Providing adapted recording parameters

Abstract

The invention relates to a computer-implemented method for providing adapted recording parameters, comprising: detecting a spatially resolved dose profile for each x-ray fluoroscopy of at least one examination object as a function of initial recording parameters by means of a photon counting x-ray detector, wherein the X-ray detector has a plurality of detector elements for photon counting, which provide a dose value in each case as a function of the number of X-ray photons detected after the X-ray radiation interacts with the at least one examination object, wherein a dose profile is formed by the dose values of the detector elements for each of the plurality of X-ray fluoroscopy; the adapted logged parameters are provided by applying a trained function to input data, where the input data is based on the initial logged parameters and the dose profile, the trained function is based on a dose-known signal quality indicator. The invention also relates to a computer-implemented method, a provision unit, a medical X-ray apparatus, a training unit and a computer program product for providing a trained function.

Description

Technical Field

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

[0002] A computed tomography system (CT system) generally has at least one receiving device that can be rotated around an object to be examined and that has an X-ray source and an X-ray detector. The rotatable part of the CT system therefore includes a recording device with the X-ray source and the X-ray detector, which can be guided, for example, in a gantry that is part of the fixed part of the CT system. In order to be able to determine a CT image data set of the object to be examined, the detected measurement data are detected after an X-ray fluoroscopy of the object to be examined with the aid of at least one X-ray detector. The CT image data set can be reconstructed from the measurement data, for example a set of slice images and / or a three-dimensional image volume.

[0003] In CT systems or generally in X-ray devices, there is a requirement that the applied X-ray radiation must always be converted into a diagnostic X-ray image. The image quality of the X-ray image is proportional to the applied X-ray dose. A disadvantage is that for examinations with the aid of X-ray radiation, an unnecessary amount of X-ray radiation is often applied in order to achieve the image quality desired for the respective diagnosis.

[0004] Dose values, for example CT dose indices (CDTI), of different CT systems, in particular CT systems of different operating sites, can be compared and their minimum and maximum X-ray doses can be determined. In the event of excessive values, the operating site can thus be informed and optimized, for example by training operating personnel. For example, a reference operating site with low X-ray dose and good image quality can be formed as the basis for the optimization. However, this process is usually very complex and can be prone to errors. Summary of the invention

[0005] It is therefore an object of the present invention to enable dose-efficient X-ray fluoroscopy of an examination object.

[0006] This object is achieved by the subject matter according to the invention. Advantageous embodiments with expedient developments are also subject matter of the invention.

[0007] In the following, the solution to this object is described in terms of a method and an apparatus for providing adapted recording parameters and in terms of a method and an apparatus for providing trained functions. Here, features, advantages and alternative embodiments of data structures and / or functions in the method and apparatus for providing adapted recording parameters can be transferred to analog data structures and / or functions in the method and apparatus for providing trained functions. Here, the features of the analog data structures can be characterized in particular by the use of the prefix "training". In addition, the trained functions used in the method and apparatus for providing adapted recording parameters can be adapted and / or provided in particular by the method and apparatus for providing trained functions.

[0008] In a first aspect, the invention relates to a computer-implemented method for providing adapted recording parameters. In a first step, a spatially resolved dose profile of a corresponding X-ray fluoroscopy of at least one object under examination is detected according to initial recording parameters by means of an X-ray detector that performs photon counting. Here, the X-ray detector has a plurality of detector elements that perform photon counting, and the plurality of detector elements respectively provide dose values ​​according to the number of X-ray photons detected after the X-ray radiation interacts with the at least one object under examination. In addition, a dose profile is formed by the dose values ​​of the detector elements of each X-ray fluoroscopy in multiple X-ray fluoroscopy. In a further step, adapted recording parameters are provided by applying a trained function to input data. Here, the input data is based on the initial recording parameters and the dose profile. In addition, the trained function is based on a signal quality indicator of known dose. In addition, adapted recording parameters can be provided as output data of the trained function.

[0009] The above steps of the proposed method may be partially or completely implemented by a computer. In addition, the above steps of the proposed method may be performed at least partially, in particular completely, sequentially or at least partially simultaneously.

[0010] The examination subject may be, for example, a male human and / or animal patient and / or a female human and / or animal patient and / or an examination phantom.

[0011] Detecting the spatially resolved dose profile may include receiving and / or recording the dose profile. Receiving the dose profile may in particular include detecting and / or reading out a computer-readable data memory and / or receiving from a data storage unit (e.g., a database). In addition, the dose profile may be provided by a providing unit of the medical X-ray device.

[0012] The initial recording parameters may represent recording parameters preset before the method is started. Here, the initial recording parameters may be received, in particular detected and / or read from a computer-readable data memory and / or received from a data storage unit (e.g. a database). Alternatively or additionally, the initial recording parameters may be detected and preset by a user input of a medical operator (e.g. by means of an input unit). The initial recording parameters may include instructions, presets, commands and / or operating parameters that cause the medical X-ray device to perform an X-ray fluoroscopy of the examination object.

[0013] The X-ray device may include an X-ray source and an X-ray detector for photon counting. The X-ray source and the X-ray detector may be arranged relative to one another in a defined arrangement, for example on a C-arm, an O-arm or a gantry. The X-ray fluoroscopy of the examination object may include the sending of X-ray radiation by means of the X-ray source according to initial recording parameters for fluoroscopy of the examination object arranged between the X-ray source and the X-ray detector.

[0014] The X-ray detector has a plurality of detector elements for photon counting, which can be arranged in rows, columns and / or grids. The detector elements can be respectively designed to count the X-ray photons sent by the X-ray source during X-ray fluoroscopy. To this end, the detector elements can respectively have a plurality of pixels, wherein the X-ray photons incident on the pixels can be counted respectively. Here, for each pixel, different energies of the X-ray photons, for example four different energies, can be distinguished by each detector element. For example, the energy, in particular the energy of the photons, can be distinguished by associating the energy of the detected X-ray photons with four intervals. For example, four different energy levels can be distinguished for each detector element, which means that the X-ray photons that fall into the energy intervals obtained thereby can be counted separately.

[0015] To this end, the detector elements can respectively receive physical signals, such as X-ray photons, wherein the corresponding data, for example, describe the X-ray photons counted for each pixel of the detector element and / or the X-ray photons counted for different energy intervals. The data recorded by the detector element can be transmitted to the circuit or read out by the detector element from the circuit. In order to further process the data recorded by the detector element, the circuit may include a storage device having at least one storage block. The data generated by the detector element can be stored at least temporarily in the storage block. The individual detector elements can preferably be designed as an application-specific integrated circuit (ASIC). Based on the X-ray photons counted by the detector element during a preset detection time period, the dose value of each detector element, in particular the dose value of each pixel, can be determined.

[0016] Conventional X-ray detectors, especially integrating X-ray detectors, have as their operating principle the integration of the detected signal over time in the presence of noise. By using X-ray detectors that count photons, the statistical statement of the signal per voxel can be expressed significantly more clearly.

[0017] In this case, a dose profile can be formed by the dose values ​​of each detector element of a plurality of X-ray fluoroscopy, in particular the dose values ​​of all detector elements. The dose profile can depict the dose values ​​of a plurality of X-ray fluoroscopy in a two-dimensional (2D) or three-dimensional (3D) spatially resolved manner.

[0018] The trained function maps input data onto output data. In this case, the output data may also be related to one or more parameters of the trained function in particular. One or more parameters of the trained function may be determined and / or adapted by training. The determination and / or adaptation of one or more parameters of the trained function may be based in particular on a pair consisting of training input data and associated training output data, in particular comparative output data, wherein the trained function is applied to the training input data to generate training mapping data. In particular, the determination and / or adaptation may be based on the training mapping data and the training output data, in particular comparative output data. In general, a trainable function, i.e. a function having one or more parameters that have not yet been adapted, is also referred to as a trained function.

[0019] Other terms for trained functions are trained mapping rules, mapping rules with trained parameters, functions with trained parameters, machine learning algorithms. An example of a trained function is an artificial neural network, wherein 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 the term "neural network". In particular, the trained function can also be a deep artificial neural network (English: deep neuronal network deep neural network, deep artificial neural network deep artificial neural network). Another example of a trained function is a "support vector machine", and in particular, other machine learning algorithms can also be used as trained functions.

[0020] In particular, the trained function can be trained by means of back propagation. First, the training mapping data can be determined by applying the trained function to the training input data. Thereafter, the deviation between the training mapping data and the training output data, in particular the comparison output data, can be determined by applying an error function to the training mapping data and the training output data, in particular the comparison output data. Furthermore, at least one parameter, in particular a weight, of the trained function can be adapted iteratively. As a result, the deviation between the training mapping data and the training output data, in particular the comparison output data, can be minimized during the training of the trained function.

[0021] Advantageously, the trained function, in particular the neural network, has an input layer and an output layer. Here, the input layer can be designed to receive input data. In addition, the output layer can be designed to provide mapping data, in particular output data. Here, the input layer and / or the output layer can each include a plurality of channels, in particular neurons.

[0022] The input data of the trained function are based on the initial recording parameters. In particular, the input data of the trained function include the initial recording parameters. In addition, the trained function provides the adapted recording parameters as output data. The trained function is based on a dose-known signal quality indicator. Here, at least one parameter of the trained function can be adapted based on a comparison of the adapted training recording parameters with the adapted comparison recording parameters. In particular, the trained function can be provided by an embodiment of the proposed method for providing a trained function, which method is described below.

[0023] The adapted recording parameters may have all the features and properties of the initial recording parameters. The trained function provides the adapted recording parameters as output data. Advantageously, the trained function may provide adapted recording parameters for each initial recording parameter, respectively. The adapted recording parameters and the initial recording parameters may differ, for example, by one or more parameter values.

[0024] Providing the adapted recording parameters may include storing on a computer-readable storage medium and / or displaying on a display unit and / or transmitting to a providing unit. Thus, the adapted recording parameters may enable a subsequent optimization of the scan log in terms of X-ray dose and signal-to-noise ratio (SNR). Alternatively or additionally, the adapted recording parameters may be used to optimize the scan log during X-ray fluoroscopy as a "real-time" dose modulation from one rotation run to the next.

[0025] In a further advantageous embodiment of the proposed method for providing adapted recording parameters, the input data of the trained function can be based on minimum values ​​and / or maximum values ​​and / or average values ​​of the dose profile.

[0026] Advantageously, a minimum value, in particular a minimum dose value, and / or a maximum value, in particular a maximum dose value, and / or an average value, in particular a weighted average value of the dose values, can be determined for the dose profile, respectively. In this case, the input data of the trained function can be based, in particular, only or additionally, on the minimum value and / or the maximum value and / or the average value of the dose profile. In particular, the input data of the trained function can include, in each case, the minimum value and / or the maximum value and / or the average value of the dose profile.

[0027] The minimum, maximum and / or average value of the dose profile can be a measure of how much X-ray radiation reaches the X-ray detector at least, at most or on average after the object is examined. Each voxel will require a certain amount of X-ray quanta in order to achieve a sufficient SNR for the corresponding diagnostic image quality. X-ray radiation exceeding this value will be an unnecessary X-ray dose burden for the object being examined. Advantageously, this embodiment enables improved adaptation of the recording parameters to minimize the X-ray dose while obtaining the image quality required for the corresponding diagnostic purpose.

[0028] In another advantageous embodiment of the proposed method for providing adapted recording parameters, the initial recording parameters can have information about an area to be mapped of at least one examination object and / or a recording trajectory and / or a recording sequence and / or operating parameters for operating an X-ray source and / or an X-ray detector.

[0029] The initial recording parameters can advantageously include information about a region, in particular a 2D or 3D region, to be mapped by means of an X-ray fluoroscopy (field-of-view) of at least one object under examination, such as spatial coordinates and / or geometrical parameters, such as the spatial position and / or orientation and / or shape and / or extension of the region to be mapped. Alternatively or additionally, the initial recording parameters can include information about a recording trajectory, such as a 2D or 3D path. Here, the recording trajectory can predetermine a plurality of positionings of the X-ray source and / or the X-ray detector, in particular relative to at least one object under examination, in order to perform the X-ray fluoroscopy. Alternatively or additionally, the initial recording parameters can include a recording sequence, in particular a temporal and / or spatial sequence of the recording positionings 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, in particular for the X-ray fluoroscopy of at least one object under examination, the initial recording parameters can include operating parameters for operating the X-ray source and / or the X-ray source, such as the tube voltage of the X-ray source and / or the detection rate of the X-ray detector.

[0030] Advantageously, the adapted recording parameters, in particular similar to the initial recording parameters, can have information about an adapted area to be mapped of at least one examination object and / or an adapted recording trajectory and / or an adapted recording sequence and / or adapted operating parameters for operating the X-ray source and / or the X-ray detector.

[0031] Advantageously, adapted recording parameters can be provided for further X-ray fluoroscopy with the aid of an X-ray source and an X-ray detector. Here, the adapted recording parameters can enable adapted control of the X-ray source and the X-ray detector for X-ray fluoroscopy of at least one examination object or another examination object.

[0032] In a further advantageous embodiment of the proposed method for providing adapted recording parameters, the initial recording parameters can comprise at least one initial reconstruction parameter for reconstructing image data from dose values.

[0033] The initial recording parameters may advantageously comprise at least one initial reconstruction parameter, in particular a plurality of initial reconstruction parameters, to reconstruct the image data from the dose values.

[0034] The at least one initial reconstruction parameter can advantageously include a pre-determination for reconstructing image data from the dose values, in particular a pre-determination for reconstructing a 2D or 3D image of the examination object from one or more X-ray transmissions, respectively. The reconstruction can, for example, include a back-projection, in particular a filtered back-projection, of the image data from the dose values.

[0035] The initial reconstruction parameters may for example define the mapping geometry and / or the slice thickness and / or the voxel size.For example, the initial reconstruction parameters may be identified based on an empirical preselection, in particular manual or automatic, of the reconstruction parameters.

[0036] In this case, the input data of the trained function can advantageously be based on at least one initial reconstruction parameter, in particular include at least one initial reconstruction parameter.

[0037] The proposed embodiment advantageously allows the adaptation of the acquisition parameters while additionally taking into account the subsequent reconstruction of the image data from the measurement data to be acquired from the X-ray fluoroscopy. In particular, this makes it possible to improve the low-noise display of small voxels with a thin slice thickness and a high spatial resolution in the slice plane at very small dose values.

[0038] In a further advantageous embodiment of the proposed method for providing adapted recording parameters, the adapted recording parameters can include at least one adapted reconstruction parameter.

[0039] Advantageously, the adapted recording parameters, in particular similar to the initial recording parameters, may comprise at least one adapted reconstruction parameter, in particular a plurality of adapted reconstruction parameters. The at least one adapted reconstruction parameter may in particular have all features and properties described with respect to the at least one initial reconstruction parameter, and vice versa. In particular, the output data of the trained function may comprise at least one adapted reconstruction parameter.

[0040] The proposed embodiment advantageously allows the adaptation of acquisition parameters including at least one reconstruction parameter. The reconstruction of image data from the measurement data to be acquired with the aid of the adapted acquisition parameters can thereby advantageously be adapted.

[0041] In a second aspect, the invention relates to a computer-implemented method for providing a trained function. In a first step, a spatially resolved training dose profile of each real or simulated X-ray fluoroscopy of at least one training examination object is detected according to different initial training recording parameters by means of an X-ray detector that counts photons. Here, the X-ray detector has a plurality of detector elements that count photons, and the plurality of detector elements provide training dose values ​​according to the number of X-ray photons detected after the X-ray radiation interacts with the at least one training examination object. In addition, the training dose profile is formed by the training dose values ​​of the detector elements of each X-ray fluoroscopy in multiple X-ray fluoroscopy. In another step, the training dose profile is first classified based on the corresponding initial training recording parameters. In another step, the classified training dose profile is second classified into compliant and deviant training dose profiles by applying a signal quality indicator with known dose to the training dose profile within the corresponding first category, and the signal quality indicator evaluates the signal quality under the consideration of the corresponding total dose. In another step, adapted comparison recording parameters are provided based on the initial training recording parameters of the training dose profile classified as compliant. In another step, adapted training recorded parameters are provided by applying the trained function to input data. Here, the input data is based on initial training recorded parameters and a training dose profile. In another step, at least one parameter of the trained function is adapted based on a comparison of the adapted training recorded parameters with an adapted comparison recorded parameter. In another step, a trained function is provided.

[0042] The above-mentioned steps of the above-mentioned method can be carried out at least partially, in particular completely, successively or at least partially simultaneously.

[0043] The at least one training examination subject can be, for example, a human and / or animal male patient and / or a human and / or animal female patient and / or an examination phantom.

[0044] Detecting the spatially resolved training dose profile may include receiving and / or recording the training dose profile. Receiving the training dose profile may in particular include detecting and / or reading out a computer-readable data memory and / or receiving from a data storage unit (e.g., a database). In addition, the training dose profile may be provided by a providing unit of the medical X-ray device. The training dose profile may in particular have all the features and properties of the dose profile described with respect to the method for providing adapted recording parameters, and vice versa.

[0045] The training dose profile can be formed by the training dose values ​​of the detector elements of each X-ray fluoroscopy in multiple real X-ray fluoroscopy, in particular, by the training dose values ​​of all detector elements. Alternatively, the training dose value can be determined by simulating multiple X-ray fluoroscopy according to different initial training record parameters. Here, the training dose profile can be formed by the training dose value of each X-ray fluoroscopy in multiple simulated X-ray fluoroscopy. The training dose profile can map the training dose values ​​of multiple real or simulated X-ray fluoroscopy in 2D or 3D space resolution.

[0046] Advantageously, in a first classification, the training dose profiles can be classified based on corresponding initial training recording parameters. The initial training recording parameters can advantageously have all the features and properties of the initial recording parameters that have been described with respect to the method for providing adapted recording parameters, and vice versa. The initial training recording parameters can represent training recording parameters that are preset before the start of the method. Here, the initial training recording parameters can, for example, be received, in particular detected and / or read out from a computer-readable data memory and / or received from a data storage unit, such as a database. Alternatively or additionally, the initial training recording parameters can be detected and preset by user input of a medical operator, for example by means of an input unit. The initial training recording parameters can include instructions, presets, commands and / or operating parameters that cause the medical X-ray device to perform X-ray fluoroscopy of the examination object.

[0047] Here, the initial training record parameters may preset the record type of the training examination object and / or the area to be mapped. The training dose profiles may be classified in a first classification according to the preset based on the initial training record parameters. Here, the first classification may be based in particular on the matching of a single or multiple initial training record parameters of multiple training dose profiles. Here, the first classification may include a first classification of the first distinction and / or a grouping of the training dose profiles based on the corresponding initial training record parameters, for example by means of pairwise comparison of the training dose profiles based on the corresponding initial training record parameters.

[0048] Furthermore, by applying a dose-known signal quality indicator, training dose profiles classified according to the first classification can be classified in the second classification into compliant training dose profiles and deviated training dose profiles. The dose-known signal quality indicator can advantageously be designed for evaluating the signal quality, such as SNR and / or contrast-to-noise ratio (CNR), taking into account the X-ray dose applied during the X-ray fluoroscopy, for example as boundary conditions. For example, the dose-known signal quality indicator can associate a quality value with each training dose profile, which quality value evaluates the signal quality of the training dose profile taking into account the X-ray dose required for its recording.

[0049] A comparison value may be preset or determined for the second category. Here, the second category may be a subcategory of the first category. Advantageously, a comparison value or a global comparison value may be preset or determined for each first category respectively. In the case of preset comparison values, the comparison value may be defined, for example, according to a user input. Alternatively, the comparison value may be determined according to the quality values ​​of the training dose profiles, for example as an average value of the quality values. In particular, the comparison value may be determined as an average value of the quality values ​​of a plurality of different operating locations.

[0050] In the second classification of the classified training dose profiles, the classified training dose profiles can be classified into compliant training dose profiles and deviant training dose profiles according to the deviation of the quality value of the response of the classified training dose profile from the comparison value. For example, the classified training dose profiles whose quality values ​​have a deviation within a preset standard deviation relative to the comparison value can be classified as compliant training dose profiles, and the remaining training dose profiles are classified as deviant training dose profiles. Therefore, the second classification can include comparing the classified training dose profiles with the global comparison value or the corresponding comparison value of the corresponding first category.

[0051] Adapted comparative recording parameters may be provided based on initial training recording parameters of training dose profiles classified as compliant. Here, adapted comparative recording parameters may be provided for each training dose profile in the first category, respectively. The adapted comparative recording parameters may be provided, for example, as initial training recording parameters of training dose profiles having, for example, the highest quality values ​​of the respective first categories. Alternatively, the adapted comparative recording parameters may be provided as an average value of initial training recording parameters of training dose profiles classified as compliant within the first category.

[0052] Advantageously, adapted training record parameters can be provided by applying a trained function to input data. Here, the input data of the trained function can be based on, in particular include, initial training record parameters and a training dose profile. The trained function can provide adapted training record parameters as output data.

[0053] By comparing the adapted training recording parameters with the adapted comparison recording parameters, at least one parameter of the trained function can be adapted. The comparison can include determining a deviation between the adapted training recording parameters and the adapted comparison recording parameters. Here, at least one parameter of the trained function can advantageously be adapted in such a way that the deviation is minimized. Adapting 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 (in particular quantifies) the deviation between the adapted training recording parameters and the adapted comparison recording parameters. In particular, adapting at least one parameter of the trained function can include a regression of the cost value of the trained function.

[0054] Providing the trained function may in particular include storing on a computer-readable storage medium and / or transmitting to a providing unit.

[0055] Advantageously, the proposed method can provide a trained function which can be used in one embodiment of the method to provide adapted recording parameters.

[0056] In a further advantageous embodiment of the proposed method for providing a trained function, input data of the trained function can be based on minimum values ​​and / or maximum values ​​and / or average values ​​of a training dose profile.

[0057] Advantageously, a minimum value, in particular a minimum training dose value, and / or a maximum value, in particular a maximum dose value, and / or an average value, in particular, for example a weighted average value, of the training dose values ​​of the training dose profile can be determined, respectively. Here, the input data of the trained function can be based, in particular, only or additionally, on the minimum value and / or the maximum value and / or the average value of the training dose profile. In particular, the input data of the trained function can include, in each case, the minimum value and / or the maximum value and / or the average value of the training dose profile.

[0058] In another advantageous embodiment of the proposed method for providing a trained function, the initial training recording parameters can have information about an area to be mapped and / or a recording trajectory and / or a recording sequence and / or operating parameters for operating an X-ray source and / or an X-ray detector of at least one examination object.

[0059] The initial training recording parameters can advantageously include information about a region (FOV), in particular a 2D or 3D region, to be mapped by means of X-ray fluoroscopy, of at least one training examination object, such as spatial coordinates and / or geometrical parameters, such as the spatial position and / or orientation and / or shape and / or extension of the region to be mapped. Alternatively or additionally, the initial training recording parameters can include information about a recording trajectory, such as a 2D or 3D path. Here, the recording trajectory can predetermine a plurality of positionings of the X-ray source and / or the X-ray detector, in particular relative to at least one training examination object, in order to perform X-ray fluoroscopy. Alternatively or additionally, the initial training recording parameters can include a recording sequence, in particular a temporal and / or spatial sequence of recording positionings 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, in particular for X-ray fluoroscopy of at least one training examination object, the initial training recording parameters can include operating parameters for operating the X-ray source and / or the X-ray source, such as the tube voltage of the X-ray source and / or the detection rate of the X-ray detector.

[0060] Advantageously, the adapted comparison recording parameters, in particular similar to the initial recording parameters, can have information about an adapted area to be mapped of at least one examination object and / or an adapted recording trajectory and / or an adapted recording sequence and / or adapted operating parameters for operating the X-ray source and / or the X-ray detector.

[0061] In a further advantageous embodiment of the proposed method for providing a trained function, the initial training recording parameters can include at least one initial training reconstruction parameter for reconstructing image data from training dose values.

[0062] The initial training recording parameters may advantageously comprise at least one initial training reconstruction parameter, in particular a plurality of initial training reconstruction parameters, to reconstruct image data from the training dose values. The at least one initial training reconstruction parameter may in particular have all features and properties of the at least one initial reconstruction parameter already described with respect to the method for providing adapted recording parameters, and vice versa.

[0063] The at least one training reconstruction parameter can advantageously include a pre-determination for reconstructing image data from the training dose values, in particular a pre-determination for reconstructing a 2D or 3D image of the training examination object from one or more X-ray transmissions. The reconstruction can, for example, include a back-projection, in particular a filtered back-projection, of the image data from the training dose values.

[0064] In this case, the input data of the trained function can advantageously be based on at least one initial trained reconstruction parameter, in particular include at least one initial trained reconstruction parameter.

[0065] In a further advantageous embodiment of the proposed method for providing a trained function, the adapted comparison recording parameters may include at least one adapted comparison reconstruction parameter and the adapted training recording parameters may include at least one adapted training reconstruction parameter.

[0066] Advantageously, the adapted comparison recording parameters, in particular similar to the initial training recording parameters, may include at least one adapted comparison reconstruction parameter, in particular a plurality of adapted comparison reconstruction parameters. The at least one adapted comparison reconstruction parameter may in particular have all features and properties described with respect to the at least one initial training reconstruction parameter, and vice versa.

[0067] In a further advantageous embodiment of the proposed method for providing a trained function, the adaptation of at least one parameter of the trained function is based on federated learning and / or a stochastic gradient method.

[0068] Federated learning can include local training, which is performed separately on different local systems (English: client systems), for example, in different hospitals. In this way, a trained function with adapted parameters can be provided to each local system respectively. The locally trained function can then be merged into the global model. This training can be carried out in several rounds, during which the global model can be distributed to the local systems for further adaptation through local training. Advantageously, this can prevent the transmission of sensitive data (such as patient data) outside a closed network, in particular a local system, although the data can still be used as input data for local training of the trained function. The training dose profile and the initial training record parameters can be advantageously detected and processed without reference to the corresponding training examination object.

[0069] Alternatively or additionally, the adaptation of at least one parameter of the trained function can be based on a stochastic gradient method. In this case, the gradient of the cost function relative to the recording parameters can be determined. Determining the gradient can include determining the partial derivative of the cost function relative to the recording parameters. In this case, the gradient of the cost function relative to the recording parameters can advantageously be designed as a vector. During the iterative adaptation of at least one parameter of the trained function, the at least one recording parameter can be iteratively adapted. Adapting at least one recording parameter can advantageously include adding and / or multiplying a previous, in particular initial, recording parameter and a gradient of the cost function relative to the recording parameters, in particular a scaled gradient. By scaling the gradient of the cost function relative to the recording parameters, an adaptation of the minimization speed of minimizing the cost value can advantageously be achieved. By adapting the recording parameters based on the gradient of the cost function relative to the recording parameters, a reduction in the cost value of the cost function can advantageously be achieved.

[0070] In a third aspect, the invention relates to a providing unit designed for performing the proposed method for providing adapted recording parameters.

[0071] In this case, the providing unit may include a computing unit, a storage unit and / or an interface. The providing unit may be designed to carry out the proposed method for providing adapted recording parameters, wherein the interface, the computing unit and / or the storage unit are designed to carry out corresponding method steps.

[0072] In particular, the interface may be designed for detecting the dose profile and providing adapted recording parameters. Furthermore, the calculation unit and / or the storage unit may be designed for applying the trained function to the input data.

[0073] The advantages of the proposed provision unit correspond substantially to the advantages of the proposed method for providing adapted recording parameters. Features, advantages or alternative embodiments mentioned here can also be transferred to other claimed subjects and vice versa.

[0074] In a fourth aspect, the invention relates to a medical X-ray device, comprising an X-ray source, an X-ray detector for photon counting and the proposed provision unit. Here, the X-ray source is designed to emit X-ray radiation. In addition, the X-ray detector has a plurality of detector elements for photon counting, which are designed to provide dose values ​​in each case as a function of the number of detected X-ray photons of the incident X-ray radiation.

[0075] Advantageously, the medical X-ray device can be designed as a computed tomography system (CT system) and / or as a C-arm X-ray device and / or as an O-arm X-ray device.

[0076] The advantages of the proposed imaging device correspond substantially to the advantages of the proposed method for providing adapted recording parameters. Features, advantages or alternative embodiments mentioned here can also be transferred to other claimed subjects and vice versa.

[0077] In a fifth aspect, the invention relates to a training unit designed for performing the proposed method for providing a trained function.

[0078] The training unit may advantageously include a training interface, a training storage unit and / or a training computing unit. The training unit may be designed to perform the method for providing a trained function in the following manner: the training interface, the training storage unit and / or the training computing unit are designed to perform corresponding method steps.

[0079] In particular, the training interface can be designed for detecting a training dose profile and providing a trained function. Furthermore, the training calculation unit and / or the training storage unit can be designed for performing the remaining method steps, in particular the first classification, the second classification, providing adapted comparison record parameters, applying the trained function to the input data and / or adapting at least one parameter of the trained function.

[0080] The advantages of the proposed training unit correspond substantially to the advantages of the proposed method for providing a trained function. Features, advantages or alternative embodiments mentioned here can also be transferred to other claimed subjects and vice versa.

[0081] In a sixth aspect, the present invention relates to a computer program product having a computer program, which can be directly loaded into a memory of a providing unit, the computer program having a program segment so that when the program segment is executed by the providing unit, all steps of the proposed method for providing a trained function are performed, and / or the computer program can be directly loaded into a training memory of a training unit, the computer program having a program segment so that when the program segment is executed by the training unit, all steps of the proposed method for providing a trained function are performed.

[0082] The invention may also relate to a computer program or a computer-readable storage medium comprising a trained function provided by the proposed method or one of its aspects.

[0083] An extensive software-based implementation has the advantage that previously used provision units and / or training units can also be easily adapted by software updates to operate in the manner according to the invention. In addition to the computer program, such a computer program product may 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. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Embodiments of the present invention are shown in the drawings and described in more detail below. In different figures, the same reference numerals are used for the same features. Among them:

[0085] Figure 1 Schematic diagram showing the proposed computer-implemented method for providing adapted recording parameters,

[0086] Figure 2 A schematic diagram showing an exemplary dose profile,

[0087] Figure 3 A schematic diagram showing the proposed computer-implemented method for providing a trained function,

[0088] Figure 4 A schematic diagram showing the proposed providing unit,

[0089] Figure 5 and Figure 6 A schematic diagram showing an advantageous embodiment of the proposed training unit,

[0090] Figure 7 A schematic diagram showing an artificial neural network,

[0091] Figure 8 A schematic diagram of a CT system is shown. DETAILED DESCRIPTION

[0092] Figure 1 A schematic diagram of an advantageous embodiment of the proposed computer-implemented method for providing adapted recording parameters aAP is shown. In a first step, a spatially resolved dose profile DP of a corresponding X-ray fluoroscopy of at least one object under examination can be detected according to initial recording parameters iAP by means of an X-ray detector that counts photons. Here, the X-ray detector can have a plurality of detector elements that count photons, which respectively provide dose values ​​according to the number of X-ray photons detected after the X-ray radiation interacts with at least one object under examination. In addition, a dose profile DP can be formed by the dose values ​​of the detector elements of each X-ray fluoroscopy in multiple X-ray fluoroscopy. In another step, an adapted recording parameter aAP is obtained by applying a trained function TF to input data. Here, the input data can be based on the initial recording parameters and the dose profile. In addition, the trained function TF can be based on a signal quality indicator with known dose. In addition, the adapted recording parameter aAP can be provided as output data of the trained function TF.

[0093] The input data of the trained function TF can be based on the minimum and / or maximum and / or average value of the dose profile DP, respectively. For example, one or more spatially resolved average values ​​of the dose values, in particular the average value of the detector signal in the X-ray detector, can be calculated as count values ​​and then used as reference values. Here, an average value of the dose values ​​can be formed. The reference value is a measure of how much X-ray radiation still reaches the X-ray detector after the object is examined. A certain amount of X-ray quanta will be required for each voxel in order to achieve a sufficient SNR for the corresponding diagnostic image quality. X-ray radiation exceeding this value will be an unnecessary X-ray dose for the object being examined.

[0094] Furthermore, the initial recording parameters iAP may have information about at least one area to be mapped of the examination object and / or a recording trajectory and / or a recording sequence and / or operating parameters for operating an X-ray source and / or an X-ray detector. Furthermore, the initial recording parameters iAP may include at least one initial reconstruction parameter for reconstructing image data from dose values. Furthermore, the adapted recording parameters aAP may include at least one adapted reconstruction parameter.

[0095] Figure 2 Schematic diagram showing exemplary training dose profiles TDP, in particular count value curves (English: count-value curves), as can be used in a computer-implemented method for providing a trained function (eg Figure 3 Here, for different measurement situations and / or training examination objects, the projection angle Plot the training dose value n. In the first classification, the training dose profile TDP can be classified, for example, according to the area to be mapped of at least one training examination object, in particular the scanning area. According to the scanning area, such as the chest, the skull and / or the limbs, the classified training dose profile CL1-TDP can be generated according to the attenuation of the training examination object. Here, curve a shows an empty record with a higher X-ray dose, in which no training examination object is arranged between the X-ray source and the X-ray detector. Curve b shows another empty record with a lower X-ray dose than the record of curve a, in which no training examination object is arranged between the X-ray source and the X-ray detector. Curve c shows an object record of medium X-ray dosing, in which a smaller training examination object (such as a head) is arranged between the X-ray source and the X-ray detector. Curve d shows another object record of high X-ray dose, in which a very large training examination object, such as a torso, is arranged between the X-ray source and the X-ray detector.

[0096] Figure 3A schematic diagram of an advantageous embodiment of the proposed computer-implemented method for providing a trained function PROV-TF is shown. In a first step, a spatially resolved training dose profile of each real or simulated X-ray fluoroscopy of at least one training examination object can be detected according to different initial training recording parameters iTAP by means of an X-ray detector CAP-TDP that counts photons. The classified training dose profile CL1-TDP can be time-normalized, in particular the photons of each reading process (English Reading). Here, the X-ray detector can have a plurality of detector elements that count photons, and these detector elements respectively provide training dose values ​​according to the number of X-ray photons detected after the X-ray radiation interacts with at least one training examination object. In addition, the training dose profile TDP can be formed by the training dose values ​​of the detector elements of each X-ray fluoroscopy in multiple X-ray fluoroscopy. In another step, the training dose profile TDP can be first classified CL1 based on the corresponding initial training recording parameter iTAP. In another step, the classified training dose profile CL1-TDP can be second classified CL2 into compliant and deviant training dose profiles CL2-TDP by applying a signal quality indicator QM with known dose to the training dose profile TDP within the corresponding first category, and the signal quality indicator evaluates the signal quality while taking into account the corresponding total dose. In another step, an adapted comparison recording parameter aVAP can be provided based on the initial training recording parameter iTAP of the training dose profile classified as compliant. Thus, an adapted training recording parameter aTAP can be provided by applying the trained function TF to input data, wherein the input data is based on the initial training recording parameter iTAP and the training dose profile TDP. Thus, at least one parameter of the trained function TF of ADJ-TF can be adapted based on a comparison of the adapted training recording parameter aTAP with the adapted comparison recording parameter aVAP. In addition, the trained function TF of PROV-TF can be provided.

[0097] The input data of the trained function TF may be based on the minimum and / or maximum and / or average value of the training dose profile TDP. Furthermore, the initial training recording parameter iTAP may have information about an area to be mapped and / or a recording trajectory and / or a recording sequence and / or operating parameters for operating an X-ray source and / or an X-ray detector of at least one inspection object. In addition, the initial training recording parameter iTAP may include at least one initial training reconstruction parameter for reconstructing image data from training dose values. Furthermore, the adapted comparison recording parameter aVAP may include at least one adapted comparison reconstruction parameter, and the adapted training recording parameter aTAP may include at least one adapted training reconstruction parameter. Advantageously, the adaptation ADJ-TF of at least one parameter of the trained function TF may be based on federated learning and / or a stochastic gradient method.

[0098] Figure 4 A schematic diagram of the proposed provision unit PRVS is shown. The provision unit PRVS may comprise a computing unit CU, a storage unit MU and / or an interface IF. The provision unit PRVS may be designed to carry out the proposed method for providing adapted recording parameters, wherein the interface IF, the computing unit CU and / or the storage unit MU are designed to carry out the corresponding method steps.

[0099] Figure 5 A schematic diagram of an advantageous embodiment of the proposed training unit TRS is shown. The training unit TRS can advantageously comprise a training interface TIF, a training storage unit TMU and / or a training calculation unit TCU. The training unit TRS can be designed to carry out a method for providing a trained function PROV-TF, wherein the training interface TIF, the training storage unit TMU and / or the training calculation unit TCU are designed to carry out corresponding method steps.

[0100] Figure 6A schematic diagram of another advantageous embodiment of the proposed training unit TRS is shown. Here, the adaptation ADJ-TF of at least one parameter of the trained function TF can be based on federated learning. Federated learning can include local training that is performed separately on different local systems 11, 12, 13, 14 to 1n (for example in different hospitals). As a result, a trained function 21, 22, 23, 24 to 2n with adapted parameters can be provided to each local system 11, 12, 13, 14 to 1n respectively. The locally trained functions 21, 22, 23, 24 to 2n can then be merged into the global model. This training can be carried out in several rounds, during which the global model can be distributed to the local systems 11, 12, 13, 14 to 1n for further adaptation by local training. Advantageously, this prevents sensitive data, such as patient data, from being transmitted outside a closed network, in particular a local system, although the data can still be used as input data for local training of the trained functions 21 , 22 , 23 , 24 to 2n.

[0101] Figure 7 It can be shown that according to Figure 2 A schematic diagram of an artificial neural network 100 used in the method of FIG. A neural network may also be referred to as an artificial neural network, an artificial neural network, or a neural network.

[0102] The neural network 100 includes nodes 120, ..., 132 and edges 140, ..., 142, wherein each edge 140, ..., 142 is a directional connection from a first node 120, ..., 132 to a second node 120, ..., 132. Generally speaking, the first node 120, ..., 129 and the second node 120, ..., 132 are different nodes, and the first node 120, ..., 132 and the second node 120, ..., 132 may also be the same. The edge 140, ..., 142 from the first node 120, ..., 132 to the second node 120, ..., 132 may also be referred to as an input edge of the second node and an output edge of the first node 120, ..., 132.

[0103] The neural network 100 responds to the input values ​​x of the plurality of input nodes 120, 121, 122 of the input layer 110. (1) 1. x (1) 2. x (1) 3. Input value x (1) 1.x (1) 2. x (1) 3 is used to generate one or more outputs x (4) 1. x (4) 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.

[0104] In this embodiment, the neural network 100 adapts the weighting factor w of each node based on the training data. i，j (weight) to learn. Weighting factor w i，j In the following, it is also referred to as weight factor. Possible input values ​​x of input nodes 120, 121, 122 (1) 1. x (1) 2. x (1) 3 may be, for example, a training dose profile TBD and initial training recording parameters iTAP.

[0105] The neural network 100 weights the input values ​​of the input layer 110 based on a learning process. The output value of the output layer 113 of the neural network 100 preferably corresponds to the adapted recording parameters or the adapted training recording parameters. The output can be transmitted via a single or multiple output nodes x in the output layer 113. (4) 1. x (4) 2 proceed.

[0106] In particular, a (real) number may be associated as a value to each node 120, ..., 132 of the neural network 100. In this case, x (n) i represents the value of the i-th node 120, ..., 132 of the n-th layer 110, ..., 113. The values ​​of the nodes 120, ..., 122 of the input layer 110 are equivalent to the input values ​​of the neural network 100. The values ​​of the nodes 131, 132 of the output layer 113 are equivalent to the output values ​​of the neural network. The artificial neural network 100 preferably includes a plurality of hidden layers 111 and 112, wherein the hidden layers include 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) 3. Here, the hidden layer 112 may use the output value of another hidden layer 111 as an input value. The nodes of the hidden layers 111 and 112 perform mathematical operations. To this end, each edge 140, ..., 142 may have a weight factor w i，j .

[0107] Node x (2) 1. x (2) 2. x (2) 3. x (2) 4.x (2) The output value of 5 corresponds to its input value x (1) 1. x (1) 2. x (1) 3 and weight factor wi，j The nonlinear function f. When obtaining the input value x (1) 1. x (1) 2. x (1) 3, node x (2) 1. x (2) 2. x (2) 3 Execute for each input value x (1) 1. x (1) 2. x (1) 3 with the help of weight factor w i，j The sum of weighted multiplications is determined by the following function:

[0108]

[0109] Weight factor w i，j In particular, it may be a real number, in particular in the interval [-1; 1] or [0; 1]. represents the weight of the edge between the i-th node of the m-th layer 110, ..., 113 and the j-th node of the n-th layer 110, ..., 113. Weight factor is the weight factor Abbreviation of .

[0110] In particular, node x (2) 1. x (2) 2. x (2) 3. x (2) 4.x (2) The output value x of 5 (3) 1. x (3) 2. x (3) 3 is formed as a function f of the node activation, such as a sigmoid function or a linear ramp function. The output value x (3) 1. x (3) 2. x (3) 3 is transmitted to one or more output nodes 128, ..., 130. Again, node x (4) 1. x (4) The output value of 2 is formed as a function f of the node activation. The output value x (4) 1. x (4) 2 is transmitted to the output nodes 131, 132.

[0111] The neural network 100 shown here is a feedforward neural network, in which all nodes 111 process the output value of the previous layer in the form of its weighted sum as input value. Of course, according to the present invention, other types of neural networks can also be used, such as feedback networks, in which the input value of a node can also be its output value.

[0112] The neural network 100 is trained to recognize patterns by means of supervised learning methods. A known process is back propagation, which can be used for all embodiments of the present 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. The mean square error (MSE) between the calculated output values ​​and the expected output values ​​is calculated iteratively, and the individual weight factors w are adapted. i，j , until the deviation between the calculated output value and the expected output value is below a predetermined threshold.

[0113] Figure 8 An exemplary embodiment of the proposed medical X-ray device is shown as a CT system 33, which includes an X-ray source 37, an X-ray detector 36 for photon counting and a proposed provision unit PRVS. Here, the X-ray source 37 and the X-ray detector 36 can be arranged opposite to each other. The X-ray source 37 can be designed to irradiate the X-ray detector 36 with X-ray radiation in the direction of X-ray incidence. The CT system 33 can also include a gantry 33 with a rotor 35. The X-ray source 37 and the X-ray detector 36 can be arranged on the rotor 35 in a defined arrangement, in particular integrated into the rotor 35 or fixed on the rotor 35. The rotor 35 can be rotatably supported about an axis of rotation 43. The examination object 39 can be supported on a patient bed 41 and can be moved through the gantry 33 along the axis of rotation 43. The provision unit PRVS can be used to control the CT system 32 and calculate slice images or volume images of the examination object 39. An input device 47 (for example a keyboard) and an output device 49 (for example a screen and / or a display) can be connected to the provision unit PRVS, in particular coupled in terms of signals. The input device 47 can advantageously be integrated into the output device 49 , for example in the case of, in particular, resistive and / or capacitive input displays.

[0114] The schematic views contained in the described figures do not represent any scale or size relationships.

[0115] Finally, it should be pointed out again that the method described in detail above and the device shown are only embodiments, which can be modified in various ways by a person skilled in the art without departing from the scope of the invention. In addition, the use of the indefinite article "a" or "an" does not exclude that the features involved can exist multiple times. Similarly, the terms "unit" and "element" do not exclude that the components involved are composed of multiple interacting subcomponents, which can be spatially distributed if necessary.

[0116] In the present application, the expression “based on” can be understood in particular as the expression “by using.” In particular, the expression that a first feature is generated (alternatively: determined, determined, etc.) based on a second feature does not exclude that the first feature is generated (alternatively: determined, determined, etc.) based on a third feature.

Claims

1. A computer-implemented method for providing adapted recording parameters (aAP), comprising: - detecting (CAP-DP) a spatially resolved dose profile (DP) for each X-ray fluoroscopy of at least one examination subject (39) according to initial recording parameters (iAP) by means of a photon counting X-ray detector (36), The X-ray detector (36) has a plurality of detector elements for photon counting, each of which provides a dose value as a function of the number of X-ray photons detected after the X-ray radiation has interacted with at least one examination object (39). wherein the dose profile (DP) is formed by the dose values ​​of the detector elements in each X-ray fluoroscopy of a plurality of X-ray fluoroscopy, - providing said adapted recording parameters (aAP) by applying a trained function (TF) to input data, wherein the input data is based on the initial recorded parameters (iAP) and the dose profile (DP), The trained function (TF) is based on a signal quality metric (QM) with known dose.

2. The method according to claim 1, Wherein the input data of the trained function (TF) is based on a minimum value and / or a maximum value and / or an average value of the dose profile (DP).

3. The method according to any one of claims 1 or 2, The initial recording parameters (iAP) contain information about the area to be mapped of the at least one examination object (39) and / or the recording trajectory and / or the recording sequence and / or operating parameters for operating the X-ray source (37) and / or the X-ray detector (36).

4. The method according to any one of the preceding claims, Wherein said initial recording parameters (iAP) comprise at least one initial reconstruction parameter for reconstructing image data from said dose values.

5. The method according to claim 4, Wherein the adapted recording parameters (aAP) include at least one adapted reconstruction parameter.

6. A computer-implemented method for providing a trained function (PROV-TF), comprising: - detecting (CAP-TDP) a spatially resolved training dose profile (TDP) for each real or simulated X-ray fluoroscopy of at least one training examination subject according to different initial training recording parameters (iTAP) by means of a photon counting X-ray detector (36), The X-ray detector (36) has a plurality of detector elements for performing photon counting, wherein the plurality of detector elements respectively provide training dose values ​​according to the number of X-ray photons detected after the X-ray radiation interacts with at least one training examination object, wherein the training dose profile (TDP) is formed by the training dose values ​​of the detector element in each X-ray fluoroscopy of a plurality of X-ray fluoroscopy, - performing a first classification (CL1) of said training dose profile (TDP) based on corresponding said initial training recorded parameters (iTAP), - a second classification (CL2) of the classified training dose profiles (CL1-TDP) into compliant and deviant training dose profiles (CL2-TDP) by applying a dose-known signal quality indicator (QM) to the training dose profiles (TDP) within the respective first category, said signal quality indicator assessing the signal quality taking into account the respective total dose, - providing adapted comparative recording parameters (aVAP) based on said initial training recording parameters (iTAP) of the training dose profile classified as compliant, - providing adapted training recording parameters (aTAP) by applying the trained function (TF) to the input data, wherein the input data is based on the initial training recorded parameters (iTAP) and the training dose profile (TDP), - adapting (ADJ-TF) at least one parameter of said trained function (TF) based on a comparison of said adapted training recording parameter (aTAP) with said adapted comparison recording parameter (aVAP), - providing (PROV-TF) said trained function (TF).

7. The method according to claim 6, Wherein the input data of the trained function (TF) is based on a minimum value and / or a maximum value and / or an average value of the training dose profile (TDP).

8. The method according to claim 6 or 7, The initial training recording parameters (iTAP) contain information about a region to be mapped of at least one examination object (39) and / or a recording trajectory and / or a recording sequence and / or operating parameters for operating an X-ray source (37) and / or the X-ray detector (36).

9. The method according to any one of claims 6 to 8, The initial training recording parameters (iTAP) include at least one initial training reconstruction parameter for reconstructing image data from the training dose values.

10. The method according to claim 9, Wherein the adapted comparison recording parameters (aVAP) include at least one adapted comparison reconstruction parameter, and the adapted training recording parameters (aTAP) include at least one adapted training reconstruction parameter.

11. The method according to any one of claims 6 to 10, Wherein the adapting (ADJ-TF) of at least one parameter of the trained function (TF) is based on federated learning and / or a stochastic gradient method. 12 . A providing unit, which is designed to carry out the method according to claim 1 .

13. A medical X-ray device comprising an X-ray source (37), an X-ray detector (36) for photon counting and a providing unit (PRVS) according to claim 12, wherein the X-ray source (37) is designed to emit X-ray radiation, The X-ray detector (36) has a plurality of detector elements that perform photon counting, and the plurality of detector elements respectively provide dose values ​​according to the number of detected X-ray photons of the incident X-ray radiation.

14. A training unit (TRS) designed to carry out the method according to any one of claims 6 to 11.

15. A computer program product comprising a computer program, The computer program can be directly loaded into the memory of the providing unit (PRVS), the computer program having program segments so that when the program segments are executed by the providing unit (PRVS), all steps of the method according to any one of claims 1 to 5 are performed, and / or The computer program can be directly loaded into a training memory (TMU) of a training unit (TRS), the computer program having program segments so that when the program segments are executed by the training unit (TRS), all steps of the method according to any one of claims 6 to 11 are performed.

Images