Method for operating a medical imaging device

Abstract

The invention relates to a method for operating a medical imaging device for generating an image data set of a measurement region of a patient which is influenced by respiratory motion, the measurement region comprising at least one z-position, the image data set comprising at least one slice image data set of the at least one z-position, the method comprising: providing a breathing curve of the patient by means of a first interface, which describes the respiratory motion of the patient over at least one breathing cycle; automatically deriving at least one measurement parameter by means of a calculation unit on the basis of the provided breathing curve, the derived measurement parameter determining a temporal resolution of the slice image data set; operating the medical imaging device to generate the image data set by means of a control unit on the basis of the automatically derived measurement parameter, wherein for the at least one z-position a plurality of projection data sets from different projection angles in a relative rotational motion between a radiation source of the medical device and the patient are detected and on the basis thereof the slice image data set of the image data set is generated.

Description

TECHNICAL FIELD

[0001] The invention relates to a method for operating a medical imaging device for generating an image data set of a patient measurement region, a device for operating a medical imaging device, a medical imaging device and a computer program product. The invention further relates to a training method for providing a trained function for use in a method for operating a medical imaging device, a training device and a computer program product. BACKGROUND

[0002] For example, for radiation planning of a patient with lung cancer or abdominal cancer, a three-dimensional, 3D, image data set of a measurement region with an extension along the z-direction is typically used. By means of a medical imaging device, in particular a computed tomography scanner, in particular a projection data set can be detected at an imaging examination, from which a 3D image data set can be reconstructed. Imaging examinations by means of a computed tomography scanner typically require ionizing X-rays. Scanning of an examination object by means of a medical imaging device, in particular for example by means of a computed tomography system (CT system), is generally known. Here, for example, a circular scan, a sequential circular scan with sequential feed or a helical scan with continuous table feed is used. Other types of scans that are not based on circular movements are also possible. By means of at least one radiation source and at least one opposing detector, absorption data of the patient from different recording angles, hereinafter also referred to as projection angles, are recorded, and the absorption data or projection data set thus collected is converted into a 3D image data set or slice image data set by means of a corresponding reconstruction method. The 3D image data set typically comprises a plurality of slice images, i.e. essentially two-dimensional image data sets, of the patient at corresponding z positions in the measurement region. The slice images typically depict axial slices of the patient in the measurement region along the z-direction, i.e. along the z-axis. Today, for reconstructing an image from the projection data set that has been detected by means of such a system, the so-called Filtered Back Projection method (FBP) is used as a standard method.

[0003] In order that the patient's anatomic structures relevant for radiation planning, in particular the patient's lung cancer or abdominal cancer, can be reconstructed in a specific breathing phase of the patient, the breathing motion of the patient can be detected during the imaging examination. Preferably, for each z position of the measurement region, a slice image over all breathing phases of a breathing cycle can be reconstructed, the breathing cycle in particular depicting the patient's inhalation and exhalation and corresponding to a segment of the periodicity of the breathing motion. Thereby, a time-resolved image data set or a breathing-related image sequence can be generated, which shows the measurement region in terms of the breathing motion at different points in time, i.e. essentially during the patient's breathing cycle, in a time-resolved manner. Thereby, it can be ensured, in particular, that the dose allocation is adapted to the planning-time ventilation in motion.

[0004] In order to depict the motion as precisely as possible, ideally, a time resolution of the image data set is desired which is as high as possible. Thus, generally, the highest possible time resolution is chosen for recording the data and for generating the image data set.

[0005] On the other hand, however, while a high image quality is required, for example, in terms of image noise, choosing a time resolution which is as high as possible leads to a higher dose to be applied to the patient or results in a higher requirement for the X-ray source used and its power reserve or leads to a limitation in the maximum possible extension of the measurement region or the duration of the measurement data dose.

[0006] At the same time, the respiratory motion of the patient is highly patient-specific. As a result, the same setting parameters for the data recording by means of the medical imaging device are not optimal for every patient, in particular not optimal in view of the relative boundary conditions for the generated image data set or image recording. SUMMARY

[0007] The object on which the present invention is based is to achieve a medical imaging device for generating an image data set of a measurement region of a patient which is influenced by respiratory motion, which is operated in an individually coordinated manner to the patient.

[0008] This object is achieved by the features of the present invention. Further advantageous and in themselves inventive embodiments and improvements of the present invention are set forth in the following description.

[0009] The present invention relates to a method for operating a medical imaging device which generates an image data set of a measurement region of a patient which is influenced by respiratory motion, the measurement region comprising at least one z-position, wherein the image data set comprises at least one slice image data set of the at least one z-position. The method comprises at least the steps of providing, automatically deriving and operating.

[0010] In the providing step, a breathing curve of the patient is provided, the breathing curve describing the respiratory motion of the patient over at least one breathing cycle.

[0011] In the automatically deriving step, at least one measurement parameter is automatically derived by means of a computing unit on the basis of the provided breathing curve, wherein the derived measurement parameter determines a time resolution of the at least one slice image data set.

[0012] In a manipulating step, the medical imaging device is manipulated to generate an image data set by means of a control unit based on the automatically derived measurement parameters, wherein for at least one z-position a plurality of projection data sets from different projection angles in a relative rotational motion between a radiation source of the medical device and the patient are detected and at least one slice image data set of the image data set is generated based on the plurality of projection data sets.

[0013] The medical imaging device is designed for a relative rotational motion between a radiation source, in particular an X-ray source, comprised by the medical imaging device and a patient. The medical imaging device comprises in particular a detector for detecting radiation, in particular X-ray radiation, emitted by the radiation source, wherein a measurement region of the patient is positioned between the radiation source and the X-ray detector. During the rotational motion, projection data sets are detected by means of the detector arranged opposite to the radiation source from different projection angles. Based on the projection data sets, finally an image data set can be reconstructed, for example by means of a filtered back-projection method or other suitable reconstruction algorithms. For example, the medical imaging device can be designed in particular as a computed tomography device.

[0014] The measurement region can extend in particular along a z-axis or comprise at least one z-position along a z-axis. The z-axis can extend in particular along a longitudinal axis extending in a longitudinal direction of the patient. The longitudinal axis of the patient can also be parallel to a longitudinal axis extending in a longitudinal direction of a patient bed. The measurement region can have at least one z-position, wherein the at least one z-position describes a position of a slice of the patient along the z-axis to be imaged. The slice image data set of the at least one z-position can depict in particular a slice of the patient at the at least one z-position. The rotational axis of the computed tomography device usually corresponds to the longitudinal axis of the patient bed and the z-axis.

[0015] The slice image data set, i.e. the image data of the z-position, can have in particular a slice thickness of the extension in the z-direction. For example, the minimum slice thickness can be determined by the pixel size of the detector used.

[0016] A plurality of projection data sets from a preferably contiguous angular range of the rotational motion of the radiation source can be associated with the slice image data set of the z-position, based on which the slice image data set can be reconstructed. The angular range can be referred to as a projection angle interval. Ideally, the projection angle interval comprises at least 180°, which is usually a minimum for reconstructing the slice image data set. The larger the projection angle interval, the more projection data sets with at least one angle within the projection angle interval are used in reconstructing the slice image data set. If the projection angle interval is at least 180°, image artifacts in the slice image can usually be minimized, which are caused in particular by a projection angle interval of less than 180°.

[0017] The projection data sets can be recorded by means of a helical acquisition by means of a computed tomography device. Here, the patient is continuously pushed through the maximum field of view of the computed tomography device. Instead of a helical acquisition, the patient bed can be pushed through the maximum field of view of the computed tomography apparatus in discrete steps. In the case of a helical acquisition, the so-called pitch, which describes the table feed, is proportional to the table feed of the patient bed and the extension of the X-ray detector in the z direction of the computed tomography apparatus. Typical values for the pitch in computed tomography apparatuses are above 0 and below 2. In order to record moving structures, i.e. a measurement region which is influenced by respiratory motion, for example, it is preferable to work with a small pitch, for example a pitch of less than 1 and / or in the case of a small rotation period time of the radiation source. The rotation period time describes the time slice required for a complete revolution of the radiation source around the patient. The shorter the rotation period time of the radiation source during the rotational motion, the higher the radiation source-detector speed, i.e. the radiation source and the detector rotate particularly quickly around the patient and / or more projection data sets can be detected per unit of time. In particular, the at least one radiation source and the at least one X-ray detector are usually continuously rotated during the imaging examination.

[0018] During the recording of the projection data sets, the angle of the at least one radiation source and the at least one X-ray detector relative to the patient bed or the patient, and the current z position of the patient bed, are usually stored, for example with reference to the projection data, during the acquisition of the projection data sets.

[0019] The temporal resolution of the slice image data set is understood in the context of the present application in such a way that the temporal resolution reflects the time slice which elapses during the detection of the projection data sets for the slice image data set. The shorter the time slice, i.e. the better the temporal resolution, the less the patient moves during the detection and the less the slice image data set is influenced by motion in the measurement region, i.e. by motion artifacts, and the more clearly structures of motion in the slice image data set are depicted. Furthermore, if a slice image data set comprising at least one z position, i.e. a time-resolved image data set, is detected for a sequence of time points, i.e. for a plurality of time points, the temporal resolution determines the minimum time spacing between two slice image data sets which temporally follow one another. The temporal resolution defines, in particular, the maximum temporal coherence of structures of the patient which move by respiratory motion, for example the skin surface, organs, cancer and / or tumors. If the temporal resolution comprises N seconds, it is preferable for the projection data sets at the at least one z position to be detected in accordance with the projection angle interval within N seconds. The maximum temporal coherence can be related to other types of patient motion. The maximum temporal coherence in the imaging examination can be optimized, in particular, when there is no motion of structures which move, in particular by respiratory motion. In other words, the less the patient moves, the better the maximum temporal coherence.

[0020] The temporal resolution can be directly proportional to the product of the rotation period time and the projection angle interval, among others.

[0021] Generally, the largest temporal resolution is chosen for the time-resolved measurement, for example by choosing the smallest possible rotation period time and the largest possible projection angle interval based on the slice image data set, in order to depict the motion of the tumor and the organs as precisely as possible.

[0022] However, the inventors have recognized that the largest temporal resolution is not always necessary for the best display. The inventors have recognized that the respiratory motion of the patient, i.e. the degree of motion per unit of time, can be patient-specific to the highest degree and, if necessary, also influenced by the current boundary conditions, so that it is not always targeted to always choose the highest possible selection temporal resolution, but instead it is advantageous to select a temporal resolution that is suitable for the current patient instead. In this way, it can be advantageously avoided that an unnecessarily high dose is administered or that limitations in the course of the scan are imposed.

[0023] According to the application, at least one measurement parameter is automatically derived on the basis of a provided respiratory curve of the patient, wherein the derived measurement parameter determines the temporal resolution of at least one slice image data set. That is to say, the image data set is preferably produced using a patient-individual temporal resolution based on the respiratory curve by means of the measurement parameter. The measurement parameter is preferably derived on the basis of a currently existing, i.e. temporally adjacent to the imaging examination, respiratory curve.

[0024] The patient respiratory curve, which reflects the respiratory motion of the patient, can be recorded, for example, by means of a sensor, for example by means of a camera or a respiratory belt. The respiratory motion generally describes the free breathing of the patient and comprises, among others, at least one breathing cycle of the patient. The respiratory motion is detected, among others, over at least one complete breathing cycle of the patient, wherein the entire breathing cycle comprises at least one inhalation and one exhalation of the patient. The provided respiratory curve preferably comprises more than one breathing cycle. The breathing cycle preferably corresponds to a periodic segment of the respiratory motion. The duration of the periodic segment corresponds to the duration of the breathing cycle. The duration of the breathing cycle or the average duration of a plurality of successive breathing cycles can be determined from the respiratory curve and, in turn, the respiratory rate, i.e. the number of inhalations or exhalations per unit of time, can be determined. The slower the patient breathes, the higher the respiratory rate.

[0025] The course of the patient respiratory curve can be patient-specific, i.e. the occurrence of a slope, the duration of an inhalation or exhalation or the occurrence of temporary rest phases with little or no motion can differ from patient to patient and, if necessary, also from imaging examination to imaging examination of the patient. This can be derived, among others, independently of the (average) respiratory rate or at least not in a simple manner from the (average) respiratory rate of the patient.

[0026] Providing the breathing curve by means of the first interface can comprise reading data from a storage unit, the patient breathing curve detected by means of the sensor being stored on the storage unit. Providing can comprise receiving measurement points of the breathing curve from a sensor unit constituting the sensor unit for detecting the patient breathing curve.

[0027] The automatic derivation of the at least one measurement parameter requires in particular no user interaction. Thereby, an erroneous determination of the at least one measurement parameter can be avoided. The automatic derivation is in particular carried out in the computing unit. To this end, the computing unit advantageously has a memory into which a program code section can be loaded. The program code section has in particular a program code which, when the breathing curve of the patient is provided, implements the automatic derivation of the at least one measurement parameter.

[0028] The automatic derivation is in particular advantageous, because the at least one measurement parameter is automatically calculated for the provided breathing curve, in turn for each patient individually, in turn the production of the image data set is in particular optimally adapted to the patient, because the at least one measurement parameter is not selected from a pre-set parameter set averaged over a patient collective or selected according to a simplified relationship which is at most on average applicable, if necessary.

[0029] The automatic derivation can comprise deriving more than only one measurement parameter.

[0030] The automatic derivation can comprise using a trained function, wherein the breathing curve of the patient or a parameter derived therefrom is included as an input parameter into the trained function. The automatic derivation can comprise an analysis of the breathing curve, the analysis analyzing the breathing curve in its course in order to extract at least one value, preferably a plurality of values, relating to the duration of the inhalation phase or the exhalation phase or the duration of a resting phase of the breathing movement based on the provided breathing curve. The values can be used as input parameters for the trained function or can also be used to query a plurality of measurement parameters stored in a database, which are associated with the input parameters.

[0031] Avoiding unnecessarily high administered dose or unnecessary limitations in terms of scan range or scan duration can be achieved by individually determining the measurement parameters with respect to the temporal resolution for the patient. Higher temporal resolution usually also means higher dose in terms of image quality of the patient remaining otherwise unchanged. Since less time for each projection data set acquisition and / or fewer projection data sets contribute to the generation of the slice image data set, this has to be compensated by higher administered dose. Higher temporal resolution usually also means higher requirements on the X-ray tube and its power reserve. This leads to limitations in terms of scan range or scan duration, for example for obese patients for whom a higher tube current has to be set if necessary in order to achieve the required image quality, resulting in limitations. If at the same time, for example, the smallest rotation period time or the fewest number of projection image data sets contributing to a slice image data set at a time point are selected, in which the image quality should be ensured to remain unchanged at the same time, the radiation source used reaches its power limits.

[0032] For example, for a patient with slow breathing, the temporal resolution can be chosen to be greater, since there is slower movement in the measurement region here, with the patient having long inhalation or exhalation phases with moderate slopes. In patients with very steep slopes in inhalation or exhalation, in contrast, a higher temporal resolution has to be chosen in order to generate a slice image data set with sufficient temporal coherence. Here, there is an important correlation between the breathing and the measurement parameters to be chosen. For example, mere contact with the average breathing rate provides insufficient results here. In contrast, automatic derivation taking into account the patient's breathing curve can ensure suitable selection, in particular also for less experienced personnel, and overall contribute to the time-efficient generation of image data sets, in which errors and thus insufficient image quality or unnecessarily administered dose can be avoided.

[0033] The medical imaging device is operated on the basis of the automatically derived measurement parameters to generate an image data set comprising at least one slice image data set. Here, for at least one z position, a plurality of projection data sets from different projection angles are detected in the relative rotational movement between the radiation source of the medical imaging device and the patient, and on the basis of this, a slice image data set is generated.

[0034] The operation can involve acquiring the projection data sets by means of the medical imaging device itself, or can also involve reconstructing the image data set on the basis of the detected projection data sets.

[0035] According to one method variant, the automatically derived measurement parameter comprises a rotation period time of the radiation source during the relative rotational motion and / or a projection angle interval, the projection angle interval comprising an angular range of projection angles, the at least one slice image data set being based on the projection angles. The automatic derivation can also comprise deriving the rotation period time and the projection angle interval. A fixed correlation between the rotation period time and the projection angle interval can also be determined, such that upon automatic derivation of one of the measurement parameters the other measurement parameter is also determined.

[0036] A time resolution coordinated to the patient and to his respiratory motion can advantageously be determined.

[0037] The manipulation can then comprise, inter alia, manipulating the radiation source-detector unit used for the acquisition, in particular manipulating the speed of the rotational motion, i.e. the rotation period time. The manipulation can then comprise selecting the detected projection data sets based on the projection angle interval, the general data comprising into the generation of the at least one slice image data set.

[0038] Based on the automatically derived one or more measurement parameters relating to the time resolution, further measurement parameters can also be derived. Correspondingly, the manipulation can also relate to further parameters. For example, further setting references can be related to or associated with the automatically derived measurement parameters relating to the time resolution. This can comprise, for example, the used tube current of the X-ray tube as the radiation source or the table feed, for example the pitch in a helical scan. A determined correlation, for example in the form of a mathematical function, between the measurement parameters can be preset. The automatic derivation can also comprise deriving a determined set of measurement parameters, which are correlated to each other, the set of measurement parameters comprising more measurement parameters than the at least one measurement parameter determining the time resolution.

[0039] According to one variant of the method, the automatic derivation comprises, inter alia, a dose parameter and / or a power parameter of the radiation source.

[0040] The power parameter of the radiation source can relate, for example, to the maximum tube power of the used X-ray tube. In the case of a very high time resolution being used, in particular in connection with a large volume to be transmitted, while a high image quality is desired, it can be necessary to set a higher tube current. If a plurality of slice image data sets should also be generated, for example when a measurement region is extended in the scan and / or during a long duration, it can likewise result in reaching the power limits of the used radiation source, which limits the maximum scannable measurement region or the maximum scan duration. In this case, a selected lower time resolution coordinated to the patient-specific respiratory curve can be advantageous. This is advantageously taken into account when the measurement parameter relating to the time resolution is automatically derived.

[0041] The dose parameters can influence the image quality, inter alia, for example in terms of image noise. At the same time, unnecessary radiation loading of the patient is to be avoided. An unnecessarily selected temporal resolution can lead to an unnecessarily increased dose applied. In particular, an automatic derivation can enable an improved selection of dose / image quality and temporal resolution. For a possibly less trained user, the important correlations between respiration, image quality, dose and tube power can not be known. An automatic derivation taking into account the current boundary conditions can advantageously contribute to avoiding faulty or suboptimal parameter selection.

[0042] According to one method variant, the automatic derivation comprises using a trained function, wherein at least one parameter of the trained function is adapted to a comparison between a training measurement parameter and a comparison measurement parameter derived on a training respiration curve of a training patient, wherein the training respiration curve and the comparison measurement parameter are associated with each other.

[0043] The derivation using the trained function can advantageously allow a derivation, in particular time-efficient, taking into account the patient-specific respiration curve.

[0044] The trained function can preferably be realized by means of an artificial intelligence system, i.e. by a method of machine learning. By carrying out the derivation based on using the trained function, all relevant influencing variables for the derivation can be considered improved, i.e. such cases for which the correlations for the derivation are not estimable or only difficult to estimate for a user, in particular for example also less trained or less experienced, are considered. An artificial intelligence system can represent a system for artificially generating knowledge from experience. The artificial system learns from examples in a training phase and can generalize after the end of the training phase. The use of such a system can include recognizing patterns and regularities in the training data. After the training phase, the artificial intelligence system can extract features or feature variables, for example in hitherto unknown measurement data, which include into the derivation. After the training phase, the optimized, i.e. trained, algorithm can estimate, for example based on a hitherto unknown measured respiration curve, measurement parameters suitable for producing an image data set. The artificial intelligence system can be an artificial neural network or can also be based on another machine learning method. In particular, after the training phase, the derivation of the measurement parameters can be realized after the training phase, in particular reliably and time-efficiently, automatically by means of the trained function based on the artificial intelligence system.

[0045] In particular, the trained function maps input data onto output data. Here, the output data can also be related to one or more parameters of the trained function, among other things. The one or more parameters of the trained function can be determined and / or adapted by training. The determination and / or adaptation of the one or more parameters of the trained function can be based, among other things, on a pair consisting of training input data and the associated, i.e. linked, comparison output data, wherein the trained function is applied to the training input data in order to generate the training output data. In particular, the determination and / or adaptation can be based on a comparison of the training output data and the training comparison data. In general, a function that can be trained, i.e. a function with one or more parameters that have not yet been adapted, is also referred to as a trained function.

[0046] Other terms for the trained function are trained mapping rule, mapping rule with trained parameters, function with trained parameters, artificial intelligence-based algorithm, machine learning algorithm. 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. Instead of the term "neural network", the term "neural net" can also be used. In particular, the trained function can also be a deep artificial neural network (English: deep neural network, deep artificial neural network). A further example of a trained function is a "support vector machine", among other things, other machine learning algorithms can also be used as trained functions.

[0047] The trained function can be trained, among other things, by means of backpropagation. First, the training output data can be determined by applying the trained function to the training input data. Thereafter, a deviation between the training output data and the training comparison data can be calculated by applying an error function to the training output data and the training comparison data. Furthermore, at least one parameter, in particular a weight, of the trained function, in particular of a neural network, can be adapted iteratively on the basis of a gradient of the error function with respect to the at least one parameter. Thereby, the deviation between the training output data and the training comparison data can advantageously be minimized during the training of the trained function.

[0048] The trained function, in particular the neural network, advantageously has an input layer and an output layer. Here, the input layer can constitute a means for receiving input data. Furthermore, the output layer can constitute a means for providing output data. Here, the input layer and / or the output layer can each comprise a plurality of channels, in particular neurons.

[0049] According to the application, the input data for the trained function can comprise a provided first breathing curve of a patient or parameters derived therefrom, preferably a plurality of parameters derived therefrom, which characterize the course of the breathing curve. According to the application, the output data can in particular comprise at least one measurement parameter for generating the at least one slice image data set, which relates to the temporal resolution of the slice image data set.

[0050] In the training phase of the trained function, according to the application, in particular a training breathing curve, preferably a plurality of training breathing curves, of a training patient can be used as training input data. On the basis thereof, training measurement parameters can be derived as training output data. Furthermore, at least one parameter of the trained function can be adapted on the basis of a comparison of the derived training measurement parameters of the training patient with comparison measurement parameters of the training patient as training comparison data.

[0051] The training input data and the training comparison data are related to one another. In particular, annotated training input data can be used, which have been annotated, for example by expert knowledge, before the start of the training phase, so that for the training input data there is advantageous training comparison data. For example, before the training phase, at least one comparison measurement parameter can be associated with a training breathing curve, respectively. The comparison measurement parameter can determine a suitable temporal resolution. The comparison measurement parameter can comprise, for example, a rotation period time or a suitable projection angle interval, which is used in a manner suitable for the respective training breathing curve. Furthermore, a dose parameter or a tube power parameter can also be taken into account here. Furthermore, an image quality parameter can be taken into account. In particular, the associated expert knowledge can be based on experimental experience and experimental research, which is obtained from actual measurements or simulations before the start of the training phase. For example, the input data can be based on real measurements on real training patients. The input data can be based on simulations or phantom measurements, which can then be annotated.

[0052] The trained function can be applied directly to the provided breathing curve. The breathing curve can correspond to the input parameter of the trained function. However, the input parameter of the trained function can also be based on values previously extracted from the breathing curve.

[0053] According to one method variant, the automatic derivation of the at least one measurement parameter by means of the computing unit comprises: extraction of at least one value relating to the duration of an inhalation phase or an exhalation phase or the duration of a resting phase of the breathing movement on the basis of the provided breathing curve and derivation of the measurement parameter on the basis thereof. For example, the slope in the breathing curve or the course of the slope in the breathing curve can be determined on the basis of a mathematical derivation.

[0054] Instead of using a trained function, the extracted values can also be used in combination with a database query. Determining the at least one measurement parameter based on the at least one extracted value can comprise querying a measurement parameter database stored in a storage unit, wherein the at least one extracted value or a parameter derived therefrom is used as a query parameter for the query.

[0055] The association of the one or preferably multiple extracted values with the at least one measurement parameter can be based on expert knowledge and experimental experience. This is to say that the measurement parameters stored in the measurement parameter database are associated with one or a set of extracted values for the query, respectively. It is also feasible to first classify, based on the preferably multiple extracted values, into one of a plurality of breathing pattern groups. The at least one measurement parameter can be associated with the breathing pattern group and can for example be present in a measurement parameter database associated with the breathing pattern group.

[0056] This embodiment can be implemented without a training phase of a trained function. However, this corresponds to a less flexible and less patient-specific implementation compared to the use of a trained function due to the classification or determination of only a predetermined breathing pattern with as much consistency as possible or a fixed association of the parameters by means of a database.

[0057] According to a variant of the method, the generated image data set comprises slice image data sets at a plurality of time points of at least one z position. Subsequently, in the step of operating the medical imaging device to generate an image data set based on the automatically derived measurement parameter of at least one z position, a plurality of projection data sets out of different projection angles are detected for each of the plurality of time points, respectively. Advantageously, the measurement region can be described in a time-resolved manner at least at the at least one z position, and the respiratory motion of the measurement region or the resulting motion of organs or structures in the measurement region can be reproduced.

[0058] According to a variant of the method, the slice image data sets of the generated image data set depict the respiratory cycle of the patient over the entire duration of the respiratory cycle. The motion can advantageously be reproduced over the entire respiratory cycle and taken into account for example in radiotherapy planning.

[0059] According to a variant of the method, during the operation of the medical imaging device, the respiratory motion of the patient is additionally detected in real time, and an image data is generated for one of the plurality of time points depending on the respiratory phase of the respiratory motion of the patient.

[0060] In particular, the respiratory phase can be associated with at least one of the time-resolved slice image data sets. The at least one of the time-resolved slice image data sets shows in particular the z position according to the respiratory phase. For example, the user can select the respiratory phase to be displayed on a monitor, and the slice image data set belonging to this respiratory phase can be displayed.

[0061] If the measurement region has an extended 3D volume in the z-direction and the measurement region comprises, for example, the chest of a patient, the chest of the patient can be displayed on the monitor in particular in accordance with a selected breathing phase of the patient. For example, in accordance with a selection of the user, the patient can be displayed once at inhalation and once at exhalation.

[0062] The generation of the image data set of the patient from at least one measurement parameter in the measurement region by means of the medical imaging device can comprise in particular the reconstruction of a breathing-related image or a sequence of breathing-related images.

[0063] The application also relates to a training method for providing a trained function for use in the method described previously. The training method comprises the following steps: providing a training breathing profile of a training patient and a comparison measurement parameter associated therewith by means of a training interface, wherein the comparison measurement parameter determines, for at least one z-position comprised by the measurement region, a temporal resolution of a training slice image data set comprised by the training image data set of the measurement region of the training patient affected by respiratory motion when the training image data set is generated by means of a training imaging device with the comparison measurement parameter from a plurality of training projection data sets from different projection angles in a relative rotational motion between a radiation source of the training imaging device and the training patient.

[0064] Furthermore, the training method comprises the following steps: applying the trained function to the provided first training breathing profile, in turn deriving a training measurement parameter by means of a training calculation unit, and the following step: adapting at least one parameter of the trained function by means of the training calculation unit on the basis of a comparison of the derived training measurement parameter and the corresponding comparison measurement parameter.

[0065] Furthermore, the training method comprises the following step: providing the trained function by means of a second training interface.

[0066] The trained function can advantageously be provided.

[0067] The provision of the training breathing profile and the training measurement parameter associated therewith can comprise in particular the detection and / or reading out of a computer-readable data storage and / or the reception from a data storage unit, for example a database. The first training breathing profile and the comparison measurement parameter can be based on real measurements of a real training patient and their associated measurement parameters. The training breathing profile can be recorded and saved by means of a sensor arranged at the training patient or by means of a camera and then provided for training. The at least one comparison measurement parameter can be stored in association with the training measurement parameter and then provided for training. In general, artificially generated, i.e. simulated, training data sets or training data sets based on phantom measurements can also be used.

[0068] The training respiratory profile and the comparative measurement parameters associated therewith are preferably based on domain-specific, measured respiratory profiles of a real patient population of the medical imaging device or of the same device group of the medical device. This means that the respiratory profiles used for the training method and on which the automatic derivation is based are preferably already determined under similar conditions or have been determined under similar conditions.

[0069] The present application also relates to a device for operating a medical imaging device for generating an image data set of a measurement region of a patient which is influenced by respiratory motion, the measurement region comprising at least one z-position, wherein the image data set comprises at least one slice image data set of the at least one z-position.

[0070] The device comprises a first interface designed to provide a respiratory profile of a patient, the respiratory profile describing a respiratory motion of the patient over at least one breathing cycle.

[0071] The device further comprises a computing unit designed to automatically derive at least one measurement parameter based on the provided respiratory profile, wherein the derived measurement parameter determines a temporal resolution of the at least one slice image data set.

[0072] The device further comprises a control unit designed to operate the medical imaging device to generate the image data set based on the automatically derived measurement parameter, wherein for the at least one z-position a plurality of projection data sets from different projection angles in a relative rotational motion between a radiation source of the medical device and the patient are detected and based on the plurality of projection data sets the at least one slice image data set of the image data set is generated.

[0073] The computing unit or the control unit can be designed to derive a control signal for operating the medical device based on the derived measurement parameter, based on which the medical imaging device can be operated. In particular, the device can be designed to automatically operate the medical imaging device based on the derived measurement parameter.

[0074] Such a device for operating a medical imaging device can in particular be designed to perform the previously described method for operating a medical imaging device according to the present application and its aspects. The device can be designed to perform the method and its aspects by the interface, the computing unit and the control unit in a way that the respective method steps are performed.

[0075] The advantages of the proposed device essentially correspond to the advantages of the proposed method for operating a medical device. The features, advantages or alternative embodiments mentioned here can likewise also be transferred to the device for operating and vice versa.

[0076] The present application also relates to a medical imaging device comprising a device for operating the medical imaging device.

[0077] The proposed medical imaging device is advantageously designed for carrying out the proposed embodiments of the method for operating a medical imaging device. The advantages of the proposed medical imaging device essentially correspond to the advantages of the proposed method for operating a medical device. The features, advantages or alternative embodiments mentioned here can likewise be transferred to the medical imaging device and vice versa.

[0078] The medical imaging device comprises in particular a radiation source and a detector opposite thereto. A patient can be positioned between the radiation source, for example an X-ray tube, and the X-ray detector. For this purpose a corresponding patient bed can be provided. The medical imaging device is designed for detecting a plurality of projection data sets from different projection angles in a relative rotational movement between the radiation source and the patient. The medical imaging device can preferably be designed as a computed tomography device. However, the medical imaging device can also be designed as a C-arm X-ray device and / or a Dyna-CT, etc.

[0079] The medical imaging device can advantageously be operated on the basis of the derived measurement parameters.

[0080] The present invention also relates to a training device for providing a trained function for use in the previously described method for operating a medical imaging device.

[0081] The training device comprises a first training interface designed for providing, by means of the training interface, a training breathing profile of a training patient and a comparison measurement parameter associated therewith. Here, when generating a training image data set by means of a training imaging device with the comparison measurement parameter, the comparison measurement parameter determines a temporal resolution of a training slice image data set of the training image data set of the measurement region affected by the breathing motion for at least one z-position comprised by the measurement region. Here, the training slice image data set is based on a plurality of training projection data sets from different projection angles in a relative rotational movement between a radiation source of the training imaging device and the training patient.

[0082] The training device further comprises a training calculation unit designed for applying the trained function to the first training breathing profile and thereby deriving a training measurement parameter.

[0083] The training calculation unit is further designed for adapting at least one parameter of the trained function on the basis of a comparison of the derived training measurement parameter and the corresponding comparison measurement parameter.

[0084] The training device further comprises a second training interface designed for providing the trained function.

[0085] Such a training device can in particular be designed to carry out the training method according to the invention for providing a trained function and its aspects as described before. The training device is designed to carry out the method and its aspects by constituting the respective method steps by means of the training interface and the training computing unit.

[0086] The invention can also relate to a computer program product having a computer program which can be directly loaded into the memory of a device for operating a medical device, the computer program having program segments in order to carry out all steps of the method for operating a medical imaging device as described before when the program segments are executed by the device.

[0087] The invention can also relate to a computer program product having a computer program which can be directly loaded into the training memory of a training device, the computer program having program segments in order to carry out all steps of the training method as described before when the program segments are executed by the training device.

[0088] The computer program product can be or comprise a computer program. Thereby, the method according to the invention can be carried out quickly, identically repeatable and robustly. The computer program product is configured in such a way that it can carry out the method steps according to the invention by means of the device or the training device. The device or the training device here must have the prerequisites, i.e. for example the respective working memory, the respective graphics card or the respective logic unit, so that the respective method steps can be carried out efficiently. The computer program product is for example stored on a computer-readable medium or on a network or a server, from where it can be loaded into the computing unit of the device or the training computing unit of the training device.

[0089] The invention can relate to a computer-readable storage medium on which program segments are stored which are readable and executable by the device in order to carry out all steps of the method for operating a medical imaging device or its aspects when the program segments are executed by the device.

[0090] An example of a computer-readable storage medium is a DVD, a magnetic tape, a hard disk or a USB stick on which electronically readable control information, in particular software, is stored.

[0091] The invention can relate to a computer-readable storage medium on which program segments are stored which are readable and executable by the training device in order to carry out all steps of the training method or one of its aspects when the program segments are executed by the training device.

[0092] The invention can also relate to a computer program or a computer-readable storage medium which comprises a trained function provided by means of the training method for providing a trained function or one of its aspects.

[0093] The implementation of the main software concept has the advantage that the processing unit and / or the training device used therewith can be retrofitted in a simple manner by means of a software upgrade in order to work in the manner according to the application. In addition to the computer program, the computer program product can also comprise additional components, i.e. for example documents and / or additional parts, and hardware components, i.e. for example hardware keys for using the software (dongles etc.), if necessary.

[0094] Furthermore, features described in relation to different embodiments of the application and / or different claim categories (method, use, device, system, apparatus etc.) can also be combined into other embodiments of the application within the scope of the application. For example, a claim relating to a device can also be improved by means of features depicted or claimed in the context of a method, and vice versa. Herein, functional features of a method can be implemented by means of correspondingly constituted entity components.

[0095] The use of the indefinite article "a" or "an" does not exclude multiple existence of the relevant feature. The use of the expression "having" does not exclude that a further feature can be added. The expression "unit" does not exclude that the respective objects to which the expression "unit" relates can comprise a plurality of components which are spatially separated from one another.

[0096] The expression "based on" can be understood in the context of the present application, in particular in the sense of the expression "using". In particular, the expression that a first feature is generated (alternatively: derived, determined etc.) on the basis of a second feature does not exclude that a third feature is used for generating (alternatively: deriving, determining etc.) the first feature. BRIEF DESCRIPTION OF DRAWINGS

[0097] In the following, the application is explained with reference to exemplary embodiments in accordance with the drawings. The representations in the drawings are schematic, strongly simplified and not necessarily to scale. Therein:

[0098] Figure 1 An exemplary medical imaging device is shown,

[0099] Figure 2 A schematic flow chart of a method for operating a medical imaging device for generating an image data set is shown,

[0100] Figure 3 and Figure 4 Exemplary breathing curves of a patient are shown,

[0101] Figure 5 A schematic flow chart of a training method for providing a trained function is shown,

[0102] Figure 6A schematic diagram of a device for controlling a medical imaging device and of a training device for providing a trained function is shown. DETAILED DESCRIPTION

[0103] Figure 1 A medical imaging device 32 in the form of a computed tomography device is shown.

[0104] The computed tomography device has a gantry 33 with a rotor 35. The rotor 35 comprises at least one radiation source 37, in particular an X-ray tube, and at least one detector 36 opposite thereto. The detector 36 and the radiation source 37 can be rotated around a common axis 43, also referred to as rotation axis 43. A patient 39 lies on a patient bed 41 and can be moved along the rotation axis 43 through the gantry 33. Typically, the patient 39 can comprise an animal patient and / or a human patient, for example.

[0105] The computed tomography device 32 comprises a processing unit 45 comprising a device for manipulating SY with a computing unit CU, an interface IF1 and a medical imaging device 32 with a control unit 51. The processing unit further comprises a storage unit MU.

[0106] Further, an input device 47 and an output device 49 are connected with the processing unit 45. The input device 47 and the output device 49 can enable an interaction by a user, for example a manual configuration, a confirmation or a triggering of a process step, for example. For example, a user can be shown computed tomography projection data sets and / or slice image data sets or three-dimensional image data sets on an output device 49 comprising a monitor.

[0107] Typically, a plurality of (raw) projection data sets of the patient 32 are recorded during a relative rotational movement between the radiation source and the patient from a plurality of projection angles during a continuous or sequential movement of the patient 39 through the gantry 33 by means of the patient bed 41.

[0108] Then, based on the projection data sets, a slice image data set of a respective z-position along the rotation axis within the measurement region can be reconstructed by means of a mathematical method, for example an iterative reconstruction method or a back projection method including filtering. The projection data sets out of a projection angle interval are associated with each slice image data set, the slice image data set being based on said projection angle interval.

[0109] The device for manipulating a medical imaging device comprised by the processing unit 45 is in particular designed for performing a method according to the present application for manipulating a medical imaging device 32 for generating an image data set of a measurement region of a patient 39 affected by respiratory motion, the measurement region comprising at least one z-position, wherein the image data set comprises a slice image data set of the at least one z-position.

[0110] Figure 2An exemplary flow chart of such a method for operating a medical imaging device is shown.

[0111] In a providing step ST1 a breathing curve S of the patient 39 is provided which describes the breathing motion of the patient 39 over at least one breathing cycle by means of a first interface IF1.

[0112] In an automatic deriving step ST2 at least one measurement parameter is derived based on the provided breathing curve S by means of a computing unit CU, wherein the derived measurement parameter determines a temporal resolution of at least one slice image data set.

[0113] In an operating step ST3 the medical imaging device 32 for generating image data sets is operated based on the automatically derived measurement parameter by means of a control unit 51, wherein for at least one z-position a plurality of projection data sets from different projection angles in a relative rotational motion between the radiation source 37 of the medical device 32 and the patient 39 are detected and based thereon at least one slice image data set of the image data sets is generated.

[0114] The method further comprises a displaying step ST4 of the at least one slice image data set. The method can further comprise detecting ST0 a breathing curve S of the patient and then providing said breathing curve for the automatic derivation by means of an interface.

[0115] Figure 3 and Figure 4 Two exemplary breathing curves S with respect to time t are shown. The shown breathing curves S(t) extend over a plurality of breathing cycles TC, in Figure 3 the case over three complete breathing cycles T C,1 , T C,2 and T C,3 . The breathing cycles have at least an inhalation phase T I,1 , T I,2 and T I,3 and an exhalation phase T E,1 , T E,2 and T E,3 , respectively. In Figure 4 the case, essentially two complete breathing cycles T C,1 , T C,2 are detected. Furthermore, Figure 4 the breathing curve S(t) of the patient 39 has a relatively long resting phase T R,1 , T R,2 between the respective exhalation and inhalation, wherein the breathing motion is hardly visible.

[0116] In case the time periods over which the shown curves S(t) extend are similar or identical, a Figure 2 significantly higher breathing rate can be associated with Figure 3exemplary respiratory curve S(t) is associated. At the same time, due to the fact that the cycle does not have a long stationary phase, the average slope is moderate despite this during inhalation or exhalation.

[0117] In contrast, Figure 4 The respiratory curve S(t) shows a clear stationary phase at a relatively low respiratory rate, whereas the average slope during inhalation and exhalation is very steep relative to the very low respiratory rate.

[0118] In the case shown, a high temporal resolution of the slice image data set is meaningful in order to produce a high-quality image data set which, in particular, shows low motion artifacts, despite the very low respiratory frequency. In contrast, in the case of a Figure 3 low respiratory frequency, a lower temporal resolution is sufficient in comparison.

[0119] Taking into account the patient-specific respiratory pattern in order to determine the temporal resolution achieves an optimum adaptation to the currently existing boundary conditions due to the patient. In particular, the automatic derivation in a time-efficient manner and without the risk of faulty settings also achieves, inter alia, taking into account unusual correlations.

[0120] The temporal resolution can be associated, inter alia, with the rotation period time and the projection angle interval used. The temporal resolution can be directly proportional, inter alia, to the product of the rotation period time and the projection angle interval. According to one method variant, the automatically derived measurement parameters include at least one rotation period time of the radiation source 37 during the relative rotational movement and / or a projection angle interval which includes an angular range of the projection angle, wherein at least one slice image data set is based on the projection angle range. The automatic derivation can also include deriving the rotation period time and the projection angle interval. A fixed correlation between the rotation period time and the projection angle interval can also be determined, such that upon automatic derivation of one of the measurement parameters, the other measurement parameter is also determined. Other parameters can also be included in this determined correlation.

[0121] Further measurement parameters can also be derived or calculated based on the automatically derived one or more measurement parameters related to the temporal resolution. Correspondingly, the manipulation can also relate to further parameters. For example, further setting parameters can be related or at least associated with the automatically derived measurement parameters related to the temporal resolution. This can for example include the used tube current of an X-ray tube used as radiation source or the pitch in a helical scan, for example in a spiral scan. Thus, for example in case of a higher rotation period time a higher pitch can be selected if necessary. For example, if at least one z-position shall be described over the entire breathing cycle, the solution can relate to the ratio of the rotation period time and the duration of the breathing cycle. If the image quality defined for example via the slice effective tube power time product for each reconstructed slice image data set shall remain constant, the tube current can be reduced in case of a reduced rotation period time. The slice effective tube current time product in particular states the degree of emitted X-ray quantity and is typically proportional to the radiation load. The higher the slice effective tube current time product, typically the higher the radiation load. The slice effective tube current time product can in particular correspond to the product of the tube current of the computed tomography apparatus at the time of the imaging examination and the tube rotation period time used at the time of the imaging examination. Likewise, for obtaining a constant image quality a lower tube current is feasible in case of a larger projection angle interval.

[0122] The automatic derivation preferably comprises using a trained function, wherein at least one parameter of the trained function is adapted based on a comparison between a training measurement parameter derived on a training breathing curve of a training patient and a comparison measurement parameter, wherein the training parameter and the comparison measurement parameter are associated with each other. The trained function preferably comprises a neural network.

[0123] The method can further comprise deriving, by means of the computing unit CU, the at least one measurement parameter ST2 comprises extracting at least one value related to the duration of an inhalation phase T I,1 , T I,2 , T I,3 or an exhalation phase T E,1 , T E,2 , T E,3 or the slope of the inhalation phase or the exhalation phase or the duration of a resting phase T R,1 , T R,2 from the provided breathing curve S and deriving the measurement parameter based on said value.

[0124] Determining the measurement parameter based on the extracted value can comprise querying a measurement parameter database stored in the storage unit MU, wherein the at least one extracted value or a parameter derived therefrom is used as a query parameter for the query.

[0125] Furthermore, a dose parameter or a power parameter of the radiation source 37 is preferably included into the automatic derivation step ST2. In particular, the dose parameter or the power parameter can be considered in the automatic derivation ST2.

[0126] For example, the dose parameter or the power parameter can be considered when annotating the training comparison data and then used for training the trained function. For example, the dose parameter or the power parameter can be considered when providing the measurement parameter database. The dose parameter or the power parameter can be included into the derivation as a boundary condition. For example, the dose parameter can relate to a maximum total dose for producing the image data set, a slice- effective tube current-time product, or also other parameters. For example, the power parameter can relate to a maximum tube power, for example a maximum tube current or a maximum tube current time product.

[0127] Figure 5 A schematic flow chart of a training method for providing a trained function for use in a method of operating a medical imaging device for producing an image data set is shown. The training method comprises a first providing step T-ST1, an applying step T-ST2, an adapting step T-ST3 and a second providing step T-ST4.

[0128] In the first providing step T-ST1, a training respiratory profile of a training patient and comparison measurement parameters associated therewith are provided by means of a training interface T-IF1.

[0129] If the training image data set is produced by means of the training imaging device using the comparison measurement parameters, the comparison measurement parameters determine a temporal resolution of a training slice image data set of at least one z-position comprised by a measurement region, the training slice image data set being comprised by the training image data set of the training patient. Here, the measurement region is affected by a respiratory motion. Here, the training slice image data set is based on a plurality of training projection data sets out of different projection angles in a relative rotational motion between a radiation source of the medical device and the training patient.

[0130] In the applying step T-ST2, the trained function is applied to the provided training respiratory profile by means of a training computing unit T-CU, wherein training measurement parameters are derived.

[0131] In the adapting step T-ST3, at least one parameter of the trained function is adapted based on a comparison of the derived training measurement parameters with the corresponding comparison measurement parameters by means of the training computing unit T-CU.

[0132] In the second providing step T-ST4, the trained function is provided by means of a second training interface T-IF2.

[0133] The method can in particular be performed repeatedly based on a plurality of training respiratory profiles and comparison measurement parameters associated therewith.

[0134] Figure 6 A schematic diagram of a device SY for controlling a medical imaging device 32 for generating a measurement region of a patient 39 influenced by respiratory motion is shown, the measurement region comprising at least one z-position, wherein the image data set comprises at least one slice image data set of the at least one z-position.

[0135] The device SY comprises an interface IF1 designed for providing a breathing curve S of the patient 39, the breathing curve describing a respiratory motion of the patient 39 over at least one breathing cycle.

[0136] The device SY comprises a computing unit CU designed for automatically deriving at least one measurement parameter based on the provided breathing curve S, wherein the derived measurement parameter determines a temporal resolution of the at least one slice image data set.

[0137] The device comprises a control unit 51 designed for controlling the medical imaging device 32 for generating the image data set based on the automatically derived measurement parameter, wherein for the at least one z-position a plurality of projection data sets from different projection angles in a relative rotational motion between the radiation source 37 of the medical device 32 and the patient 39 are detected and based thereon at least one slice image data set of the image data set is generated.

[0138] To this end, the device SY is signal-coupled with the medical imaging device 32, so that the medical imaging device 32 can be controlled based on the at least one measurement parameter or by means of a control signal based on the at least one measurement parameter.

[0139] The device SY is in particular designed for carrying out the proposed method for controlling the medical imaging device 32. The proposed device SY for controlling the medical imaging device 32 can be designed in such a way by means of the computing unit CU, the interface IF and the control unit 51 to constitute the respective steps of the method for carrying out an implementation variant of the proposed method for controlling the medical imaging device 32.

[0140] Figure 6 A training device T-SY for providing a trained function is also shown.

[0141] The training device T-SY advantageously comprises a first training interface T-IF1, a training computing unit T-CU, a second training interface T-IF2 and a training storage unit T-MU.

[0142] The first training interface T-IF is designed to provide the training patient's training respiratory profile and the comparative measurement parameters associated therewith, wherein the comparative measurement parameters are for determining, if the comparative measurement parameters are also used in combination with training the imaging device to generate the training image data set, a temporal resolution of the training slice image data set of the measurement region of the training patient 39 affected by the respiratory motion for at least one z-position comprised by the measurement region, and wherein the training slice image data set is based on a plurality of training projection data sets from different projection angles in a relative rotational motion between a radiation source of the training imaging device and the training patient.

[0143] The training computing unit T-CU is designed to apply the trained function to the training respiratory profile for deriving the training measurement parameters. Furthermore, the training computing unit T-CU is designed to adapt at least one parameter of the trained function based on a comparison of the derived training measurement parameters and the corresponding comparative measurement parameters.

[0144] The second training interface T-IF2 is designed to provide the trained function.

[0145] The illustrated training unit T-SY is advantageously designed to execute the proposed method for providing a trained function. The training device T-SY can in particular be designed in such a way that it constitutes the respective steps of the method for executing the method by means of the training interfaces T-IF1, T-IF2 and the training computing unit T-CU. The training device T-SY can in particular be designed to execute an implementation variant of the method for providing a trained function.

[0146] The device SY, the processing unit 45 and / or the training device T-SY can in particular be a computer, a microcontroller or an integrated circuit. Here, alternatively, it can be a real or virtual cluster of computers (the English professional term for a real cluster is "Cluster" and the English professional term for a virtual cluster is "Could"). The device SY and / or the training device T-SY can also be designed as a virtual system (English virtualization) constituted on a real computer or a real or virtual cluster of computers.

[0147] The interface IF and / or the training interfaces T-IF1, T-IF2 can be hardware or software interfaces (e.g. PCI bus, USB or Firewire). The computing unit CU and / or the training computing unit T-CU can have hardware elements or software elements, e.g. a microprocessor or a so-called FPGA (abbreviation of the English term "Field Programmable Gate Array"). The storage unit MU and / or the training storage unit T-MU can be implemented as non-permanent working memory (RAM) or as permanent mass storage device (hard disk, USB stick, SD card, solid state disk).

[0148] The interface IF and / or the training interfaces T-IF1, T-IF2 can comprise, inter alia, a plurality of sub-interfaces. In other words, the interface IF and / or the training interface T-IF can also comprise a plurality of interfaces IF or a plurality of training interfaces T-IF. The computing unit CU and / or the training computing unit T-CU can comprise, inter alia, a plurality of sub-computing units which execute different steps of the respective method. In other words, the computing unit CU and / or the training computing unit T-CU can also be understood as a plurality of computing units CU or a plurality of training computing units T-CU.

[0149] In the embodiment shown, the device SY is connected to the training device T-SY via the network NETW. For example, the function trained by means of the training device is transmitted to the device via the network NETW. Furthermore, in the example shown, the device SY is coupled directly to the medical device 32. In particular, the device can also be comprised by the medical device 32. However, the device SY can also be connected to the medical imaging device 32 by means of the network NETW.

[0150] Furthermore, the communication between the device SY and the training device T-SY can also take place offline, for example by exchanging data carriers.

[0151] The communication between the device SY and the training device T-SY can, for example, transmit further training data from the device SY to the training device T-SY or the training device T-SY transmits the trained function to the device SY. Furthermore, the training device T-SY can also be connected to further data sources.

[0152] The network NETW can be a local area network (English: "Local Area Network", abbreviated "LAN") or can be a large network (English: "Wide Area Network", abbreviated "WAN"). An example of a local area network is an intranet, an example of a large network is the Internet. The network NETW can also be formed wirelessly, in particular as a WLAN (for "Wireless LAN", the abbreviation "WiFi" is common in English) or as a Bluetooth connection. The network NETW can also be formed as a combination of the above examples.

[0153] The device SY can also comprise a database of measurement parameters stored on the storage unit MU and retrievable, which can also be accessed by means of a query. The database of measurement parameters can also be stored on an external storage unit. For example, the external storage unit can be linked with the device via the network NETW.

Claims

1. A method for manipulating a medical imaging device (32) for generating an image dataset of a measurement region of a patient (39) affected by respiratory motion, the measurement region including at least one z-position, wherein the image dataset includes at least one slice image dataset of the at least one z-position, the method comprising the steps of: • By means of a first interface (IF1), a respiratory curve (S) of the patient (39) is provided (ST1), the respiratory curve describing the respiratory movements of the patient (39) within at least one respiratory cycle. • Based on the provided respiration curve (S), at least one measurement parameter is automatically derived (ST2) by means of a computing unit (CU), wherein the derived measurement parameter determines the temporal resolution of the at least one slice image dataset, and wherein the automatic derivation of the at least one measurement parameter includes: Based on the provided respiratory curve (S), the resting phase (T) involving the respiratory movement is extracted. R,1 T R,2 The duration of at least one value, and the at least one measurement parameter is derived based on the extracted at least one value. • The medical imaging device (32) is manipulated (ST3) to generate the image dataset by means of a control unit (51) based on the automatically derived measurement parameters, wherein for the at least one z position, multiple projection datasets from different projection angles are detected in the relative rotational motion between a radiation source (37) of the medical imaging device (32) and the patient (39), and the at least one slice image dataset of the image dataset is generated based on the multiple projection datasets.

2. The method according to claim 1, wherein the measurement parameters include the rotation period time of the radiation source (37) during the relative rotational motion, and / or the projection angle interval, the projection angle interval including the angular range of the projection angle, and the at least one slice image dataset is based on the angular range.

3. The method according to claim 1 or 2, wherein the automatic derivation (ST2) further includes the dose parameter or power parameter of the radiation source (37).

4. The method of claim 1 or 2, wherein the automatic derivation (ST2) includes using a trained function, wherein at least one parameter of the trained function is adapted based on a comparison between training measurement parameters and comparison measurement parameters derived on a training respiratory curve of a training patient, wherein the training respiratory curve and the comparison measurement parameters are correlated with each other.

5. The method of claim 4, wherein the trained function comprises a neural network.

6. The method according to claim 1 or 2, wherein the extracted at least one value further comprises: Inhalation phase (T I,1 T I,2 T I,3 The duration of the inhalation phase or the slope of the exhalation phase; or the duration of the exhalation phase (T). E,1 T E,2 T E,3 The duration or slope of the exhalation phase.

7. The method of claim 6, wherein deriving the measurement parameter based on the extracted at least one value comprises: The query is stored in a measurement parameter database in a storage unit (MU), wherein the at least one value extracted or a parameter derived from the at least one value is used as the query parameter for the query.

8. The method according to claim 1 or 2, wherein the generated image dataset comprises a plurality of slice image datasets at a plurality of time points at the at least one z position, and in the step of manipulating (ST3) the medical imaging device (32) to generate the image dataset based on the measurement parameters automatically derived for the at least one z position, a plurality of projection datasets from different projection angles are detected at each of the plurality of time points.

9. The method of claim 8, wherein the resulting slice image dataset depicts the respiratory cycle of the patient (39) throughout the entire duration of the respiratory cycle.

10. A training method for providing a trained function, the trained function being used in any one of claims 4 to 6, the training method comprising the following steps: • A training interface (T-IF1) provides training respiratory curves for (T-ST1) trained patients and comparative measurement parameters associated with said training respiratory curves, wherein... - When a training image dataset of a measurement region is generated using the comparative measurement parameters with the aid of a training imaging device, the comparative measurement parameters determine the temporal resolution of the training slice image dataset of the measurement region, which is affected by respiratory motion of the training patient (39), for at least one z-position included in the measurement region, and The training slice image dataset is based on multiple training projection datasets from different projection angles generated during the relative rotational motion between a radiation source of the training imaging device and the training patient. • The trained function is applied to the training breathing curve provided by (T-ST2), and then training measurement parameters are derived using a training computation unit (T-CU) based on extracted values ​​of the duration of the resting phase involving the breathing motion. • By means of the training computation unit (T-CU), at least one parameter of the trained function is adapted (T-ST3) based on a comparison of the derived training measurement parameters and the corresponding comparison measurement parameters. • The trained function (TF) described in (T4) is provided by means of the second training interface (T-IF2).

11. An apparatus (SY) for manipulating a medical imaging device (32), the medical imaging device being used to generate an image dataset of a measurement region of a patient (39) affected by respiratory motion, the measurement region including at least one z-position, wherein the image dataset includes at least one slice image dataset of the at least one z-position, the apparatus comprising: • An interface (IF1) is designed to provide a respiratory curve (S) of the patient (39) describing the respiratory movements of the patient (39) within at least one respiratory cycle. • A computing unit (CU) is designed to automatically derive at least one measurement parameter based on the provided respiratory curve (S), wherein the derived measurement parameter determines the temporal resolution of the at least one slice image dataset, and the computing unit is also designed to extract the resting phase (T) involving the respiratory motion based on the provided respiratory curve (S). R,1 T R,2 The duration of at least one value, and the at least one measurement parameter is derived based on the extracted at least one value. • A control unit (51) is designed to manipulate the medical imaging device (32) to generate the image dataset based on the automatically derived measurement parameters, wherein multiple projection datasets from different projection angles are detected for the at least one z position in the relative rotational motion between a radiation source (37) of the medical imaging device (32) and the patient (39), and the at least one slice image dataset of the image dataset is generated based on the multiple projection datasets.

12. A medical imaging device (32) comprising a device (SY) for operating the medical imaging device according to claim 11.

13. A training apparatus (T-SY) for providing a trained function, said trained function for use in any one of claims 4 to 6, said training apparatus comprising: • The first training interface (T-IF1) is designed to provide training respiratory curves for (T-ST1) training patients and comparative measurement parameters associated with said training respiratory curves, wherein - When a training image dataset of a measurement region is generated using the comparative measurement parameters with the aid of a training imaging device, the comparative measurement parameters determine the temporal resolution of the training slice image dataset of the measurement region, which is affected by respiratory motion of the training patient (39), for at least one z-position included in the measurement region, and The training slice image dataset is based on multiple training projection datasets from different projection angles generated during the relative rotational motion between a radiation source of the training imaging device and the training patient. • A training computation unit (T-CU) is designed to apply a trained function to the training breathing curve and thereby derive training measurement parameters based on extracted values ​​of the duration of the resting phase involving the breathing motion, and is also designed to adapt at least one parameter of the trained function based on a comparison of the derived training measurement parameters and the corresponding comparison measurement parameters. • The second training interface (T-IF2) is designed to provide the trained functions.

14. A computer program product having a computer program that can be directly loaded into a storage unit (MU) of a device (SY) for operating a medical imaging device (32), the computer program having program segments that, when the program segments are executed by the device (SY), perform all the steps of the method according to any one of claims 1 to 9.

15. A computer program product having a computer program that can be directly loaded into a training memory (T-MU) of a training device (T-SY), the computer program having program segments such that when the program segments are executed by the training device (T-SY), all steps of the method according to claim 10 are performed.

Images