DEVICE AND METHOD FOR DETERMINING AN ACCOUNTING PARAMETER AND / OR FOR PROVIDING A MAINTENANCE RECOMMENDATION FOR A COMPUTER TOMOGRAPHY FACILITY

Description

[0001] The present invention relates to the technical field of computed tomography, and in particular to computed tomography systems. Exemplary embodiments of the present invention relate to a device, a method, and a computer program for determining at least one acquisition parameter for acquiring X-ray data of a test object using a computed tomography system and / or for providing a maintenance recommendation for a computed tomography system. Some specific exemplary embodiments of the present invention relate to a cognitive CT add-on. Background of the invention

[0002] Industrial computed tomography (CT) or industrial micro-CT allows for the non-destructive imaging and analysis of internal structures in various objects. Image quality and measurement duration depend on the selected acquisition parameters, i.e., the parameterization of the CT system.

[0003] Addressing complex issues is generally very time-consuming: parameters must be manually determined and adapted for different components. Measurements with insufficient parameters must be repeated. To avoid repeat measurements, more measurement time than theoretically necessary is often planned as a precaution, thus tying up valuable and expensive working time of qualified specialists with unnecessary tasks. Parameterization is usually performed by a user; therefore, the optimization of CT measurements is heavily influenced by the user's experience.

[0004] Furthermore, there is practically no comparability between CT service measurements taken under different conditions. For example, CT service measurements are performed on different CT scanners from different manufacturers, by different users with varying levels of experience, or on a CT scanner in different maintenance states. This leads to sometimes different results within a single company when measurements are taken at different locations, on different CT systems, or by different CT operators. Moreover, the assessment of the resulting image quality is subjective.

[0005] Another problem concerns the maintenance and care of the X-ray components. An important aspect of predictive maintenance is, for example, the technical maintenance of the X-ray tube. The stability of the X-ray radiation can be improved in the short term by refocusing the X-ray beam. However, further measures are necessary in long-term operation. For instance, in the case of transmission tubes, after heavy use of the X-ray target, it may be necessary to rotate it slightly to increase the radiation intensity again and prevent physical target failure. Furthermore, if instability increases, it is helpful to condition the tube for several hours.

[0006] Currently, the determination of when maintenance is required is based on experience. However, neglected maintenance can negatively impact image quality and thus the analysis result, necessitating repeat measurements. Furthermore, neglected maintenance can shorten the lifespan of CT system components, leading to more frequent repairs, resulting in repair costs and system downtime.

[0007] Both experience-based parameterization and experience-based maintenance of CT systems therefore have a negative impact on the availability, throughput, and analysis results of CT systems, resulting in high costs and poor results.

[0008] Approaches already exist to address the parameterization, for example of acquisition parameters, or the quantification of image quality of CT measurements [1], [2]. The vast majority of these approaches are software-driven, for example through simulation-based approaches [3], or use existing X-ray hardware [4].

[0009] Some approaches attempt to use deep neural networks for the processing, enhancement, analysis, and interpretation of CT volume data. These methods are currently used in medical CT. Typical examples include radiation dose reduction [5], the detection and identification of abnormalities in the human body [6], and the classification of benign and malignant regions [7].

[0010] However, few activities exist for the application and use of such methods in the field of industrial CT, such as the detection of voids [8] in wheels, the segmentation of optical fibers [9], or the detection of foreign bodies in dual-energy images

[10] .

[0011] As mentioned at the outset, the maintenance of CT systems is also an important aspect that significantly influences the quality of the images. However, due to their diversity, the maintenance of entire CT systems cannot be standardized. Similarly, there is no uniform maintenance plan for individual components. Components are typically serviced when the scan quality noticeably deteriorates. Therefore, the decision to perform maintenance measures is based on the experience of an expert. On the other hand, the causes of wear are numerous, making a standardized maintenance approach, for example in the form of a guideline, difficult to envision. Publication

[11] presents a learning diagnostic system that detects faults in a machine by analyzing data acquired by the machine and suggesting repairs.In this case, the data is generated through a test procedure according to a test protocol. Such a test procedure is time-consuming and requires planned intervals for implementing a test protocol.

[0012] The publication

[12] proposes a method and system for user-specific parameterization of an X-ray system. In a medical context, this aims to optimize image quality for a defined user.

[0013] In the field of CT scans of humans, it is known to use cameras to determine the position of characteristic body parts of a patient with respect to a coordinate system of the CT scanner, cf. US 2015 / 0104092 A1.

[0014] DE 10 2018 105 709 A1 further describes a method for examining a workpiece using computed tomography, in which it is planned to take pre-radiographic images before a subsequent actual computed tomography in order to determine an acquisition parameter.

[0015] Further prior art relating to CT scans is known from EP 1 690 115 A1, EP 3 451 284 A1, US 2006 / 285645 A1, US 2019 / 056338A1, DE 10 2005 026940 A1 and US 2007 / 162190 A1.

[0016] While all known approaches deliver relatively usable results for optimizing CT systems, depending on their respective applications, the implementation of these techniques and procedures is sometimes very time-consuming and complex, resulting in a still rather unsatisfactory efficiency of CT systems.

[0017] Therefore, it would be desirable to improve known devices and methods for optimizing CT systems in such a way that the efficiency, i.e., the throughput, of the computed tomography system and also the quality of the results delivered by the computed tomography system can be increased, while at the same time making the operation of the CT system more user-friendly and simpler, and enabling measurements or recordings to be carried out with significantly reduced time expenditure.

[0018] The present invention is defined in independent claims 1 and 13. Further embodiments are defined in the dependent claims.

[0019] Therefore, a device for determining at least one recording parameter for recording one or more

[0020] X-ray projections of a workpiece using a computed tomography (CT) system, for example, a micro-CT system. The device includes a measuring device configured to provide measurement data, for example, multimodal data, about the workpiece, wherein the measurement data from the measuring device includes at least the average physical density of the workpiece and / or its mass, for example, through suitable sensors. The measuring device can be a standalone unit, or it can be integrated into a CT system, or at least it can be configured to be integrated into a CT system. Furthermore, the measuring device can, for example, be configured to process measurement data about the test object, for example, to combine one or more measured values ​​with one or more other measured values ​​or one or more values ​​otherwise available about the test object.Furthermore, the device includes a data analyzer which, using the measurement data from the measuring device and a prediction model, provides at least one acquisition parameter, for example an object-specific acquisition parameter, for the computed tomography system.The prediction model features a self-learning algorithmic process and is based on a multitude of datasets, each containing at least one of the following pieces of information: a test task, an object shape of a workpiece, a density distribution of a workpiece, a physical density of a workpiece, a mass of the workpiece, an X-ray projection of a workpiece, a CT volume dataset of a workpiece, an acquisition parameter of an X-ray projection and / or a CT volume dataset, status information concerning the computed tomography system, one or more measurement data points about a workpiece, one or more X-ray data points about a workpiece, information about the image quality of X-ray data of a workpiece.

[0021] The combination of obtaining measurement data and using the predictive model offers the advantage of not having to rely on prior information about the test object and the test task (i.e., the inspection task) to determine the acquisition parameters. By providing measurement data about the test object via the measuring device, information about the test object is available even without prior knowledge before a CT scan of the test object using the CT system. This makes it possible to determine at least one acquisition parameter with prior knowledge of the measurement data about the test object. The additional information about the test object allows an acquisition parameter to be specifically adapted to the test object.By adjusting an acquisition parameter to the test object, it is possible to save time and protect CT system components by, for example, not using more measurement time or a higher intensity than necessary for the test object. Furthermore, the result of the X-ray data acquisition using the CT system can be improved from the very first scan, thus avoiding repeat measurements and saving time. Compared to parameterization measurements of the test object using the CT system itself, this device offers the advantage that various measurement data about the test object, such as its shape in addition to its mass, can be incorporated into the determination of at least one acquisition parameter from the X-ray data. This increases the variety of information about the test object, thereby improving the accuracy of the prediction of at least one acquisition parameter.Furthermore, the device can provide measurement data significantly faster than a CT scanner. By combining the device's measurement data with a predictive model in the data analyzer, the quality of providing one or more acquisition parameters can be further improved. This is because the predictive model's prior knowledge allows it to adapt at least one acquisition parameter to the test object. As a result, an acquisition parameter does not need to be selected by a user based on their experience, but can be chosen based on the predictive model's prior knowledge. The device thus offers a cross-application solution for determining acquisition parameters, since the predictive model enables the data analyzer to provide acquisition parameters for various applications.Due to the use of the predictive model, the selection of the data analyzer's acquisition parameters is applicable both across users and for different CT systems. This allows for improvements in the quality of X-ray data acquisition and in the comparability between X-ray data acquired by different users and / or on different CT systems. The device according to the invention is thus able to shorten and optimize the phase before the start of the CT measurement, establish comparability between CT service measurements, optimize the utilization of service systems by avoiding long setup times and frequent repeat measurements, and eliminate the need for lengthy training of new employees in the complex field of computed tomography.Thus, the device according to the invention is able to increase the throughput and quality of the results supplied by the CT system.

[0022] Embodiments and further advantageous aspects of this device are mentioned in the respective dependent patent claims.

[0023] According to the invention, the data analyzer further uses an X-ray projection of the test object to provide the acquisition parameter for the computed tomography system.

[0024] The additional use of an X-ray projection of the test object increases the range of information available, thereby improving the quality of the provision of at least one acquisition parameter. Furthermore, combining the measurement data from the measuring device with the X-ray projection of the test object allows for the generation of additional information. For example, combining the (average) physical density of the test object with an X-ray projection can generate a density distribution. Knowledge of the density distribution of the test object is advantageous for determining an acquisition parameter, such as an exposure time or a projection number.

[0025] According to another embodiment, the data analyzer can also use a defined inspection task to provide the acquisition parameter for the computed tomography system. The inspection task could, for example, be the detection of defects or a target / actual comparison of the object's shape. Using the defined inspection task to determine the at least one acquisition parameter offers the advantage that an acquisition parameter can be adapted to the inspection task; that is, the acquisition parameter can be task-specific. Depending on the inspection task, one or more acquisition parameters can be optimized with regard to task-specific requirements. This allows, for example, a good compromise between measurement time and required image quality to be achieved. This, in turn, is advantageous for increasing the throughput of the CT system and for protecting components.

[0026] According to a further embodiment, the device can be designed to operate independently of the computed tomography system, for example, to provide an acquisition parameter for the system. This can mean, for instance, that the device can be operated in parallel with the computed tomography system, for example, simultaneously or to examine a different test object than one being examined concurrently by the computed tomography system. The device or measuring device can, for example, be operated without a hardware connection to the computed tomography system. For instance, the device or measuring device can be designed to operate in a location separate from the computed tomography system.This allows the device to operate independently of the computed tomography (CT) system. The time required to determine an acquisition parameter for one test object can then be used concurrently to acquire X-ray data for a second test object using the CT system. This parallelization saves time and increases the throughput of the CT system, i.e., the number of test objects examined. Furthermore, because the device is designed to be independent of the CT system, it can be used with any CT system from any manufacturer, as it does not depend on the CT system's specifications allowing it to operate. For example, the device can be operated regardless of whether a hardware interface to the CT system is available.For example, by using the device for different CT systems, the comparability of CT scans taken using the different CT systems can be further increased.

[0027] According to the invention, the device is designed ,to obtain an acquisition parameter for an X-ray projection and / or at least one acquisition parameter for the computed tomography system using the physical density of the test object. The physical density is an average physical density, for example, a quotient of the mass and volume of the test object. The physical density of the test object can be advantageously used to determine an X-ray spectrum suitable for the test object. By using the physical density of the test object to determine an acquisition parameter, such as an X-ray spectrum, the acquisition parameter can be particularly well adapted to the test object. For example, an X-ray spectrum can be selected to ensure good penetration of the test object by the X-rays.

[0028] According to another embodiment, the measurement data of the measuring device can include the physical density of the test object and / or the shape of the test object and / or its mass. The combination of the mass and shape of the test object is particularly well suited for accurately determining its physical density. This combination of mass and shape—for example, the evaluation of this measurement data—can be performed within the measuring device itself, so that the measurement data provided by the device includes the physical density of the test object.

[0029] According to a further embodiment, the measuring device can include a sensor for determining mass, for example, a scale, an accelerometer, or a force sensor, in order to determine the mass of the test object. Alternatively or additionally, the measuring device can include a stereoscopic sensor to determine the shape of the test object. Alternatively or additionally, the measuring device can be configured to determine the physical density of the test object based on its mass and shape. A stereoscopic sensor can, for example, be an imaging sensor capable of determining a three-dimensional image of the test object. The shape of the test object can, for example, be a convex hull or a spatial extent.A design of the measuring device according to this embodiment offers a particularly advantageous compromise between speed, effort and accuracy for providing the physical density of the test object.

[0030] According to a further embodiment, the device can be configured to determine a parameter relating to an X-ray spectrum, such as X-ray energy, spectral width, a physical pre-filter, or the voltage of an X-ray tube, based on the physical density of the test object. The physical density of the test object is a suitable parameter for adapting the X-ray spectrum used for X-ray projection or for acquiring X-ray data from the test object using the computed tomography system to the test object. This adaptation allows, for example, a good compromise to be found where the test object is sufficiently well illuminated while simultaneously protecting hardware components, such as the X-ray detector and the X-ray tube. This increases the service life of the corresponding components of the computed tomography system.

[0031] According to a further embodiment, the device can be configured to determine an image scale for the computed tomography system and / or for an X-ray projection using the shape of the test object. By adapting the image scale to the shape of the test object, good image quality of the X-ray data of the test object can be ensured while simultaneously protecting components of the computed tomography system, such as the X-ray detector.

[0032] According to a further embodiment, the device can be configured to determine the density distribution of the test object using its physical density and X-ray projection, and to provide an acquisition parameter, such as an exposure time or a projection number, for the computed tomography system based on this density distribution. By combining the physical density of the test object with the X-ray projection, the density distribution of the test object can be determined very accurately. Precise information about the density distribution of the test object, in turn, allows for a very good prediction of the acquisition parameters for the computed tomography system, for example, with regard to an exposure time or a projection number, so that good image quality of the X-ray data of the test object can be achieved while simultaneously saving measurement time and resources.

[0033] According to another embodiment, the measuring device can include an X-ray fluorescence sensor. Alternatively or additionally, the device can be configured to provide an acquisition parameter for the computed tomography system using data from the X-ray fluorescence sensor. The use of an X-ray fluorescence sensor (XRF) can provide information about the material composition of the test object, from which additional information for the X-ray spectrum, or the X-ray spectrum to be selected, can be obtained. A more precise determination of the X-ray spectrum allows for further optimization of the choice of acquisition parameters.

[0034] According to a further embodiment, the device can be configured to obtain an acquisition parameter relating to an axis system of the computed tomography system using the object shape of the test object and / or the definition of a test task. An axis system, for example a polyCT, can be a device that enables the simultaneous acquisition of X-ray data from multiple test objects with a CT system, for example by having multiple centers of rotation. Combining the measurement data with the predictive model and using a polyCT as an additional acquisition parameter to increase throughput makes it possible to optimize the use of such an axis system, avoiding time-consuming test scans and / or repeat measurements, across multiple tasks and not just for a single specific question.For example, an optimal angular resolution can be selected, thus increasing the throughput of the computed tomography system, while ensuring the image quality appropriate for a testing task.

[0035] According to the invention, the predictive model comprises a self-learning algorithmic method, for example, a method using machine learning techniques. For instance, the predictive model can be generated using machine learning methods trained on datasets from various test scenarios and test objects. The predictive model includes, for example, a multitude of datasets, each containing one or more pieces of information about a test object and / or information about X-ray data of the test object and / or information about how the X-ray data of the test object were acquired. The use of a self-learning algorithmic method offers the advantage that results from previous acquisitions of X-ray data of test objects can be used to improve the determination of at least one acquisition parameter for the computed tomography system.For example, to determine at least one acquisition parameter, information from X-ray data originating from a variety of different test objects, acquired using a variety of different CT systems, and / or acquired by a variety of different users can be combined. The device thus utilizes machine learning capabilities to shorten and optimize the phase before the start of the CT measurement. Furthermore, by employing the self-learning algorithm, the device or CT system exhibits a learning capability with regard to the acquisition parameters for well-known questions.

[0036] According to the invention, the prediction model is based on a plurality of data sets, each data set containing at least one of the following information: a test task, the shape of the test object, the density distribution of a test object, the physical density of a test object, the mass of a test object, an X-ray projection of a test object, a CT volume data set of a test object, at least one acquisition parameter of an X-ray projection and / or a CT volume data set, one or more status information relating to the computed tomography system, one or more measurement data about a test object, one or more X-ray data about a test object, information about the image quality of X-ray data about a test object.By evaluating one or more of these pieces of information from the multitude of datasets, the predictive model is able to adapt at least one acquisition parameter for the computed tomography system to the test object or test task and / or the computed tomography system and / or another framework condition, such as the state of the computed tomography system. The use of a self-learning algorithm offers the advantage that the determination of at least one acquisition parameter results from the evaluation of previous datasets and is not dependent on the individual user. This improves the comparability of CT data acquired by different users. Furthermore, the predictive model, through the self-learning algorithm, is able to establish particularly good comparability between different computed tomography systems. The use of a self-learning algorithm creates a learning capability of the CT system.the device according to the invention with regard to the recording parameters for well-known issues, e.g. testing tasks.

[0037] According to a further embodiment, the data analyzer can be configured to provide a maintenance recommendation for the computed tomography system using an X-ray projection of the test object generated by the computed tomography system and the predictive model. The predictive model can incorporate a self-learning algorithmic process and can be based on information from a multitude of X-ray projections of the test object and / or other test objects previously generated by the computed tomography system. The maintenance recommendation can, for example, include calibration aspects of a detector and / or axis system belonging to the CT system, or aspects of predictive maintenance, such as those relating to the X-ray tube.Analyzing the maintenance status of a computed tomography (CT) system using a predictive model offers the advantage that the system's condition can be incorporated into determining at least one acquisition parameter. Providing a maintenance recommendation using the predictive model and the X-ray projection of the test object enables the device according to the invention to identify situationally required maintenance steps and prevent defects in X-ray components. Furthermore, using an X-ray projection of the test object to provide the maintenance recommendation avoids the need for additional X-ray projections for testing or calibration purposes. This prevents unnecessary downtime due to premature or excessive maintenance measures or costly service calls.

[0038] According to a further embodiment, the data analyzer can use the X-ray projection of the test object generated by the computed tomography system to determine information about image quality, such as the homogeneity of the X-ray projection or the number of defective pixels in the X-ray projection generated by the computed tomography system. Alternatively or additionally, the data analyzer can use this information about the image quality of the X-ray projection generated by the computed tomography system to determine information about the current maintenance status of the computed tomography system, such as the number of defective pixels in an X-ray detector. A defective pixel could, for example, be a pixel with reduced or no dynamic range compared to the normal state.Degradation of the X-ray detector can manifest as an increased number of defective pixels, meaning pixels that exhibit reduced dynamic range compared to normal operation or even no longer carry any image information at all. For example, defective pixels may be detected when determining maintenance recommendations. These can be masked out or compensated for either through maintenance or by applying a median filter during image acquisition, which is an acquisition parameter. The overall sensitivity and noise floor of the detector can change, which may necessitate the acquisition of new calibration images, for example, for gain or offset adjustments, to obtain X-ray data or projections with as little detector-induced interference as possible.An axis system can become misaligned through frequent use and mechanical or thermal stress, causing the assumed geometry to deviate from the actual scan geometry and resulting in artifacts in the reconstructed volume. Information about the current maintenance status of the computed tomography system can include, for example, the sensitivity or background noise of an X-ray detector, or an axis system calibration. Alternatively or additionally, the device can use this information to provide maintenance recommendations for the computed tomography system. Using the image quality of the X-ray projection of a test object to provide maintenance recommendations offers the advantage that these recommendations can be generated during the operation of the computed tomography system and in response to X-ray data from the tested objects.

[0039] Based on the embodiments described above, a further embodiment of the present invention relates to a corresponding device for providing a maintenance recommendation for a computed tomography system, wherein the device comprises a data analyzer which provides the maintenance recommendation for the computed tomography system using an X-ray projection of a test object generated by the computed tomography system and using a predictive model. The predictive model incorporates a self-learning algorithmic method. Furthermore, the predictive model is based on information from a plurality of X-ray projections of the test object and / or other test objects generated previously by the computed tomography system and / or other computed tomography systems.Using a computed tomography (CT) scan image of an object for maintenance recommendations allows for predicting maintenance needs without requiring separate CT scan data acquisition. Instead, existing data acquired during the actual inspection is used. This increases the CT scan's availability for testing objects. Furthermore, providing maintenance recommendations based on the CT scan image enables timely detection of maintenance needs.This enables timely maintenance, which has a beneficial effect on components of the computed tomography system, such as an X-ray tube or an axis system, and extends the service life of these components. Instead of relying on the subjective experience of a user, the device according to the invention provides objective indications of when maintenance of the X-ray components is necessary, i.e., when maintenance measures are required. Timely maintenance therefore avoids costly repairs and, consequently, unnecessary downtime. The device according to the invention is able to recognize the maintenance steps required in each individual situation and prevent defects in X-ray components. Unnecessary downtime due to premature or excessive maintenance measures or costly service calls is avoided.Thus, the device according to the invention is able to avoid costly repairs and unnecessary downtime due to neglected or excessive maintenance of the X-ray components, thereby increasing the throughput and quality of the results delivered by the CT system.

[0040] According to a further embodiment, the device for providing a maintenance recommendation for a computed tomography system can be designed such that the data analyzer receives information about the image quality of the X-ray projection of the test object using the computed tomography system, and the data analyzer receives information about the current maintenance status of the computed tomography system using the information about the image quality of the X-ray projection, and the device provides the maintenance recommendation for the computed tomography system using the information about the current maintenance status of the computed tomography system.

[0041] Another embodiment relates to a computed tomography system with a device described above for determining at least one imaging parameter for recording X-ray data of a test object using a computed tomography system, and / or with a device described above for providing a maintenance recommendation for a computed tomography system.

[0042] The corresponding computed tomography system is based on the same principles as the devices described above. Furthermore, it should be noted that the computed tomography system can be supplemented with all the features, functions, and details described herein with regard to the device according to the invention. The computed tomography system can be supplemented with the aforementioned features, functionalities, and details both individually and in combination.

[0043] According to the invention, a corresponding method for determining at least one recording parameter for recording X-ray data of a test object using a computed tomography system is defined according to claim 13.

[0044] Another embodiment of the present invention relates to a corresponding method for providing a maintenance recommendation for a computed tomography system, wherein the method comprises the following steps: obtaining an X-ray projection of a test object produced by the computed tomography system and providing the maintenance recommendation for the computed tomography system using the X-ray projection of the test object and using a predictive model based on the image quality of the X-ray projection of the test object, wherein the predictive model comprises a self-learning algorithmic method and wherein the predictive model is based on information from a plurality of X-ray projections of the test object and / or of further test objects previously produced by the computed tomography system and / or other computed tomography systems.

[0045] Another embodiment relates to a computer program with program code for carrying out at least one of the aforementioned methods, when the program runs on a computer.

[0046] Some non-limiting embodiments are shown in the drawing and explained below. They show: Fig. 1 a schematic representation of a device for determining an acquisition parameter for a computed tomography system according to one embodiment, Fig. 2 a schematic representation of a device for determining an acquisition parameter for a computed tomography system according to another embodiment, Fig. 3 a schematic representation of a device for providing a maintenance recommendation for a computed tomography system according to one embodiment, Fig. 4A a schematic representation of a computed tomography system with a device for determining at least one acquisition parameter for acquiring X-ray data of a test object using the computed tomography system according to one embodiment, Fig. 4B a schematic representation of a computed tomography system with a device for providing a maintenance recommendation for the computed tomography system according to one embodiment.Figure 5 shows a block diagram of a method for determining an acquisition parameter for a computed tomography system according to an embodiment, and Figure 6 shows a block diagram of a method for providing a maintenance recommendation for a computed tomography system according to an embodiment. Detailed description of the invention

[0047] Exemplary embodiments are described in more detail below with reference to the figures, wherein elements with the same or similar function are provided with the same reference numerals. Elements shown with dashed lines are either optional elements or elements that are not part of the invention but are shown in the drawing for a better understanding of the invention.

[0048] Process steps depicted in a block diagram and explained with reference to it can also be executed in a different sequence than the one shown or described. Furthermore, process steps relating to a specific feature of a device are interchangeable with that very feature of the device, and vice versa.

[0049] Figure 1Figure 1 shows a schematic representation of an embodiment of a device 100 for determining at least one acquisition parameter 150 for acquiring X-ray data 192 of a test object 110 using a computed tomography system 190. The device 100 includes a measuring device 120 configured to provide measurement data 122 about the test object 110. The device 100 also includes a data analyzer 130. Using the measurement data 122 from the measuring device 120 and a predictive model 140, the data analyzer 130 provides the at least one acquisition parameter 150 for the computed tomography system 190.

[0050] The in Fig. 1The arrangement of the device 100 shown in relation to the CT system 190 is to be understood schematically; that is, the device 100 can be arranged spatially separately, next to or within the CT system 190, or integrated into the CT system 190. Likewise, parts of the device 100, for example, the measuring device 120 and / or the data analyzer 130, can be arranged spatially separately, next to or within the CT system 190, or integrated into the CT system 190. It is also possible that components of the device 100 and the CT system 190 can be shared.

[0051] Test object 110, for example, is a workpiece whose flawless manufacturing can be verified. Predictive model 140 may, for example, consist of a neural network.

[0052] Figure 2 shows a schematic representation of another embodiment of the in Fig. 1Device 100 shown for determining at least one recording parameter 150 for recording X-ray data 192 of a test object 110 using a computed tomography system 190.

[0053] An acquisition parameter 150 for acquiring the X-ray data 192 can, for example, be an X-ray spectrum, i.e., a spectral distribution of the X-rays used to acquire the X-ray data 192, or a parameter that influences the X-ray spectrum, such as an X-ray voltage (a voltage for an X-ray tube) or physical pre-filtering, i.e., spectral filtering of the X-ray light. An acquisition parameter 150 can relate to an exposure for the acquisition. An acquisition parameter 150 can be a current-time product or an intensity. An acquisition parameter 150 can be a projection number, a spatial resolution of the overall image, or an image scale. An acquisition parameter 150 can also be a measurement mode, i.e., a focal spot size, which can, for example, influence the intensity and the spatial resolution of the overall image.A recording parameter 150 can also relate to an axis system, for example a polyCT system; for example, a recording parameter 150 can be a suggestion for the use of an axis system and / or a parameter for an axis system, for example an angle.

[0054] The data analyzer 130 can, for example, be a computer. The data analyzer 130 can be connected to the measuring device 120 for the purpose of electronic data transmission. The data analyzer 130 can also be connected to the computed tomography system 190 for the purpose of electronic data transmission.

[0055] The data analyzer 130 uses an X-ray projection 160 of the test object 110 to provide at least one recording parameter 150 for the computed tomography system 190.

[0056] An X-ray projection 160 of the test object 110 can, for example, include a fluoroscopy of the test object 110 or a two-dimensional X-ray image of the test object 110.

[0057] The data analyzer 130 can obtain the X-ray projection 160, for example, directly from the X-ray detector, manually by the operator, or via a "watchdog" mechanism. Various input options for the X-ray projection ensure the manufacturer-independent applicability of the device 100. The data analyzer 130 can use a defined test task 135 to provide the at least one acquisition parameter 150 for the computed tomography system 190. This means that the data analyzer 130 can use the measurement data 122, for example, multimodal measurement data, the X-ray projection 160, and the prediction model 130 to decide on the at least one acquisition parameter 150 for acquiring the X-ray data 192 for the test object 110 specifically for the previously defined test task 135. The test task 135 can, for example, be defined by a user or selected from a range of test tasks or test scenarios.Test task 135 can, for example, involve finding defects or comparing the object's shape to its actual shape. Test task 135 can be a task from which criteria for acquiring the X-ray data 192 (e.g., a measurement time) or for the X-ray data 192 to be acquired (e.g., a spatial resolution) can be derived. Test task 135 can, for example, define one or more criteria for optimizing the acquisition of the X-ray data 192. For a metrological question, such as comparing the object's shape to its actual shape, a different spatial resolution may be required than for finding defects with a minimum diameter of one millimeter. The data analyzer 130 can use test task 135 to define an optimization task for the prediction model 140; that is, the optimization task can depend on test task 135, i.e., on a specific task.The optimization task can relate, for example, to a minimum measurement time to increase throughput or to a required image quality. The data analyzer 130 can select at least one acquisition parameter 150 differently depending on the test object 110 and the test task 135.

[0058] The device 100 can be designed in such a way that it can be operated independently of the operation of the computed tomography system 190. For example, the device 100 can be a stand-alone unit.

[0059] Alternatively, the device 100 or the measuring device 120 can be integrated into the computed tomography system 190, for example, by installing measuring sensors for acquiring the measurement data 122. This allows the measurement data 122 to be provided via the test object 110 at the same location as the acquisition of the X-ray data 192 via the test object 110. Thus, after determining at least one acquisition parameter 150, the test object 110 no longer needs to be moved or installed in another device. Alternatively or additionally, the data analyzer 130 can be integrated into the CT system 190, for example, by using a shared computer.

[0060] The device 100 can be configured to obtain, using a physical density of the test object 110, at least one recording parameter 150 for the computed tomography system 190 and / or an alternative or additional recording parameter for an X-ray projection.

[0061] The measuring device 120 can include a sensor 124 for determining a mass in order to obtain the mass of the test object 110. Furthermore, the measuring device 120 can include a stereoscopic sensor 126 to determine the shape of the test object 110. The measuring device 120 can also be configured to provide the physical density of the test object 110 based on its mass and shape.

[0062] The measurement data 122 of the measuring device 120 include at least the physical density of the test object 110 and / or a mass of the test object 110 and / or the object shape of the test object 110.

[0063] The device 100 can determine one or more acquisition parameters 150 based on the measurement data 122. For example, the device 100 can be configured to obtain a parameter relating to an X-ray spectrum based on the physical density of the test object 110. An X-ray spectrum can, for example, be provided as an acquisition parameter 150 for the computed tomography system 190. Furthermore, an X-ray spectrum can be provided as an acquisition parameter for an acquisition of an X-ray projection 160 to determine the at least one acquisition parameter 150 for the computed tomography system 190.

[0064] Furthermore, the device 100 can be designed to provide an image scale for the computed tomography system 190 and / or for an X-ray projection using the object shape of the test object 110.

[0065] The device 100 can be configured to obtain a density distribution of the test object 110 using the physical density and the X-ray projection, and to provide an acquisition parameter 150 for the computed tomography system 190 using the density distribution of the test object 110.

[0066] The measuring device 120 can additionally include an X-ray fluorescence sensor (XRF sensor). The device 100 can be configured to provide an acquisition parameter 150 for the computed tomography system 190 using data from the X-ray fluorescence sensor. An XRF sensor can provide information about the material composition of the test object 110, which the data analyzer 130 can use to determine the at least one acquisition parameter 150, for example, to determine an X-ray spectrum suitable for the test object 110. The device 100 can be configured to obtain an acquisition parameter 150 relating to an axis system of the computed tomography system 190 using the object shape of the test object 110 and / or the definition of a test task 135.Furthermore, the computed tomography system 190 can have an axis system, for example, a poly-CT system, which allows several test objects 110 to be examined simultaneously, thus further increasing efficiency. The axis system can be designed, for example, to be scaled to be used with any CT system. The axis system can, for example, be equipped with variable rotation centers. The data analyzer 130 can determine an acquisition parameter 150 regarding the use of the axis system using the test task 135, whereby the prediction model 140 can employ machine learning methods. Through appropriate use of the axis system, it may be possible to achieve a throughput increase of a factor of 2 to 5, depending on the shape of the test object 110 or multiple test objects.

[0067] To determine the acquisition parameter 150, the data analyzer 130 uses a prediction model 140. The prediction model 140 has a self-learning algorithmic procedure. The prediction model 140 is based on a multitude of data sets, each of which can contain at least one of the following pieces of information: a test task, the shape of a test object, the density distribution of a test object, the physical density of a test object, the mass of the test object, an X-ray projection of a test object, a CT volume data set of a test object, an acquisition parameter of an X-ray projection and / or a CT volume data set, one or more status information pieces relating to the computed tomography system, one or more measurement data pieces about a test object, one or more X-ray data pieces about a test object, or an image quality of X-ray data about a test object.

[0068] The predictive model 140 can be generated using machine learning methods, which can be trained on datasets of different X-ray images of a test object 110. Such a dataset of an X-ray image of a test object 110 can, for example, contain the acquired X-ray data of the test object 110 and / or an evaluation or assessment of the image quality of the acquired X-ray data of the test object 110.Furthermore, a data set concerning an acquisition of X-ray data of a test object 110 may contain information related to the acquisition of the X-ray data, for example, an acquisition parameter, in particular the acquisition parameter to be provided by the prediction model 140, or the data set may contain information concerning the tested test object 110, for example, information of the type provided by the measurement data 122 supplied by the measuring device 120, such as a mass, an object shape, or a physical density. Furthermore, a data set concerning an acquisition of X-ray data of a test object 110 may contain information concerning the computed tomography system 190, for example, information about a manufacturer, a model, or a component of the computed tomography system 190, or a condition, such as an operating condition or a maintenance condition, of the computed tomography system 190.

[0069] The results of the image quality evaluation, or the maintenance status itself, can be incorporated into the parameterization for determining the maintenance status. This could, for example, involve a joint prediction model 140 for the acquisition parameters 150 and the maintenance recommendation.

[0070] The prediction model 140 can receive as input for determining at least one recording parameter 150 the measurement data 122 generated by the measuring station 120, and / or the X-ray projection 160 and / or a categorization of the test task 135 for predicting the parameters to be used.

[0071] The at least one acquisition parameter 150 provided by the data analyzer 130 can, for example, include at least one of the following acquisition parameters: an X-ray spectrum, an X-ray voltage, a physical pre-filtering, a current-time product, an image scale, a projection number, a parameter relating to an axis system, a measurement mode, a focal spot size, an intensity, a spatial resolution of the overall image.

[0072] In addition to determining the acquisition parameter 150, the data analyzer 130 can further be configured to provide a maintenance recommendation 180 for the computed tomography system 190, i.e., for example, a maintenance recommendation for the X-ray components of the computed tomography system 190, using an X-ray projection 160 of the test object 110 created by the computed tomography system 190 and using the predictive model 140. The predictive model 140 can have a self-learning algorithmic procedure and can be based on information about a large number of X-ray projections of the test object 110 and / or other test objects previously created by the computed tomography system 190 and / or other computed tomography systems.

[0073] The predictive model 140 can also be trained to provide recommendations for the maintenance of X-ray components. A maintenance recommendation 180, for example, might include calibration aspects of a detector, such as an X-ray detector, or an axis system, such as a polyCT system. A maintenance recommendation 180 might address degradation, sensitivity, or background noise of an X-ray detector. Furthermore, a maintenance recommendation 180 might concern an X-ray tube, such as focusing an X-ray beam, rotating an X-ray target, or conditioning the X-ray tube. The maintenance recommendation 180 can ensure that these measures are carried out in a timely manner.

[0074] The maintenance recommendation 180 for the computed tomography system 190 can optionally be based exclusively on the prediction model 140 and the X-ray projection 160 of the test object 110.

[0075] The data analyzer 130 can obtain information about the image quality of the X-ray projection 160 of the test object 110, generated by the computed tomography system 190. This allows the data analyzer 130 to determine the current maintenance status of the computed tomography system 190 using this information about the image quality of the X-ray projection 160. Based on this information about the current maintenance status of the computed tomography system 190, the device 100 can then provide a maintenance recommendation 180 for the computed tomography system 190.

[0076] The following describes the function and interaction of the individual components of the device 100.

[0077] Device 100 can be designed to be used with various computed tomography systems. Accordingly, computed tomography system 190 can be any computed tomography system from any manufacturer.

[0078] The device 100 can be configured to electronically transmit the acquisition parameter 150 to the computed tomography system 190. However, the acquisition parameter 150 can also be transmitted manually by a user. For example, the device 100 can be configured to display the acquisition parameter 150, allowing a user to enter the acquisition parameter 150 into the computed tomography system 190 or into a program for operating the computed tomography system. This ensures that the device 100 can be used with any computed tomography system 190, regardless of the manufacturer of the computed tomography system 190.

[0079] The data analyzer 130 can use the X-ray projection 160 to determine a density distribution of the test object 110 using a physical density of the test object 110, which may be contained in the measurement data 122 or calculated from the measurement data 122, which can, for example, serve as a basis for presetting the exposure time and projection number.

[0080] Optionally, the measuring device 120 can include an X-ray detector to record the X-ray projection 160 of the test object 110. This allows the X-ray projection 160 to be recorded while the computed tomography system 190 is available for acquiring X-ray data of a test object different from the test object 110. Furthermore, this allows the determination of at least one acquisition parameter 150 to be completed before the test object 110 is placed in the computed tomography system 190 for the acquisition of the X-ray data 192.

[0081] Optionally, the X-ray projection 160 of the test object 110 can be acquired using the computed tomography system 190, thus eliminating the need for additional X-ray hardware, such as an X-ray detector or X-ray tube, for the measuring device 120. This makes the measuring device more cost-effective, easier to operate, and exempt from radiation protection regulations. Alternatively or additionally to a self-learning algorithm, the prediction model 140 can be a physical, X-ray-physical, or analytical model, or an algorithmic method.

[0082] Fig. 3Figure 3 shows a further embodiment of a device 300 according to another aspect of the invention. This device 300 is configured to provide a maintenance recommendation 180 for a computed tomography system 390. The device 300 has a data analyzer 330 which, using an X-ray projection 360 of a test object 310 generated by the computed tomography system 390 and a predictive model 340, provides the maintenance recommendation 380 for the computed tomography system 390. The predictive model 340 has a self-learning algorithmic method. Furthermore, the predictive model 340 is based on information from a plurality of X-ray projections of the test object 310 and / or other test objects generated in advance by the computed tomography system 390 and / or other computed tomography systems.

[0083] Device 300 can replace device 100 Fig. 1 or Fig. 2The data analyzer 330 can be replaced by the data analyzer 130. Fig. 1 or Fig. 2 The prediction model 340 can correspond to the prediction model 140. Fig. 1 or Fig. 2 The 360° X-ray projection can be compared to the 160° X-ray projection. Fig. 2 Maintenance recommendation 380 can be replaced by maintenance recommendation 180. Fig. 2 The CT scanner 390 can replace the CT scanner 190. Fig. 1 or Fig. 2 The test object 310 can correspond to the test object 110 from Fig. 1 or Fig. 2 are equivalent to.

[0084] The data analyzer 330 can obtain information about the image quality of the X-ray projection 360 of the test object 310, which was created by the computed tomography system 390. Furthermore, the data analyzer 330 can obtain information about the current maintenance status of the computed tomography system 390, using this information. The device 300 can then provide a maintenance recommendation 380 for the computed tomography system, using this information.

[0085] Fig. 4AFigure 1 shows an embodiment of a complete system comprising a computed tomography system 490 and a device 100 according to the invention. The device 100 serves to determine at least one imaging parameter 450 for acquiring X-ray data 492 of a test object 410 using the computed tomography system 490.

[0086] Maintenance recommendation 480 can be replaced by maintenance recommendation 180. Fig. 2 The CT system 490 can be replaced by the CT system 190. Fig. 1 or Fig. 2 Therefore, for details on determining at least one recording parameter 450, please refer to the description of the Figure 1 and 2 referred.

[0087] Alternatively or in addition to the one with reference to the Figure 1 and 2 The described device 100 can be used in Figure 4A The device shown 100 also provides a maintenance recommendation 480 for the computed tomography system.

[0088] The in Fig. 4AThe arrangement of the device 100 shown in relation to the CT system 490 is to be understood schematically; that is, the device 100 may be arranged wholly or partially separately, next to or within the CT system 490, or integrated into the CT system 490. It is also possible that components of the device 100 and the CT system 490 may be shared.

[0089] Fig. 4B Figure 1 shows a further embodiment of a complete system comprising a computed tomography system 491 and a device 300 according to the invention, wherein the device here serves to provide a maintenance recommendation 481 for the computed tomography system 491.

[0090] Maintenance recommendation 481 can be replaced by maintenance recommendation 380. Fig. 3 CT scanner 491 can be replaced by CT scanner 390. Fig. 3 Therefore, for details on providing a maintenance recommendation, please refer to the description at Figure 3 referred.

[0091] The in Fig. 4B The arrangement of the device 300 shown in relation to the CT system 491 is to be understood schematically; that is, the device 300 can be arranged wholly or partially separately, next to or within the CT system 491, or integrated into the CT system 191. It is also possible that components of the device 300 and the CT system 491 may be shared.

[0092] Fig. 5Figure 500 shows a block diagram of a method 500 for determining an acquisition parameter 150, 450 for a computed tomography system 190, 490 according to an exemplary embodiment. The method 500 for determining at least one acquisition parameter 150, 450 for acquiring X-ray data 192, 492 of a test object 110, 410 using a computed tomography system 190, 490 comprises a step 501. In step 501, measurement data 122 are generated, for example, by measurement or acquisition, wherein the measurement data 122 include information about the physical density of the test object 110, 410. In step 502, the at least one acquisition parameter 150, 450 is provided to the computed tomography system 190, 490 using a prediction model 140.

[0093] Fig. 6Figure 600 shows a block diagram of a method 600 for providing a maintenance recommendation 180, 380, 480, 481 for a computed tomography system 190, 390, 490, 491 according to an exemplary embodiment. In step 601, an X-ray projection 160, 360 of a test object 110, 310, 410, created using the computed tomography system 190, 390, 490, 491, is obtained.In step 602, the maintenance recommendation 180, 380, 480, 481 for the computed tomography system 190, 390, 490, 491 is provided using the X-ray projection 160, 360 of the test object 110, 310, 410 and using a prediction model 140, 340 based on the image quality of the X-ray projection 160, 360 of the test object 110, 310, 410, wherein the prediction model 140, 340 has a self-learning algorithmic procedure and wherein the prediction model 140, 340 is based on information from a multitude of X-ray projections of the test object 110, 310, 410 and / or other computed tomography systems previously created using the computed tomography system 190, 390, 490, 491 and / or other computed tomography systems. Test objects are based on this.

[0094] The present invention, comprising the apparatus 100, 300, and the method 500 and 600, will now be summarized again in other words: The invention comprises a cognitive computed tomography system 190, which can be implemented as an add-on suitable for micro-CT systems from all manufacturers. By simultaneously evaluating multimodal sensor data in combination with recorded projections from a standard CT system 190, optimal acquisition parameters can be determined using machine learning (ML) methods 140, and recommendations for the maintenance of the X-ray components can be provided.

[0095] According to one embodiment, the data analyzer 130 can output at least the following parameters: the appropriate X-ray spectrum (X-ray energy, physical pre-filtering), current-time product, image scale, projection number and the use of PolyCT.

[0096] According to one embodiment, the data analyzer 130 suggests a measurement mode, i.e., a focal spot size, which also influences the intensity and spatial resolution of the overall image. By selecting a suitable measurement mode, the X-ray source of the computed tomography system 190 can be conserved.

[0097] The invention provides a measuring station that can be used with micro-CT systems 190 from all manufacturers and that can provide multimodal data about the test object 110. Furthermore, the invention includes an algorithmic method that can use both the projection data from a CT system 190 and the measurement data from the measuring station to determine one or more object- and test-task-specific acquisition parameters for a CT measurement using machine learning (ML) methods 140 and to provide information on the maintenance of the X-ray components. This self-calibration results in an overall "cognitive" CT add-on.

[0098] The effects of the cognitive CT add-on include the determination of CT acquisition parameters and the indication of maintenance for X-ray components. The advantages of the cognitive CT add-on, for example, device 100, 300, or procedure 500 or 600, are as follows: Firstly, it establishes comparability between CT service measurements. This strengthens confidence in CT as a quality assurance tool and facilitates the use of CT for standardized testing procedures.

[0099] On the other hand, the efficiency of CT service providers is increased through the use of cognitive CT add-ons, such as Device 100, 300, or Procedure 500 or 600. Currently, the utilization of service facilities is limited not by a lack of orders, but by long setup times, avoidable repeat measurements, and a lack of or insufficiently experienced personnel. Device 100, 300, and Procedure 500, 600 address these issues: setup times are shortened, repeat measurements are reduced, and newly hired specialists can begin work sooner, i.e., after less intensive training. Therefore, more inspection orders can be processed in the same amount of time, with consistently high quality.

[0100] At the same time, devices 100 and 300, as well as methods 500 and 600, offer the advantage of identifying necessary maintenance steps in a timely manner, thus reducing unnecessary downtime and costly repairs. This not only increases the efficiency of operations but also enables the sustainable use of material resources.

[0101] As an alternative to using the device 100 or the method 500 according to the invention, a rapid prescan prior to the actual CT measurement can enable an estimation of image quality [4]. This is time-consuming and does not offer any learning capability. Moreover, this approach is only feasible if the interfaces to the X-ray components are open. Therefore, unlike the device 100 and the method 500, this approach is not applicable across manufacturers for all micro-CT systems.

[0102] As an alternative to using the device 100, 300 or the method 500, 600 according to the invention, sensor data can be used as input for parameterization methods that are not based on AI methods, but rather on simulation or X-ray physics principles. However, in this case, a disadvantage lies in the system's lack of learning capability.

[0103] According to the invention, the method described herein can be used for image acquisition planning or for the purpose of predictive maintenance of CT systems or X-ray components. This method can be equipped with a sensor connection, enabling the determination of object / material properties and their evaluation using AI methods. According to the invention, this can be achieved, for example, by equipping a CT system 190 with a measuring device (e.g., a laboratory balance), and / or a force sensor and / or other sensors, from which, for example, the mean density or surface data of the object under investigation 110 can be determined.

[0104] In the prior art, inspection personnel typically rely on precise knowledge of the object being inspected and the specific issue (i.e., the inspection task). However, not being dependent on this information is a significant advantage of the present invention. Using the present invention, CT service providers can directly increase their throughput of CT measurements because a shorter measurement time, reduced manual parameterization, and the elimination of repeat measurements allow more orders to be processed in the same amount of time. Furthermore, the advancement of additive manufacturing processes will lead to a decentralization of production in the medium term. To meet the logistical challenges, there will therefore be an increased need for quality assurance tools at different locations while maintaining consistent inspection quality.At the same time, quality assurance using CT is necessary, as there is currently no long-term empirical data regarding the quality of additively manufactured components. Since the sheer number of objects to be examined would exceed the logistical capabilities of service laboratories, CT must be made accessible to a broad range of users.

[0105] Research institutions use CT to solve problems within their own disciplines, such as geology, glaciology, process engineering, space physics, or crystallography. By using cognitive CT add-ons, for example, devices 100 and 300 or procedures 500 and 600, improved image quality and thus better data analysis can be achieved.

[0106] The embodiments described above merely illustrate the principles of the present invention. It is understood that modifications and variations of the arrangements and details described herein will be obvious to other people skilled in the art. Therefore, it is intended that the invention be limited only by the scope of protection set forth in the following claims and not by the specific details presented herein by way of description and explanation of the embodiments.

[0107] Although some aspects have been described in connection with a device, it is understood that these aspects also constitute a description of the corresponding process, so that a block or component of a device is also to be understood as a corresponding process step or as a feature of a process step. Similarly, aspects described in connection with or as a process step also constitute a description of a corresponding block, detail, or feature of a corresponding device.

[0108] Some or all of the process steps can be performed by (or using) a hardware apparatus, such as a microprocessor, a programmable computer, or an electronic circuit. In some embodiments, some or more of the key process steps can be performed by such an apparatus.

[0109] Depending on specific implementation requirements, embodiments of the invention can be implemented in hardware or in software, or at least partially in hardware or at least partially in software. The implementation can be carried out using a digital storage medium, for example, a floppy disk, a DVD, a Blu-ray disc, a CD, a ROM, a PROM, an EPROM, an EEPROM, a FLASH memory, a hard disk, or another magnetic or optical storage medium, on which electronically readable control signals are stored. These control signals can interact with, or interact with, a programmable computer system in such a way as to execute the respective method. Therefore, the digital storage medium can be computer-readable.

[0110] Some embodiments according to the invention therefore include a data carrier which has electronically readable control signals which are able to interact with a programmable computer system in such a way that one of the methods described herein is carried out.

[0111] In general, embodiments of the present invention can be implemented as a computer program product with a program code, wherein the program code is effective in carrying out one of the methods when the computer program product runs on a computer.

[0112] The program code can also be stored on a machine-readable medium, for example.

[0113] Other embodiments include a computer program for carrying out one of the methods described herein, wherein the computer program is stored on a machine-readable medium. In other words, an embodiment of the method according to the invention is thus a computer program that includes program code for carrying out one of the methods described herein when the computer program is executed on a computer.

[0114] Another embodiment of the methods according to the invention is thus a data carrier (or a digital storage medium or a computer-readable medium) on which the computer program for carrying out one of the methods described herein is recorded. The data carrier or the digital storage medium or the computer-readable medium is typically tangible and / or non-volatile.

[0115] Another embodiment of the method according to the invention is thus a data stream or a sequence of signals that represents the computer program for carrying out one of the methods described herein. The data stream or sequence of signals can be configured, for example, to be transferred via a data communication connection, such as the Internet.

[0116] Another embodiment comprises a processing device, for example a computer or a programmable logic device, which is configured or adapted to perform one of the methods described herein.

[0117] Another embodiment comprises a computer on which the computer program for performing one of the procedures described herein is installed.

[0118] Another embodiment of the invention comprises a device or system designed to transmit a computer program for carrying out at least one of the methods described herein to a receiver. The transmission can be, for example, electronic or optical. The receiver can be, for example, a computer, a mobile device, a storage device, or a similar device. The device or system can, for example, include a file server for transmitting the computer program to the receiver.

[0119] In some embodiments, a programmable logic device (for example, a field-programmable gate array, an FPGA) can be used to perform some or all of the functionalities of the methods described herein. In some embodiments, a field-programmable gate array can interact with a microprocessor to perform one of the methods described herein. Generally, in some embodiments, the methods are performed by any hardware device. This can be general-purpose hardware such as a computer processor (CPU) or method-specific hardware such as an ASIC.

[0120] The embodiments described above merely illustrate the principles of the present invention. It is understood that modifications and variations of the arrangements and details described herein will be obvious to other people skilled in the art. Therefore, it is intended that the invention be limited only by the scope of protection set forth in the following claims and not by the specific details presented herein by way of description and explanation of the embodiments. References

[0121] [1] R. Schielein, "Analytical Simulation and Acquisition Planning for Industrial X-ray Computed Tomography," Dissertation, University of Würzburg, 2018. [2] Fraunhofer-Gesellschaft, "Device and Method for Calculating an Acquisition Trajectory." Patent Publication No. DE 102016213403 A1, 2018. [3] S. Reisinger, "Simulation-Based Acquisition Planning in Industrial Computed Tomography," Master's Thesis, University of Bamberg, 2010. [4] M. Reiter et al., "Evaluation of a histogram-based image quality measure for X-ray computed tomography," iCT, 2012. [5] [Online]. Available: https: / / arxiv.org / ftp / arxiv / papers / 1609 / 1609.08508.pdf. [6] [Online]. Available: https: / / www.researchgate.net / publication / 315753315_Lesion_ detection_in_CT_images_using_Deep_Learning_semantic_segmentation_ tachnique. [7] [Online]. Available: https: / / core.ac.uk / download / pdf / 76948049.pdf, [8] [Online]. Available: https: / / ws680.nist.gov / publication / get_pdf.cfm?pub_id=924455. [9] [Online].Available: https: / / arxiv.org / ftp / arxiv / papers / 1901 / 1901.01211.pdf.

[10] S. Rüger, "Comparison of different methods for the detection of bone in tissues using artificial neural networks based on dual-energy X-ray images." Master's thesis, University of Erlangen-Nuremberg, 2018.

[11] GE Company, "Diagnostic System with learning capabilities." Patent US6442542, 2002.

[12] Siemens AG, 2007. Method and device for user-specific parameterization of an x-ray device. Patent US 7298823.

Claims

1. A device (100) for determining at least one capturing parameter (150) for capturing an X-ray projection (192) of a workpiece (110) by means of a computed tomography system (190), the device (100) comprising: a measuring device (120) configured to provide measurement data (122) about the workpiece (110), wherein the measurement data (122) of the measuring device (120) include an average physical density of the workpiece (110) and / or a mass of the workpiece (110); and a data analyzer (130) configured to determine, using the measurement data (122) of the measuring device (120) and using a prediction model (140), the at least one capturing parameter (150) for the computed tomography system (190), wherein the prediction model (140) comprises a self-learning algorithmic method and is based on a plurality of data sets, wherein each of the data sets comprises at least one of the following pieces of information: a test task, an object shape of a workpiece, a density distribution of a workpiece, a physical density of a workpiece, a mass of the workpiece, an X-ray projection of a workpiece, a CT volume data set of a workpiece, a capturing parameter of an X-ray projection and / or of a CT volume data set, a status information relating to the computed tomography system, one or more measurement data about a workpiece, one or more X-ray data about a workpiece, a piece of information about an image quality of X-ray data of a workpiece.

2. The device (100) according to claim 1, wherein the data analyzer (130) is configured to further use an X-ray projection (160) of the workpiece (110) and / or a definition of a test task (135) for determining the at least one capturing parameter (150) for the computed tomography system (190).

3. The device (100) according to one of the preceding claims, wherein the device (100) is operable independently of the computed tomography system (190).

4. The device (100) according to one of the preceding claims, wherein the measurement data (122) of the measuring device (120) include an object shape of the workpiece (110).

5. The device (100) according to one of the preceding claims, wherein the device (100) is configured to determine a physical density of the workpiece (110) and, based thereon, to determine the at least one capturing parameter (150) for the computed tomography system and / or a capturing parameter for an X-ray projection of the workpiece (110) and / or a parameter relating to an X-ray spectrum, and / or wherein the device (100) is configured to determine, based on the physical density and an X-ray projection of the workpiece (110), a density distribution of the workpiece (110) and to determine, based on the density distribution of the workpiece (110), the at least one capturing parameter (150) for the computed tomography system.

6. The device (100) according to one of the preceding claims, wherein the measuring device (120) comprises a sensor (124) for determining a mass in order to determine the mass of the workpiece (110); and wherein the measuring device (120) further comprises a stereoscopic sensor (126) in order to determine the object shape of the workpiece (110); wherein the measuring device (120) is configured to determine, based on the mass and the object shape of the workpiece (120), the physical density of the workpiece (110).

7. The device (100) according to one of the preceding claims, wherein the device (100) is configured to determine the object shape of the workpiece (110) and to provide, using the object shape of the workpiece (110), a reproduction scale for the computed tomography system (190) and / or for an X-ray projection, and / or wherein the device (100) is configured to obtain, using the object shape of the workpiece (110) and / or the definition of a test task (135), a capturing parameter (150) relating to an axis system of the computed tomography system (190).

8. The device (100) according to one of the preceding claims, wherein the measuring device (120) comprises an X-ray fluorescence sensor; and wherein the device (100) is configured to determine, using data of the X-ray fluorescence sensor, the at least one capturing parameter (150) for the computed tomography system (190).

9. The device (100) according to one of the preceding claims, wherein the at least one capturing parameter (150) determined by the data analyzer (130) comprises at least one of the following parameters: an X-ray spectrum, an X-ray voltage, a physical pre-filtering, a current-time product, a reproduction scale, a projection number, a parameter relating to an axis system, a measurement mode, a focal spot size, an intensity, a spatial resolution of the overall imaging.

10. The device (100) according to one of the preceding claims, wherein the data analyzer (130) is configured to provide, using an X-ray projection (160) of the workpiece (110) created by means of the computed tomography system (190) and using the prediction model (140), a maintenance recommendation (180) for the computed tomography system (190), wherein the prediction model (140) comprises a self-learning algorithmic method and wherein the prediction model (140) is based on information about a plurality of X-ray projections of the workpiece (110) and / or of further workpieces, wherein the X-ray projections are created beforehand by means of the computed tomography system (190) and / or other computed tomography systems.

11. The device (100) according to claim 10, wherein the data analyzer (130) is configured to determine, based on the X-ray projection (160) of the workpiece (110) created by means of the computed tomography system (190), a piece of information about an image quality of the X-ray projection (160) created by means of the computed tomography system (190), and wherein the data analyzer (130) is configured to determine, based on the piece of information about the image quality of the X-ray projection (160) created by means of the computed tomography system (190), a piece of information about a current maintenance state of the computed tomography system (190), and wherein the device (100) is configured to provide, based on the piece of information about the current maintenance state of the computed tomography system (190), the maintenance recommendation (180) for the computed tomography system (190).

12. A computed tomography system (490) comprising a device (100) for determining at least one capturing parameter (450) for capturing X-ray data (492) of a workpiece (410) by means of the computed tomography system (490) according to any one of claims 1-11.

13. A method (500) for determining at least one capturing parameter (150, 450) for capturing an X-ray projection (192, 492) of a workpiece (110, 410) by means of a computed tomography system (190, 490), the method (500) comprising: determining (501) measurement data (122) about the workpiece (110, 410), wherein the measurement data (122) include information about an average physical density of the workpiece (110, 410) and / or include a mass of the workpiece (110); and determining (502) the at least one capturing parameter (150, 450) for the computed tomography system (190, 490) based on the measurement data (122) and using a prediction model (140), wherein the prediction model (140) comprises a self-learning algorithmic method and is based on a plurality of data sets, wherein each of the data sets comprises at least one of the following pieces of information: a test task, an object shape of a workpiece, a density distribution of a workpiece, a physical density of a workpiece, a mass of the workpiece, an X-ray projection of a workpiece, a CT volume data set of a workpiece, a capturing parameter of an X-ray projection and / or of a CT volume data set, a status information relating to the computed tomography system, one or more measurement data about a workpiece, one or more X-ray data about a workpiece, a piece of information about an image quality of X-ray data of a workpiece.

14. A computer program with a program code for carrying out the method (500) according to claim 13, when the program runs on a computer.