Method and system for obtaining a location and / or representation of a three-dimensional object on a two-dimensional image obtained by x-ray angiography

Abstract

Method and device for obtaining a representation of a three-dimensional object on a two-dimensional image obtained by X-ray angiography, comprising: - Positioning a divergent X-ray angiographic radiation source at a source position on a first side of an object (a patient) in a physical space; - Positioning a receptor surface in the physical space on a second side of the object (the patient) opposite the first side; - Obtaining a first digital two-dimensional projection image of the object on the receptor surface through the radiation from the source; - Obtaining a digital three-dimensional representation of at least part of the same or the same object; - Positioning the first digital two-dimensional projection image in a digital space that is a representation of at least part of the physical space; - Providing a projection of the three-dimensional representation on a virtual projection screen in digital space from a position corresponding to the source position, which virtual projection screen is arranged parallel to the two-dimensional projection image and which projection forms a second digital two-dimensional projection image; - Providing a combined display of the first and second digital two-dimensional projection image.

Description

[0001] Method and system for obtaining a location and / or representation of a three-dimensional object on a two-dimensional image obtained by X-ray angiography

[0002] The present invention relates to a method and system for obtaining a location and / or representation of a three-dimensional object on a two-dimensional image obtained by X-ray of fluoroscopy (X-ray movie) like, but not limited to angiography. This invention can be used in all areas where a minimally invasive intervention is performed that is visualized in a live fashion through fluoroscopic imaging. The 2D fluoroscopic images are annotated with a 2D image that represents a 3D object of the organ that is treated. Use cases are the implantation of a pacemaker in the heart, performing tissue biopsies or injections in the heart or in the bladder, performing implantations of valves or stents or performing tissue ablations in the organ of interest. X-ray (Rdntgen) and fluoroscopy (E.G. angiography) imaging, is well known in the art. In X-Ray Angiography (XA), a radioactive source is placed behind an object and x-rays project through the object onto an image receptor (detector) plate to produce a projection image. Objects are thus imaged, by irradiating them with a light source with light to which they are at least partly translucent, and receiving the light transmitted through the object on a light sensitive plane, capable of forming a visible image.

[0003] Because most irradiating sources provide a divergent light source, the received image provides a distorted two-dimensional view of the object. Additionally, the object may in fact be depicted from a side that is in fact a backside, because a working space or area is required at the opposite side, which is then to be seen as a front side. This is a disadvantage when the image is used to provide information when working on the object.

[0004] Another disadvantage of images provided via X-ray angiography in general, is that only limited aspects of the actual object can be made visible. These may, dependent of the purpose of the image, not provide enough information.

[0005] For this latter reason it may be desired to provide the image obtained by X-ray angiography with annotations. The annotations may have various forms and may be generated based on various inputs. One form may be a three dimensional representation of the object. Such representation may be obtained by calculations or estimates, or be based on earlier measurements of the object, such as another (type of) imaging method. One possible way is to use an image formed by another imaging modality like, but not limited to, MRI or CT imaging.

[0006] The representation that is used as a base for the annotation, is usually not formed by X-ray angiography, and / or may not be distorted as a result, but rather be provided in for instance a three dimensional (3D) mesh of measurement points described in Cartesian coordinates that represent its actual shape and format. It may further not be inverted by being irradiated from an actual backside. This makes it difficult to merge the images in a visual representation.

[0007] On top of this, (commercially) available image processing software may not be able to perform all transformations or mathematics required for this purpose.

[0008] Yet another disadvantage of 2D X-ray angiography is that the locations of anatomical structures or interventional devices / instruments cannot be determined on a single image. Another acquisition from at least one different angle is necessary to obtain all information necessary to calculate the 3D position.

[0009] It is therefore a first goal of the present invention to provide a method and system for obtaining a representation of a three-dimensional object on a two-dimensional image obtained by X-ray angiography, and a further goal to provide in particular a method and system to calculate a 3D location based on two distorted 2D acquisitions taken at two different angles, that take away the disadvantages of the prior art, and / or to provide a useful alternative to systems and methods accordingto said prior art.

[0010] The invention thereto proposes a method for obtaining a representation of a three-dimensional object on a two-dimensional image obtained by X-ray angiography, comprising the steps of:

[0011] Positioning a divergent X-ray angiographic radiation source at a source position on a first side of an object in a physical space;

[0012] Positioning a receptor surface in the physical space on a second side of the object opposite the first side;

[0013] Obtaining a first digital two-dimensional projection image of the object on the receptor surface through the radiation from the source;

[0014] Obtaining a digital three-dimensional representation of at least part of the same or the same object;

[0015] Positioning the first digital two-dimensional projection image in a digital space that is a representation of at least part of the physical space;

[0016] Providing a projection of the three-dimensional representation on a virtual projection screen in digital space from a position corresponding to the source position, which virtual projection screen is arranged parallel to the two-dimensional projection image and which projection forms a second digital two-dimensional projection image;

[0017] Providing a combined display of the first and second digital two-dimensional projection image. By providing a projection of the three-dimensional representation on a virtual projection screen in digital space from a position corresponding to the source position, which virtual projection screen is arranged parallel to the two-dimensional projection image and which projection forms a second digital two-dimensional projection image, the method according to the invention projects the three dimensional representation in the same perspective as the first two dimensional representation, thus obtaining a second two-dimensional representation, that is rendered and distorted in the same way as the image obtained by X-ray angiography. As a result, the combined display of the first and second digital two-dimensional projection provides both images in a consistent and comparable manner.

[0018] A further advantage of the method according to the present invention is that it can be actually implemented in practice, by adjusting and tweaking parameters that are or can be made available in image processing software, and that allow to perform required calculations on a real-time basis. Especially three dimensional image processing and rendering may require extensive processing power. The method according to the invention is suitable to be implemented by parameter settings and allocating and using calculation power that is normally available on a graphic card of image processing and / or displaying devices. The object according to the invention may be any object that is suitable to be imaged by means of X-ray angiography, such as (part of) a human body.

[0019] In a preferred embodiment of the invention, the method comprises determining the focal point of the two-dimensional representation and the three-dimensional representations. As the focal point, the centroid, may be used. In mathematics and physics, the centroid, also known as geometric center or center of figure, of a plane figure or solid figure is the arithmetic mean position of all the points in the surface of the figure.

[0020] The focal point may be used as a single point indication of where the object is positioned, and it allows to position or to initialize the positioning of the first and second digital two-dimensional projection with respect to each other. The method according to the invention may comprise providing a user with the possibility to position the first and second digital two-dimensional projection relative to each other, and / or to provide a user interface with means for mutually moving or positioning the first and second digital two-dimensional projections.

[0021] As they represent the same, in most cases it may be beneficial to position the two-dimensional and three-dimensional representations in digital space in such a way that both theirfocal points on the combined image coincide. This may be done by the aforementioned user interface, orthe focal points may be recognised in an automated manner, directly or indirectly for instance by recognising fiducial markers or organ surfaces or similar landmarks in the images, and the first and second digital two-dimensional projections may be automatically positioned with respect to each other.

[0022] Furthermore, in order to provide a useful combination, the method may comprise the step of adjusting the scale of the second digital two-dimensional projection image by changing the distance in the digital space from the virtual projection screen to the source position. The further the virtual projection screen is from the source position, the larger the depicted image is.

[0023] Preferably, the position of the virtual projection screen is chosen to coincide with the position of the first digital two-dimensional projection image in the digital space. In that case, both scale and perspective of both images are the same. The positioning of the virtual projection screen may be done automatically, based on recognition of common features in the images, or manual positioning means may be provided in a user interface.

[0024] In a further preferred embodiment, both the first and second digital two-dimensional projection images are displayed from a field of view from a side of the receptor surface remote from the object and the virtual projection screen. This may comprise displaying the second digital two- dimensional projection image from a field of view from a side of the virtual projection screen remote from the object comprises mirroring a projection image from a field of view from the source position.

[0025] The obtained digital, three-dimensional representation of at least part of the same or a comparable object is for instance arranged in Cartesian coordinates, and in particular consists of a group of measuring points, such as a mesh. Such digital, three-dimensional representation may for instance be obtained by an MRI scan of the same or of a similar object.

[0026] A further embodiment of the invention may comprise translating or rotating the three- dimensional representation in digital space before or during projection or colouring surfaces in the second digital two-dimensional projection image, whether or not partially transparent.

[0027] In yet a further embodiment, the invention comprises mirroring the digital three-dimensional representation of at least part of the same or the same object in a plane parallel to the two- dimensional projection image and through the focal point. Herewith it is assured that objects are displayed in such way that their front and rear side actually match the viewing direction from a side of the plane(s), remote from the source. This is perceived intuitive, and may also be used to base rotational directions of the image on. That is, when rotating the two dimensional projection of the three dimensional object on a display, a front and rear side are depicted in such way that they correspond to the viewing direction from the user.

[0028] A further embodiment of the invention comprises additional steps to identify an exact 3D location based on two first digital two-dimensional projection images obtained from two different angles around the object without the exact knowledge of the distance between the source position and the first digital two-dimensional projection images and the exact size of the first digital two-dimensional projection images in physical space. After positioning the two- dimensional and three-dimensional representations in digital space in such a way that both their focal points on the combined image coincide, a location of an anatomical landmark or interventional device can be determined on the first of two acquired first digital two-dimensional projection images. This determined location can be a representation of an anatomical landmark or an interventional device on the image (in the following referred to as the first projection point). To determine this location in virtual 3D space, a virtual two-dimensional projection image is positioned at any distance from the source position in a direction perpendicular to the first and second digital two-dimensional projection images , and a projection of the coinciding focal points of the two-dimensional and three-dimensional representations is done on this virtual plane (in the following referred to as the second projection point). Subsequently, the difference between the two projection points 1 and 2 in the virtual two-dimensional projection image are determined by taking the difference in their locations in image coordinates. These offsets determine a first estimated 3D point. Next, this first estimated 3D point is reprojected to the virtual two-dimensional projection image plane by a line from the source location to the virtual 2D image to represent it on the virtual two-dimensional projection image (in the following referred to as the third projection). Then, the estimated 3D point is moved in space such that the third projection point matches with the first projection point 1 on the virtual two-dimensional projection image. As a result, a virtual line between the source position and the true position of the 3D point is obtained without knowledge of the exact distance between the source position and the first digital two-dimensional projection images and the exact size of the first digital two- dimensional projection images in physical space. When the abovementioned steps are repeated for two first digital two-dimensional projection images obtained from two different angles around the object, but with coinciding focal point positions, the exact 3D location of the determined location of an anatomical landmark or interventional device in two first digital two- dimensional projection images obtained from two different angles around the object can be determined on the intersection of the two virtual lines between the source position and the projection of the determined locations in the two first digital two-dimensional projection images obtained from two different angles around the object.

[0029] Herein, the first projection point 1 may be chosen or selected or indicated manually, or be automatically determined in the 2D image plane using feature detection software. Standard functions for template matching based on which a device like a catheter can be recognized within a few frames in a 2D image may be used.

[0030] In an embodiment of the invention both images acquired at different angles around the object, but with identical focal point positions (origin of rotation) are obtained simultaneously by using a so-called bi-plane fluoroscopy system.

[0031] The invention further relates to a device for obtaining a representation of a three-dimensional object on a two-dimensional image. Such device may in general comprise a computing or processing apparatus configured for performing the above described method according to the invention.

[0032] Specifically, the invention relates to a device for obtaining a representation of a three- dimensional object on a two-dimensional image obtained by means of X-ray angiography, comprising:

[0033] A data processing unit, such as a computer or cloud service, designed for receiving a first digital two-dimensional projection image of a diverging X-ray angiographic radiation source object placed on a first side of an object in a physical space on a receptor surface on a second side of the object opposite the first side, a digital three-dimensional representation of at least part of the same or similar object, positioning the first digital two-dimensional projection image in a digital space that is a representation of at least part of the physical space, providing a projection of the three-dimensional representation on a virtual projection screen in digital space from a position corresponding to the source position, which virtual projection screen is arranged parallel to the two-dimensional projection image and which projection forms a second digital two-dimensional projection image and providing a combined display of the first and second digital two-dimensional projection image. Additionally, the device may be configured for providing the combined display on a screen.

[0034] In a practical embodiment, the open source VTK toolkit may be used for visualization. This toolkit enables users to load meshes and to handle placement of lights, to set mesh properties like opacity and color. VTK uses a camera class object, vtkCamera, to render the meshes in the 3D scene. To ensure that the projection of the 3D meshes representing the object is the same as that of the X Ray Angiography of the object, according to the invention, the following approach may be followed.

[0035] A projection camera is moved to the same position as the X Ray Angiography Source. When a camera Focal Point is at the center of the object, this means that the camera is moved to a position on the diametrically opposite side of the Focal Point of the object to be depicted, at a distance equal to the Source-Object distance. This can be done by subtracting the Focal Point position, which is considered as 0, 0, 0, from the Camera Position, or in otherwords by translating the camera position to 0,0,0, negating the values of the resulting position, or in other words rotating the camera around 0,0,0 and then adding the Focal Point to that translated and rotated position. Mathematically, this can be described in a form of a projection matrix of a camera, which also may give the correct depth scaling (S) of projected mesh vertices.

[0036] Applying the below given matrix may have the result that left and right of the 3D meshes appear flipped, and the same goes for the back and the front of the meshes. A so called homogeneous User Transform perspective matrix on the camera may look as follows.

[0037] Herein, [0,0] and [2,2] are negated to perform the X / Y and View Axis flips, ‘S’ is a scaling factor to scale the resulting projection. This is required, because moving the camera to the source position results in quite a small projection.

[0038] The invention will now be elucidated into more detail with reference to the following figures.

[0039] Herein:

[0040] Figure 1 shows a schematic view of an X-Ray Angiography setup;

[0041] Figure 2A shows schematic view of a projection from a regular camera perspective;

[0042] Figure 2B shows schematic view of a projection from an X-Ray Angiography perspective;

[0043] Figure 3 shows the localization of a 3D point projected on detector images at two view angles;

[0044] Figure 4 shows iterative registration of a digital camera system to match with a scan that is made on the X-ray imager; Figure 5 shows locating a line drawn between a camera position and a selected point on the image; and

[0045] Figure 6 shows a step in obtaining the closest point between two lines in 3D.

[0046] Figure 1 shows a schematic view 1 of an X-Ray Angiography setup. In this specific example, the object 3 to be represented is a persons 2 or patients 2 heart 3. The setup is suitable for use in a method for obtaining a representation of a three-dimensional object on a two-dimensional image obtained by X-ray angiography according to the present invention. In the image, a divergent X-ray angiographic radiation source 4 is arranged at a source position 5 on a first side A of the object 3 in a physical space 6. A receptor surface 7 is positioned in the physical space on a second side B of the object 3 opposite the first side A. The receptor surface 7 is used for obtaining a first digital two-dimensional projection image of the object 3 on the receptor surface 7 through the radiation from the source 4. Multiple aspects of the projected image are determined by the Source to Object Distance (SOD) and Source to Image Receptor Distance (SID).

[0047] In figure 1 , the following terms and equations are applicable:

[0048] Figure 2A shows schematic view 20 of a projection from a regular camera 21 perspective. In this case, parts of the object 22 that are on the front from the perspective of the camera appear F larger than parts of the object that are on the back of the object appear B. In addition, parts that are closer to the camera 21 seemingly appear on a near clipping plane 24 in front of those closer to the far clipping plane 23.

[0049] Figure 2B shows schematic view of a projection from an X-Ray Angiography perspective. When comparingwith figure 2A, it becomes clearthat with a camera position 21 ’ that coincides with the position 5 of the X-Ray Angiography source, parts of the object 22 that are on the front from the perspective of the camera now appear F’ smaller than parts of the object that are on the back of the object appear B’. In addition, parts that are closer to the camera 21 ’ seemingly appear on a near clipping plane 24 behind those closer to the far clipping plane 23.

[0050] Figures 3-6 show the additional steps to identify an exact 3D location based on for two first digital two-dimensional projection images obtained from two different angles around the object without knowledge of the exact distance between the source position and the first digital two- dimensional projection images and the exact size of the first digital two-dimensional projection images in physical space.

[0051] Figure 3 illustrates an overview of a system to image an object with fluoroscopy by capturing images from two view angles and calculating the 3D position of any point projected onto those images. This can be applied with a single C-arm system by rotating at two different angles, or with a biplane system equipped with two C-arms. Images are captured at two source positions (Caand Cb) corresponding to two detector / image positions (la, lb). When two 2D points are selected on the images — corresponding to the projected locations (P'aand P'b) of a target point P — the method according to the invention allows to calculate the 3D location of the target point P in space, without knowledge about the distances between the source location and the first digital two-dimensional projection images (SID fig 1 ) as well as the image size (#Pxfig 1 .)

[0052] A combined display of the first and second digital two-dimensional projection image in which the positions of the two-dimensional and three-dimensional representations in digital space are done in such a way that both their focal points on the combined image coincide allows the calculation of a 3D location.

[0053] In this figure, calculation includes the following steps:

[0054] 1 . Iterative registration of the two-dimensional representation of the 3D object displayed on the first digital two-dimensional projection image for two views.

[0055] 2. For each of the two views, an anatomical location or a location of an interventional device is determined by a point. A line will be drawn between the determined point and the source position.

[0056] 3. The intersection between the two lines obtained in step 2 is calculated.

[0057] 4. This intersection point is then displayed in a 3D workspace for visualization.

[0058] Figure 4 shows the positioning steps (arrows) of the three-dimensional representations in digital space in such a way that in both acquisitions from different angles, the focal points of the two- and three-dimensional representations of the object on the combined image coincide, that ensures that the location and orientation of the two-dimensional representation of the 3D object matches with the object in physical space. The relative positions of the source and the second digital two-dimensional projection images are rigid and fixed. The whole system of Ca-la-Cb-lb is iteratively repositioned such that what is seen in the image la, lb correctly overlaps the two- dimensional representation of the 3D object with the digital two-dimensional projection image.

[0059] This is done consecutively between the two views and repeated until the best match is achieved. When in both the la and lb images a correct overlap of the two-dimensional representation of the 3D object with the digital two-dimensional projection images are obtained, the registration process is stopped.

[0060] The two sources are directed toward the same focal point (FP) — a point in space where both sources are directed, serving as a common reference point for the imaging system. This point helps establish spatial relationships between the sources, virtual cameras, two-dimensional projection image planes, and the object being imaged.

[0061] Initially, the digital source system and the object model are not at the same relative position as in the physical system. The real scan refers to the physical system consisting of the C-arm imager + image plane (detector). A scan leads to a 2D fluorescence image of the real (physical) 3D object.

[0062] At this time, the focal point of the digital 3D object is at an initial position (FPinit). During positioning steps (arrows fig 2) of the three-dimensional representations in digital space, the 3D object and the focal point is repositioned (translated) along the plane parallel to the detector plane such that in both acquisitions from different angles, the focal points of the two- and three- dimensional representations of the object on the combined image coincide and the location and orientation of the two-dimensional representation of the 3D object matches with the two- dimensional projection of the object in the two first digital two-dimensional projection images acquired from different angles. As illustrated in Figure 4, when following an iterative approach each repositioning step moves the focal point closer to the correct focal point. When in both the la and lb images an optimal overlap between the two-dimensional representation of the 3D object and the two-dimensional projection of the 3D object in the two first digital two- dimensional projection images la and lb are obtained, the repositioning process is stopped.

[0063] When the two first digital two-dimensional projection images are acquired subsequently in time, with a single C-arm, it may be important to keep the physical object unmoved between the two acquisitions or a motion correction needs to be applied in such case. . Figure 5 shows a marking of a point on an image. The unknown 3D location which the invention seeks to find is the location of the target point P. As a result, several steps may be done automatically:

[0064] 1 . The point (P') is determined on the first digital two-dimensional projection image.

[0065] 2. An initial first guess (dummy) point Po is chosen on the plane (C, FP, Po).. The reprojection of Po on the first digital two-dimensional projection image can be anywhere on the cross-section between the first digital two-dimensional projection image and plane C-FP-Po. This cross-section is the bold diagonal line in Figure 5.

[0066] 3. This point Po is reprojected to the first digital two-dimensional projection image and displayed on the first digital two-dimensional projection image as P'o.

[0067] 4. Point P'o is matched to P' by matching line CP'Oto line CP'. The result of this step is a correct line drawn through the source position C and the 3D target point P.

[0068] The reason behind this algorithm is that the virtual camera position C and focal point FP are always projected in the center of the image, and the projected point P'ois always on the line intersection between plane (C, FP, Po) and the first digital two-dimensional projection image. Any initial point on this plane may be chosen, it does not need to be on the exact location of the physical detector plane — therefore it is not necessary to know or determine a source-to-image distance (SID) to know where the detector is exactly. No image pixel size is required either. This is also robust to source-to-object-distance (SOD) — a change in SOD can be compensated effectively by a zoom factor that may be controlled via a user interface or other automated method.

[0069] The reprojection of point Po must be done exactly the same way as the two-dimensional projection of the 3D object in the two first digital two-dimensional projection images, and a positioning step such that in both acquisitions from different angles, the focal points of the two- and three-dimensional representations of the object on the combined image coincide as described in the previous step will take care of it. The algorithm ensures that the location marked in the first digital two-dimensional projection images is the same as where the obtained line is projected on the first digital two-dimensional projection images.

[0070] An intersection point between the two lines obtained from two first digital two-dimensional projection images obtained from two different angles around the object as described in the previous steps is calculated. The two lines can cross each other at a point, they can be skew, or they can be parallel. The two lines may be skew. For this situation (Figure 6), the closest points on the two lines are located, and the midpoint between those two points is the "intersection" point between the lines.

[0071] The obtained point is then displayed in 3D in exactly the same way as the object model, thereby locating the point relative to the model.

Claims

Claims1 . Method for obtaining a representation of a three-dimensional object on a two-dimensional image obtained by X-ray angiography, comprising:Positioning a divergent X-ray angiographic radiation source at a source position on a first side of an object (a patient) in a physical space;Positioning a receptor surface in the physical space on a second side of the object (the patient) opposite the first side;Obtaining a first digital two-dimensional projection image of the object on the receptor surface through the radiation from the source;Obtaining a digital three-dimensional representation of at least part of the same or the same object;Positioning the first digital two-dimensional projection image in a digital space that is a representation of at least part of the physical space;Providing a projection of the three-dimensional representation on a virtual projection screen in digital space from a position corresponding to the source position, which virtual projection screen is arranged parallel to the two-dimensional projection image and which projection forms a second digital two-dimensional projection image;Providing a combined display of the first and second digital two-dimensional projection image.

2. Method according to claim 1 , comprising determining the focal point of the first digital two- dimensional representation and the focal point in the three-dimensional representation.

3. Method accordingto claim 2, comprising positioning the two-dimensional and three- dimensional representation in digital space in such a way that both theirfocal points on the combined image coincide.

4. Method accordingto any of the preceding claims, comprising adjusting the scale of the second digital two-dimensional projection image by changing the distance in the digital space from the virtual projection screen to the source position.

5. Method accordingto claim 4, wherein the position of the virtual projection screen is chosen to coincide with the position of the first digital two-dimensional projection image in the digital space.

6. Method accordingto any of the preceding claims, comprising positioningthe digital two- dimensional projection images in such a way that organ contours or other landmarks for registration visible in both images coincide.

7. Method accordingto any of the preceding claims, wherein both the first and second digital two-dimensional projection images are displayed from a field of view from a side of the receptor surface remote from the object and the virtual projection screen.

8. Method accordingto claim 7, wherein displaying the second digital two-dimensional projection image from a field of view from a side of the virtual projection screen remote from the object comprises mirroring a projection image from a field of view from the source position.

9. Method accordingto any of the preceding claims, wherein the obtained digital, three- dimensional representation of at least part of the same or a comparable object is arranged in Cartesian coordinates, and in particular consists of a group of measuring points, such as a mesh.

10. Method accordingto claim 4, wherein the digital, three-dimensional representation is obtained by an MRI scan of the same or the same object.11 . Method accordingto any of the previous claims, comprising translating or rotating the three- dimensional representation in digital space before or during projection.

12. Method accordingto any of the preceding claims, comprising colouring surfaces in the second digital two-dimensional projection image, whether or not partially transparent.

13. Method according to any of the preceding claims and claim 2, comprising mirroring the digital three-dimensional representation of at least part of the same or the same object in a plane parallel to the two-dimensional projection image and through the focal point.

14. Method accordingto any of the preceding claims including claims 2 and 3, comprising: rotating the X-ray angiographic radiation source and the receptor surface in the physical space about the same angle around an axis:• that goes through the focal point, and• that is perpendicular to a line from the radiation source to the receptor surface, which line is preferably also perpendicular to the receptor surface, and subsequently repeating the steps of: o Positioning a divergent X-ray angiographic radiation source at a source position on a first side of an object (a patient) in a physical space; o Positioning a receptor surface in the physical space on a second side of the object (the patient) opposite the first side; o Obtaining a first digital two-dimensional projection image of the object on the receptor surface through the radiation from the source; o Positioning the first digital two-dimensional projection image in a digital space that is a representation of at least part of the physical space; o Providing a projection of the three-dimensional representation on a virtual projection screen in digital space from a position corresponding to the source position, which virtual projection screen is arranged parallel to the two- dimensional projection image and which projection forms a second digital two- dimensional projection image; o Providing a combined display of the first and second digital two-dimensional projection image.

15. Method accordingto any of the preceding claims, wherein a digital two-dimensional projection image plane of infinite size is positioned at any distance from the virtual camera position in a direction parallel to the first digital two-dimensional projection image.

16. Method accordingto any of the preceding claims wherein the relative location of an anatomical landmark or an interventional device with respect to the centre of the digital two- dimensional projection image is determined by a dummy point.

17. Method accordingto any of the preceding claims wherein the 3D location of the dummy point is reprojected onto the digital two-dimensional projection image and its relative location is updated such that the location of a target point is obtained.

18. Method according to any of the preceding claims in which the cross-section of the lines between the virtual camera position and the target point in at least two angles around the object is stored in 3D.

19. Device for obtaining a representation of a three-dimensional object on a two-dimensional image obtained by means of X-ray angiography, comprising:A data processing unit, such as a computer or cloud service, designed for: receiving: o a first digital two-dimensional projection image of a diverging X-ray angiograpic radiation source object placed on a first side of an object (a patient) in a physical space on a receptor surface on a second side of the object (the patient) opposite the first side ; o a digital three-dimensional representation of at least part of the same or similar object; positioning the first digital two-dimensional projection image in a digital space that is a representation of at least part of the physical space; providing a projection of the three-dimensional representation on a virtual projection screen in digital space from a position corresponding to the source position, which virtual projection screen is arranged parallel to the two-dimensional projection image and which projection forms a second digital two-dimensional projection image; providing a combined displayof the first and second digital two-dimensional projection image.

20. Device accordingto claim 14, comprising providingthe combined display on a screen.

Images