Technique For Determining A Position of One Or More Imaged Markers In An Image Coordinate System

Abstract

A method and a device for determining a respective position of one or more markers in a 3D image coordinate system are provided. A plurality of image data sets taken from a 3D volume in which an object and one or more markers are disposed. The 3D volume comprises a central volume containing at least a portion of the object and further comprises a peripheral volume adjacent to the central volume and containing the one or more markers. The image data sets have been taken from at least one of different positions and different orientations relative to the object. A first subset comprises image data sets that each includes at least one dedicated marker of the one or more markers and a second subset comprises at least one image data set that does not include the at least one dedicated marker. The method further comprises determining, from the image data sets, a position of the at least one dedicated marker in a 3D image coordinate system of a 3D reconstruction of the central volume with the object portion.

Claims

1. A computer-implemented method of determining a respective position of one or more markers in a 3D image coordinate system, the computer-implemented method comprising: obtaining a plurality of image data sets taken from a 3D volume in which an object and one or more markers are disposed, wherein the 3D volume comprises a central volume containing at least a portion of the object and further comprises a peripheral volume adjacent to the central volume and containing the one or more markers, wherein the image data sets have been taken from at least one of different positions and different orientations relative to the object, wherein a first subset comprises image data sets that each includes at least one dedicated marker of the one or more markers and a second subset comprises at least one image data set that does not include the at least one dedicated marker; and determining, from the image data sets, a position of the at least one dedicated marker in a 3D image coordinate system of a 3D reconstruction of the central volume containing the object portion.

2. The computer-implemented method of claim 1, further comprising selecting, from the image data sets, at least two image data sets of the first subset, wherein the position of the at least one dedicated marker in the 3D image coordinate system of the 3D reconstruction of the central volume containing the object portion is determined from the selected image data sets.

3. The computer-implemented method of claim 1, wherein the at least one dedicated marker is included in a portion of the peripheral volume that is not included in the at least one image data set of the second subset.

4. The computer-implemented method of claim 1, wherein the central volume is covered entirely in all image data sets and the peripheral volume is covered less than entirely in at least some of the image data sets.

5. The computer-implemented method of claim 1, wherein the image data sets are representative of projections of the 3D volume on an imaging line or imaging plane, or wherein the computer-implemented method further comprises visualizing the 3D reconstruction of the central volume, wherein the peripheral volume is excluded from visualization.

6. The computer-implemented method of claim 1, further comprising processing the position of the at least one dedicated marker for surgical navigation.

7. The computer-implemented method of claim 6, wherein the at least one dedicated marker is associated with a known position in a 3D navigation coordinate system, and wherein processing the position of the at least one dedicated marker for surgical navigation comprises registering the 3D reconstruction with the 3D navigation coordinate system.

8. The computer-implemented method of claim 1, wherein the one or more markers are one or more artefacts of one or more devices trackable by a surgical navigation system.

9. The computer-implemented method of claim 8, wherein the one or more trackable devices comprise at least one of: (i) one or more light emitting diodes each having a solder joint to comprise the artefact; (ii) one or more passive elements each configured to reflect radiation and to comprise the artefact; and (iii) one or more coils each configured to detect electromagnetic radiation and to comprise the artefact.

10. The computer-implemented method of claim 1, wherein the object portion is an interior portion of the object.

11. The computer-implemented method of claim 1, wherein the one or more markers are loosely arranged relative to the object.

12. The computer-implemented method of claim 1, wherein the one or more markers are attached to an exterior portion of the object.

13. The computer-implemented method of claim 1, wherein the one or more markers comprise two or more markers arranged together on a carrier.

14. The computer-implemented method of claim 13, wherein at least a portion of the carrier is flexible such that a first one of the markers on the portion is movable relative to a second one of the markers on the portion.

15. The computer-implemented method of claim 1, wherein the central volume has a longitudinal axis and wherein the image data sets have been taken by one of (i) moving an imaging device to different positions about or along the longitudinal axis of the central volume and (ii) moving the object and the one or more markers about or along the longitudinal axis; and/or wherein the image data sets have been generated by one of computer tomography and an X-ray device coupled to a C-arm.

16. A device for determining a respective position of one or more markers in a 3D image coordinate system, the device being configured to: obtain a plurality of image data sets taken from a 3D volume in which an object and the one or more markers are disposed, wherein the 3D volume comprises a central volume containing at least a portion of the object and further comprises a peripheral volume adjacent to the central volume and containing the one or more markers, wherein the image data sets have been taken from at least one of different positions and different orientations, wherein a first subset comprises image data sets that each includes at least one dedicated marker of the one or more markers and a second subset comprises at least one image data set that does not include the at least one dedicated marker; and determine, from the image data sets, a position of the at least one dedicated marker in a 3D image coordinate system of a 3D reconstruction of the central volume containing the object portion.

17. The device of claim 16, wherein the device is further configured to: select, from the image data sets, at least two image data sets of the first subset; and wherein the position of the at least one dedicated marker in the 3D image coordinate system of the 3D reconstruction of the central volume containing the object portion is determined from the selected image data sets.

18. The device of claim 16, wherein the at least one dedicated marker is included in a portion of the peripheral volume that is not included in the at least one image data set of the second subset.

19. The device of claim 16, wherein the central volume is covered entirely in all image data sets and the peripheral volume is covered less than entirely in at least some of the image data sets.

20. A navigation system comprising: a device configured to: obtain a plurality of image data sets taken from a 3D volume in which an object and the one or more markers are disposed, wherein the 3D volume comprises a central volume containing at least a portion of the object and further comprises a peripheral volume adjacent to the central volume and containing the one or more markers, wherein the image data sets have been taken from at least one of different positions and different orientations, wherein a first subset comprises image data sets that each includes at least one dedicated marker of the one or more markers and a second subset comprises at least one image data set that does not include the at least one dedicated marker; determine, from the image data sets, a position of the at least one dedicated marker in a 3D image coordinate system of a 3D reconstruction of the central volume containing the object portion; select, from the image data sets, at least two image data sets of the first subset, wherein the position of the at least one dedicated marker in the 3D image coordinate system of the 3D reconstruction of the central volume containing the object portion is determined from the selected image data sets; and a processing device configured to process the position of the at least one dedicated marker for surgical navigation.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Further details, advantages and aspects of the present disclosure will become apparent from the following embodiments taken in conjunction with the drawings, wherein:

[0031] FIG. 1A shows a C-arm configured to capture image data sets of an object;

[0032] FIG. 1B shows a C-arm in three different tilting angles according to a prior art approach;

[0033] FIG. 1C shows a C-arm in three different tilting angles according to an embodiment;

[0034] FIG. 2A shows a cross-section of a patient within a 3D volume with a schematic illustration of an associated central volume and peripheral volume;

[0035] FIG. 2B shows examples of projecting X-ray beams through the 3D volume;

[0036] FIG. 3 shows image data sets resulting from different X-ray beam projections;

[0037] FIG. 4A shows a flow diagram of an embodiment of a method aspect of the present disclosure; and

[0038] FIG. 4B shows a flow diagram of an embodiment of a variant of the method aspect of the present disclosure; and

[0039] FIG. 5 shows an embodiment of a surgical navigation system realizing a device aspect of the present disclosure.

DETAILED DESCRIPTION

[0040] In the following description, exemplary embodiments of a technique for determining a respective position of one or more markers in a 3D image coordinate system will be explained with reference to the drawings. The same reference numerals will be used to denote the same or similar structural features. It will be appreciated that the technique can be embodied in a method, a device, a system, a computer program product, or otherwise.

[0041] The technique described herein comprises obtaining a plurality of image data sets. For an easier understanding of how the image data sets are obtained, an exemplary process of capturing these image data sets is described first. While the technique can be used in medical and non-medical fields, such as material science, it will in the following be explained with reference to medical imaging.

[0042] Two common projection-based medical imaging technologies are computer tomography (CT) and X-ray imaging using a C-arm. In both technologies, image data sets of an object are taken from different orientations and/or positions relative to the imaged object. CT usually uses an imaging line, i.e., projections onto a substantially one dimensional sensor, whereas X-ray imaging using a C-arm typically uses an imaging plane, i.e., projections onto a two dimensional sensor. The technique herein is not limited to X-ray imaging and may be used with any other radiation.

[0043] FIG. 1A shows an exemplary C-arm 10 configured to capture image data sets of an object 12 to be imaged in accordance with certain embodiments of the present disclosure. The object 12 may be any biological or non-biological specimen. The object 12 may, for example, be a patient or a body part of a patient.

[0044] The C-arm 10 shown in FIG. 1A comprises an imaging device in form of an X-ray sensor 14 and an X-ray source 16, which both are movable simultaneously in rotatory motion around a longitudinal axis 22. The object 12 is arranged to be intersected by (e.g., co-linear with) the longitudinal axis 22 and is therefore located between the X-ray sensor 14 and the X-ray source 16. Due to its rotatory movability, the C-arm 10 shown in FIG. 1A is capable of capturing image data sets of the object 12 from different positions and different orientations. Alternatively, an imaging device may be provided that is configured to capture image data sets only from different positions (e.g., using an X-ray sensor and an X-ray detector that both move in a straight path) or only from different orientations (e.g., using a pivoting X-ray sensor or X-ray source).

[0045] In the process of capturing image data so as to obtain an image data set, X-rays are emitted from the X-ray source 16 that reach the X-ray sensor 14 after passing through the object 12. A portion of the X-rays is absorbed or scattered by the object 12. The resulting image data captured by the X-ray sensor 14 is indicative of a spatial distribution of differently absorbing matter in the object 12 and therefore of a projection of a volume between the X-ray source 16 and the X-ray sensor 14 onto an imaging plane, wherein the image plane is a sensor area of the X-ray sensor 14. Due to the finite sensor area of the X-ray sensor 14, the X-rays that reach the X-ray sensor 14 are confined to a probing X-ray beam 18 that covers a limited solid angle (e.g., in a cone-like manner).

[0046] An image data set may consist of or comprise the image data generated by a capturing process from a given position and/or orientation. In some variants, an image data set may additionally comprise information about the position and/or orientation underlying the associated image data capturing process.

[0047] The C-arm 10 shown in FIG. 1A is configured to generate a plurality of image data sets from different angles by rotating the X-ray source 16 together with the X-ray sensor 14 around the object 12.

[0048] FIG. 1B shows an abstract representation of the C-arm 10 in three different tilting angles around the longitudinal axis 22 relative to a set of markers 30. In this example, which reflects the prior art cited in the introductory portion of this description, the markers 30 are contained in every image taken since the markers are located in a central volume 24 imaged in each rotational position of the C-arm 10.

[0049] FIG. 1C illustrates an embodiment in which the central volume 24 has been chosen smaller compared to the illustration in FIG. 1B, so that the markers 30 will only be contained in a subset of the images taken by the C-arm 10.

[0050] In more detail, FIG. 1C shows an abstract representation of three different tilting angles of the C-arm 10 around the longitudinal axis 22 relative to the object 12. In practice, the tilting angle is usually changed in smaller increments as shown in FIG. 1C. The C-arm may rotate in angular increments such as 1°, 2°, 3° or more before capturing the next image data set. Over the course of capturing the plurality of image data sets, the C-arm 10 may move for a full rotation of 360°, a fraction thereof (e.g., 180°) or more than one rotation (e.g., 540°). For example, the C-arm 10 may move for a full rotation of 360° with 3.6° increments for each image data set. In such a case, the C-arm 10 captures 100 image data sets.

[0051] It is not required that an imaging device 10 moves relative to the object 12 in order to capture the image data sets. Alternatively, the object 12 may be moved relative to the imaging device 10. Such movement may comprise moving the object 12 about or along a longitudinal axis, such as the longitudinal axis 22 shown in FIG. 1A that is arranged between the X-ray sensor 14 and the X-ray source 16.

[0052] Any volume that can be covered by the imaging device 10 may be defined as a 3D volume 20. In the case shown in FIG. 1C, the exemplary C-arm 10 moves around the entire angular range of 360° so that the 3D volume 20 essentially describes a cylindrical shape arranged coaxially with the longitudinal axis 22 of the C-arm 10. The radius of the 3D volume 20 reaches up to the X-ray sensor 14 and X-ray source 16. In the case of FIGS. 1A and 1C, the X-ray sensor 14 and X-ray source 16 are spaced in an equal distance away from the longitudinal axis 22. In the case of different distances, the 3D volume 20 reaches up to the component with the larger distance relative to the longitudinal axis 22.

[0053] As can be seen in FIG. 1C, the 3D volume 20 comprises in its centre around the longitudinal axis 22 a central volume 24. The central volume 24 is covered by probing X-ray beams 18 (and the associated projections on the X-ray sensor 14) for any tilting angle of the C-arm 10. The central volume 24 can therefore be defined as a volumetric intersection of all probing X-ray beams 18 (and the associated projections). Consequently, any feature arranged in the central volume 24 is included in all of the captured image data sets.

[0054] The 3D volume 20 further comprises a peripheral volume 26 adjacent to the central volume 24. In FIG. 1C the peripheral volume 26 is located radially adjacent relative to the central volume 24 and therefore essentially describes the shape of a hollow cylinder. The central volume 24 and the peripheral volume 26 may have other shapes, which may depend, for example, on the paths of the X-ray sensor 14 and the X-ray source 16, the covered angular range, and the size of the angular increments. Features located in the peripheral volume 26, such as the markers 30, are not necessarily covered by all probing X-ray beams 18. Therefore, such features may not be included in all of the captured image data sets.

[0055] It should be noted that defining the central volume 24 as the volumetric intersection of all probing X-ray beam 18 is one of many ways to define the central volume 24. Alternatively, the central volume 24 may be smaller or larger than the volumetric intersection of all probing X-ray beams 18. The size of the central volume 24 will be discussed further below in the context of a 3D reconstruction that is based on the central volume 24.

[0056] The C-arm 10 comprises or is connected to at least one processor and at least one storage medium (not shown). The X-ray sensor 14 is configured to generate the image data sets based on X-rays received at the X-ray sensor 14. The processor is configured to receive the image data sets from the X-ray sensor 14 and store the image data sets on the storage medium.

[0057] Based on the image data sets, the processor is configured to generate a 3D reconstruction of at least a part of the 3D volume. The 3D reconstruction may be performed by a method known in prior art. The 3D reconstruction is arranged in a 3D image coordinate system and may, for example, be represented in form of voxels or a point cloud.

[0058] The 3D reconstruction recreates features that are included in the image data sets taken by the C-arm 10. Evidently, accuracy of the 3D reconstruction of a feature improves with the number of image data sets that include said feature. As explained above, features that are located in the central volume 24 shown in FIGS. 1A and 1C are included in every image data set. On the other hand, features that are arranged in the peripheral volume 26 are not included in all image data sets. Therefore, a 3D reconstruction of features inside the central volume 24 has a higher accuracy than 3D reconstruction of features inside the peripheral volume 26.

[0059] Since accuracy of a 3D reconstruction of the peripheral volume 26 is comparably lower, 3D reconstruction is conventionally only performed for the central volume 24. This means the output of a C-arm or a similar imaging system is, or permits, a 3D reconstruction of the central volume 24 only, whereas any information regarding the peripheral volume is typically discarded.

[0060] In the example of FIG. 1C, the central volume 24 is defined as a volumetric intersection of all probing X-ray beams 18, which is essentially the largest portion of the 3D volume 20 that is covered by every X-ray projection (i.e., every probing X-ray beam 18). Alternatively, a smaller portion of the volumetric intersection may be selected as central volume 24. A 3D reconstruction based on such a volume contains less data and requires less computing resources. Further alternatively, a portion of the 3D volume 20 that is larger than the volumetric intersection may be selected as the central volume 24.

[0061] FIG. 2A shows a cross section of a patient 28 to be imaged. Also illustrated in FIG. 2A are a central volume 24 and a peripheral volume 26. These volumes 24, 26 are defined as explained above with reference to FIGS. 1A and 1C.

[0062] The central volume 24 is focused onto an object within the patient's body of interest for surgical navigation. The object may be the patient's spine. In the example shown in FIG. 2A, the central volume 24 is focused on an object in form of one or more vertebrae 13. Therefore, 3D reconstruction will be performed for the central volume 24 containing the vertebra 13. The 3D reconstruction could also be performed for the central volume 24 and at least a part of the peripheral volume 26. Further, the central volume 24 may alternatively cover any other body part of the patient, such as an organ, a bone, or skin tissue.

[0063] For the purpose of surgical navigation, the 3D reconstruction needs to be registered with a 3D navigation coordinate system of a surgical navigation system. Such registration is based on markers that can be imaged by the C-arm 10 or other imaging device (i.e., that are detectable in the image data sets). As the markers are arranged on, in or in the vicinity of the object 12 to be imaged, the plurality of image data sets are captured (e.g., pre- or intra-operatively) by the imaging device such as the C-arm 10 as described with reference to FIGS. 1A and 1C above. During surgical navigation, the markers are still arranged at, in or in vicinity of the object 12. Consequently, the markers themselves or tracking elements having a fixed and predefined spatial relationship relative to an associated marker can be identified by the surgical navigation system (e.g., by a camera thereof) and assigned to coordinates in the 3D navigation coordinate system. Based on a determination of positions of the markers in a 3D image coordinate system (i.e., relative to the 3D reconstruction of the central volume 24) and a determination of associated positions of the markers (or the associated tracking elements) in the 3D navigation coordinate system, the 3D reconstruction can be registered relative to the 3D navigation coordinate system by aligning the marker coordinates in the 3D image coordinate system with the marker (or tracking element) coordinates in the 3D navigation coordinate system (optionally taking into account the relative position between each marker and its associated tracking element).

[0064] Having thus explained utilization of markers for exemplary registration purposes in a surgical navigation context, FIG. 2A further shows a trackable device 29 comprising a flexible carrier 32 that supports a plurality of such markers 30 (e.g., three markers 30A, 30B, and 30C), wherein the carrier 32 is attached via an adhesive to a skin surface on the back of the patient 28. The carrier 32 shown in FIG. 2A is flexible and therefore adapts to any movement of the patient's skin, wherein the markers 30 may be movable relative to each other. A large number of markers 30 may be located on a frame like flexible carrier 32 having a central opening through which surgery is performed, such as the SpineMask™ tracker distributed by Stryker Navigation. The SpineMask™ tracker may thus be considered an example of the trackable device 29.

[0065] Optionally, the carrier 32 may at least partially be rigid. In such a case the markers may be rigidly located relative to each other. In further embodiments, the trackable device 29 may comprise an interface for attaching the carrier 32 to the patient 28. The interface may comprise at least one of a clamp, a screw, and a magnet.

[0066] The markers 30 shown in FIG. 2A are defined by LEDs on the carrier 32 and, more precisely, by the solder joints of these LEDs. The solder joints are easily detectable as artefacts in the image data sets as they substantially block any X-rays from reaching the X-ray sensor 14 (see FIGS. 1A and 1C). The LEDs on the carrier 32, in turn are easily detectable by a camera of a surgical navigation system. As such, the surgical navigation system may determine the positions of the LEDs in a 3D navigation coordinate system, and the positions of the associated solder joints may be determined in a 3D image coordinate system. Based on a predetermined spatial relationship between the solder joints and the associated LEDs (e.g., a predetermined spatial vector including the null vector), the 3D image coordinate system can thus be registered relative to the 3D navigation coordinate system.

[0067] In still further embodiments, the markers may be provided on a carrier that is configured to be loosely arranged relative (e.g., besides) the patient 28. See, for example, the registration device of U.S. Pat. No. 8,483,434 B2, that could also be used in embodiments of the present disclosure. The registration device comprises, in addition to the one or more markers, tracking elements (such as LEDs) capable of being tracked by a camera of a navigation system, wherein the markers may have a known spatial relationship relative to the tracking elements.

[0068] As illustrated in FIG. 2A, the central volume 24 is focused onto the vertebrae 13 inside the patient 28. As such, the central volume 24 does not cover the markers 30, or at least not all of them. Therefore, any 3D reconstruction of the central volume 24 with the vertebrae 13 will likewise not contain the markers 30 (or at least not all of them). Consequently, the 3D reconstruction of the central volume 24 alone cannot be used for registration purposes in the context of surgical navigation.

[0069] FIG. 2B shows examples of probing X-ray beams 18 projected through the 3D volume 20, wherein some beams 18 do and some do not cover markers 30. For example, probing X-ray beams 18A and 18B cover marker 30B, but probing X-ray beam 18C does not cover marker 30B. More specifically, probing X-ray beam 18A covers marker 30B. Probing X-ray beam 18B covers markers 30B and 30C. And probing X-ray beam 18C does not cover any of markers 30A, 30B, or 30C.

[0070] FIG. 3 shows image data sets 34 generated as projections resulting from the different X-ray beams 18A, 18B, 18C. Image data set 34A comprises image data 36A captured using X-ray beam 18A. Image data set 34B comprises image data 36B captured using X-ray beam 18B and image data set 34C comprises image data 36C captured using X-ray beam 18C. Optionally, each of the image data sets 34 may comprise additional parameters 37A, 37B, 37C, respectively, such as the tilting angle of the C-arm 10 (as will be described further below).

[0071] The following explanations exemplarily focus on one dedicated marker, namely marker 30B. However, these explanations may be applied to any other marker 30 and any combination of markers 30.

[0072] With continued reference to FIG. 3, two subsets 38, 40 of image data sets can be defined. A first subset 38 comprises all of the captured image data sets 34 that each include the dedicated marker 30B (i.e., image data sets 34A and 34B with image data 36A and 36B captured using the probing X-ray beams 18A and 18B, respectively). A second subset 40 comprises at least one image data set that does not include dedicated marker 30B (i.e., image data set 34C with image data 36C captured with the probing X-ray beam 18C). In practice, 100 and more image data sets may be taken, so that at least the second subset 40 will typically comprise dozens of data sets.

[0073] In the schematic example illustrated in FIG. 3, all three image data sets 34 include the object 12 of interest, i.e., the vertebra 13 shown in FIGS. 2A and 2B. The vertebra 13 is included in all image data sets 34 because it is located in the central volume 24. Since the markers 30 are located in the peripheral volume 26, not all image data sets 34 include all the markers 30. Whereas image data sets 34A and 34B include the dedicated marker 30B, image data set 34C does not includes the dedicated marker 30B. Consequently, image data sets 34A and 34B are assigned to the first subset 38, whereas image data set 34C is assigned to the second subset 40.

[0074] In the above example, first and second subsets 38, 40 of image data sets are defined for a dedicated marker 30B. Corresponding subsets may be defined for other markers, such as marker 30C. It will be appreciated that the same image data set may be assigned to the first subset for a dedicated first marker (as it includes the dedicated first marker) and to the second subset for a dedicate second marker (as it does not include the dedicated second marker).

[0075] FIGS. 4A and 4B show a flow diagram of an embodiment of a method aspect of the present disclosure and an optional variant thereof.

[0076] Due to the positioning of the markers 30 in the peripheral volume 26, the 3D reconstruction of the central volume 24 does not contain information about the positions of the markers 30 in a 3D image coordinate system of the 3D reconstruction. The information about the marker positions has to be determined from the image data sets as shown in FIGS. 4A and 4B.

[0077] Each one of FIGS. 4A and 4B shows a flow diagram 100 of a computer implemented method embodiment of determining the respective position of each of the markers 30 in the 3D coordinate system of the 3D reconstruction of the central volume 24.

[0078] The method comprises obtaining in step 102 the plurality of image data sets 34 taken from the 3D volume 20 in which the object 12 (e.g., the vertebrae 13) and the one or more markers 30 are disposed. The image data sets 34 may have been captured using the C-arm 10 of FIGS. 1A and 1C and may be obtained by the computer system via a hardware interface that couples the computer system to the C-arm 10.

[0079] As illustrated in FIG. 2A, the 3D volume 20 comprises the central volume 24 containing at least a portion of the object 12 and further comprises the peripheral volume 26 adjacent to the central volume 24 and containing the one or more markers 30. The image data sets 34 have been taken from at least one of different positions and different orientations relative to the object 12 (see FIGS. 1C and 2B for the C-arm example). As illustrated in FIG. 3, a first subset 38 of image data sets 34 comprises image data sets 34A, 34B that each includes the dedicated marker 30B of the one or more markers 30. A second subset 40 of image data sets 34 comprises at least one image data set 34C that does not include the dedicated marker 30B.

[0080] The plurality of image data sets 34 may be provided in form of captured image data 36 per image data set 34. Optionally, each of the image data sets 34 comprises additional parameters, such as parameters 37A, 37B, 37C shown in FIG. 3. The additional parameters 37 may include an orientation (e.g., a tilting angle of the C-arm 10) and/or a position of the X-ray sensor 14 for which corresponding image data 36 were captured. The parameters 37 may further include at least one of patient data, time, date, energy and/or wavelength of the emitted and/or detected X-rays, X-ray intensity, type of imaging device, and contrast agent injected into the patient. The image data sets 34 may conform to any standard, such as the DICOM standard

[0081] Returning to FIG. 4A, the method also comprises determining in step 106, from the image data sets, the position of the at least one dedicated marker in the 3D image coordinate system of the 3D reconstruction of the central volume with the object portion. The step 106 may comprise performing a 3D reconstruction of the central volume 24 and at least a part of the peripheral volume 26 that includes the one or more (e.g., all) markers. Step 106 may also comprise using at least two image data sets of the first subset 38.

[0082] Using the different subsets 38, 40 of image data sets 34, the respective positions of one or more of the markers 30 in the 3D image coordinate system may be determined using an approach as illustrated in FIG. 4B. In the computer implemented method embodiment shown in FIG. 4B, the method also comprises selecting in step 104, from the image data sets 34, at least two image data sets 34 of the first subset 38. The selection of the at least two image data sets 34 may comprise a determination of whether an image data set 34 comprises the at least one dedicated marker 30, such as the marker 30B (see 3), and therefore is to be assigned to the first subset 38. This determination may be performed based on automatic image recognition performed for each set of image data 36. The markers 30 may be identified by their shape and/or detected intensity. As explained above, the markers 30 may be artefacts of tracking elements. The trackable device 29 shown in FIG. 2A comprises LEDs connected with electrical circuits via solder joints. These solder joints have comparably low transmittance for X-rays and therefore appear on in a visual representation of the image data 36 as distinguishable dark dots, as indicated for markers 30B and 30C in 3. Therefore, the markers 30 may be identified as artefacts that are constituted by the solder joints for each of the associated LEDs.

[0083] The first subset 38 comprises all image data sets 34 that include the dedicated marker 30B. A larger number of image data sets 34 increases the accuracy of the position determination for the dedicated marker 30B. Alternatively, the first subset 38 may comprise only a selection of the image data sets 34 that include the dedicated marker 30B. For example, the first subset 38 may only comprise image data sets 34 that additionally fulfil a quality criterion. Such a quality criterion may define a threshold of number of identifiable markers 30 or a intensity and/or contrast threshold for the markers 30 included in the image data sets 34. A limited number of image data sets 34 also reduces the amount of data that is required to be processed.

[0084] The method further comprises determining in step 106, from the selected image data sets 34, a position of the at least one dedicated marker 30B in the 3D image coordinate system of the 3D reconstruction of the central volume 24 with the object 12.

[0085] The position of the at least one dedicated marker 30B may be determined according to approaches known in prior art. One example of a method for determining the position is briefly explained below. The example will use two image data sets 34, which is the lowest number of image data sets 34 required to determine the position of the at least one dedicated marker 30B.

[0086] In a first step, a position and orientation of the X-ray detector 14 is determined for each of the two image data sets 34 that include the dedicated marker 30B. The position and orientation may be determined from feature recognition inside both images 36, wherein a comparison of the recognized features between both images 36 allows determining the tilting angles of the C-arm under which the two images 36 were taken. Based on a known distance of the X-ray detector 14 relative to the longitudinal axis 22, a position and orientation of the X-ray detector 14 for both images 36 can be determined. Alternatively, the orientation and/or position of the X-ray detector 14 may be provided in form of additional parameters 37 within each image data set 34, as explained above with reference to FIG. 3.

[0087] The orientation and/or position are typically given in a 3D image coordinate system in which also the central volume 24 will be reconstructed. This means that the position of the dedicated marker 30B will also be obtained in that 3D image coordinate system of the 3D reconstruction of the central volume 24.

[0088] In a second step, two virtual lines, or projections, are determined between the X-ray source 16 and the positions in the sensor area of the X-ray detector 14 at which the dedicated marker 30B was respectively detected.

[0089] In a third step, a point of intersection (or smallest distance) of the two virtual lines is determined. The point of intersection is the position of the dedicated marker 30B in the 3D image coordinate system.

[0090] The method may also be performed with more than two image data sets 34 containing the at least one dedicated marker 30B. The position of the dedicated marker 30B may then be determined, for example, as an average of a plurality of intersection points or using a least squares method.

[0091] The position of the at least one dedicated marker 30B may be determined without determining a position and/or orientation of the X-ray sensor 14 as explained above. A different approach comprises recognizing different features (e.g., features located in the central volume 24) in different image data 36 and determining a distance of the markers 30 relative to recognized features.

[0092] Once the positions of the marker 30B and of the other markers 30A, 30C have been determined in the 3D image coordinate system, the positions can be processed for surgical navigation purposes. In this regard, FIG. 5 shows an embodiment of a surgical navigation system 42. The surgical navigation system 42 comprises a device 44 for determining a respective position of the markers 30 in the 3D image coordinate system as explained above with reference to FIG. 4B.

[0093] In the embodiment shown in FIG. 5, the position determination device 44 comprises a processor 46 and a computer program product 48 comprising program code portions configured to perform the method described above with reference to FIG. 4B when executed by the processor 46. The computer program product 48 may be stored on a computer-readable recording medium, such as a disk drive, compact disk, flash drive, or a server.

[0094] The surgical navigation system 42 shown in FIG. 5 further comprises a camera 52 that is configured to detect a trackable device 29. The trackable device 29 may the trackable device 29 shown in FIG. 2A or any other type of tracker described above or known in the art. The surgical navigation system 42 may also use non-optical tracking technologies such as electromagnetic tracking based on tracking elements in the form of coils. As will be appreciated, such coils also create artefacts in the image data sets created by a C-arm or similar imaging device.

[0095] A data connection is provided between the position determination device 44 and the C-arm 10 (e.g., the C-arm 10 as described above). The data connection may be a wired connection as shown in FIG. 5. Additionally or alternatively, the data connection may comprise one or both of a wireless connection and an intermediate data storage such as a memory stick, local or remote server for storing data. Over the data connection, the position determination device 44 obtains a plurality of image data sets 34 taken by the C-arm 10 (see, e.g., FIG. 3). The data connection may also be used to send a 3D reconstruction generated from image data sets 34 taken by the C-arm 10 for navigation (e.g., visualisation) purposes to the surgical navigation system 42.

[0096] The C-arm 10 or a processor connected to the C-arm 10 generates the 3D reconstruction (in a 3D image coordinate system) based on the plurality of image data sets 34 taken by the C-arm 10. The 3D reconstruction is then sent to the surgical navigation system 42. The 3D reconstruction can consequently be registered with a 3D navigation coordinate system if a sufficient amount of markers 30 can be determined by the position determination device 44 in the 3D image coordinate system and the corresponding tracking elements can be determined by the surgical navigation system 42 in the 3D navigation coordinate system. The sufficient amount of markers 30 may depend on the amount of markers of the trackable device 29 and the amount of degrees of freedoms that the trackable device 29 needs to be tracked in. Typically, the positions of at least three markers 30 need to be determined.

[0097] Once the surgical navigation system 42 has access to the determined positions of the three markers 30 (determined by the C-arm 10 or the surgical navigation system 42), the surgical navigation system 42 is configured to register the 3D image coordinate system with a 3D navigation coordinate system. For example, the marker 30B (and the other markers 30A, 30C) is associated with a known position in the 3D navigation coordinate system, for example, based on images captured by the camera 52 of a tracking element (e.g., an LED or reflecting sphere) having a predetermined spatial relationship with the marker 30B (and similar for the markers 30A, 30C). Based on the determined marker positions in the 3D image coordinate system and the associated position in the 3D navigation coordinate system, the 3D image coordinate system and, thus, the 3D reconstruction of the central volume 24 can be registered with the 3D navigation coordinate system. To this end, the surgical navigation system 42 may be configured to determine a transformation that translates the 3D reconstruction inside the 3D navigation coordinate system in such a way that the position of each of the markers in the 3D image coordinate system coincides with the corresponding position associated with each of the markers 30 in the 3D navigation coordinate system. After the registration has been performed, the 3D reconstruction of the central volume 24 can be used for navigation purposes, for example by displaying the position of a surgical instrument tracked by the surgical navigation system 42 relative to the 3D reconstruction including the vertebrae 13 of FIG. 2A.

[0098] The above described technique allows processing the 3D reconstruction independently from whether or not markers 30 are located inside the central volume 24. Therefore, the central volume 24 can be defined more freely. In the case of medical imaging this enables limiting the central volume 24 to essential body parts and consequently reduces radiation exposure to the patient.

[0099] The features described in relation to the exemplary embodiments shown in the drawings can be readily combined to result in different embodiments. It is apparent, therefore, that the present disclosure may be varied in many ways. Such variations are not to be regarded as a departure from the scope of the invention as defined by the claims appended hereto.