EP1328209A1

EP1328209A1 - Method and device for navigation during medical interventions or for localising a non-osseous structure

Info

Publication number: EP1328209A1
Application number: EP01988559A
Authority: EP
Inventors: Gerald Glombitza; Marcus Vetter; Peter Hassenpflug; Carlos Cardenas; Matthias Thorn; Ivo Wolf; Volker Braun; Christoph Giess; Harald Evers; Wolfram Lamade; Hans-Peter Meinzer
Original assignee: Deutsches Krebsforschungszentrum DKFZ
Current assignee: Deutsches Krebsforschungszentrum DKFZ
Priority date: 2000-10-23
Filing date: 2001-10-23
Publication date: 2003-07-23
Also published as: DE10194616D2; WO2002034153A1; AU2002215823A1

Abstract

The invention relates to a method for navigating during medical interventions in non-osseous tissue, whereby, prior to the intervention, static image data of the non-osseous material is recorded and stored, during the intervention, dynamic image data of the non-osseous material is recorded and stored and sub-structures are extracted from said image data. The position of at least one interesting partial region of the non-osseous material can be determined and given during the medical intervention, by means of a comparison of the substructure in the static image data with the substructure in the dynamic image data, whereby a very precise navigation is made possible. Said substructures are preferably vessels. According to the invention, the integration of the invention in medical procedures may be guaranteed, whereby the effective scope of medical instruments and the image field of the image producing sensors are kept in a relationship to each other.

Description

Method and device for navigation during medical interventions or for fixing a non-bony structure

The invention relates to a method and a device for navigation during medical interventions on non-osseous structures or for the fixation of a non-osseous structure. The invention relates in particular to the field of medical interventions on soft tissue structures of a body.

So far, the person responsible for such interventions was essentially dependent on his sense of touch and sight. In particular, this means that safety distances to vital areas of the body can often not be maintained. In addition, the orientation in such a procedure is not transparent for other people involved in the medical intervention. This proves to be extremely difficult, in particular, since soft tissue structures of this type generally change their position during the procedure, so that vital areas come to a different position than at the beginning of the procedure. It is an object of the present invention to enable more objective and precise navigation during medical interventions on non-bony structures.

As a solution, the invention proposes on the one hand to record and save static image data of the non-bony structure before an intervention, to record dynamic image data of the non-bony structure during the intervention, to extract substructures from the image data and by comparing the substructure in the static image data with the substructure in the dynamic image data to assign the position of at least one area of interest to the non-bony structure during the medical intervention.

The method according to the invention is particularly suitable for non-bony structures, such as soft tissue, which are not adequately fixed by bony structures for navigation purposes. This method is preferably suitable for organs such as the liver, spleen, heart, lungs and kidneys or large-volume muscle mass.

The method according to the invention enables a geometrical assignment of image data recorded before the intervention to the current position of the partial area, so that safety distances can be maintained even in the case of adverse optical and palpatory conditions. By determining the position of the current cutting plane in the static and dynamic image data, the possibilities for orientation are improved and third parties gain insight into the navigation. It goes without saying that the solution proposed according to the invention is preferably implemented with computer support. As a result, the steps of the structural recording and the suitable assignment that are necessary according to the invention can be implemented relatively easily. It is correspondingly easy to save the necessary data and prepare it for output. On the other hand, according to the invention, the necessary information can be recorded and stored in any other conceivable way. This can also be achieved, for example, by creating a suitable model or the like. Likewise, the data can be output in any conceivable form, output of image data, in particular of three-dimensionally prepared image data, which give a therapist an overall spatial impression, is particularly suitable. On the other hand, the corresponding information can also be output to other senses of a therapist, such as the sense of touch or hearing, in order to convey to him an idea of the position of the non-bony structure.

At least one vessel, preferably a blood or lymph vessel or a hollow vessel, is preferably used as the substructure used for navigation. For example, bronchial tubes or the esophagus come into consideration as hollow vessels. Vessels, especially blood vessels, differ relatively clearly from the tissue surrounding them. This is due on the one hand to the vessel walls themselves and on the other hand to the properties of the medium present in these vessels, such as liquids or gases. In this regard, in particular with blood vessels Flow direction and speed of the blood are relatively easy to detect. A corresponding vascular tree can then be used as a substructure.

In this respect, by comparing the substructure in the static image data with the substructure in the dynamic image data, a transformation can preferably be determined which can be applied to the subarea of interest of the non-bony structure in the static image data in order to indicate the current position of the subarea. In this way, the current position of the partial area can be specified relatively safely and yet with reasonable computing effort, so that, in particular with this configuration, navigation which is fast and thus timely during a medical intervention is possible.

In this context, it goes without saying that the transformation does not necessarily have to be determined from the actual image data, but that this can also take place on the basis of other data corresponding to the image data, for example on the basis of measurement data. In this respect, the term “image data” in the present context refers to any data record that, due to a spatial assignment, enables a non-bony structure to be characterized in terms of its spatial extent.

It goes without saying that the image data in the present context can be stored and processed in a corresponding data processing system, the representation of the image data also depending on the creter design, can be selected accordingly by screens or printers or other aids as desired.

Static image data are preferably selected which have at least three spatial dimensions, so that the entire volume of the partial structure or of the overall structure can be made available visually or in another suitable form, in particular also in preparation for the medical intervention. In particular, from a three-dimensional image data set, suitable sections and viewing angles can be easily selected, which can be used to prepare for the medical intervention or for comparison with the dynamic image data. On the other hand, a higher-dimensional image data set can also be selected, which assigns further information to certain locations, such as flow directions or flow velocities, the strength of certain signal responses, frequencies or frequency spectra, chemical compositions or temperatures or the like. In particular, information or dimensions are selected that can be recorded intraoperatively quickly or reliably. For example, flow directions and flow velocities can be detected very quickly and reliably using Doppler ultrasound measurements, so that the position of a vessel or a vessel tree can thus be determined quickly and reliably. In particular, such a low-dimensional structure, such as a vascular tree, can then be used for navigation and a corresponding data acquisition and transfer is required. mation can only be based on this low-dimensional and therefore smaller amount of data.

To determine the static image data, for example, the intraoperative geometry of the site before the actual resection or other interventions can be compared with planning data obtained in a first operative step.

The dynamic image data is preferably recorded in two dimensions. This has the advantage that data acquisition can then be carried out correspondingly quickly. In particular, if a cut is to be carried out, a two-dimensional data set lying in the cutting plane can already be sufficient to enable sufficiently precise navigation. A plurality of two-dimensional data records are preferably recorded as dynamic image data. On the one hand, this can be used to find a certain level. On the other hand, these data sets can be combined to form a three-dimensional image, so that a volumetric image of the corresponding structure is obtained.

It goes without saying that three-dimensional data acquisition can also take place, the main criterion for the selection of the acquisition type for the dynamic image data being the ability to process the data as quickly as possible and, if possible, for real-time display. A method is preferably used to record the static image data, which maps the three-dimensional anatomical structure with sufficient accuracy for the purposes of the medical practitioner. The dynamic image data, on the other hand, can originate from an arbitrarily dimensional method, it being particularly important that the substructures of interest can be reproduced in the dynamic image data with sufficient accuracy for a medical intervention.

In order in particular to be able to implement the necessary transformations or calculations in real time or almost in real time, at least some of the dynamic image data or the intraoperative image data is preferably acquired only in two dimensions. In this way, the amount of data to be processed and in particular also the number of computation steps required for registration or analysis can be considerably reduced, so that real-time analysis or an analysis or display performed almost in real time can be ensured with relatively little effort.

On the one hand, sufficient 2-dimensional data, which may have been subjected to an analysis or segmentation, can be used to provide sufficient navigation capability. For example, this data can be used to check whether a certain level, for example a cutting level for a planned resection, has been reached or has been achieved with sufficient accuracy by appropriately analyzing the corresponding 2-dimensional data record and determining the data from it Data are compared with the static image data. Such a comparison can be made, for example, by finding a transformation and comparing the transformed pattern with the corresponding static data, the similarity of the patterns on the two image planes being a measure of the extent to which a corresponding plane has been reached.

In particular, it is also possible, in addition or alternatively, to subject the 2-dimensional data to a first analysis, with which essential partial data, which for example represent substructures, is extracted and then further processed. This alone results in a significant reduction in data for further processing and a consequent increase in processing speed. In particular, it is also possible to successively combine such 2-dimensionally recorded and preprocessed data into 3-dimensional and higher-dimensional representations, should this be necessary for further processing steps. In this respect, for example, 2-dimensional recordings with regard to vessel walls, flow directions or velocities, vessel contents such as air or liquids, or the like can be analyzed and, during or afterwards, successively combined into 3-dimensional data records.

It goes without saying that the recording of two-dimensional data sets and their use in the manner described above is also advantageous independently of the other features of the present invention. Preferably, at least one sensor is used to record the static image data and is also used to record the dynamic image data. In this way, the corresponding measurement data correspond in terms of their type with regard to the tissue parameters determined with the sensor, and the necessary calculations, such as the generation of corresponding image data and the determination of the corresponding transformation, can be carried out relatively easily, and accordingly also in a correspondingly short time. so that - if possible - data can be output in real time or almost in real time.

Although static and dynamic data of the same modality can usually be processed more easily and quickly, in particular compared and adjusted, it can often be advisable to use different sensors and to subject both sets of sensor outputs to suitable intermediate processing steps which solve the problem of matching the entirety reduce the data of both sets to their essential partial data, which represent substructures, for example, and thereby simplify and accelerate them. In this respect, a substructure in the two image data sets can preferably be analyzed in each case with regard to at least one location-dependent feature, such as a flow direction, a fabric wall or a flow velocity, and a comparison can be made or a transformation can be determined between the two image data sets. Accordingly, it can be advantageous, for example, to record the static image data by means of angio-CT and the dynamic image data by means of Doppler ultrasound an adequate description of the vessels as the substructures relevant for the comparison is achieved. The static image data are then preferably transformed such that flow velocities are also assigned to them and data processing can take place correspondingly quickly after the dynamic image data has been recorded.

With regard to the transformation, the procedure is preferably such that the corresponding substructure is extracted both from the static image data and from the dynamic image data and the respective data records are mapped to one another. A corresponding transformation can also be determined between two dynamic image data records recorded at different times or under different conditions. The transformation determined in this way is then applied to the data record, which contains information that cannot be found in the other data record. As a rule, the transformation is applied to the static image data in order to be able to adapt the information missing in the dynamic image data or to present it to the therapist or surgeon according to the current position or position of the non-bony structure.

The determined transformation, which - depending on the specific design - was then determined only from a comparison of part of the static image data with the dynamic image data, can then be applied to the entire static image data set, in order to determine the position of the structure also for tissue areas, the one with the sensor for dynamic Data record is not, only inaccurate, takes a lot of time or is difficult to record. A surgical tool or instrument can also be represented in this way, especially if it extends deep into the non-bony structure and can no longer be seen from the outside. In this context, it goes without saying that - as already explained above - a transformation does not necessarily have to be determined from the actual image data shown. Data corresponding to the image data, such as measurement data, can also be used directly accordingly. For example, the flow direction and speed can be used directly as an output variable for a data comparison to determine a transformation.

Depending on the course of the intervention, at least second dynamic image data can be recorded and the substructures of interest can also be extracted from the second dynamic image data. By comparing the substructure in the static image data and / or the first dynamic image data with the substructure in the second dynamic image data, the position of the subarea of interest of the non-osseous structure can then be determined and output during the medical intervention. As a result, the movement sequence of the structure or different deformations of the structure can be tracked during a medical intervention and the exact position can be determined in each case. With regard to the overall process, an intervention is preferably preceded by a computer-assisted operation planning, in which at least one three-dimensional data record is processed, which was previously determined using various measurement methods, such as ultrasound, CT, MR and the like. In this case, the respective resection areas and the like can also be defined from this. During the surgical intervention, the dynamic image data are then first determined by the non-bony structure essentially contained in the static image data being determined again as part of the dynamic image data acquisition and the data obtained in this way being compared with the static data.

For this purpose, the non-bony structure is preferably fixed, so that a data comparison can be carried out quickly and reliably. This can be done in particular by means of suitable anesthetic measures or surgical measures. These include, in particular, jet ventilation (ventilation technology to immobilize the liver) and the abdominal cavity with cloths or the fixation of the non-osseous structure using a suitable gripper arm.

In addition, a fixation shell can also be used, by means of which the non-osseous structure is fixed, the fixation preferably taking place largely in the natural form. In particular, the fixation shell can be used for this purpose on the basis of the image data obtained preoperatively, for example by rapid prototyping or similar processes. ren, be prepared or preformed. If the non-bony structure is largely fixed in its natural form, a data comparison between preoperative data and first intraoperative data to determine the stationary data can be carried out relatively quickly and reliably.

In a next step, the preoperatively obtained planning data can then be compared as static image data with the geometry of the intraoperative site and supplemented if necessary. For this purpose, for example, the fixed non-bony structure is scanned by means of ultrasound and the image data record obtained in this way is mapped to the data record obtained preoperatively using a suitable mathematical transformation.

It goes without saying that such a fixation device or fixation shell and the corresponding procedure are also advantageous independently of the other features of the present invention. It is also understood that the method according to the invention can possibly also be carried out without a fixation.

In addition, the invention proposes two different procedures by which it is ensured that the detection field of the sensor used contains the field of action of the medical instrument.

On the one hand, tracking methods can be used which determine the detection field and the field of action and ensure that the imaging sensor essentially detects the field of action of the medical device. In this way it can be displayed or output detected. In this way, it can be displayed or output where the corresponding area of action or the corresponding area of action currently lies, so that the effect of the medical instrument can be appropriately anticipated by the therapist or surgeon. It is not absolutely necessary for the sensor to cover the entire field of action of the medical instrument. Rather, the relevant areas of the field of activity should preferably be recorded. For example, in the case of a jet cutter, the upper region of the water jet, which as a rule is not involved in the cutting process anyway, can be arranged outside the field of action.

On the other hand, the invention provides a device for navigation during medical interventions on non-osseous structures, which comprises a sensor for recording image data of the non-osseous structure during the medical intervention, which is connected to a medical instrument. In this way, dynamic image data can be recorded with this device, which are directly correlated with the respective medical instrument. On the one hand, this enables the image data to be acquired without having to take another medical instrument. On the other hand, the method according to the invention makes it relatively easy to determine the exact position of the medical instrument in relation to the structures of interest. The medical instrument thus has a dual function. The arrangements described above are particularly suitable for a cutter, such as a scalpel or a jet cutter (water jet cutter). The cutting plane can be detected by the sensor so that a doctor can easily determine or search for the correct cutting plane before the actual cut is made. In particular, in interaction with an automatically adjustable cutting tool, it can be provided that the cutting performance is regulated in accordance with specifications made during the preparation for the operation. In particular, the cutting performance can be reduced if there is a deviation from a cut surface defined during the preparation for the operation. It goes without saying that this procedure or arrangements of this type are also advantageous independently of the other features of the present invention.

Accordingly, it is advantageous if the sensor has a detection field that includes a direction of action or an area of action of the operating tool, so that the corresponding adjustment can take place immediately.

The sensor preferably comprises an ultrasonic sensor. Such a sensor has a relatively simple structure, is usually available in every clinic and can also be used during an operation without any additional measures, since no essential safety precautions need to be taken, as is the case with X-ray apparatuses or similar radiographic devices. In particular, a Doppler ultrasonic sensor without additional effort is used which allows the measurement of flows, in particular flows in vessels.

Furthermore, MR (magnetic resonance) devices, mobile CTs (computer tomographs), online X-ray devices and the like, in particular for vascular recognition, can be used as sensors, insofar as the necessary safety measures or technical developments permit use during an operation ,

Further advantages, goals and properties of the present invention are explained with reference to the following description of the attached drawing, in which an organ to be operated and a corresponding surgical tool as well as process sequences according to the invention are exemplified. The drawing shows:

FIG. 1: a schematic view of the resection of a liver with a tumor,

FIG. 2 shows a schematic view of an alternative for resection of a liver with a tumor,

FIG. 3 shows a schematic view of a registration or transformation method,

FIG. 4 shows a schematic view of a second registration or transformation method, FIG. 5 shows a schematic view of a third registration or transformation method,

FIG. 6 shows a schematic view of an instrument according to the invention,

FIG. 7 shows a schematic view of a further instrument according to the invention,

Figure 8 shows an embodiment for the underside of the instrument of Figure 7 and

9 shows a further embodiment for the underside of the instrument according to FIG. 7.

1 shows a liver 1 with a deep-lying tumor 2 in the right half of the liver. The right hepatic vein 3 is fully covered by the tumor 2, while the middle hepatic vein 4 is within a safety distance when the tumor-bearing half of the liver is removed. The left hepatic vein 5 is apparently not affected and is essential for survival.

The corresponding operation is preceded by a computer-assisted operation planning, in which a three-dimensional data set is determined using various measurement methods, such as ultrasound, CT, MR and the like, from which the respective resection areas are defined. In a preprocessing step, a symbolic representation of the Vascular structures are calculated in which different morphological, geometric and functional properties of the vascular tree 3, 4, 5 ,. such as shape features, lengths, diameters, angles and positions of bifurcations, flow direction and speed of the blood and the like. Ä., are saved.

In a next step in this example, after mobilizing the liver, dynamic image data of the intraoperative geometry of the site are compared with planning data obtained preoperatively as static image data. For this purpose, the liver is scanned three-dimensionally using ultrasound, for example. This data record is then mapped to the data determined during operation planning, for example an MR data record.

Preferably, after the liver 1 has been mobilized, that is to say after the liver 1 has been exposed in the building cavity, the latter is initially fixed. This can be done in particular by suitable anesthetic measures or surgical measures. These include in particular the jet ventilation and the design of the abdominal cavity with cloths or the fixation of the liver 1 by means of a suitable gripping arm. In addition, a fixation shell 12 (FIG. 1) can also be used, by means of which the liver 1 is fixed, the liver 1 preferably being largely fixed in its natural form. For this purpose, in particular the fixation shell 12 can be prepared or preformed using rapid prototyping on the basis of the preoperatively obtained image data. In a next step, the planning data obtained preoperatively are compared as static image data with the geometry of the intraoperative site and, if necessary, supplemented. For this purpose, for example, the fixed liver is scanned using ultrasound and the image data set obtained in this way is mapped to the data set obtained preoperatively using a suitable mathematical transformation.

In addition, this enables a simple and quick comparison of the dynamic data obtained intraoperatively with the static data set.

The resection is carried out by means of a jet cutter 6, which can cut using a water jet 7. The jet cutter 6 also includes a Doppler ultrasound sensor with a detection field 8. The current position of the jet cutter 6 with respect to the sectional plane defined in the operation planning is determined by means of the measurement data obtained from this, by the position of the vessels 3, 4, 5 determined in the dynamic image data, a corresponding transformation is determined and this transformation is applied to the static image data. A corresponding image display is made available to the surgeon.

The cutting performance of the jet cutter 6 is reduced according to the preoperative planning when the jet cutter 6 leaves the previously determined cutting surface. If a further tumor is detected by the intraoperative ultrasound 8, a new resection proposal can be worked out using the existing preoperative data and then implemented during the ongoing operation.

In a concrete implementation, depending on the requirements, a two-dimensional (FIGS. 3 and 4) or a three-dimensional (FIG. 5) measurement of the dynamic data can take place.

If a two-dimensional, dynamic image is used (see FIG. 3), a rapid extraction of the vascular structures 3, 4, 5 and their relevant properties can be carried out due to the small amount of data. In this way, in particular, a real-time method can be easily implemented in this way, by means of which a possibly previously planned cutting plane defined or definable in the static data can be localized in the deformed organ 1. By extracting the vascular structures 3, 4, 5 from the dynamic images in real time, a direct comparison of the two geometric arrangements can be carried out to compare the position of the two image planes. The similarity of the vascular patterns on the two image planes is therefore a measure of the extent to which the two image planes are positioned in accordance with their vascular features 3, 4, 5.

As shown in particular in FIG. 3, in this exemplary embodiment, the liver 1 is recorded using preoperative data recordings, such as, for example, contrast medium CT recordings 30. Based on this A three-dimensional representation 31 of the liver 1 can be taken, which supports in particular the planning of the operation and can also be used in particular for the construction of the fixation shell 12. In the present exemplary embodiment, a symbolic vascular tree 33 can be extracted from these preoperative data using a vascular tree analysis 32.

In the exemplary embodiment shown in FIG. 3, a resection area 34 is preferably selected first, a flow simulation 35 then being carried out on the vascular tree 33 and a pattern 36 of the vessels on the resection area 34 with simulated flow directions being able to be determined therefrom.

Intraoperatively, Doppler ultrasound recordings 37 are recorded as a sequence of dynamic 2D images 38, which are segmented in real time and analyzed 39 for their characteristics. A representation 40 of the vessels and their directions of flow is thus also determined from this in this exemplary embodiment. Intraoperatively, a similarity measure can then be determined by means of a transformation between the pattern 36 and the representation 40, which gives a therapist or surgeon information about when he has reached the resection area.

Two-dimensional dynamic image data, corresponding to the exemplary embodiment shown in FIG. 4, which essentially corresponds to the exemplary embodiment shown in FIG. 3, so that the same method steps are also provided with an identical reference number. It is expanded that relevant image features, such as vessel patterns and flow information, are extracted from the individual two-dimensional image data 38 and the extracted features 41 are successively supplemented to a higher-dimensional dynamic decision space 42. Accordingly, a three-dimensional pattern space 44 of the rivers and the flow directions is determined preoperatively from the vascular tree 33 by a flow simulation 43, which is then available for transformation calculations 45, 46. Preferably, especially for reasons of accuracy, but also because of shorter computing times, in this exemplary embodiment, a dynamic data set is initially recorded intraoperatively with the liver 1 as undeformed as possible (registration 45) and used to determine the static image data before resection areas are searched and resections are performed by then further transformations 46 are carried out and the therapist or surgeon is given an up-to-date picture of the liver 1 or the position of its instruments. In terms of effort and accuracy, this procedure thus lies between the evaluation of the purely two-dimensional and three-dimensional image data (FIG. 5).

This procedure largely corresponds to the exemplary embodiment according to FIG. 5, so that here, too, the same method steps are provided with identical reference numerals. In this exemplary embodiment, however, 3D images 47 are recorded immediately, so that spatial data sets 48 are created, which are extracted and processed 49 accordingly in three dimensions. If three-dimensional dynamic image data (FIGS. 4 and 5) are available, they can be subjected to the same preprocessing step as the static image data. This creates a further symbolic description of the now deformed vessel tree 3, 4, 5. The two image data sets can then be mapped onto one another with the aid of the branching structures shown, even if the dynamic image data only partially reproduce the vessel tree.

The instrument according to FIG. 6 comprises a puncture needle 50 which is to be inserted in a navigated manner into a substructure 52 of a non-bony structure 51, such as a liver, for example. For this purpose, the puncture needle 50 is provided with a holder 53, which in turn is attached to an ultrasound head 54. The angle of the holder 53 can be adjusted so that different positions can be reached under the ultrasound head 54. In addition, the puncture needle 50 can be moved in the holder 53, as indicated by the double arrows. The angle is preferably set exactly before the insertion of the needle 50 and then fixed, so that the needle 50 can be brought to its place of use in a targeted and navigated manner.

In a similar way, the arrangement according to FIG. 7 enables navigated work. This is a jet cutter 55 which can be used for resection of structure 51, for example for resection of substructure 52. The jet cutter 55 comprises a water supply 56 known per se and ejects a water jet 57 which is used for a Section can be used. In addition, the jet cutter 55 is provided with an ultrasound array 58, which in turn is connected to a data processing system via a data connection 59. The data processing system regulates the water jet 57 and, in particular, ensures that the water jet is switched off when the planned resection area is left or parts such as veins that are left intact are reached.

In this exemplary embodiment, as shown in FIG. 8, the ultrasound array 58 is linear, so that 2-dimensional images can be generated and used for navigation as described above. Here, the jet cutter 55 is preferably configured such that the water exit point 60 is framed by the array 58. The same applies to the jet cutter 61 according to FIG. 9, in which a 3-dimensional ultrasound array 62 frames the water exit point 63.

Claims

claims:

1. Method for navigation during medical interventions on non-osseous structures, characterized in that before

Intervention static image data of the non-bony structure are recorded and stored, that dynamic image data of the non-bony structure are recorded during the intervention, that substructures are extracted from the image data and that by comparing the substructure in the static image data with the substructure the position of at least one sub-area of interest of the non-bony ones in the dynamic image data

Structure is located and assigned during the medical intervention in the various image data sets used.

2. The method according to claim 1, characterized in that as

Substructure at least one vessel, preferably a blood vessel, is used.

3. The method according to any one of claims 1 or 2, characterized in that a transformation is determined by comparing the substructure in the static image data with the substructure in the dynamic image data, which transforms to the subarea of interest of the non-bony structure in the static image data is used to indicate the current location of the section.

4. The method according to any one of claims 1 to 3, characterized in that the static image data have at least three spatial dimensions.

5. The method according to any one of claims 1 to 4, characterized in that the dynamic image data can have any dimension.

6. The method according to claim 5, characterized in that the dynamic image data comprise several data sets of any dimension.

7. The method according to claim 5 or 6, characterized in that at least a part of the dynamic image data or the intraoperative image data is only acquired in two dimensions.

8. The method according to claim 7, characterized in that the 2-dimensional data is subjected to an analysis or segmentation and then compared with the static image data.

9. The method according to claim 7 or 8, characterized in that the 2-dimensional data is subjected to an analysis with which essential partial data, which for example represent substructures, are extracted and then further processed, and that the data analyzed in this way is successively increased to higher dimensions - Len feature spaces are put together.

10. The method according to any one of claims 1 to 9, characterized in that at least one sensor is used to record the static image data, which is also used to record the dynamic image data.

11. The method according to any one of claims 1 to 10, characterized in that at least one first sensor with a first modality is used to record the static image data and at least one second sensor with a different modality is used to record the dynamic image data.

12. The method according to claim 11, characterized in that a substructure in the two image data sets is analyzed in each case with respect to at least one location-dependent feature and a comparison is made or a transformation is determined between the two image data sets.

13. The method according to any one of claims 1 to 12, characterized in that the non-bony structure is fixed before recording the static or dynamic image data, preferably by a fixing device.

14. The method according to any one of claims 1 to 13, characterized in that at least second dynamic image data is recorded and that the substructures are also extracted from the second dynamic image data and that by comparing the substructures Structure in the static image data and / or the first dynamic image data with the substructure in the second dynamic image data, the position of the subarea of interest of the non-osseous structure is determined and output during the medical intervention.

15. The method according to any one of claims 1 to 14, characterized in that the shape of the non-bony structure is determined preoperatively and a fixation device is constructed on the basis of this data.

16. A method for the fixation of a non-bony structure, characterized in that the shape of the non-bony structure is determined preoperatively and a fixation device is constructed on the basis of this data.

17. The method according to claim 15 or 16, characterized in that the fixation device is constructed by rapid prototyping.

18. Device for fixing a non-bony structure, characterized by a shell, the shape of which is designed such that the non-bony structure is approximately fixed in its natural position in the shell.

19. Device for navigation during medical interventions on non-osseous structures, characterized in that at least during the medical intervention the field of action of a medical instrument is essentially contained in the detection field of a sensor for recording image data of the non-bony structure.

20. Navigation device according to claim 19, characterized in that the spatial relationship between the detection field and the field of action is determined by suitable methods.

21. Device for navigation during medical interventions on non-osseous structures, characterized in that the field of action of a medical instrument and the acquisition field of a sensor for recording image data of the non-osseous structure are fixed to one another at least during the medical intervention.

22. Navigation device according to claim 21, characterized in that the spatial relationship of the detection field and the

Field of action is fixed in that the sensor and the medical instrument are integrated into one device.

23. Navigation device according to one of claims 19 to 22, characterized in that the sensor comprises an ultrasonic sensor, preferably a Doppler ultrasonic sensor.