecessarily high X-ray doses are often applied to an examination in order to achieve the desired image quality for a corresponding diagnosis.

[0004] Dose values, such as a CT dose index (CTDI), from different CT systems, particularly those from different sites, can be compared, and their minimum and maximum X-ray doses can be determined. This allows sites to be informed of excessively high values ​​and optimized, for example, through operator training. Reference sites with low X-ray doses and good image quality can serve as a basis for optimization. However, this approach is often complex and potentially prone to errors.

[0005] Document US 2019 / 0150864A1 discloses a method for controlling an X-ray imaging device. The method comprises, for an image acquisition process of a patient, determining at least one input parameter relating to an attenuation property of the patient and / or a purpose of the image acquisition, determining or adjusting at least one operating parameter of the X-ray detector as a function of the determined input parameter, and performing the image acquisition using the determined or adjusted operating parameter, wherein the at least one operating parameter may be derived at least partially from the at least one input parameter in an optimization procedure.

[0006] Document CN 1 16 725 567 A discloses a method for determining scan parameters for photon-counting computed tomography comprising acquiring feature information of the target object, determining the size of the area of ​​the target object to be scanned based on the feature information, determining a constant factor, wherein this constant factor is determined on the basis of a social group with which the target object is associated, determining an absorbed dose corresponding to the initial scan parameters of the target object, taking into account the size of the area to be scanned and the constant factor, and adjusting the initial scan parameters of the target object based on the absorbed dose.

[0007] Publication US 2015 / 0199478A1 discloses a method comprising receiving a patient record containing one or more reconstructions, one or more preliminary examinations or patient information, and one or more acquisition parameters; calculating one or more patient features based on one or both of the preliminary examinations and the patient information; calculating one or more image features related to the reconstructions; grouping the patient record with one or more other patient records using the patient features; and identifying one or more suitable image acquisition parameters for the patient record using the image features, grouping with other patient records, or a combination thereof.

[0008] The publication DE 10 2021 203 273 A1 discloses a method for providing a classification of an X-ray image into a diagnostically usable X-ray image or a diagnostically unusable X-ray image by applying a trained function, wherein the trained function has been adapted to the X-ray system.

[0009] It is therefore the object of the present invention to enable dose-efficient X-ray fluoroscopy of an object under investigation.

[0010] The problem is solved according to the invention 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.

[0011] The inventive solution to the problem is described below with regard to both methods and devices for providing adapted recording parameters and methods and devices for providing a trained function. Features, advantages, and alternative embodiments of data structures and / or functions in methods and devices for providing adapted recording parameters can be transferred to analogous data structures and / or functions in methods and devices for providing a trained function. Analogous data structures can be characterized, in particular, by the use of the prefix "trainings". Furthermore, the trained functions used in methods and devices for providing adapted recording parameters can be adapted and / or provided, in particular, by methods and devices for providing a trained function.

[0012] The invention relates in a first aspect to a computer-implemented method for providing a trained function according to claim 1. The method and embodiment variants thereof are described in more detail below.

[0013] In a second aspect, the invention relates to a computer-implemented method for providing adapted acquisition parameters using the trained function provided by the aforementioned method. In a first step, spatially resolved dose profiles for each X-ray fluoroscopy of at least one object under investigation are acquired according to initial acquisition parameters using a photon-counting X-ray detector. The X-ray detector has several photon-counting detector elements, each of which provides a dose value as a function of the number of detected X-ray photons after an interaction of the X-ray radiation with the at least one object under investigation. Furthermore, the dose profiles are formed by the dose values ​​of the detector elements of each of the several X-ray fluoroscopy scans.In a further step, the adapted acquisition parameters are provided by applying the trained function to input data. This input data is based on the initial acquisition parameters and dose profiles. Furthermore, the trained function is based on a dose-aware signal quality metric. Finally, the adapted acquisition parameters can be provided as output data from the trained function. The steps of the proposed procedure described above can be partially or fully computerized. Furthermore, the steps of the proposed procedure described above can be executed at least partially, and in particular completely, sequentially or at least partially simultaneously.

[0014] The object of study can be, for example, a human and / or animal patient and / or a study phantom.

[0015] Acquiring spatially resolved dose profiles can involve receiving and / or recording them. Receiving the dose profiles can specifically include recording and / or reading them from a computer-readable data storage device and / or receiving them from a data storage unit, such as a database. Furthermore, the dose profiles can be provided by a delivery unit of a medical X-ray device.

[0016] The initial acquisition parameters can refer to parameters that are predefined before the procedure begins. These initial parameters can be received, for example, from a computer-readable data storage device and / or read from a data storage unit, such as a database. Alternatively or additionally, the initial acquisition parameters can be entered and specified by a medical operator, for example, using an input device. The initial acquisition parameters can include instructions, specifications, commands, and / or operating parameters that instruct a medical X-ray device to perform an X-ray examination of the subject.

[0017] The X-ray unit can comprise an X-ray source and a photon-counting X-ray detector. The X-ray source and the X-ray detector can be arranged opposite each other in a defined configuration, for example, on a C-arm, an O-arm, or a gantry. X-ray fluoroscopy of the object under investigation can involve emitting X-rays from the X-ray source according to the initial exposure parameters to fluoresce the object positioned between the X-ray source and the X-ray detector.

[0018] The X-ray detector comprises several photon-counting detector elements, which can be arranged in rows, columns, and / or grids. Each detector element can be configured to count X-ray photons emitted by the X-ray source during fluoroscopy. For this purpose, each detector element can have a number of pixels, and the X-ray photons arriving at each pixel can be counted. Each detector element can distinguish between different energies of X-ray photons per pixel, for example, four different energies. The differentiation of energies, particularly photon energies, can be achieved, for example, by assigning the energies of the detected X-ray photons to four intervals.For example, four different energy levels can be distinguished per detector element, meaning that the X-ray photons falling into the resulting energy intervals can be counted separately.

[0019] The detector elements can each receive physical signals, such as X-ray photons, with corresponding data describing, for example, the X-ray photons counted per pixel of the detector element and / or the X-ray photons counted for different energy intervals. The data acquired by the detector elements can be transmitted to an electrical circuit or read from the detector elements by the circuit. To further process the data acquired by the detector elements, the electrical circuit can include a storage device with at least one memory block. The data generated by the detector elements can be stored, at least temporarily, in this memory block. The individual detector elements can preferably be implemented as specialized, application-oriented integrated circuits (ASICs).Based on the X-ray photons counted by the detector elements over a specified detection period, a dose value per detector element, in particular a dose value per pixel, can be determined.

[0020] Conventional X-ray detectors, especially integrating ones, operate on the principle of a noisy integration of the detected signals over time. By using a photon-counting X-ray detector, a statistically accurate statement of signal per voxel can be made.

[0021] The dose profiles are generated by the dose values ​​of the detector elements of each of the multiple fluoroscopic examinations. They can also be generated by the dose values ​​of all detector elements of each of the multiple fluoroscopic examinations. The dose profiles can represent the dose values ​​of the multiple fluoroscopic examinations in two dimensions (2D) or three dimensions (3D) with spatial resolution.

[0022] The trained function maps input data to output data. The output data may, in particular, depend on one or more parameters of the trained function. These one or more parameters of the trained function can be determined and / or adjusted through training. Determining and / or adjusting the one or more parameters of the trained function can, in particular, be based on a pair of training input data and associated training output data, especially comparison output data, where the trained function is applied to the training input data to generate training mapping data. Specifically, determining and / or adjusting the parameters can be based on a comparison of the training mapping data and the training output data, especially comparison output data.In general, a trainable function, that is, a function with one or more parameters that have not yet been fitted, is also referred to as a trained function.

[0023] Other terms for trained functions include trained mapping rule, mapping rule with trained parameters, function with trained parameters, and machine learning algorithm. An example of a trained function is an artificial neural network, where the edge weights of the artificial neural network correspond to the parameters of the trained function. The term "neural network" can also be used instead of "neural network." In particular, a trained function can also be a deep artificial neural network. Another example of a trained function is a support vector machine; furthermore, other machine learning algorithms can also be used as trained functions.

[0024] The trained function can be trained, in particular, using backpropagation. First, training mapping data can be determined by applying the trained function to the training input data. Then, any deviation between the training mapping data and the training output data, especially the comparison output data, can be determined by applying an error function to both the training mapping data and the training output data, particularly the comparison output data. Furthermore, at least one parameter, in particular a weight of the trained function, can be iteratively adjusted. This allows the deviation between the training mapping data and the training output data, especially the comparison output data, to be minimized during the training of the trained function.

[0025] Advantageously, the trained function, particularly the neural network, has an input layer and an output layer. The input layer can be configured to receive input data. Furthermore, the output layer can be configured to provide mapping data, particularly the output data. Both the input layer and / or the output layer can each comprise multiple channels, particularly neurons.

[0026] The input data of the trained function is based on the initial acquisition parameters. In particular, the input data of the trained function includes the initial acquisition parameters. Furthermore, the trained function provides the adapted acquisition parameters as output data. The trained function is based on a dose-aware signal quality metric. At least one parameter of the trained function is adapted based on a comparison of adapted training acquisition parameters with adapted comparison acquisition parameters. The trained function is provided by an embodiment of the proposed method for providing a trained function, which is described below.

[0027] The adapted acquisition parameters can have all the features and properties of the initial acquisition parameters. The trained function provides the adapted acquisition parameters as output data. Advantageously, the trained function can provide one adapted acquisition parameter for each of the initial acquisition parameters. The adapted acquisition parameters and the initial acquisition parameters can differ, for example, by one or more parameter values.

[0028] Providing the adjusted acquisition parameters can involve saving them to a computer-readable storage medium, displaying them on a display unit, and / or transmitting them to a delivery unit. The adjusted acquisition parameters can thus enable subsequent optimization of scan protocols with regard to X-ray dose and signal-to-noise ratio (SNR). Alternatively or additionally, the adjusted acquisition parameters can be used to optimize scan protocols during fluoroscopy as "real-time" dose modulation from one rotation to the next.

[0029] In a further advantageous embodiment of the proposed method for providing adapted recording parameters, the input data of the trained function are based on a minimum and / or maximum value and / or mean value of the dose profiles.

[0030] Advantageously, a minimum value, in particular a minimum dose value, and / or a maximum value, in particular a maximum dose value, and / or a mean value, in particular a weighted mean value of the dose values, is determined for each dose profile. The input data of the trained function can be based, in particular exclusively or additionally, on the minimum values ​​and / or maximum values ​​and / or mean values ​​of the dose profiles. Specifically, the input data of the trained function can each include a minimum value and / or maximum value and / or mean value for the dose profiles.

[0031] The minimum, maximum, and / or mean values ​​of the dose profiles can represent a measure of how much X-ray radiation, after passing the object under investigation, reaches the X-ray detector at least, at most, or on average. A certain number of X-ray quanta per voxel may be required to achieve a sufficient signal-to-noise ratio (SNR) for diagnostic image quality. Excess X-ray radiation can represent an unnecessary X-ray dose exposure for the object under investigation. Advantageously, the present embodiment allows for improved adjustment of the acquisition parameters to minimize the X-ray dose while simultaneously maintaining the image quality necessary for diagnostic purposes.

[0032] In a further advantageous embodiment of the proposed method for providing adapted recording parameters, the initial recording parameters include information on an area to be imaged of the at least one object under investigation and / or a recording trajectory and / or a recording sequence and / or an operating parameter for operating the X-ray source and / or the X-ray detector.

[0033] The initial acquisition parameters can advantageously include information, such as spatial coordinates and / or geometric parameters, about an area to be imaged by X-ray fluoroscopy (field-of-view), in particular a 2D or 3D area of ​​at least one object under investigation, for example, a spatial position and / or orientation and / or shape and / or extent of the area to be imaged. Alternatively or additionally, the initial acquisition parameters can include information about an acquisition trajectory, for example, a 2D or 3D path. The acquisition trajectory can specify several positions for the X-ray source and / or the X-ray detector, particularly with respect to the at least one object under investigation, for performing the X-ray fluoroscopy.Alternatively or additionally, the initial acquisition parameters may include an acquisition sequence, in particular a temporal and / or spatial sequence of acquisition positions of the X-ray source and the X-ray detector, and / or a temporal sequence of operating parameters for the operation of the X-ray source and / or the X-ray detector. Alternatively or additionally, the initial acquisition parameters may include an operating parameter, for example, a tube voltage of the X-ray source and / or a detection rate of the X-ray detector, for the operation of the X-ray source and / or the X-ray detector, in particular for the X-ray fluoroscopy of the at least one object under investigation.

[0034] Advantageously, the adapted recording parameters, in particular analogous to the initial recording parameters, can include information on an adapted imaging area of ​​the at least one object under investigation and / or an adapted recording trajectory and / or an adapted recording sequence and / or an adapted operating parameter for the operation of the X-ray source and / or the X-ray detector.

[0035] Advantageously, the adjusted acquisition parameters can be provided for further X-ray fluoroscopy using the X-ray source and the X-ray detector. These adjusted parameters can then enable customized control of the X-ray source and the X-ray detector for X-ray fluoroscopy of at least one or a further examination object.

[0036] In a further advantageous embodiment of the proposed method for providing adapted acquisition parameters, the initial acquisition parameters include at least one initial reconstruction parameter for reconstructing image data from the dose values.

[0037] The initial acquisition parameters advantageously include at least one initial reconstruction parameter. In particular, they can also include several initial reconstruction parameters for reconstructing image data from the dose values.

[0038] The at least one initial reconstruction parameter can advantageously include a specification for reconstructing the image data from the dose values, in particular a 2D or 3D image of the object under investigation from one or more of the fluoroscopic examinations. The reconstruction can, for example, include a backprojection of the image data from the dose values, particularly a filtered one.

[0039] The initial reconstruction parameters can define, for example, an image geometry and / or slice thickness and / or voxel size. For instance, the initial reconstruction parameters can be identified based on an empirical preselection of reconstruction parameters, particularly whether manual or automatic.

[0040] The input data of the trained function can advantageously be based on at least one initial reconstruction parameter, in particular comprising at least one initial reconstruction parameter.

[0041] The proposed embodiment can advantageously allow for adjustment of the acquisition parameters, taking into account the subsequent reconstruction of image data from the X-ray fluoroscopy measurement data. In particular, this can improve low-noise imaging of small voxels at very low dose values, where these voxels have a thin slice thickness and high spatial resolution within the slice plane.

[0042] In a further advantageous embodiment of the proposed method for providing adapted recording parameters, the adapted recording parameters comprise at least one adapted reconstruction parameter.

[0043] Advantageously, the adapted acquisition parameters, particularly analogous to the initial acquisition parameters, can comprise at least one adapted reconstruction parameter, and in particular several adapted reconstruction parameters. The at least one adapted reconstruction parameter can, in particular, have all the features and properties that were described in relation to the at least one initial reconstruction parameter, and vice versa. In particular, the output data of the trained function can include the at least one adapted reconstruction parameter.

[0044] The proposed embodiment can advantageously allow for the adjustment of the acquisition parameters, including at least one reconstruction parameter. This allows for the advantageous adaptation of the reconstruction of the image data from the measurement data acquired using the adapted acquisition parameters.

[0045] The invention relates, in its first aspect, to a computer-implemented method for providing a trained function. In a first step, spatially resolved training dose profiles for each real or simulated X-ray fluoroscopy of at least one training object are acquired using a photon-counting X-ray detector, according to various initial training parameters. The X-ray detector comprises several photon-counting detector elements, each of which provides a training dose value as a function of the number of detected X-ray photons following an interaction of the X-ray radiation with the at least one training object. Furthermore, the training dose profiles are formed by the training dose values ​​of the detector elements for each of the several X-ray fluoroscopy sessions.In a further step, the training dose profiles are initially classified based on their respective initial training intake parameters. In a subsequent step, these classified training dose profiles are further subdivided into compliant and non-compliant profiles by applying a dose-aware signal quality metric, which assesses signal quality considering the total dose, to the classified training dose profiles within each first class. Finally, adjusted reference intake parameters are provided based on the initial training intake parameters of the compliant training dose profiles.

[0046] In a further step, adapted training input parameters are provided by applying the trained function to training input data. This training input data is based on the initial training input parameters and the training dose profiles. In a further step, at least one parameter of the trained function is adjusted based on a comparison of the adapted training input parameters with the adapted reference input parameters. Finally, the trained function is deployed.

[0047] The steps of the proposed procedure described above can be carried out at least partially, and in particular completely, sequentially or at least partially simultaneously.

[0048] The at least one training subject can be, for example, a human and / or animal patient and / or a study phantom.

[0049] The acquisition of spatially resolved training dose profiles can include receiving and / or recording the training dose profiles. Receiving the training dose profiles can specifically include recording and / or reading from a computer-readable data storage device and / or receiving from a data storage unit, such as a database. Furthermore, the training dose profiles can be provided by a provisioning unit of a medical X-ray device. The training dose profiles can, in particular, exhibit all the features and properties of the dose profiles that were described in relation to the procedure for providing adapted exposure parameters, and vice versa.

[0050] The training dose profiles can be generated from the training dose values ​​of detector elements, in particular from the training dose values ​​of all detector elements, of each of the several real fluoroscopic examinations. Alternatively, the training dose values ​​can be determined by simulating several fluoroscopic examinations according to the various initial training parameters. In this case, the training dose profiles can be generated from the training dose values ​​of each of the several simulated fluoroscopic examinations. The training dose profiles can spatially represent the training dose values ​​of the several real or simulated fluoroscopic examinations in 2D or 3D.

[0051] Advantageously, in the first classification, the training dose profiles are classified based on the respective initial training intake parameters. The initial training intake parameters can advantageously possess all the features and properties of the initial intake parameters described in relation to the procedure for providing adapted intake parameters, and vice versa. The initial training intake parameters can be predefined training intake parameters prior to the start of the procedure. These initial training intake parameters can be received, for example, by being acquired and / or read from a computer-readable data storage device and / or received from a data storage unit, such as a database.Alternatively or additionally, the initial training parameters can be entered and specified by a medical operator, for example, using an input unit. These initial training parameters can include instructions, specifications, commands, and / or operating parameters that instruct a medical X-ray machine to perform an X-ray examination of the patient.

[0052] The initial training parameters can specify the type of recording and / or the area of ​​the training subject to be imaged. In the first classification, the training dose profiles can be categorized according to these specifications, based on the initial training parameters. This first classification can be based, in particular, on a match between one or more initial training parameters across multiple training dose profiles. The initial classification can include a preliminary differentiation and / or grouping of the training dose profiles based on their respective initial training parameters, for example, by means of a pairwise comparison of the training dose profiles based on these parameters.

[0053] Furthermore, the training dose profiles classified according to the first classification are classified in a second classification into compliant and non-compliant training dose profiles by applying the dose-aware signal quality metric. The dose-aware signal quality metric can advantageously be designed to evaluate signal quality, for example, signal-to-noise ratio (SNR) and / or contrast-to-noise ratio (CNR), taking into account the X-ray dose applied during fluoroscopy, for example, as a boundary condition. For instance, the dose-aware signal quality metric can assign a quality value to each of the training dose profiles, which evaluates the signal quality of the training dose profile considering the X-ray dose that would be required for its acquisition.

[0054] For the second classification, a benchmark value can be specified or determined. This second classification can be a subclassification of the first. Advantageously, a benchmark value or a global benchmark value can be specified or determined for each of the first classifications. If the benchmark value is specified, it can be defined, for example, based on user input. Alternatively, the benchmark value can be determined based on the quality values ​​of the training dose profiles, for example, as the mean of the quality values. In particular, the benchmark value can be determined as the mean of the quality values ​​from several different operating locations.

[0055] In the second classification of the classified training dose profiles, the classified training dose profiles can be categorized as compliant or non-compliant based on the deviation of their respective quality score from the reference value. For example, those classified training dose profiles whose quality score deviates within a predefined standard deviation of the reference value can be classified as compliant training dose profiles, and the remaining training dose profiles as non-compliant training dose profiles. Thus, the second classification can involve comparing the classified training dose profiles with the global reference value or the respective reference value of the corresponding first class.

[0056] The adjusted reference parameters are provided based on the initial training parameters of the training dose profiles classified as compliant. Adjusted reference parameters can be provided for each of the first classes of the training dose profiles. For example, the adjusted reference parameters can be provided as the initial training parameters of the training dose profile with the highest quality score for each first class. Alternatively, the adjusted reference parameters can be provided as the mean of the initial training parameters of the training dose profiles classified as compliant within each first class.

[0057] Advantageously, the adapted training parameters are provided by applying the trained function to the training input data. The training input data of the trained function is based on the initial training parameters and the training dose profiles; in particular, it can include the initial training parameters and the training dose profiles. The trained function can provide the adapted training parameters as output data.

[0058] By comparing the adjusted training input parameters with the adjusted comparison input parameters, at least one parameter of the trained function is adjusted. The comparison can include determining a deviation between the adjusted training input parameters and the adjusted comparison input parameters. Advantageously, the at least one parameter of the trained function can be adjusted such that the deviation is minimized. Adjusting the at least one parameter of the trained function can include optimizing, in particular minimizing, a cost value of a cost function, wherein the cost function characterizes, and in particular quantifies, the deviation between the adjusted training input parameters and the adjusted comparison input parameters. In particular, adjusting the at least one parameter of the trained function can include a regression of the cost value of the trained function.

[0059] Providing the trained function may include, in particular, storing it on a computer-readable storage medium and / or transferring it to a provisioning unit.

[0060] Advantageously, the proposed method can provide a trained function which, in one embodiment of the method, can be used to provide adapted recording parameters.

[0061] In a further advantageous embodiment of the proposed method for providing a trained function, the training input data of the trained function are based on a minimum and / or maximum value and / or mean value of the training dose profiles.

[0062] Advantageously, a minimum value, in particular a minimum training dose value, and / or a maximum value, in particular a maximum dose value, and / or a mean value, in particular a weighted mean value of the training dose values, are determined for each training dose profile. The training input data of the trained function can be based, in particular exclusively or additionally, on the minimum values ​​and / or maximum values ​​and / or mean values ​​of the training dose profiles. Specifically, the training input data of the trained function can each include a minimum value and / or maximum value and / or mean value for the training dose profiles.

[0063] In a further advantageous embodiment of the proposed method for providing a trained function, the initial training acquisition parameters include information on an area to be imaged of the at least one object under investigation and / or an acquisition trajectory and / or an acquisition sequence and / or an operating parameter for operating the X-ray source and / or the X-ray detector.

[0064] The initial training parameters can advantageously include information, such as spatial coordinates and / or geometric parameters, about an area (FOV) to be imaged by fluoroscopy, in particular a 2D or 3D area, of at least one training object, for example, a spatial position and / or orientation and / or shape and / or extent of the area to be imaged. Alternatively or additionally, the initial training parameters can include information about an image acquisition trajectory, for example, a 2D or 3D path. The image acquisition trajectory can specify several positions for the X-ray source and / or the X-ray detector, in particular with respect to the at least one training object, for performing the fluoroscopy.Alternatively or additionally, the initial training parameters can include an acquisition sequence, in particular a temporal and / or spatial sequence of acquisition positions of the X-ray source and the X-ray detector, and / or a temporal sequence of operating parameters for operating the X-ray source and / or the X-ray detector. Alternatively or additionally, the initial training parameters can include an operating parameter, for example, a tube voltage of the X-ray source and / or a detection rate of the X-ray detector, for operating the X-ray source and / or the X-ray detector, in particular for the X-ray fluoroscopy of the at least one training subject.

[0065] Advantageously, the adapted comparison parameters, in particular analogous to the initial training parameters, can include information on an adapted imaging area of ​​the at least one object under investigation and / or an adapted imaging trajectory and / or an adapted imaging sequence and / or an adapted operating parameter for the operation of the X-ray source and / or the X-ray detector.

[0066] In a further advantageous embodiment of the proposed method for providing a trained function, the initial training acquisition parameters include at least one initial training reconstruction parameter for reconstructing image data from the training dose values.

[0067] The initial training acquisition parameters advantageously include at least one initial training reconstruction parameter. In particular, they can also include several initial training reconstruction parameters for reconstructing image data from the training dose values. The at least one initial training reconstruction parameter can, in particular, have all the features and properties of the at least one initial reconstruction parameter that were described in relation to the method for providing adapted acquisition parameters, and vice versa.

[0068] The at least one training reconstruction parameter can advantageously include a specification for reconstructing the image data from the training dose values, in particular a 2D or 3D image of the at least one training object from one or more of the fluoroscopic examinations. The reconstruction can, for example, include a backprojection of the image data from the training dose values, particularly a filtered one.

[0069] The training input data of the trained function can advantageously be based on at least one initial training reconstruction parameter, in particular comprising at least one initial training reconstruction parameter.

[0070] In a further advantageous embodiment of the proposed method for providing a trained function, the adapted comparison acquisition parameters comprise at least one adapted comparison reconstruction parameter and the adapted training acquisition parameters comprise at least one adapted training reconstruction parameter.

[0071] Advantageously, the adapted comparison parameters, particularly analogous to the initial training parameters, can comprise at least one adapted comparison reconstruction parameter, and in particular several adapted comparison reconstruction parameters. The at least one adapted comparison reconstruction parameter can, in particular, have all the features and properties that were described in relation to the at least one initial training reconstruction parameter, and vice versa.

[0072] In a further advantageous embodiment of the proposed method for providing a trained function, the adjustment of at least one parameter of the trained function is based on federated learning and / or a stochastic gradient method.

[0073] Federal learning can include local training, which is executed separately on different local systems (client systems), for example, in various hospitals. This allows a trained function with adapted parameters to be provided to each local system. Subsequently, the locally trained functions can be combined into a global model. Such training can be performed in several rounds, with the global model being distributed to the local systems in between for further refinement through local training. Advantageously, this prevents the transmission of sensitive data, such as patient data, outside a closed network, particularly the local system, while still allowing the data to be used as training input for the local training of the function.The training dose profiles and the initial training intake parameters can advantageously be recorded and processed without reference to the respective training study object.

[0074] Alternatively or additionally, the adjustment of at least one parameter of the trained function can be based on a stochastic gradient method. In this case, a gradient of the cost function with respect to the input parameters can be determined. Determining the gradient can include calculating partial derivatives of the cost function with respect to the input parameters. Advantageously, the gradient of the cost function with respect to the input parameters can be expressed as a vector. During an iterative adjustment of the at least one parameter of the trained function, at least one input parameter can be adjusted iteratively. The adjustment of the at least one input parameter can advantageously include adding and / or multiplying the previous, especially initial, input parameters and the, especially scaled, gradient of the cost function with respect to the input parameters.Scaling the gradient of the cost function with respect to the input parameters advantageously allows for adjusting the minimization rate of the cost value. Adjusting the input parameters based on the gradient of the cost function with respect to the input parameters advantageously allows for a reduction in the cost value of the cost function.

[0075] In a third aspect, the invention relates to a provisioning unit which is designed to carry out a proposed method for providing adapted recording parameters.

[0076] The provisioning unit can comprise a computing unit, a storage unit, and / or an interface. The provisioning unit can be configured to execute a proposed procedure for providing customized acquisition parameters, in which the interface, computing unit, and / or storage unit are configured to perform the corresponding procedure steps.

[0077] In particular, the interface can be configured to capture dose profiles and provide the adapted uptake parameters. Furthermore, the processing unit and / or the storage unit can be configured to apply the trained function to the input data.

[0078] The advantages of the proposed provisioning unit essentially correspond to the advantages of the proposed method for providing adapted recording parameters. Features, advantages, or alternative embodiments mentioned here can likewise be transferred to the other claimed items and vice versa.

[0079] In a fourth aspect, the invention relates to a medical X-ray device comprising an X-ray source, a photon-counting X-ray detector, and a proposed delivery unit. The X-ray source is configured to emit X-rays. Furthermore, the X-ray detector has several photon-counting detector elements, each configured to provide a dose value depending on the number of detected X-ray photons of incident X-rays.

[0080] Advantageously, the medical X-ray machine can be designed as a computed tomography system (CT system) and / or C-arm X-ray machine and / or O-arm X-ray machine.

[0081] The advantages of the proposed imaging device essentially correspond to the advantages of the proposed method for providing customized acquisition parameters. Features, advantages, or alternative embodiments mentioned here can likewise be transferred to the other claimed subject matter and vice versa.

[0082] In a fifth aspect, the invention relates to a training unit which is configured to carry out a proposed method for providing a trained function.

[0083] The training unit can advantageously comprise a training interface, a training memory unit, and / or a training processing unit. The training unit can be configured to execute a procedure for providing a trained function by configuring the training interface, the training memory unit, and / or the training processing unit to perform the corresponding procedure steps.

[0084] In particular, the training interface can be configured to capture the training dose profiles and to provide the trained function. Furthermore, the training processing unit and / or the training storage unit can be configured to execute the remaining process steps, in particular the first classification, the second classification, the provision of the adapted comparison parameters, the application of the trained function to the training input data, and / or the adaptation of at least one parameter of the trained function.

[0085] The advantages of the proposed training unit essentially correspond to the advantages of the proposed method for providing a trained function. Features, advantages, or alternative embodiments mentioned here can likewise be transferred to the other claimed items and vice versa.

[0086] In a sixth 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 adapted recording parameters when the program sections are executed by the provisioning unit, and / or which can be directly loaded into a training memory of a training unit, with program sections to execute all steps of a proposed method for providing a trained function when the program sections are executed by the training unit.

[0087] The invention may further relate to a computer program or computer-readable storage medium comprising a trained function provided by a proposed method or one of its aspects.

[0088] A largely software-based implementation has the advantage that previously used deployment units and / or training units can be easily retrofitted via a software update to operate in accordance with 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.

[0089] 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. They show: Fig. 1 a schematic representation of a proposed computer-implemented method for providing adapted recording parameters, Fig. 2 a schematic representation of exemplary dose profiles, Fig. 3 a schematic representation of a proposed computer-implemented procedure for providing a trained function, Fig. 4 a schematic representation of a proposed deployment unit, Fig. 5 and Fig. 6 schematic representations of advantageous embodiments of proposed training units, Fig. 7 a schematic representation of an artificial neural network, Fig. 8 a schematic representation of a CT system.

[0090] Fig. Figure 1 shows a schematic representation of an advantageous embodiment of a proposed computer-implemented method for providing adapted acquisition parameters aAP. In a first step, spatially resolved dose profiles DP for each X-ray fluoroscopy of at least one object under investigation are acquired according to initial acquisition parameters iAP using a photon-counting X-ray detector. The X-ray detector has several photon-counting detector elements, each of which provides a dose value as a function of the number of detected X-ray photons after an interaction of the X-ray radiation with the at least one object under investigation. Furthermore, the dose profiles DP are formed by the dose values ​​of the detector elements of each of the several X-ray fluoroscopy sessions.In a further step, the adjusted acquisition parameters aAP are applied to input data by applying a trained function TF, which was provided using the proposed computer-implemented method for providing a trained function TF. The input data is based on the initial acquisition parameters and the dose profiles. Furthermore, the trained function TF is based on a dose-aware signal quality metric. Finally, the adjusted acquisition parameters aAP are provided as output data from the trained function TF.

[0091] The input data for the trained function TF can be based on a minimum, maximum, and / or mean value of the dose profiles DP. For example, one or more spatially resolved mean values ​​of the dose values, particularly mean values ​​of the detector signals in the X-ray detector, can be determined as count values ​​for each fluoroscopy and then used as a reference value. The mean dose values ​​can then form the reference value. The reference value represents a measure of how much X-ray radiation reaches the X-ray detector after passing the object under investigation. A certain number of X-ray quanta per voxel may be required to achieve a sufficient signal-to-noise ratio (SNR) for diagnostic image quality. Excess X-ray radiation can represent an unnecessary dose exposure for the object under investigation.

[0092] Furthermore, the initial acquisition parameters iAP can include information on an area of ​​the at least one object to be imaged and / or an acquisition trajectory and / or an acquisition sequence and / or an operating parameter for the operation of the X-ray source and / or the X-ray detector. In addition, the initial acquisition parameters iAP can include at least one initial reconstruction parameter for reconstructing image data from the dose values. Furthermore, the adapted acquisition parameters aAP can include at least one adapted reconstruction parameter.

[0093] Fig. Figure 2 shows a schematic representation of exemplary training dose profiles (TDP), in particular count-value curves, as they can be acquired in a computer-implemented procedure for providing a trained function (illustrated in Figure 2). Fig. 3) The training dose values ​​n are plotted against the projection angle φ for various measurement situations and / or training objects. In the first classification, the training dose profiles TDP can be classified, for example, according to a region of the at least one training object to be imaged, in particular a scan region. Depending on the scan region, for example, a thorax, skull, and / or extremities, classified training dose profiles CL1-TDP can be generated after attenuation by the training object. Curve a shows an outdoor image, without a training object positioned between the X-ray source and the X-ray detector, with a high X-ray dose. Curve b shows another outdoor image, without a training object positioned between the X-ray source and the X-ray detector, with a lower X-ray dose compared to the image in curve a.Curve c shows an object image where a small training object, for example a head, is positioned between the X-ray source and the X-ray detector, with a medium X-ray dose. Curve d shows another object image where a very large training object, for example a torso, is positioned between the X-ray source and the X-ray detector, with a high X-ray dose.

[0094] Fig. Figure 3 shows a schematic representation of an advantageous embodiment of a proposed computer-implemented method for providing a trained function PROV-TF. In a first step, spatially resolved training dose profiles TDP are acquired for each real or simulated X-ray fluoroscopy of at least one training object according to various initial training acquisition parameters iTAP using a photon-counting X-ray detector CAP-TDP. The classified training dose profiles CL1-TDP can be time-normalized, in particular to represent photons per read. The X-ray detector has several photon-counting detector elements, each of which provides a training dose value depending on the number of detected X-ray photons after an interaction of the X-ray radiation with the at least one training object.Furthermore, the training dose profiles (TDP) are generated by the training dose values ​​of the detector elements of each of the multiple fluoroscopic examinations. In a further step, an initial classification (CL1) of the training dose profiles (TDP) is performed based on the respective initial training exposure parameters (iTAP) to create classified training dose profiles (CL1-TDP). In a further step, a second classification (CL2) of the classified training dose profiles (CL1-TDP) is performed into compliant and non-compliant training dose profiles (CL2-TDP) by applying a dose-aware signal quality metric (QM), which evaluates the signal quality considering the respective total dose, to the classified training dose profiles (CL1-TDP) within each first class. In a further step, adapted comparison exposure parameters (aVAP) are provided based on the initial training exposure parameters (iTAP) of the training dose profiles classified as compliant.Next, adapted training intake parameters aTAP are provided by applying the trained function TF to training input data, where the training input data is based on the initial training intake parameters iTAP and the training dose profiles TDP. Then, at least one parameter of the trained function TF is adjusted ADJ-TF based on a comparison of the adapted training intake parameters aTAP with the adapted comparison intake parameters aVAP. Finally, the trained function TF PROV-TF is provided.

[0095] The training input data for the trained function TF can be based on a minimum and / or maximum value and / or mean value of the training dose profiles TDP. Furthermore, the initial training acquisition parameters iTAP can include information about an area to be imaged of at least one object under investigation and / or an acquisition trajectory and / or an acquisition sequence and / or an operating parameter for the operation of the X-ray source and / or the X-ray detector. In addition, the initial training acquisition parameters iTAP can include at least one initial training reconstruction parameter for reconstructing image data from the training dose values. Furthermore, the adapted comparison acquisition parameters aVAP can include at least one adapted comparison reconstruction parameter, and the adapted training acquisition parameters aTAP can include at least one adapted training reconstruction parameter.Advantageously, the ADJ-TF adjustment of at least one parameter of the trained function TF can be based on federated learning and / or a stochastic gradient method.

[0096] Fig. Figure 4 shows a schematic representation of a proposed provisioning unit PRVS. The provisioning unit PRVS comprises a computing unit CU, a storage unit MU, and / or an interface IF. The provisioning unit PRVS is configured to execute a proposed procedure for providing adapted acquisition parameters by configuring the interface IF, the computing unit CU, and / or the storage unit MU to perform the corresponding procedure steps.

[0097] Fig. Figure 5 shows a schematic representation of an advantageous embodiment of a proposed training unit TRS. The training unit TRS advantageously comprises a training interface TIF, a training storage unit TMU, and / or a training processing unit TCU. The training unit TRS is configured to execute a method for providing a trained function PROV-TF by configuring the training interface TIF, the training storage unit TMU, and / or the training processing unit TCU to perform the corresponding method steps.

[0098] Fig. Figure 6 shows a schematic representation of another advantageous embodiment of a proposed training unit TRS. In this model, the adaptation of at least one parameter of the trained function TF is based on federated learning. The federated learning can include local training, which is performed separately on different local systems 11, 12, 13, 14 to 1n, for example, in different hospitals. This allows a trained function 21, 22, 23, 24 to 2n with adapted parameters to be provided to each local system 11, 12, 13, 14 to 1n. Subsequently, the locally trained functions 21, 22, 23, 24 to 2n can be combined into a global model. Such training can be performed in several rounds, with the global model being distributed to the local systems 11, 12, 13, 14 to 1n in between for further adaptation through local training.Advantageously, this prevents the transmission of sensitive data, such as patient data, outside a closed network, especially the local system, while still allowing the data to be used as training input data for the local training of the trained function 21, 22, 23, 24 to 2n.

[0099] Fig. Figure 7 shows a schematic representation of an artificial neural network 100, as implemented in a procedure according to Fig. 2. The neural network can also be referred to as an artificial neural network, artificial neural network, or neural network.

[0100] Neural network 100 comprises nodes 120, ..., 132 and edges 140, ..., 142, where each edge 140, ..., 142 is a directed connection from a first node 120, ..., 132 to a second node 120, ..., 132. Generally, the first node 120, ..., 129 and the second node 120, ..., 132 are distinct nodes; however, it is also possible for the first node 120, ..., 132 and the second node 120, ..., 132 to be identical. An edge 140, ..., 142 from a first node 120, ..., 132 to a second node 120, ..., 132 can also be referred to as an incoming edge for the second node and an outgoing edge for the first node 120, ..., 132.

[0101] The neural network 100 responds to input values ​​x (1) 1, x (1) 2, x (1) 3 to a multitude of input nodes 120, 121, 122 of input layer 110. The input values ​​x (1) 1, x (1) 2, x (1)3 are applied to generate one or a multitude of outputs x (4) 1, x (4) To create 2. For example, node 120 is connected to node 123 via edge 140. For example, node 121 is connected to node 123 via edge 141.

[0102] In this embodiment, the neural network 100 learns by applying the weighting factors w i,j (weights) of the individual nodes are adjusted based on training data. The weighting factors w i,j These are also referred to as weighting factors below. Possible input values ​​x (1) 1, x (1) 2, x (1) Three of the input nodes 120, 121, 122 can be, for example, the training dose profiles TBD and the initial training intake parameters iTAP.

[0103] Neural network 100 weights the input values ​​of input layer 110 based on the learning process. The output values ​​of output layer 113 of neural network 100 preferentially correspond to the adapted input parameters or the adapted training input parameters. The output can be distributed via a single or a multitude of output nodes x. (4) 1, x (4) 2 in output layer 113.

[0104] In particular, each node 120, ..., 132 of the neural network 100 can be assigned a (real) number as a value.

[0105] where x denotes (n) iThe value of the i-th node 120, ..., 132 of the n-th layer 110, ..., 113. The values ​​of nodes 120, ..., 122 of input layer 110 are equivalent to the input values ​​of neural network 100. The values ​​of nodes 131, 132 of output layer 113 are equivalent to the output values ​​of the neural network. The artificial neural network 100 preferably comprises several hidden layers 111 and 112, which contain a plurality of nodes x. (2) 1, x (2) 2, x (2) 3, x (2) 4, x (2) 5 and x (3) 1, x (3) 2, x (3) The three layers are interconnected. A hidden layer 112 can use output values ​​from another hidden layer 111 as input values. The nodes of hidden layers 111 and 112 perform mathematical operations. For this purpose, each edge 140, ..., 142 can have a weighting factor w. i,j exhibit.

[0106] An output value of a node x (2) 1, x (2) 2, x (2) 3, x (2)4, x (2) 5 corresponds to a non-linear function f of its input values ​​x (1) 1, x (1) 2, x (1) 3 and the weighting factors W i, j After receiving input values ​​x (1) 1, x (1) 2, x (1) 3 leads to a node x (2) 1, x (2) 2, x (2) 3 a summation of one with the weighting factors w i,j weighted multiplication of each input value x (1) 1, x (1) 2, x (1) 3 by, as determined by the following function: xj(n+1)=f(∑ixi(n)⋅wi,j(n)).

[0107] The weighting factor w i,j It can be, in particular, a real number, especially in the interval [-1;1] or [0;1]. The weighting factor wi,j(m,n) denotes the weight of the edge between the i-th node of an m-th layer 110, ..., 113 and a j-th node of the n-th layer 110, ..., 113. The weighting factor wi,j(m,n) is an abbreviation for the weighting factor wi,j(n,n +1).

[0108] In particular, an output value x (3) 1, x (3) 2, x (3) 3 of a node x (2) 1, x (2) 2, x (2) 3, x (2) 4, x (2) 5 is formed as a function f of a node activation, for example a sigmoidal function or a linear ramp function. The output values ​​x (3) 1, x (3) 2, x (3) 3 are transferred to output node(s) 128,..., 130. Another output value is then sent to node x. (4) 1, x (4) 2 is formed as a function f of a node activation. The output values ​​x (4) 1, x (4) 2 are transferred to output node(s) 131 and 132.

[0109] The neural network 100 shown here is a feedforward neural network in which all nodes 111 process the output values ​​of a previous layer as their weighted sum as input values. Naturally, other neural network types can also be used according to the invention, for example, feedback networks in which an input value of a node can simultaneously be its output value.

[0110] The neural network 100 is trained using a supervised learning method to recognize patterns. A known approach is backpropagation, which can be applied to all embodiments of the invention. During training, the neural network 100 is applied to input training data or values ​​and must generate corresponding, previously known output training data or values. Mean square errors (MSE) between calculated and expected output values ​​are iteratively calculated, and individual weighting factors w are determined. i,j The values ​​are adjusted until the deviation between calculated and expected output values ​​is below a predetermined threshold.

[0111] Fig.Figure 8 shows an exemplary embodiment of a proposed medical X-ray device as a CT system 33, comprising an X-ray source 37, a photon-counting X-ray detector 36, and a proposed PRVS delivery unit. The X-ray source 37 and the X-ray detector 36 are arranged opposite each other. The X-ray source 37 is configured to expose the X-ray detector 36 to X-rays along an X-ray incidence direction. The CT system 33 also includes a gantry 33 with a rotor 35. The X-ray source 37 and the X-ray detector 36 are arranged in a defined configuration on the rotor 35, in particular integrated into or attached to the rotor 35. The rotor 35 is rotatably mounted about an axis of rotation 43. The examination object 39 is mounted on a patient table 41 and is movable along the axis of rotation 43 through the gantry 33. The CT system 32 is controlled and computed with cross-sectional images.The PRVS delivery unit can be used to generate volumetric images of the object under investigation 39. An input device 47, for example a keyboard, and an output device 49, for example a screen and / or display, are connected to the PRVS delivery unit, in particular by means of a signal connection. The input device 47 can advantageously be integrated into the output device 49, for example in the case of an input display, in particular a resistive and / or capacitive display.

[0112] The schematic representations contained in the described figures do not depict any scale or size ratios.

[0113] Finally, it should be noted once again that the preceding detailed methods and the illustrated devices 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.

[0114] In the context of this application, the expression "based on" can be understood in particular as meaning "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 trained function (PROV-TF), comprising: - Acquisition (CAP-TDP) of spatially resolved training dose profiles (TDP) for each real or simulated X-ray fluoroscopy of at least one training subject according to different initial training acquisition parameters (iTAP) using a photon-counting X-ray detector (36), wherein the X-ray detector (36) has several photon-counting detector elements, each of which provides a training dose value depending on a number of detected X-ray photons after an interaction of the X-ray radiation with the at least one training object, the training dose profiles (TDP) are formed by the training dose values ​​of the detector elements of each of the several fluoroscopic examinations, - first classification (CL1) of the training dose profiles (TDP) based on the respective initial training intake parameters (iTAP) to classified training dose profiles (CL1-TDP), - Second classification (CL2) of the classified training dose profiles (CL1-TDP) into compliant and deviating training dose profiles (CL2-TDP) by applying a dose-aware signal quality metric (QM), which assesses the signal quality taking into account a respective total dose, to the classified training dose profiles (CL1-TDP) within each first class, - Providing adapted comparison uptake parameters (aVAP) based on the initial training uptake parameters (iTAP) of the training dose profiles classified as compliant, - Providing adapted training input parameters (aTAP) by applying the trained function (TF) to training input data, where the training input data is based on the initial training input parameters (iTAP) and the training dose profiles (TDP), - Adjusting (ADJ-TF) at least one parameter of the trained function (TF) based on a comparison of the adjusted training input parameters (aTAP) with the adjusted comparison input parameters (aVAP), - Deploying (PROV-TF) the trained function (TF). [2] Method according to claim 1, wherein the training input data of the trained function (TF) are based on a minimum and / or maximum value and / or mean value of the training dose profiles (TDP). [3] Method according to one of claims 1 or 2, wherein the initial training acquisition parameters (iTAP) include information on an area to be imaged of the at least one object of investigation (39) and / or an acquisition trajectory and / or an acquisition sequence and / or an operating parameter for operating the X-ray source (37) and / or the X-ray detector (36). [4] Method according to any one of claims 1 to 3, wherein the initial training acquisition parameters (iTAP) comprise at least one initial training reconstruction parameter for reconstructing image data from the training dose values. [5] Method according to claim 4, wherein the adapted comparison acquisition parameters (aVAP) comprise at least one adapted comparison reconstruction parameter and the adapted training acquisition parameters (aTAP) comprise at least one adapted training reconstruction parameter. [6] Method according to any one of claims 1 to 5, wherein the fitting (ADJ-TF) of the at least one parameter of the trained function (TF) is based on federated learning and / or a stochastic gradient method. [7] Computer-implemented method for providing adapted recording parameters (aAP), comprising: - Acquisition (CAP-DP) of spatially resolved dose profiles (DP) for each X-ray fluoroscopy of at least one object (39) according to initial acquisition parameters (iAP) by means of a photon-counting X-ray detector (36), wherein the X-ray detector (36) has several photon-counting detector elements, each of which provides a dose value as a function of a number of detected X-ray photons after an interaction of the X-ray radiation with the at least one object (39), wherein the dose profiles (DP) are formed by the dose values ​​of the detector elements of each of the several X-ray fluoroscopy sessions, - Providing the adapted uptake parameters (aAP) by applying the trained function (TF) provided according to any one of claims 1 to 6 to input data, wherein the input data is based on the initial uptake parameters (iAP) and the dose profiles (DP). [8] Method according to claim 7, wherein the input data of the trained function (TF) are based on a minimum and / or maximum value and / or mean value of the dose profiles (DP). [9] Method according to one of claims 7 or 8, wherein the initial acquisition parameters (iAP) include information on an area to be imaged of the at least one object of investigation (39) and / or an acquisition trajectory and / or an acquisition sequence and / or an operating parameter for operating the X-ray source (37) and / or the X-ray detector (36). [10] Method according to any one of claims 7 to 9, wherein the initial acquisition parameters (iAP) comprise at least one initial reconstruction parameter for reconstructing image data from the dose values. [11] Method according to claim 10, wherein the adapted recording parameters (aAP) comprise at least one adapted reconstruction parameter. [12] Provisioning unit which is configured to carry out a method according to any one of claims 7 to 11. [13] Medical X-ray device comprising an X-ray source (37), a photon-counting X-ray detector (36) and a provisioning unit (PRVS) according to claim 12, wherein the X-ray source (37) is designed to emit X-ray radiation, wherein the X-ray detector (36) has several photon-counting detector elements which are configured to each provide a dose value depending on a number of detected X-ray photons of incident X-ray radiation. [14] Training unit (TRS) which is configured to perform a method according to any one of claims 1 to 6. [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 the method according to any one of claims 7 to 11 when the program sections are executed by the provisioning unit (PRVS) and / or which can be directly loaded into a training memory (TMU) of a training unit (TRS), comprising program sections to execute all steps of the method according to any one of claims 1 to 6 when the program sections are executed by the training unit (TRS).

Images