Computer-implemented method, processing device, imaging device and computer program for specifying the position and / or thickness of a layer to be imaged in a patient

Abstract

Computer-implemented method for specifying the position (1) and / or thickness (2) of a layer (3) to be imaged of a patient (4) in a tomosynthesis imaging procedure (5) using an imaging device (6), comprising the steps: - Obtaining three-dimensional image data (7) from a prior tomosynthesis imaging (8) or other three-dimensional imaging of the patient (4), wherein the three-dimensional image data (7) depict at least a part (9) of an object (10) located at least partially within the patient (4), - Evaluating the three-dimensional image data (7) to determine the position (11) of a longitudinal axis (12) of the object (10) along which the part (9) of the object (10) extends, - Determining the position (1) and / or the thickness (2) of the layer (3) to be imaged as a function of the determined position (11) of the longitudinal axis (12) such that the layer (3) to be imaged represents, on the one hand, a predetermined target position (13) within the patient (4) and, on the other hand, at least a longitudinal section (14) of the object (10) when the longitudinal axis (12) of the object (10) is in the determined position (11).

Description

[0001] The invention relates to a computer-implemented method for specifying a position and / or thickness of a patient slice to be imaged. The invention also relates to a processing device, an imaging system, a computer program, and a data storage medium.

[0002] Interventional imaging can monitor and / or assist the automatic or manual guidance of an elongated, essentially rigid object, such as a biopsy needle, to specific positions within a patient, for example, the location of a suspected tumor. Optimal detection of the position and orientation of the object would in principle be possible based on three-dimensional imaging of the entire potentially relevant area of ​​the patient. For this purpose, high-speed computed tomography (CT) can be performed, for example. However, the use of CT for real-time imaging can severely restrict access to the patient and is also typically associated with a relatively high radiation exposure for the patient.

[0003] Therefore, fluoroscopy is often used to monitor or control object guidance, providing continuous or quasi-continuous two-dimensional imaging with a relatively low X-ray dose. However, in this case, it is usually necessary to repeatedly switch between two imaging geometries offset by 90° during object guidance: one with fluoroscopy whose projection direction is approximately parallel to the object's longitudinal axis, and the other with fluoroscopy whose projection directions are approximately perpendicular to the longitudinal axis. However, the required pivoting of the imaging geometry, and thus, for example, of a C-arm, by 90° during object guidance can significantly disrupt the workflow.

[0004] The document DE 10 2016 210 093 A1 discloses a method for planning an examination of an examination subject using a tomosynthesis device. Furthermore, the document US 2008 / 0 198 966 A1 discloses a device for three-dimensional imaging and extraction of tissue from a human breast, wherein an orientation and position of a needle for tissue extraction can be determined from recorded or reconstructed image data using an image analysis unit. Furthermore, the document US 2016 / 0 089 090 A1 discloses a radiation imaging system that can capture multiple projection images of a breast with a biopsy needle arranged therein, wherein an image of the biopsy needle from one of the projection images can be inserted into a reprojection image generated from tomographic images, wherein the tomographic images are generated from the projection images.

[0005] The invention is therefore based on the object of providing an improved approach to detecting a relative position of an object with respect to a target position and, in particular, to support the guidance of the object to the target position.

[0006] The object is achieved according to the invention by a computer-implemented method for specifying a position and / or a thickness of a slice of a patient to be imaged in a tomosynthesis imaging to be parameterized by means of an imaging device, which comprises the following steps: - Obtaining three-dimensional image data from a previous tomosynthesis imaging or another three-dimensional imaging of the patient, wherein the three-dimensional image data depict at least a part of an object located at least partially within the patient, - Evaluating the three-dimensional image data to determine the position of a longitudinal axis of the object along which the part of the object extends, - Determining the position and / or the thickness of the slice to be imaged as a function of the determined position of the longitudinal axis such that the slice to be imaged, on the one hand, images a predetermined target position within the patient and, on the other hand, images at least a longitudinal section of the object when the longitudinal axis of the object is in the determined position.

[0007] As long as tomosynthesis images both a sufficient longitudinal section of the object and the target position, for example, the area of ​​a tumor to be biopsied, the movement of the object to guide it to the target position can be visualized and / or controlled just as well as with computed tomography. However, compared to conventional computed tomography, access to the patient is less restricted, and the irradiated X-ray dose is typically significantly lower at the same frame rate.

[0008] However, since tomosynthesis imaging can only image relatively thin slices, conventional tomosynthesis imaging, which uses a fixed position and thickness of the slice to be imaged, is not suitable for robustly guiding an object to the target position in many applications. Even small deviations from a planned object movement path can cause the end of the object facing the target position to leave the slice to be imaged. Such deviations can occur, for example, during manual guidance of the object or due to tolerances in automated guidance of the object and / or when the object is deflected from a desired path due to patient tissue properties or similar.

[0009] However, within the scope of the invention, it was recognized that this problem can be avoided by the explained adaptation of the slice to be imaged by the subsequent tomosynthesis imaging depending on the position of the longitudinal axis, so that the explained procedure can combine robust targeting of the object with a fluid workflow, good accessibility of the patient and low radiation exposure of the patient.

[0010] The longitudinal section of the object is, in particular, a section that extends in the direction of the longitudinal axis. The direction in which the longitudinal axis extends is, in the object's coordinate system, the direction in which the object, or at least its imaged part, has its maximum extent.

[0011] The preceding tomosynthesis imaging or the other three-dimensional imaging and / or the tomosynthesis imaging to be parameterized may be performed outside of the claimed method. For example, the claimed method may be performed entirely within a time interval between the completion of the preceding tomosynthesis imaging or the other three-dimensional imaging and the start of the tomosynthesis imaging to be parameterized.

[0012] However, the method can alternatively also include performing the preceding and / or the tomosynthesis imaging to be parameterized, or the other three-dimensional imaging. For example, it can be performed as part of interventional imaging to track the slice to be imaged depending on the determined position of the longitudinal axis, so that the longitudinal section of the object can be robustly imaged. In particular, this can ensure that the end of the object facing the target position does not leave the slice to be imaged.

[0013] The tomosynthesis imaging to be parameterized and the preceding tomosynthesis imaging preferably use the same imaging device and are in particular carried out immediately one after the other or at a short time interval, for example at a time interval of less than 5 minutes or less than 1 minute or less than 30 seconds.

[0014] However, if the tomosynthesis imaging to be parameterized and the preceding tomosynthesis imaging or the other three-dimensional imaging are performed by different imaging devices, their coordinate systems can first be registered to each other in the usual way.

[0015] The respective tomosynthesis imaging comprises, in particular, the acquisition of multiple X-ray images of the patient from different angles by the imaging device and the reconstruction of three-dimensional image data of an imaged slice from the X-ray images. The other three-dimensional imaging can, for example, comprise computed tomography, magnetic resonance imaging, positron emission tomography, or ultrasound imaging.

[0016] The slice to be imaged is spanned, in particular, by a length and a width and the thickness of the slice to be imaged. The length, width, and thickness can each correspond to dimensions of the slice to be imaged in mutually perpendicular spatial directions, namely the longitudinal direction, the width direction, and the thickness direction of the slice to be imaged. The plane spanned by the length and width, or the longitudinal direction and the width direction, of the slice to be imaged can also be considered the image plane, which is, in particular, perpendicular to the projection direction of the imaging at the mean acquisition angle of an angular range to be covered within the scope of the tomosynthesis imaging to be parameterized.

[0017] In particular, an orientation of the layer to be imaged or of the image plane or of the normal to the image plane and / or a position of the layer to be imaged can be determined as the position of the layer to be imaged.

[0018] The position of the layer to be imaged is preferably selected such that the longitudinal axis of the object is parallel to an image plane spanned by a length and a width of the layer to be imaged, or is angled to the image plane at an angle of less than 45°, less than 30°, or less than 15°. In particular, the longitudinal axis of the object can lie in the image plane or intersect it, particularly within the layer to be imaged. The image plane can be arranged centrally within the layer to be imaged, particularly in the direction of the thickness of the layer to be imaged.

[0019] As a rule, for example, when the object is a needle, especially a biopsy needle, it can be assumed that the object moves at least approximately parallel to the object's longitudinal axis between successive imaging sequences. If the slice to be imaged is selected such that the image plane runs at least roughly parallel to the longitudinal axis, a movement of the object thus leads to only a relatively small displacement of the object perpendicular to the image plane or in the thickness direction of the slice to be imaged.

[0020] Since the thickness of the slice to be imaged in tomosynthesis imaging is typically significantly smaller than the slice extension perpendicular to the thickness direction or within the image plane, this can minimize the probability that relevant parts of the object, for example the tip of a biopsy needle, lie outside the slice to be imaged in the tomosynthesis imaging to be parameterized and are therefore not imaged.

[0021] In particular, the position of the slice to be imaged can be selected so that both the longitudinal axis of the object and the target position lie within the image plane. This enables optimal imaging of the relative position and orientation of the object with respect to the target position, even with very thin slices or, as will be explained later, when only two-dimensional imaging is temporarily performed.

[0022] Using a thinner layer to be imaged is typically advantageous because, in this case, the imaging system only needs to cover a narrower angular range during imaging than with a thicker layer to be imaged. This allows higher imaging rates to be achieved with decreasing layer thickness, or a given imaging rate can be achieved with lower accelerations of imaging system components.

[0023] The thickness of the layer to be imaged is preferably selected such that the target position and / or an end of the object in the direction of the longitudinal axis, the position of which is determined as a function of the three-dimensional image data, each have at least a predetermined minimum distance from boundary surfaces of the layer to be imaged in the direction of the thickness.

[0024] By specifying the minimum distance, it can be ensured that even in the event of a certain movement of the patient and / or the object with respect to the imaging device and thus with respect to the slice to be imaged, the target position or the end of the object lie within the slice to be imaged during the tomosynthesis imaging to be parameterized, whereby the position and orientation of the object with respect to the target position can be well visualized or evaluated on the basis of the parameterized tomosynthesis imaging.

[0025] To determine the position of the layer to be imaged, at least one of the following quantities or a weighted sum comprising several or all of the following quantities as summands can be minimized: - an angle between the longitudinal axis of the object in the determined position of the longitudinal axis and the image plane or an image plane spanned by a length and width of the layer to be imaged, - a distance of the target position from the center of the layer to be imaged in the direction of the thickness of the layer to be imaged, and / or - a distance of the or one end of the object in the direction of the longitudinal axis of the object, the position of which is determined as a function of the three-dimensional image data, from the center of the layer to be imaged in the direction of the thickness of the layer to be imaged.

[0026] This optimization of the position of the slice to be imaged allows the selection of a relatively small thickness of the slice to be imaged without risking that relevant features in the tomosynthesis imaging to be parameterized lie outside the slice to be imaged.

[0027] To specify an orientation of the slice to be imaged, a mean acquisition angle of an angular range to be covered by the imaging device within the scope of the tomosynthesis imaging to be parameterized can be specified.

[0028] The minimization of at least one variable or the weighted sum for determining the position of the slice to be imaged can be performed under the constraint that the mean acquisition angle is restricted to a predefined angular interval. For example, such a restriction may be necessary due to limited mobility of the imaging device, such as a C-arm, or it may be expedient to avoid, or at least minimize, disruption or restriction of medical personnel.

[0029] To specify the position of the slice to be imaged, the orientation and / or position of the slice to be imaged can be specified. The position of the slice to be imaged can be specified, in particular, by selecting the isocenter of the tomosynthesis imaging to be parameterized. The position specification can be made in such a way that the distances of relevant features, in particular the target position and the end of the object, from the edges of the slice to be imaged in the thickness direction and / or the longitudinal direction and / or the width direction of the slice to be imaged and / or a weighted sum that includes these distances as summands are maximized.

[0030] Two-dimensional X-ray imaging can always be performed prior to the tomosynthesis imaging to be parameterized, or when an imaging condition is met. The imaging geometry of the two-dimensional X-ray imaging is specified depending on the determined position of the object's longitudinal axis. Two-dimensional imaging can be performed once, several times, or continuously, for example, as fluoroscopy, prior to the tomosynthesis imaging to be parameterized.

[0031] To specify the imaging geometry, a C-arm carrying the X-ray source and the X-ray detector can be controlled, for example. Two-dimensional imaging can be performed in particular such that the projection plane of the two-dimensional image is parallel to the image plane of the tomosynthesis imaging being parameterized and / or such that the projection direction of the two-dimensional image is perpendicular to this image plane.

[0032] For example, after each tomosynthesis imaging, at least if a suitable imaging geometry can be selected based on this and the imaging condition is thus met, two-dimensional imaging can initially be performed for a specified time interval or until a trigger condition is met, for example, to allow higher imaging rates and / or reduce the patient's radiation exposure. For example, the tomosynthesis imaging to be parameterized can then be performed at regular intervals or when the trigger condition is met to acquire three-dimensional image data, based on which the position of the object or its longitudinal axis can be determined again in order to track the imaging.

[0033] In the method according to the invention, a preliminary target position can be specified in a reference coordinate system of a three-dimensional reference image data set, wherein the reference image data set at least partially images the patient, wherein the reference coordinate system of the three-dimensional reference image data set is registered to a device coordinate system defined with respect to the imaging device in order to transfer the preliminary target position into the device coordinate system and thus to specify the target position.

[0034] The registration of the reference coordinate system with the device coordinate system can be performed on the basis of the reference image dataset and at least two two-dimensional X-ray images of the patient by the imaging device, which were acquired at acquisition angles that differ from one another by at least 30° or at least 60°. In particular, the acquisition angles or projection directions of the two-dimensional X-ray images can be at least approximately perpendicular to one another. For example, for one of the X-ray images, the projection direction can run at least approximately parallel to the longitudinal axis of the object, and for the other of the X-ray images, it can be at least approximately perpendicular to the longitudinal axis.

[0035] The reference image dataset can be, for example, a computed tomography (CT) scan or a magnetic resonance imaging scan of the patient. The reference image dataset can be used, for example, to automatically or manually identify a target position to which the object is to be guided during image data analysis and to mark it in the reference image dataset to specify the preliminary target position.

[0036] The reference image dataset can, in particular, be acquired by a different imaging device than the one used for tomosynthesis imaging. The reference image dataset can, in particular, be acquired significantly before the tomosynthesis imaging, for example, one or more days before.

[0037] In particular, the two acquisition geometries can be used to capture the X-ray images used for registration. As explained above, the state of the art repeatedly switches between these geometries when monitoring a biopsy or similar procedures using fluoroscopy. In contrast to the procedure described there, however, the associated significant change in acquisition geometry, for example, pivoting a C-arm by 90°, is only required once for registration and does not need to be repeated during monitoring.

[0038] To specify the thickness of the slice to be imaged, the width of the angular range to be covered by the acquisition angle during the tomosynthesis imaging to be parameterized can be specified. In addition or alternatively, the position of an isocenter of the tomosynthesis imaging to be parameterized can be specified.

[0039] The projection direction for the mean acquisition angle of the angular range corresponds, in particular, to the thickness direction of the slice to be imaged. The slice thickness that can be imaged with a given image quality is determined by the angular range covered. The isocenter of the tomosynthesis imaging to be parameterized can be shifted to shift the imaged area. Tilting the imaged area or specifying the orientation of the slice to be imaged can be done as explained above.

[0040] In the method according to the invention, a respective tomosynthesis imaging can be performed repeatedly, wherein the three-dimensional image data of the respective most recently performed tomosynthesis imaging can be used to parameterize the respective subsequent tomosynthesis imaging as the tomosynthesis imaging to be parameterized in order to specify the position and / or thickness of the respective slice of the patient to be imaged. As already explained above, this allows for tracking of the imaging.

[0041] The position of the longitudinal axis of the object and / or further image data acquired as part of the tomosynthesis imaging to be parameterized can be evaluated in order to provide at least one control parameter for controlling an actuator for the automated movement of the object to the target position and / or can be visualized for a user via a display device in order to assist the user in guiding the object to the target position.

[0042] After determining the position and / or thickness of the slice to be imaged, an imaging device and / or components of the imaging device can be controlled so that three-dimensional image data describing the slice to be imaged are acquired by the imaging device as part of a tomosynthesis imaging process parameterized by the position and / or thickness. This is also possible if the acquisition of the patient's three-dimensional image data by the preceding tomosynthesis imaging occurs outside the procedure or if these three-dimensional image data are acquired by a different three-dimensional imaging process than the three-dimensional tomosynthesis imaging process.

[0043] In the method according to the invention, the image of a cuboid layer with constant thickness can be parameterized.

[0044] However, it is also possible for the slice to be imaged to be a wedge-shaped slice. This can be useful because, for example, in a biopsy, even small deviations in the puncture geometry can lead to significant deviations in the position of the tip of the biopsy needle during deeper punctures. This can be compensated for, for example, by imaging a wedge-shaped slice that becomes thicker with increasing distance from the puncture site. By appropriately selecting the acquisition geometries used in tomosynthesis imaging, the same maximum slice thickness can potentially be achieved with a wedge-shaped slice using a smaller angular range covered by the acquisition angle than would be the case for a cuboid-shaped slice. Thus, imaging a wedge-shaped slice can potentially reduce the required movement of components used for the acquisition, such as a C-arm.

[0045] Preferably, the position of the wedge-shaped layer to be imaged is determined such that the thickness of the layer to be imaged varies in a direction of variation that runs parallel to the longitudinal axis of the object or is angled at an angle of less than 45°, less than 30°, or less than 15° to the longitudinal axis. Additionally or alternatively, the position of the wedge-shaped layer can be determined such that its thicker end faces away from the puncture site of a biopsy needle or other needle used as the object.

[0046] If a wedge-shaped layer is imaged, the determined thickness of the layer to be imaged can be determined, for example, as the maximum layer thickness, the minimum layer thickness, and / or an average layer thickness. If only one of the specified thicknesses is specified by the method explained above, one of the other thicknesses can, for example, be fixed or specified via a fixed relationship with the determined layer thickness.

[0047] In addition to the computer-implemented method according to the invention, the invention relates to a processing device which is configured to carry out the computer-implemented method according to the invention.

[0048] The invention further relates to an imaging device configured to acquire three-dimensional image data of a patient slice to be imaged using tomosynthesis imaging, wherein the imaging device is configured to carry out the computer-implemented method according to the invention. The imaging device can, in particular, be an X-ray device. In this case, imaging can be performed using at least one X-ray source and at least one X-ray detector, for example, by emitting X-rays using the X-ray source to illuminate the patient and detecting the X-rays using the X-ray detector after their interaction with the patient. In particular, an examination volume of the imaging device can be accessible during imaging. This can be achieved, for example, by using a C-arm or O-arm X-ray device as the imaging device.

[0049] In particular, the imaging device or a control device of the imaging device can be configured, as explained above, to control individual components of the imaging device, for example an X-ray source, an X-ray detector and / or an actuator for moving these components or, for example, a C-arm carrying these components, in order to carry out the tomosynthesis imaging parameterized by the thickness and / or position of the slice to be imaged.

[0050] Furthermore, the invention relates to a computer program comprising program instructions which are designed to implement the computer-implemented method according to the invention when they are executed on a data processing device.

[0051] The invention also relates to a data carrier comprising the computer program according to the invention.

[0052] Further advantages and details of the invention will become apparent from the following exemplary embodiments and the accompanying drawings. These schematically show: Fig. 1 shows an embodiment of a processing device according to the invention, which is designed to implement an embodiment of the computer-implemented method according to the invention for specifying a position and a thickness of a slice to be imaged in a tomosynthesis imaging, Fig. 2 a schematic representation of a by the in Fig. 1 used method to parameterize layer, and Fig. 3 a flow chart of the Fig. 1 method used to parameterize the layer.

[0053] Fig. Figure 1 schematically shows tomosynthesis imaging of a patient 4 by an imaging device 6. In a conventional manner, multiple X-ray images of the patient 4 are acquired from different angles within an angular range 24 covered by the imaging, and a processing device 39 reconstructs three-dimensional image data from these images. Based on the three-dimensional image data, a visualization 32 of relevant features can then be performed on a display device 34, for example, by outputting individual slices or a generated artificial projection image.

[0054] In the example, the imaging is intended to assist a user (not shown) in guiding an object 10, for example, a biopsy needle, within the patient 4 to a target position 13, for example, to a potential tumor 40. Alternatively, it would also be possible, for example, to control an actuator (not shown) based on the three-dimensional image data in order to automatically guide the object 10 to the target position 13.

[0055] A problem here, however, is that the swept angular range 24 should typically be kept as small as possible to avoid disturbing medical personnel in the area of ​​the patient 4 and to enable high imaging rates. However, in tomosynthesis imaging, this results in only relatively thin slices being imaged with sufficient quality. Therefore, with a fixed specification of the slice to be imaged, as is used in conventional tomosynthesis imaging, even a relatively slight movement of the patient 4 or the object 10 with respect to the imaging device 6 can result in features relevant for guiding the object 10 being outside the imaged slice and thus not being able to be imaged.

[0056] Therefore, the processing device 39 implements a method for specifying the position and thickness of the slice to be imaged for parameterizing tomosynthesis imaging on the basis of three-dimensional image data from a previous tomosynthesis imaging, which method is described below with additional reference to Fig. 2 and Fig. 3. This method, in particular, allows the parameterization of the slice to be adjusted in such a way that a relevant longitudinal section 14 of the object 10 and the target position 13 can typically be imaged robustly. For the sake of completeness, it should be noted that it may also be sufficient to dynamically adjust only the position or the thickness of the slice to be imaged, with an adjustment of both variables typically being advantageous.

[0057] Fig. Figure 2 shows a schematic representation of a layer 3 to be parameterized and features that are taken into account when parameterizing layer 3, while Fig. 3 shows a flowchart of the procedure used.

[0058] The concrete, in Fig. The flowchart shown in Figure 3 represents a particularly advantageous embodiment of the method and also includes preparatory or downstream steps that are not necessarily part of the method. Therefore, for better understanding, the central process steps will first be briefly explained before the entire process is discussed in detail.

[0059] The parameterization of the tomosynthesis imaging 5 to be parameterized, or specifically the specification of the position 1 and the thickness 2 of the slice 3 to be imaged, is based on three-dimensional image data 7 of a previous tomosynthesis imaging 8 of the patient 4, which image at least a part 9 of the object 10 located at least partially within the patient 4.

[0060] Since in the example a tomosynthesis imaging 5, 8 which takes place at relatively short time intervals is to be parameterized during operation, the three-dimensional image data 7 in the example are image data 7 acquired in step S7 by the imaging device 6. In principle, however, the image data 7 in step S7 could also originate from another source, for example read out from a database which stores image data 7 of a tomosynthesis imaging 8 which was completed before the start of the method.

[0061] The three-dimensional image data 7 are subsequently evaluated in step S8 by the processing device 39 in order to determine at least the position 11 of the longitudinal axis 12 of the object 10. In step S9, the position 1 and the thickness 2 of the slice 3 to be imaged are then specified, as will be explained in more detail later, depending on the determined position 11 of the longitudinal axis 12 in such a way that the slice 3 to be imaged images, on the one hand, the specified target position 13 within the patient 4 and, on the other hand, the longitudinal section 14 of the object 10 when the longitudinal axis 12 of the object 10 is in the determined position 11.

[0062] In other words, slice 3 is parameterized in particular such that, at least for the current object position, all relevant features for guiding object 10 to target position 13 can be imaged. In particular, if the parameterization of slice 3 takes into account an anticipated direction of movement of object 10, which typically coincides with the longitudinal axis for needles, catheters, and the like, for example, it can be achieved that even with relatively large movements of object 10 between tomosynthesis images 5, 8, the relevant features can also be robustly imaged in the subsequent tomosynthesis imaging 5 to be parameterized. Thus, the method for specifying the position 1 and thickness 2 of slice 3 to be imaged could, for example, already be implemented by steps S7 to S9.

[0063] The following describes the Fig. 3 is explained in detail, whereby for a better understanding the representation of the parameterized layer 3 in Fig. 2 is used additionally.

[0064] In step S1, a three-dimensional reference image data set 29 is first acquired, which at least partially depicts the patient 4. The reference image data set 29 can, for example, be an image data set acquired by computed tomography, magnetic resonance imaging, positron emission tomography, ultrasound imaging, or the like.

[0065] In step S2, a preliminary target position 28 is defined in the reference image data set 29, to which the object 10 is to be guided at a later time. For example, a potential tumor 40 to be biopsied can be identified automatically or by medical personnel.

[0066] The reference image data set 29 is typically acquired by a different imaging device than the imaging device 6 used for tomosynthesis imaging 5, 8, or at least the patient 4 is often not in the exact same position relative to the imaging device 6 when the reference image data set is acquired as was the case when the reference image data set 29 was acquired. Therefore, the reference coordinate system of the three-dimensional reference image data set 29 should first be registered to the device coordinate system of the imaging device 6.

[0067] For this purpose, after positioning the patient 4 in the imaging device 6, a two-dimensional X-ray image 30, 31 is first acquired in steps S3 and S4. Significantly different angles are used to acquire the X-ray images 30 and 31.

[0068] For example, the X-ray images 30 can be acquired using an imaging geometry in which the projection direction is substantially parallel to the presumed or pre-procedure planned position of the longitudinal axis 12 of the object 10. A projection direction can then be used for the X-ray image 31 that is substantially perpendicular to the projection direction for the X-ray image 30.

[0069] The use of significantly different imaging geometries may, for example, require a 90° pivoting of a C-arm, which may be disruptive for medical personnel in the area of ​​the patient 4. In contrast to the repeated switching between corresponding imaging geometries explained above, which is typically required in fluorescence imaging to monitor the guidance of the object 10 to the target position 13, such a significant change in the imaging geometry is only required once in the described method to prepare for the procedure.

[0070] In step S5, the reference coordinate system can then be registered with the device coordinate system based on the reference image data set 28 and the two-dimensional X-ray images 30, 31. Approaches for registering three-dimensional image data with two-dimensional image data acquired from multiple perspectives are known per se and will therefore not be explained in detail.

[0071] In step S6, since the coordinate systems are now registered to each other, the target position 13 in the facility coordinate system can be determined from the provisional target position 28 in the reference coordinate system.

[0072] In step S7, a first tomosynthesis imaging 8 can then be performed in order to provide the three-dimensional image data 7 of the slice imaged by this tomosynthesis imaging 8.

[0073] The initial selection of the position 1 and thickness 2 of the slice 3 to be imaged can be based, for example, on an intervention plan, which can be carried out, for example, using the reference image dataset 29 and can thus be transferred to the device coordinate system due to the prior registration of the coordinate systems in step S5. For example, in addition to the target position, a puncture site for the object 10 can also be specified in the reference image dataset, and within the framework of the parameterization of the tomosynthesis imaging, it can initially be assumed that the longitudinal axis 12 of the object 10 runs at least approximately parallel to the connecting line between the puncture site and the target position 13.

[0074] In step S8, the three-dimensional image data 7 from the preceding tomosynthesis imaging 8 are evaluated to determine, on the one hand, the position 11 of the longitudinal axis 12 of the object 10 and, in the example, also the position of the end 49 of the object 10 in the direction of the longitudinal axis 12. Thus, if the object 10 is, for example, a biopsy needle, its orientation and the position of its tip are determined. Artificial objects, such as biopsy needles, generally generate sufficient contrast during tomosynthesis imaging to enable, for example, fully automated segmentation of such an object 10, so that the aforementioned dimensions can be easily determined.

[0075] In step S9, based on the position 11 of the longitudinal axis 12 and, in the example, also the position of the end 13 of the object 10, the position 1 and the thickness 2 of the slice 3 to be imaged in the tomosynthesis imaging to be parameterized are determined. In the example, the position 1 of slice 3 or of an image plane 17, which is spanned by the longitudinal direction 42 and the width direction 43 and is located centrally with respect to the thickness 2 of slice 3, is first optimized. With respect to the position 1, an orientation 22 of slice 3 or of the image plane 17 and, on the other hand, the position 47 of slice 3 can be specified.

[0076] As In Fig. 2 is schematically represented by the arrows 44 to 46, the slice 3 to be imaged can be rotated about several rotation axes to specify the orientation 22, which are each parallel to the thickness direction 41, the longitudinal direction 42 or the width direction 43. The orientation can be set by selecting the mean recording angle 23 of an angular range 24 to be covered by the imaging device 6 within the scope of the tomosynthesis imaging 5 to be parameterized. Depending on the design of the imaging device 6, a pivoting of the slice 3 to be imaged about all three rotation axes may be possible, or only parts of these rotations may be possible. The position 47 of the slice 3 can be adjusted by shifting the isocenter 48 of the tomosynthesis imaging.

[0077] To determine the position 1 of the layer 3 to be imaged, a weighted sum comprising the following variables as summands is minimized in the example: - the angle 18 between the longitudinal axis 12 of the object 10 and the image plane 17, - the distance of the target position 13 from the center 21 of the layer 3 to be imaged in the direction of the thickness 2 of the layer 3 to be imaged, and - the distance of the end 49 of the object 10 from the center 21 of the layer 3 to be imaged in the direction of thickness 2.

[0078] As already explained in detail in the general part, such a choice of the position of layer 3 enables the use of particularly thin layers in imaging and thus, for example, high imaging rates or small swept angle ranges in imaging.

[0079] Alternatively or additionally, within the scope of the optimization, the distances of relevant features, i.e. in particular the target position 13 and the end 49 of the object 10, from the edge surfaces of the layer to be imaged in the longitudinal direction 42 and / or the width direction 43 of the layer to be imaged can be maximized, for example by including them with negative weights in the weighted sum to be minimized.

[0080] Due to the design of the imaging device 6, or to ensure that medical personnel have easy access to the patient 4 during imaging, it may be expedient to restrict the imaging geometry for the tomosynthesis imaging 5 to be parameterized. Therefore, in the example, the weighted sum is minimized under the boundary condition that the mean acquisition angle 23 is restricted to a predefined angular interval 25.

[0081] As in Fig. 2 is shown schematically, the position 1 or the orientation 22 of the layer 3 to be imaged can be selected such that the longitudinal axis 12 of the object 10 is angled at an angle 18 of less than 30° to the image plane 17. In the example according to Fig. 2, a relatively large angle 18 is used for clearer representation.

[0082] Typically, within the scope of the optimization explained above, it can be achieved that the angle 18 is significantly smaller than in Fig. 2, and the longitudinal axis 12 can, in particular, be at least approximately parallel to the image plane 17. As already explained in the general section, this makes it possible for the object 10 to be displaced over long distances without leaving the imaged slice 3. Tracking slice 3 is thus less time-critical, allowing for longer time intervals between the tomosynthesis images 5, 8.

[0083] In the example, the thickness 2 of the slice 3 to be imaged is selected such that the target position 13 and the end 49 of the object 12 each have a predetermined minimum distance from the interfaces 20 of the slice 3 to be imaged, the thickness direction 41 of the slice 3 to be imaged. In the example, the thickness of the slice to be imaged is predetermined by the width of the angular range 24 to be covered by the acquisition angle within the scope of the tomosynthesis imaging 5 to be parameterized.

[0084] In principle, the tomosynthesis imaging 5 to be parameterized could be performed immediately after step S9. However, in the example, as far as possible, a two-dimensional X-ray imaging 27, for example, a fluoroscopy, should be performed temporarily between the tomosynthesis imaging 5, 8, which can lead to a lower radiation exposure for patient 4 at the same imaging rates.

[0085] Therefore, in step S10, it is first checked whether an imaging condition 26 is fulfilled, the fulfillment of which implies the admissibility of the two-dimensional X-ray imaging 27. In the example, the imaging condition 26 evaluates, on the one hand, the angle 18 between the image plane 17 and the longitudinal axis 12 of the object 10 and the distances of the target position 13 and the previously determined positions 19 of the end 49 of the object 10 from the interfaces of the previously parameterized slice 3, and, on the other hand, a time interval since the execution of the previous tomosynthesis imaging 8. In this case, the time interval can, in particular, be compared with a limit value that depends on the specified angle at the specified distances.In other words, two-dimensional imaging 27 may be permissible for a certain time if, due to sufficiently large distances and a sufficiently flat angle 18, it is very unlikely for sufficiently long periods of time that relevant features will leave the imaged layer 3.

[0086] If the imaging condition 26 is met in step S17, a two-dimensional X-ray imaging 27 can first be carried out in step S11, after which step S10 is repeated, for example to check whether the limit value for the time interval to the last tomosynthesis imaging 8 has already been exceeded.

[0087] If the imaging condition 26 is not or no longer met in step S10, the imaging device 6 is controlled in step S12 to carry out the tomosynthesis imaging 5 to be parameterized. This is used, on the one hand, to output a visualization 32 of the slice or of relevant image content to the user via the display device 34 in step S13 and / or, on the other hand, to output control signals 33 for an actuator system (not shown) for the automated movement of the object 10.

[0088] On the other hand, based on the three-dimensional image data of the tomosynthesis imaging 5 performed in step S12, the method steps from step S8 can be repeated in order, for example, to iteratively track the imaged layer 3 over several tomosynthesis imagings.

[0089] The method steps explained are implemented in the example by the processing device 39. This is implemented in the example by programming a programmable data processing device 35 with a processor 36 and a memory 37 using a suitable computer program 38. The processing device 39 can, for example, be a workstation computer, be integrated into the imaging device 6, be provided by a server, or be implemented as a cloud solution.

[0090] Although the invention has been illustrated and described in detail by the preferred embodiment, the invention is not limited to the disclosed examples and other variations may be derived therefrom by those skilled in the art without departing from the scope of the invention.

[0091] Regardless of the grammatical gender of a particular term, persons with male, female or other gender identities are included.

Claims

[1] Computer-implemented method for specifying a position (1) and / or a thickness (2) of a slice (3) of a patient (4) to be imaged in a tomosynthesis imaging (5) to be parameterized by means of an imaging device (6), comprising the steps: - Obtaining three-dimensional image data (7) of a previous tomosynthesis imaging (8) or another three-dimensional imaging of the patient (4), wherein the three-dimensional image data (7) depict at least a part (9) of an object (10) located at least partially within the patient (4), - evaluating the three-dimensional image data (7) to determine the position (11) of a longitudinal axis (12) of the object (10) along which the part (9) of the object (10) extends, - Determining the position (1) and / or the thickness (2) of the slice (3) to be imaged as a function of the determined position (11) of the longitudinal axis (12) such that the slice (3) to be imaged, on the one hand, images a predetermined target position (13) within the patient (4) and, on the other hand, images at least one longitudinal section (14) of the object (10) when the longitudinal axis (12) of the object (10) is in the determined position (11). [2] Computer-implemented method according to claim 1, characterized by that the position (1) of the layer (3) to be imaged is selected such that the longitudinal axis (12) of the object (10) is parallel to an image plane (17) spanned by a length (15) and a width (16) of the layer (3) to be imaged or is angled at an angle (18) of less than 45° or less than 30° or less than 15° to the image plane (17). [3] Computer-implemented method according to claim 1 or 2, characterized bythat the thickness (2) of the layer (3) to be imaged is selected such that the target position (13) and / or an end (49) of the object (12) in the direction of the longitudinal axis (12), the position (19) of which is determined as a function of the three-dimensional image data (7), each have at least a predetermined minimum distance from boundary surfaces (20) of the layer (3) to be imaged in the direction of the thickness (2). [4] Computer-implemented method according to one of the preceding claims, characterized by that to determine the position (1) of the layer (3) to be imaged, at least one of the following variables or a weighted sum comprising several or all of the following variables as summands is minimized: - an angle (18) between the longitudinal axis (12) of the object (10) in the determined position (11) of the longitudinal axis (12) and the or an image plane (17) spanned by a length (15) and width (16) of the layer (3) to be imaged, - a distance of the target position (13) from the center (21) of the layer (3) to be imaged in the direction of the thickness (2) of the layer (3) to be imaged, and / or - a distance of the or an end (49) of the object (10) in the direction of the longitudinal axis (12) of the object (10), the position of which is determined as a function of the three-dimensional image data (7), from the center (21) of the layer (3) to be imaged in the direction of the thickness (2) of the layer (3) to be imaged. [5] Computer-implemented method according to one of the preceding claims, characterized by in that, in order to specify an orientation (22) of the slice (3) to be imaged, a mean recording angle (23) of an angular range (24) to be covered by the imaging device (6) within the scope of the tomosynthesis imaging (5) to be parameterized is specified. [6] Computer-implemented method according to claims 4 and 5, characterized bythat the minimization of the at least one variable or the weighted sum for determining the position (1) of the layer (3) to be imaged takes place under the boundary condition that the mean recording angle (23) is restricted to a predetermined angular interval (25). [7] Computer-implemented method according to one of the preceding claims, characterized by that a two-dimensional X-ray imaging (27) is always carried out before the tomosynthesis imaging (5) to be parameterized or when an imaging condition (26) is fulfilled, wherein an imaging geometry of the two-dimensional X-ray imaging (27) is predetermined as a function of the determined position (11) of the longitudinal axis (12) of the object (10). [8] Computer-implemented method according to one of the preceding claims, characterized byin that a provisional target position (28) is specified in a reference coordinate system of a three-dimensional reference image data set (29), wherein the reference image data set (29) at least partially images the patient (4), wherein the reference coordinate system of the three-dimensional reference image data set (29) is registered with a device coordinate system defined with respect to the imaging device (6) in order to transfer the provisional target position (28) into the device coordinate system and thus to specify the target position (13). [9] Computer-implemented method according to claim 8, characterized bythat the registration of the reference coordinate system to the device coordinate system is carried out on the basis of the reference image data set (29) and of at least two two-dimensional X-ray images (30, 31) of the patient (4) by the imaging device (6), which are recorded at recording angles that differ from one another by at least 30° or at least 60°. [10] Computer-implemented method according to one of the preceding claims, characterized by in that, in order to specify the thickness (2) of the slice (3) to be imaged, the width of the angular range (24) to be covered by the recording angle within the scope of the tomosynthesis imaging (5) to be parameterized is specified and / or in that, in order to specify the position (47) of the slice to be imaged, the position of an isocenter (48) of the tomosynthesis imaging (5) to be parameterized is specified. [11] Computer-implemented method according to one of the preceding claims, characterized bythat a respective tomosynthesis imaging (5, 8) is carried out repeatedly, wherein the three-dimensional image data (7) of the respective most recently carried out tomosynthesis imaging (8) are respectively used to parameterize the respective subsequent tomosynthesis imaging (5) as tomosynthesis imaging (5) to be parameterized in order to specify the position (1) and / or the thickness (2) of the respective slice (3) of the patient (4) to be imaged. [12] Computer-implemented method according to one of the preceding claims, characterized bythat the position (11) of the longitudinal axis (12) of the object (10) and / or further image data which are acquired as part of the tomosynthesis imaging (5) to be parameterized are evaluated in order to provide at least one control parameter (33) for controlling an actuator for the automated movement of the object (10) to the target position (13) and / or are visualized for a user via a display device (34) in order to assist the user in guiding the object (10) to the target position (13). [13] Computer-implemented method according to one of the preceding claims, characterized bythat after determining the position (1) and / or the thickness (2) of the layer (3) to be imaged, an imaging device (6) and / or components of the imaging device (6) are controlled, so that three-dimensional image data describing the layer (3) to be imaged are acquired by the imaging device (6) within the framework of a tomosynthesis imaging (5) parameterized by the position (1) and / or the thickness (2). [14] Processing device, characterized by that it is designed to carry out the computer-implemented method according to one of the preceding claims. [15] Imaging device which is designed to acquire three-dimensional image data of a slice (3) of a patient (4) to be imaged by means of tomosynthesis imaging (5), characterized by that the imaging device is configured to carry out the computer-implemented method according to one of claims 1 to 13. [16] A computer program comprising program instructions adapted to implement the computer-implemented method according to any one of claims 1 to 13 when executed on a data processing device (35). [17] A data carrier comprising the computer program (38) according to claim 16.

Images