Method and device for assisting in the treatment of bone fractures

Abstract

A method for assisting a user in the treatment of bone fractures with a stabilization implant, wherein a database is used in which a plurality of comparison data sets are stored, each comprising at least one 3D image data set of at least one bone, wherein each comparison data set is assigned to an examination object in which a comparison implant was implanted before the 3D image data set was acquired, comprising the steps: - Taking at least one two-dimensional X-ray image of a target bone to which the stabilisation implant is to be attached using an X-ray machine, - 2D-3D registration of the X-ray image with 3D image data sets of a selection of the comparison data sets determined by a selection criterion or of all comparison data sets and calculation of a deviation value, which is a measure of the deviation of the 3D image data set registered to the X-ray image from the X-ray image, for each comparison data set for whose 3D image data set a 2D-3D registration was carried out, by a computing device, - Determination of a reference data set by selecting the comparison data set with the lowest deviation value by the computing device, - on the one hand, determining at least one implantation parameter which describes a parameter of the implant and / or of a fastening element for fastening the implant, using the reference data set, wherein the implantation parameter or at least one of the implantation parameters is determined by adopting a comparison parameter stored in the reference data set which describes a parameter of the comparison implant or of a comparison fastening element, and / or by calculation from the 3D image data set of the reference data set, and / or on the other hand, displaying data of the reference data set on a display device by the computing device.

Description

[0001] The invention relates to a method and a device for assisting a user in the treatment of bone fractures with a stabilization implant.

[0002] Bone fractures are often treated by attaching an implant to a bone. One advantage of using implants in the treatment of bone fractures is that the bone can be loaded again quickly. In addition, the bone fragments are fixed in the correct places, which allows for reliable bone repair, especially in complex fractures. The implants often consist of plates that are attached to broken bones. During implantation, a treating physician must decide where an implant will be placed and how fixation screws will be applied to secure individual bone fragments. With many implants, the angle at which fasteners are attached to the bone can be selected.At the same time, it is important to avoid a fastener penetrating the bone and penetrating the tissue or a joint.

[0003] The standard method for determining the position of the implant and the position of the fixation elements is to take several x-ray images intraoperatively and use these to plan and verify the implantation. However, the surgeon must not only estimate the optimal position of the implant from a few individual two-dimensional x-ray images, but also which fixation elements, in particular how long the fixation elements, should be used to secure the implant and at what angle they should be inserted into the implant. Modern implants often allow the screw angle to be varied within a range of approximately + / - 15°. While this improves the possibility of attaching bone fragments to the implant, it also increases the risk of penetrating the bone and damaging the surrounding tissue or a joint.In addition, it can be difficult to take X-ray images from different perspectives during surgery, so that often only a single projection of the bone is available for planning and verifying the implantation.

[0004] Published applications DE 10 2007 034221 A1 and DE 10 2006 048451 A1 each disclose a method for the virtual adaptation of an object to a patient's body part. The object can be taken from a database for this purpose. Published application WO 2012 / 113030 A1 discloses a computer-implemented method for the virtual adaptation of an orthopedic implant to a patient's joint. Published application US 2011 / 0093108 A1 discloses a method for producing an implant or surgical instrument adapted to a patient's body part.

[0005] It is therefore the object of the method according to the invention to provide an improved method for assisting a user in the treatment of bone fractures with a stabilization implant.

[0006] The object is achieved according to the invention by using a database in which a plurality of comparison data sets are stored, each of which comprises at least one 3D image data set of at least one bone, wherein each comparison data set is assigned to an examination object in which a comparison implant was implanted before the 3D image data set was recorded, comprising the steps: - Taking at least one two-dimensional X-ray image of a target bone to which the stabilisation implant is to be attached using an X-ray machine, - 2D-3D registration of the X-ray image with 3D image data sets of a selection of the comparison data sets determined by a selection criterion or of all comparison data sets and calculation of a deviation value which is a measure of the deviation of the 3D image data set registered to the X-ray image from the X-ray image, for each comparison data set for whose 3D image data set a 2D-3D registration was carried out by a computing device, - Determination of a reference data set by selecting the comparison data set with the lowest deviation value by the computing device, - on the one hand, determining at least one implantation parameter which describes a parameter of the implant and / or of a fastening element for fastening the implant, using the reference data set, wherein the implantation parameter or at least one of the implantation parameters is determined by adopting a comparison parameter stored in the reference data set which describes a parameter of the comparison implant or of a comparison fastening element, and / or by calculation from the 3D image data set of the reference data set, and / or on the other hand, displaying data of the reference data set on a display device by the computing device.

[0007] The invention is based on the idea that stabilization implants for the treatment of bone fractures are often arranged in a relatively similar manner, since human bones often do not differ greatly in the typical fracture areas within certain patient groups, and the areas where fractures occur are also similar. It can therefore generally be assumed that a database containing data on a large number of previous implantation procedures already contains data on an implantation that corresponds to the implantation to be performed, or that was at least performed on a similar bone.

[0008] If such a database of previous implantations exists, a suitable data set for the current implantation must be selected, and it must be checked which implantation parameters can be adopted from this data set. To determine a suitable comparison data set from a previous examination, the method according to the invention first takes a two-dimensional X-ray image of the target bone to which the stabilization implant is to be attached. This X-ray image can be acquired using any X-ray device, for example, a C-arm. Of course, it is also possible to acquire multiple images in this step to improve the selection and registration of the comparison data set. However, this is not absolutely necessary in the method according to the invention.

[0009] The next step of the method according to the invention is to determine which previous implantation the implantation to be performed most closely corresponds to. This is equivalent to selecting a comparison data set. The comparison data sets used for this purpose have already been stored in a database. It is essential that for each examination object for which a comparison data set is available, an implantation of a stabilization implant was performed before the 3D image data set was acquired. At least one 3D image data set should be available for each comparison data set. The 3D image data set can, for example, be a CT scan of the examination object. However, the 3D image data set can also be pre-segmented image data or image data from other imaging procedures.

[0010] To ensure high quality of the comparison data sets, only data sets from examination subjects in which the implantation was classified as successful should be saved as comparison data sets. The 3D image data sets can be verification data sets, i.e., images of examination subjects for three-dimensional verification of the correct fit of the stabilization implant. In this case, the 3D image data set depicts both the target bone and the stabilization implant and the attachment elements. However, it is often advantageous to use data sets acquired after the stabilization implant has already been removed as comparison data sets.

[0011] In addition to the 3D image datasets, the comparison datasets can contain a variety of additional information. For example, additional segmentation information, the type and / or other details, such as size information, of the implant or the fastening elements, the position and / or orientation of the implant and / or the fastening elements, or other information about the implantation, the patient's age, the patient's gender, or similar information can be stored in the comparison dataset.

[0012] This additional information can be used, in particular, to preselect the comparison data sets whose 3D image data set is to be used for 2D-3D registration with the X-ray image. 2D-3D registrations can be computationally demanding. Since a database containing numerous stabilization implantation procedures can potentially be very large, it is advantageous to reduce the number of 3D image data sets used for 2D-3D registration with the X-ray image. For example, only those data sets can be used in which the same bone was fractured and / or in which the age of the other examination subject is close to the age of the patient whose bone fracture is to be treated.

[0013] 2D-3D registrations are an active field of research. Therefore, 2D-3D registration will only be explained here using a few exemplary embodiments. For example, it is possible to first generate a projection image in a specific projection direction from the 3D image dataset that is to be registered to the 2D X-ray image. The projection direction can be determined from a known acquisition direction of the 2D X-ray image and a known orientation of the 3D image dataset. However, it is also possible to determine it during the registration process. Once such a projection image has been created, 2D-3D registration can be performed like a normal 2D-2D registration.

[0014] In this case, it is possible, for example, to directly use a correlation function between these two images. Alternatively, an edge detector can be used first and the images can then be correlated. It is also possible to detect features, for example, using a scale-variant feature transformation, in one of the images and then detect these features in the second image. Compared to a direct correlation of the images, approaches based on feature, corner, or edge detection have the advantage that distortions between the images can be detected, and these distortions can potentially be used to determine a correction of the projection direction.

[0015] Alternatively or additionally, it is also possible to perform 2D-3D registration using segmentation algorithms or existing segmentation data. In this case, only the segmented regions or their boundaries need to be aligned. Methods for both two-dimensional and three-dimensional registration are known for this purpose.

[0016] For each 3D image dataset, and thus for each comparison dataset that includes this dataset, a deviation value is determined after registration or even during registration. Many registration methods already utilize the minimization of a deviation value to optimize registration. In these cases, it is possible to use this value as the deviation value for optimal registration. However, it is also possible to first perform image registration and then, for example, perform a 2D projection of the 3D image dataset and calculate a cost function between the two 2D images, such as a standard deviation. It is often advantageous to apply this cost function to processed image data, for example, data that has already been processed with an edge detector or similar.

[0017] After a deviation value has been determined for each comparison data set, the comparison data set with the lowest deviation value is designated as the reference data set. It should be noted that although this step is described as occurring after the 2D-3D registration step and the determination of the deviation value, a mixture of these steps is of course also possible. Particularly on systems with limited memory, it can be advantageous to keep only a few deviation values ​​and data sets in memory. In this case, for example, after each registration it can be checked whether a deviation value is smaller than a previously calculated deviation value and, in this case, the data set can be saved as a potential reference data set. The potential reference data set saved after all 3D image data sets have been processed is then the actual reference data set.Likewise, a multitude of other implementations are conceivable, but they all lead to the same result, namely the determination of the comparison data set, which includes the 3D image data set with the lowest deviation value, as the reference data set.

[0018] In the simplest case, the reference dataset can be used to provide the user with a representation of the 3D image dataset of the reference dataset. This shows the user a bone very similar to the bone into which the implant is to be inserted, with or without an implant attached. Of course, in this case, virtual implant models or models of attachment elements can also be displayed and positioned along with the 3D image dataset. This enables implant planning in three dimensions, which can significantly improve planning quality and safety.

[0019] Alternatively or additionally, one or more implantation parameters can also be determined directly. For example, the position of the implant, the implant type, the length or type of a fastening element, or even the insertion angle of a fastening element can be suggested directly. In particular, the previously described representation of the 3D image dataset creates a very easy-to-use and very powerful planning system, as a multitude of implantation-relevant parameters can be suggested to the user. These parameters can be immediately visually displayed on a 3D model based on data from the comparison dataset, and can be interactively adjusted by the user.

[0020] It is possible for a comparison implant depicted in the 3D image data set to be segmented by the computing device before the comparison data set comprising the 3D image data set is stored in the database or when this comparison data set is retrieved from the database, wherein in particular the segmented comparison implant is not taken into account during registration and when calculating the deviation value. This is particularly advantageous if verification images are stored in the database, i.e. images that were taken to check the correct fit of an implant, but the method according to the invention is to be used in a phase of implantation planning in which the implant has not yet been arranged on the target bone. In this case, the 3D image data sets also include the implant. In particular, when CT data sets are used, such implants, which are often made of metal, display a very strong contrast.These 3D image datasets are now to be registered with a 2D X-ray image that does not depict an implant. This can lead to significant registration errors. Therefore, it is advantageous to determine the position and extent of the implant and not use the points of the 3D image dataset covered by the implant for registration with the X-ray image and to determine the deviation value.

[0021] However, segmenting the implant also has the advantage of allowing implantation parameters to be obtained directly. For example, angles of fastening elements can be easily identified during segmentation and, after selecting a comparison dataset as a reference dataset, can be immediately made available as implantation parameters. The same applies to the position and length of the fastening elements, as well as the position and orientation of the implant itself.

[0022] The determination of the implantation parameters can be carried out in a variety of ways. For example, it is possible that the implantation parameters or at least one of the implantation parameters can be determined by - Adoption of a comparison parameter stored in the reference data set, which describes a parameter of the comparison implant or a comparison fastening element, and / or - Calculation from the 3D image data set of the reference data set, whereby in particular a segmentation of the 3D image data set is used and / or a bone model is adapted to the 3D image data set, and / or - Output of at least one overlay of a two-dimensional representation of the 3D image data of the reference data set and a two-dimensional representation of a three-dimensional implant model and / or at least one three-dimensional fastening element model on the display device and interactive adaptation of the implant model and / or the fastening element model by a user.

[0023] In the simplest case, comparison parameters are already stored in the reference dataset. This makes it possible for implantation parameters such as the implant type, the implant position, the type of fastening elements used, or similar, that were used during the previous implantation to be saved. This is possible if these parameters were entered manually. For example, it is possible for software used to plan the implantation to add these parameters to a patient dataset, and for this patient dataset to be anonymized and included in the comparison dataset when the comparison dataset is created. In addition or alternatively, it is also possible for the implantation parameters to have been obtained from a previous segmentation of the 3D image data, for example, during the verification of the implantation.

[0024] Of course, implantation parameters can also be obtained directly from the 3D image dataset of the reference dataset. For example, if an implant is depicted in this dataset, the implant can be segmented in the dataset, as already described, and from this segmentation, the parameters, or at least parts of the parameters, of the implant and the fastening elements can be obtained. Furthermore, a bone model can be obtained from the 3D image dataset. The bone model can be calculated by segmenting the dataset or by adapting an existing bone model to the 3D image dataset. The bone model thus determined can then be used to optimize the implantation parameters. For example, the angles and lengths of fastening elements can be adjusted to achieve optimal hold without penetrating the bone and endangering the tissue or joints.These parameters can also be used to determine particularly advantageous implant positions or the implant type. Once the complete bone model is known, implant adaptation can be viewed and solved simply as a multidimensional optimization problem, with optimization parameters such as maximum fixation element length and the greatest possible coverage of the fracture area by the implant.

[0025] As mentioned at the beginning, an overlay of the 3D image data of the reference dataset with an implant model can also be displayed. If the parameters of the implant model or fastener models can be interactively adjusted in this overlay, this can also be used for implantation planning, whereby the implantation parameters can be determined automatically. Of course, the described methods for determining implantation parameters can also be combined, and individual parameters can be specified by a user.

[0026] Performing a 2D-3D registration can often be computationally very complex if the direction of the 2D image relative to the 3D image dataset is unknown. It is therefore advantageous to use an orientation and / or location information stored in the comparison dataset during the 2D-3D registration step to restrict the search space for the registration parameters. In this case, the projection direction and the advance of the image data can be optimized within a small angular range or for small local shifts, which significantly accelerates registration.

[0027] The method can be further significantly accelerated by limiting the number of comparison data sets examined. As explained, this is possible with the aid of a selection criterion. The selection criterion for selecting the comparison data sets can be a comparison of at least one piece of additional information stored with the comparison data set, in particular a patient's age, the imaged bone, a bone diameter and / or information about the comparison implant, or an image parameter determined by the computing device from the respective 3D image data set, in particular a parameter determined by segmenting the 3D image data set, with a predetermined or predeterminable value, in particular the upper or lower limit of a value interval. For example, a maximum and minimum bone diameter for the comparison data sets can be determined from the diameter of a bone in the X-ray image.A search is quick if such a parameter is already stored in the comparison data sets. However, this is not always the case, so such a parameter can also be determined directly from the 3D image data sets.

[0028] It is possible for the implantation parameters to be displayed on the display device, with the representation being particularly graphical by displaying an implant model and / or at least one fastening element model. Such a representation makes the parameters particularly easy for a user to recognize. Furthermore, such a representation can be used for interactive adjustment of the parameters, as explained.

[0029] It is often not possible to transfer information obtained from a smaller bone to a larger bone, or vice versa. Therefore, in the method according to the invention, it is advantageous if size information is available for the X-ray image and the 3D image data sets. Such size information is automatically available if a known implant is depicted in the X-ray image or in the 3D image data sets. However, if the method according to the invention is to be used to plan an implantation, an implant is generally not located in the image area. Therefore, it is advantageous if a calibration body for size calibration is arranged in the image area of ​​the X-ray image before the X-ray image is taken. This calibration body should advantageously be arranged at the height of the target bone. Of course, other methods for determining a size scale can also be used.In some circumstances, it may also be advantageous to use a calibration body to acquire the 3D image data sets. However, it is also possible, for example, to obtain a size scale using a stabilization implant present in the image area or to use known calibrations of the acquisition device.

[0030] In the human body, many bones are mirror-symmetric. To achieve greater efficiency in the database used, it is therefore possible to use comparison data sets for target bones that are present twice in the body of the subject under examination and that are mirror-symmetrical. These data sets may also be used for registration and calculation of the deviation value. Their 3D image data sets do not depict the target bone, but rather the bone that is mirror-symmetrical to the target bone. In this case, the X-ray image or the 3D image data set is mirrored.

[0031] Registering the X-ray image with the 3D image dataset, and thus determining the correct reference dataset or the correct comparison parameters, is particularly easy if the stabilization implant is depicted in both the 3D image dataset and the X-ray image. Furthermore, it can be advantageous to determine additional implantation parameters for an implant already positioned in the bone. Therefore, after the implant has been inserted into the body of the subject, at least one additional X-ray image can be taken, and at least one implantation parameter can be determined again and / or another implantation parameter can be determined.

[0032] The method according to the invention can be used to determine all or a selection of the implantation-relevant parameters. Thus, the implantation parameter or one of the implantation parameters can be an implant type, an implant position, a fastening element type, a fastening element length, and / or an insertion angle of at least one fastening element. In particular, the fastening elements can be screws. In this case, parameters describing the screw head, the thread, or the like can also be determined.

[0033] Often, only limited computing capacity is available at the site of surgery, and results must be presented quickly to the procedure user. Therefore, it is advantageous to segment a 3D image dataset before saving the comparison dataset containing the 3D image dataset in the database, and to store the segmentation data in the comparison dataset. The segmentation data is used for 2D-3D registration and / or the calculation of the deviation value and / or the implantation parameter. In particular, it is also possible to review and / or correct such segmentation before saving the comparison dataset.

[0034] In addition, before saving the comparison dataset in the database or retrieving the comparison dataset from the database by the computing device, the comparison implant can be segmented in the 3D image dataset and registered with a model implant, and the coordinate system of the 3D image dataset can be transformed into a coordinate system of the model implant. This is particularly advantageous if a stabilization implant is depicted in both the 3D image dataset and the X-ray image in such a way that it is already in a final position. In this case, 2D-3D registration is no longer necessary, as the 3D image dataset is already registered to the implant. Of course, a subsequent 2D-3D registration with a restricted parameter space can be performed to further improve the registration result.

[0035] In addition, the invention relates to a device for assisting a user in the treatment of bone fractures with a stabilization implant comprising - a storage device for providing a database in which a plurality of comparison data sets are stored, each comprising at least one 3D image data set of at least one bone, each comparison data set being associated with an examination subject in which a comparison implant was implanted before the 3D image data set was recorded, - a registration device for 2D-3D registration of a previously recorded X-ray image with 3D image data sets of a selection of the comparison data sets or all comparison data sets determined by a selection criterion, - a calculation device for calculating a deviation value, which is a measure of the deviation of the 3D image data set registered to the X-ray image from the X-ray image, for each comparison data set for whose 3D image data set a 2D-3D registration was carried out, - a reference determination device for determining a reference data set by selecting the comparison data set with the lowest deviation value, - an implant parameter determination device for determining at least one implantation parameter that describes a parameter of the implant and / or a fastening element for fastening the implant, using the reference data set and - a display device for outputting the representation of the data of the reference data set or the determined implantation parameters, wherein the device is particularly designed to carry out at least one of the methods described above. All statements regarding the method according to the invention can be applied analogously to the device according to the invention, with which the aforementioned advantages are also achieved.

[0036] As is generally known, the components of the device can be implemented by one or more computing devices and corresponding hardware and / or software components.

[0037] Further advantages and details of the process are evident from the following examples and the accompanying drawings. They show: Fig. 1 shows an example of the arrangement of an implant on a radius bone, Fig. 2 a flowchart for generating a comparison data set for the database used in the method according to the invention, Fig. 3 a flowchart of an embodiment of the method according to the invention, Fig. 4 a flowchart of a further embodiment of the method according to the invention, Fig. 5 a flowchart of a third embodiment of the method according to the invention, and Fig. 6 a schematic representation of an embodiment of a device according to the invention for assisting a user in the treatment of bone fractures with a stabilization implant.

[0038] Fig. 1 shows the prior art arrangement of a stabilizing implant 1 on a bone 2. The stabilizing implant 1 essentially has the shape of a plate with multiple holes and can be angled to better adapt to the bone. The angulation is usually predetermined and is not changed during adaptation to the patient. The stabilizing implant 1 has a plurality of openings 3 and 4 for inserting fastening elements. Some of the openings 3 determine the position of the fastening element, but allow a variation of the angle of the fastening element within a range of, for example, 30°. However, after the fastening element has been inserted, this angle is fixed and can no longer be changed. Other openings 4 for inserting fastening elements allow a displacement of the fastening element in one or two dimensions.The fastening elements serve, on the one hand, to attach the stabilization implant 1 to the unbroken part of the bone 2 and, on the other hand, to secure bone fragments in such a way that they can grow back together with the bone. When planning the positioning and attachment of the implant, it is essential to ensure the best possible hold of the implant on the bone and the bone fragments, while avoiding penetrating the bone and thus damaging the tissue and / or a joint. Often, only X-ray images from one perspective are available during surgery to plan the positioning and attachment of the stabilization implant 1. The described procedure is intended to improve this.

[0039] The described procedure uses comparison data sets from previous implantations. Fig. Figure 2 shows a flow chart of how such a comparison data set can be created. Typically, this is Fig. The method shown in Figure 2 is only used after an implantation has been classified as successful. However, it is also possible to use the method during the verification phase of the implantation and subsequently discard data sets from implantations that were assessed as unsuccessful or poor.

[0040] After the method begins in step S10, a three-dimensional image dataset of a bone in which a stabilizing implant was previously implanted is created in step S11. It is possible that the stabilizing implant is still in place in the bone at the time of image acquisition in step S11, but an image of an already healed bone without an implant can also be acquired. The 3D image data is acquired, for example, using a computed tomography scanner.

[0041] In step S12, it is determined whether a stabilization implant is located in the image area of ​​the 3D image dataset acquired in step S11. It is possible for a user to manually set whether a stabilization implant is present on the bone before or after the image acquisition in step S11. Alternatively, an image recognition algorithm can also be used in step S12 to determine whether the image dataset acquired in step S11 includes a stabilization implant. Such a stabilization implant is easily recognizable, at least in a CT image, since such implants are often made of metal and therefore have a high X-ray absorption coefficient.

[0042] If no stabilization implant is depicted in the 3D image dataset, the method continues directly in step S15. If a stabilization implant is included in the 3D image dataset, it is segmented in step S13. Due to the high absorption coefficient of most stabilization implants for X-rays, segmentation of the supporting implant is usually easy, robust, and possible with high resolution. Therefore, during segmentation of the implant in step S13, further information about the implant can also be obtained, such as the implant position and orientation, as well as at least the length, location, and orientation of the fastening elements. This data can be saved in the comparison dataset in a later step.

[0043] In particular, if the comparison data set is also to be used for registration with X-ray images that also depict a stabilization implant, the implant segmented in step S13 can also be registered with a model implant in step S14. This registration with a model implant can be used, on the one hand, to determine implantation parameters; on the other hand, after registration, the 3D image data set can be transformed into a coordinate system that corresponds to the coordinate system of the model implant. This has numerous advantages. Firstly, for a known type of model implant, this registration and coordinate transformation can be used to determine a size scale for the 3D image data set and ensure that all 3D image data sets stored in comparison data sets have a common size scale.This common size scale significantly facilitates the use of comparative data sets from different sources. Furthermore, this preparatory transformation can significantly accelerate 2D-3D registration or the determination of a deviation parameter for cases where a stabilization implant is depicted in the X-ray image. This will be discussed below with reference to . Fig. 5 is explained in more detail.

[0044] Regardless of whether a stabilization implant is located in the image area of ​​the 3D image dataset, a segmentation of at least the bone where the stabilization implant is located is subsequently performed in step S15. Additionally, other bones or other organs can also be segmented to facilitate later registration of the 3D image dataset with the X-ray image. Segmenting the 3D image data before saving the comparison dataset has the particular advantage that the intraoperative part of the procedure is less computationally intensive. This allows for a reduction in the computing power required by the device in the operating room, and also for the procedure to be accelerated, resulting in a shorter operating time.Similar to the segmentation of the implant in step S13, parameters of the bones, such as the diameter of a bone or the like, can also be determined during the segmentation of the bones in step S15.

[0045] In step S16, all acquired data, i.e., at least the 3D image dataset and other information relating to bone and / or implant segmentation, patient information, bone parameters, or implantation parameters, are stored together in the comparison dataset. The comparison dataset can be saved as a single document, but it is also possible to store the comparison dataset, for example, in a database. For example, a core dataset can contain only parameters that require relatively little memory, such as bone type, age, bone diameter, or the like, while relatively extensive data, such as the 3D image dataset, can be stored only as references. However, a variety of other formats are also possible.In particular, the comparison data set can also be saved in a format that allows parts of the comparison data set to be displayed in common treatment planning systems or other systems for displaying medical data sets from different data sources.

[0046] After saving the comparison data set, the method is terminated in step S17. As already mentioned, it is possible that a comparison data set is already created before the surgical result has been evaluated. In this case, the comparison data set can also be used to evaluate the surgical result, and the comparison data set can be discarded if the implantation result is not considered satisfactory.

[0047] Fig. 3 shows a first exemplary embodiment of the method for supporting the treatment of bone fractures with a stabilization implant. At the beginning of the method in step S21, a database with a large number of comparison data sets is already available. In addition, a treatment object whose bone fracture is to be treated is prepared such that X-ray images of the treatment object can be taken and the stabilization implant can be implanted in the treatment object. In step S22, an X-ray image of the treatment object is then first taken. The image area of ​​the X-ray image is selected such that the bone to which the stabilization implant is to be attached is located within the image area. In the exemplary embodiment described here, it is assumed that the stabilization implant is still outside the body and thus not within the image area of ​​the X-ray image.In the further course of the procedure, the aim is to identify a bone from the comparison data sets that is as similar as possible to the target bone of the treatment object.

[0048] To facilitate the determination of the reference data set, the X-ray image can be preprocessed in step S23. In this step, the X-ray image is processed in such a way that subsequent registration in step S27 is facilitated. For example, edge detection can be performed, in particular by applying a Sobel operator to the X-ray image, or corner detection or scale-invariant features can be performed.

[0049] Parallel to the acquisition and processing of the X-ray image, comparison data sets can be selected and prepared for registration. In step S24, the relevant comparison data sets are first selected. This makes it possible for comparison data sets for a large number of bones to be stored in a database. In this case, only the comparison data sets that relate to a bone that corresponds to the target bone should be selected. In addition, a variety of other selection criteria can be applied. For example, the age of the examination object from which the comparison data set originates can be restricted to an age range that corresponds to the age range of the treatment object. Under certain circumstances, information obtained from the X-ray image acquired in step S22 can also be used in step S24.For example, in step S24, a bone diameter of the X-ray image taken in step S22 can be recognized and only comparison data sets can be selected that relate to a bone having a similar bone diameter.

[0050] In addition, a variety of other selection criteria are conceivable to restrict the selection of relevant comparison data sets before registration and can be applied in step S24.

[0051] After determining a selection of comparison data sets in step S24, a 2D projection of the 3D image data set contained in the comparison data set is generated in step S25. Generating a 2D projection allows for particularly simple registration of the 3D image data with the X-ray image, since only two-dimensional data needs to be registered. If the orientation of the 3D image data set is known, the projection direction can be adapted to the acquisition direction of the X-ray image. If the orientation is unknown, a predefined number of projection images can be generated that correspond to different projection directions. During the registration process, the projection direction can then be adjusted, and a new projection can be calculated for the newly determined projection direction.

[0052] After calculating the 2D projection in step S25, a processing step equivalent to the one applied to the X-ray image in step S23 is applied to the 2D projection image in step S26. For example, edge detection or scale-invariant feature recognition can also be used here. In step S27, the X-ray image preprocessed in step S23 and the 2D projection prepared in step S26 are registered, and a deviation value for the registration is calculated. Many registration methods already use a deviation value during registration. This deviation value can be directly adopted as the deviation value.

[0053] In step S28, a check is then made to determine whether the registration and determination of the deviation value has already been performed for all comparison data sets determined in step S24. If this is not the case, the process is repeated from step S25.

[0054] After steps S25 - S27 have been performed for all comparison data sets, in step S29 the data set whose 3D image data set has the smallest deviation value when registered with the X-ray image is determined as the reference data set.

[0055] In step S30, at least one implantation parameter is then determined. The implantation parameter can be determined simply by adopting a parameter stored in the reference data set as the implantation parameter. It is also possible, however, to segment an implant depicted in the 3D image data set and to obtain implantation parameters from the segmentation of the implant. Segmentation of the bone in the 3D image data set or adaptation of a bone model to the 3D image data set can also be used to determine an optimal value for an implantation parameter. In this case, a three-dimensional bone model is available as an intermediate result, and an optimization algorithm can be used to determine the implantation parameters depending on predefined optimization parameters, such as maximizing the length of the fastening elements.The implantation parameters can also be determined interactively by overlaying the representation of an implant model with the 3D image data or the segmentation data of the 3D image dataset and manipulating the representation by a user. After determining the implantation parameters, the method ends in step S31.

[0056] Fig. 4 shows another embodiment of a method for assisting in the treatment of bone fractures. In Fig. The method presented in Figure 4 takes advantage of the fact that segmentation information is already stored in the comparison data sets in the database. This can potentially result in a faster and more robust method.

[0057] After the method begins in step S41, an x-ray image is acquired in step S42. The beginning of the method thus corresponds to the previously described method. However, for further processing of the x-ray image, at least the bone to which the stabilization implant is to be attached is segmented in the x-ray image in step S43. After the bone has been segmented in the x-ray image, the comparison data sets to be considered are selected in step S44. This selection is carried out in a similar way to step S24. It is particularly advantageous here to also use parameters determined during the segmentation of the bones in step S43 to select the comparison data sets.

[0058] In step S45, a 2D-3D registration of the segmented X-ray image with the segmented 3D image dataset is performed. Such registration is particularly simple with segmented data, as significantly less data needs to be considered. In addition, a deviation parameter is determined in step S45. This can also be determined from the segmentation data; however, after successful registration, it is also possible to determine the deviation parameter directly from the X-ray image and the 3D image datasets, for example, by generating a suitable projection of the 3D image dataset and then using a cost function, in particular a standard deviation, to compare these images. Of course, the use of further processed data to calculate the cost function is also possible. In particular, edge detectors can be applied to the data and / or offsets can be subtracted.

[0059] The steps S46 - S49, i.e. the checking whether the registration and deviation determination have been carried out for all comparison data sets, the determination of a reference data set, the determination of the implantation data and the end of the method, correspond to the steps S28 - S31 in the method with reference to Fig. 3 described procedures.

[0060] Fig. 5 shows a flow diagram of a third embodiment of a method for assisting a user in the treatment of bone fractures. A key difference from the methods already described is that in this embodiment, an image of a stabilizing implant is already present in the x-ray image. This embodiment can therefore be used primarily to redetermine implantation parameters after an implant has been initially positioned on the bone or to determine further implantation parameters. For example, the type of implant and an initial positioning can be carried out using one of the methods already described, and then the method described below can be used to determine, in particular, the angle and length of the fastening elements as well as the exact position of the implant.

[0061] Alternatively or additionally, it is also possible to use the procedure described below before a stabilization implant is inserted into the body of a patient. As explained in more detail later, a virtual implant can be superimposed on the X-ray image as part of the procedure. The resulting overlay image can be used in the subsequent procedure like an X-ray image that already includes an implant. In this case, for example, interactive optimization of a given implant is possible.

[0062] After the method begins in step S51, an x-ray image is taken in step S52. As described at the beginning, the following description assumes that a stabilization implant that is already located in the body of the treatment subject is depicted in the image area of ​​this x-ray image. If there is no stabilization implant in the body of the treatment subject, a virtual implant can be superimposed on the x-ray image after the x-ray image has been taken in step S52 in a step not shown. To do this, a user selects an implant they wish to place in the x-ray image, taking into account the x-ray image taken in step S52, and determines the position and orientation of the implant. The computing device can then calculate a superimposed representation of the implant model and the x-ray image, which resembles an x-ray image with an implant at this location.This is particularly possible because stabilization implants generally exhibit significantly greater absorption than the rest of the tissue. This allows the image area covered by the virtual stabilization implant to be replaced with an image of the stabilization implant in a simple approximation. This allows a virtual X-ray image to be used in the subsequent procedure. In all subsequent procedural steps, whenever reference is made to an X-ray image, this reference can also be interpreted as a reference to a virtual X-ray image.

[0063] After the X-ray image has been acquired, the implant is segmented in the X-ray image in step S53. In step S54, the implant segmented in step S53 is registered to a model implant. In particular, a coordinate transformation into the coordinate system of the implant model can also be performed here. Subsequently, in step S55, a selection of comparison data sets is determined that are to be registered with the X-ray image. This selection is essentially carried out as in step S24 in Fig. 3, in step S55, however, it is particularly advantageous to use parameters to select the comparison data sets that describe the implant or the implant position.

[0064] Advantageously, the comparison data sets are already transferred to the coordinate system of a model implant. This also eliminates the need to register the 3D image data sets, as both data sets already have the same coordinate system. A deviation value can therefore be determined directly in step S56. In particular, segmentation information about bones stored in the comparison data set and a segmentation of the bones in the 3D image can be used for this purpose. However, a variety of other methods are also available, as described, for example, with reference to step S27 in Fig. 3 are described.

[0065] The further steps, the check whether all comparison data sets have been processed in step S57, the determination of a reference data set in step S58, the determination of at least one implantation parameter in step S59 and the end of the method in step S60 correspond to steps S28 - S31 in Fig. 3.

[0066] Fig.6 schematically shows an embodiment of a device 5 for assisting a user in the treatment of bone fractures with a stabilization implant. The device has a storage device 6 for providing a database in which a plurality of comparison data sets are stored, each comprising at least one 3D image data set of at least one bone, wherein each comparison data set is associated with an examination subject in which a comparison implant was implanted before the 3D image data set was acquired. A registration device 7 serves for the 2D-3D registration of a previously acquired X-ray image with 3D image data sets of a selection of the comparison data sets determined by a selection criterion, or of all comparison data sets.In addition, the device comprises a calculation device 8 for calculating a deviation value, which is a measure of the deviation of the 3D image data set registered with the X-ray image from the X-ray image, for each comparison data set for whose 3D image data set a 2D-3D registration was performed. A reference determination device 9 is used to determine a reference data set by selecting the comparison data set with the lowest deviation value. An implant parameter determination device 10 serves to determine at least one implantation parameter, which describes a parameter of the implant and / or a fastening element for fastening the implant, using the reference data set. In addition, the device 5 has a display device 11 for outputting the representation of the data of the reference data set or the determined implantation parameters.

[0067] The device 5 may comprise further components, in particular an X-ray device 12 and / or an interface for providing X-ray images, in particular via a network, as well as one or more input devices 13 for capturing user inputs.

[0068] The device 1 is designed in particular to carry out one of the methods described above.

[0069] Of course, several of the components of the device 5 can also be formed by the same elements. For example, a single computing device 14 can fulfill the functions of the recording device 7, the calculation device 8, the reference determination device 9, and the implantation parameter determination device 10. However, the components can also be designed separately. In particular, the storage device 6 can be spatially spaced from the other components and connected via a network, in particular the Internet. The recording device 7, the calculation device 8, the reference determination device 9, and the implantation parameter determination device 10 can also be connected to the display device 11 and the other components via a network.

[0070] Of course, the features of the described embodiments can be combined in many different ways.

[0071] 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.

Claims

[1] A method for assisting a user in the treatment of bone fractures with a stabilization implant, using a database in which a plurality of comparison data sets are stored, each comprising at least one 3D image data set of at least one bone, each comparison data set being associated with an examination object in which a comparison implant was implanted before the 3D image data set was acquired, comprising the steps: - Taking at least one two-dimensional X-ray image of a target bone to which the stabilisation implant is to be attached using an X-ray machine, - 2D-3D registration of the X-ray image with 3D image data sets of a selection of the comparison data sets determined by a selection criterion or of all comparison data sets and calculation of a deviation value, which is a measure of the deviation of the 3D image data set registered with the X-ray image from the X-ray image, for each comparison data set for whose 3D image data set a 2D-3D registration was carried out, by a computing device, - Determination of a reference data set by selecting the comparison data set with the lowest deviation value by the computing device, - on the one hand, determining at least one implantation parameter which describes a parameter of the implant and / or of a fastening element for fastening the implant, using the reference data set, wherein the implantation parameter or at least one of the implantation parameters is determined by adopting a comparison parameter stored in the reference data set which describes a parameter of the comparison implant or of a comparison fastening element, and / or by calculation from the 3D image data set of the reference data set, and / or on the other hand, displaying data of the reference data set on a display device by the computing device. [2] Method according to claim 1, wherein a comparison implant depicted in the at least one 3D image data set is segmented by the computing device before the comparison data set comprising the at least one 3D image data set is stored in the database or when this comparison data set is retrieved from the database, wherein in particular the segmented comparison implant is not taken into account in the registration and the calculation of the deviation value. [3] Method according to claim 1 or 2, wherein the implantation parameter or at least one of the implantation parameters is determined by calculation from the 3D image data set of the reference data set, wherein a segmentation of the 3D image data set is used and / or a bone model is adapted to the 3D image data set. [4] Method according to one of the preceding claims, wherein the implantation parameter or at least one of the implantation parameters is determined by outputting at least one superposition of a two-dimensional representation of the 3D image data of the reference data set and a two-dimensional representation of a three-dimensional implant model and / or at least one three-dimensional fastening element model on the display device and interactive adaptation of the implant model and / or the fastening element model by a user. [5] Method according to one of the preceding claims, wherein in the step of 2D-3D registration an orientation or recording location information stored in the comparison data set is used to restrict a search space of the registration parameters. [6] Method according to one of the preceding claims, wherein the selection criterion for selecting the comparison data sets is a comparison of at least one item of additional information stored with the comparison data set, in particular a patient age, the imaged bone, a bone diameter and / or information about the comparison implant, or an image parameter determined by the computing device from the respective at least one 3D image data set, in particular a parameter determined by segmenting the at least one 3D image data set, with a predetermined or predeterminable value. [7] Method according to one of the preceding claims, wherein the implantation parameter is displayed on the display device, the representation being graphical by displaying an implant model and / or at least one fastening element model. [8] Method according to one of the preceding claims, wherein a calibration body for size calibration is arranged in the image area of ​​the X-ray image before the X-ray image is taken. [9] Method according to one of the preceding claims, wherein for target bones which are present twice in the body of the examination subject and are mirror-symmetrical, comparison data sets are also used in the registration and the calculation of the deviation value, the 3D image data set of which does not depict the target bone, but the bone which is mirror-symmetrical to the target bone, in which case the X-ray image or the 3D image data set is mirrored. [10] Method according to one of the preceding claims, wherein after insertion of the implant into the body of the examination subject, at least one further X-ray image is taken and at least one implantation parameter is determined again and / or a further implantation parameter is determined. [11] Method according to one of the preceding claims, wherein the implantation parameter or one of the implantation parameters is an implant type, an implant position, a fastener type, a fastener length or an insertion angle of at least one fastener. [12] Method according to one of the preceding claims, wherein a 3D image data set is segmented before storing the comparison data set comprising the at least one 3D image data set in the database, and the segmentation data are stored in the comparison data set, wherein the segmentation data are used in the 2D-3D registration and / or the calculation of the deviation value and / or the implantation parameter. [13] Method according to one of the preceding claims, wherein, before storing the comparison data set in the database or upon retrieval of the comparison data set from the database by the computing device, the comparison implant is segmented in the at least one 3D image data set and registered with a model implant, and the coordinate system of the at least one 3D image data set is transformed into a coordinate system of the model implant. [14] Device for assisting a user in the treatment of bone fractures with a stabilizing implant comprising - a storage device (6) for providing a database in which a plurality of comparison data sets are stored, each comprising at least one 3D image data set of at least one bone, each comparison data set being associated with an examination object in which a comparison implant was implanted before the 3D image data set was recorded, - a registration device (7) for 2D-3D registration of a previously recorded X-ray image with 3D image data sets of a selection of the comparison data sets or of all comparison data sets determined by a selection criterion, - a calculation device (8) for calculating a deviation value, which is a measure of the deviation of the 3D image data set registered to the X-ray image from the X-ray image, for each comparison data set for whose 3D image data set a 2D-3D registration was carried out, - a reference determination device (9) for determining a reference data set by selecting the comparison data set with the lowest deviation value, - an implant parameter determination device (10) for determining at least one implantation parameter which describes a parameter of the implant (1) and / or of a fastening element for fastening the implant (1), using the reference data set and - a display device (11) for outputting the representation of the data of the reference data set or of the determined implantation parameters, wherein the device is designed to carry out a method according to one of the preceding claims.

Images