DE102007039207A1 - Object collision detection method for dimensioning of medical y-stent utilized during endcoscopic procedure in patient, involves determining whether point of frame lying in limitation capacity of another frame is lying within closed surface - Google Patents

Abstract

The method involves calculating a limitation capacity that completely receives one of two line segments of a wire frame of objects (9, 9'). A closed surface is formed from the two line segments by performing triangulation. Determination is performed for each of the whether one of points lie within limitation capacity of another wire frame. Determination is performed whether each point of the former wire frame lying in the capacity of the latter wire frame is lying within the closed surface. The limitation capacity is determined by axis aligned bounding box method. An independent claim is also included for a device for dimensioning of a medical y-stent.

Description

The The invention relates to a calculation method for detection collisions of at least two by means of a respective wireframe model represented spatial objects.

through a wireframe model can be a real spatial Display object on a virtual level. This virtual level includes in particular a computer system on which software for virtualization spatial objects is set up. In such a For example, software is one for three-dimensional Representations suitable CAD program. With the help of such wireframe models can be technical components or technical elements on one simulate virtual level in a simple way. Several of these technical Components are virtually one another to a complex technical Device can be combined. Components can also be connected to each other abut or abut each other. For a realistic Simulation of a complex technical device is therefore important, the touch of the individual components, so-called Collisions to capture each other.

The aim of the invention is therefore to provide a calculation method with which the touches can be detected as precisely as possible by means of wireframe models of represented spatial objects. The creation of wireframe models is for example in the technical article of Per-Olof Persson and Gilbert Strang, "A Simple Mesh Generator in Matlab," SIAM Review, Volume 46 (2), pp. 329-345, July 2004 , explained.

The This invention makes use of the fact that the wireframe models the spatial objects each have a plurality of ge closed and by means of connecting lines of interconnected polylines.

In a first method step is in each case two adjacent Line up a wireframe model with both polylines fully accommodating bounding volume calculated. In other words, both lines are complete included in this bounding volume.

In a second process step becomes from the two Linienügen using triangulation a closed surface Triangular surfaces formed. This closed surface triangular areas are completely within the limiting volume and encloses a partial area this limit volume. The enclosed by the triangular surfaces Volume is that approximated by the wireframe model from the spatial object between the two lines of lines occupied volume.

In a third process step is closed for each Polyline of a wireframe model checked, whether at least one of its points is within a bounding volume a wireframe model of another object is located. Such points of the considered Linienzuges are therefore within the of the Triangular surfaces enclosed volume or in its immediate vicinity. These points thus represent candidates for their position a collision with another spatial Object takes place.

For each of these points will continue to be reviewed in a fourth step, whether he also within the closed surface of triangular surfaces lies. Exactly in this case lies a collision of the two spatial Objects in front. The sum of the points where there is a collision provides the virtual collision area of the two considered Objects. Leave the collisions between more than two objects always identify yourself by always colliding each time in turn between two of these objects.

All four process steps are based on simple geometrical considerations, which can be implemented with the basics of vector analysis. Therefor Compared to complex calculations, this is only a comparative one low computer power necessary, so that all perform four process steps relatively quickly in time to let.

There continue only for points of line trains, the in bounding volumes of other spatial objects whether it is also in the belonging of triangular surfaces enclosed volumes, can be the computational Effort for detecting a collision of two spatial Lower objects further. The collisions between spatial Objects can thus be detected quickly and easily.

The Sum of all spatial points in which mathematically a touch of two objects is the virtual Collision area of the two objects. Because by means of representation of spatial objects as wireframe models the reality good approximation, can be based on of the determined virtual collision area also on the actual Close the collision area.

This minimizes the risk of incorrect dimensioning of individual components of a complex device. It can be determined whether the individual components of the device fit together dimensionally. A production of a prototype whose individual components are not dimensionally related fit and therefore costly us time-consuming changes must be made, is therefore excluded. This is for example in the development of an automotive engine, which is composed of many, sometimes very complex items, a great advantage. It is thus possible to virtually assemble the motor vehicle engine in all individual parts with the aid of such a CAD program. In this case, it can be determined, for example, whether the motor vehicle engine can be installed in its assembled state with its virtually predetermined dimensions in an installation space of a motor vehicle body. Thus, attachments or structures of the motor vehicle engine can be designed in advance so that it does not come later during installation to any difficulties.

In In medical technology it is possible with this method, one Model prosthesis with such a wireframe model and They virtually at the destination in the body of a patient use. For this, the body region of the patient, in which the prosthesis is to be inserted, by means of a medical imaging device, for example with a computer tomograph or detected with an ultrasonic measuring device. To a spatial Obtaining information about the body region becomes image information recorded from different measuring positions and offset against each other. The captured image information is then used with the Computer system prepared so that the insertion of the prosthesis virtual is adjustable.

Of the Seat of the prosthesis at its destination can thus be on a screen immediately followed by a medical doctor. at The parameters of the prosthesis can be easily met be modified until the seat of the virtual prosthesis is satisfactory. The actual prosthesis is based the parameter determined in this way or made of a Set of standardized prostheses selected. Thus, at one Inserting the prosthesis in a surgical procedure the best possible Ensures fit of the prosthesis. The risk of a repeat operation for replacing a faulty sized prosthesis is therefore low.

In an expedient development, the limiting volume accommodating the two polylines is formed according to the AABB method. AABB stands for Axis-Aligned Bounding Box or German axis-oriented bounding volume. The AABB method is described in the technical article of G. van den Bergen, "Efficient Collision Detection of Complex Deformable Models Using AABB Trees", Journal of Graphics Tools, 2 (4): 1-14, 1997 , described. The first vertex is formed from the minimal Cartesian x, y, and z coordinates of both polylines. The second vertex is formed from the maximum Cartesian x, y and z coordinates of both polylines. Subsequently, straight lines running through the two determined vertices and parallel to the three axes of the Cartesian coordinate system are constructed. The remaining six points of the bounding volume are the intersections of these lines with each other. This method for creating bounding volumes is characterized by the fact that only six coordinates need to be stored per bounding volume. Furthermore, the calculation method for setting the two vertices and the other six intersections is very simple. Thus, a fast and the computer system with regard to memory requirements or processor utilization little incriminating calculation of the limitation volumes is possible. Furthermore, it is very easy to check whether a point is contained in a bounding volume formed according to the AABB method. For this, only the Cartesian coordinates of the point are compared with the Cartesian coordinates of the bounding volume. An arithmetically complex conversion of the coordinates of the point is not necessary for this purpose.

In an expedient refinement, a point of a further wireframe model lying within a delimiting volume and representing a candidate for a collision of two spatial objects taking place in its coordinates is connected to a further point outside this delimiting volume. Based on the number of intersections with the triangular surfaces of the closed surface, it is then decided whether the point lies inside or outside the closed surface. One method by which it can be decided whether a straight line intersects a triangle is in the technical article of T. Möller and B. Trumbore, "Fast, Minimum Storage Ray-Triangle Intersecution", Journal of Graphics Tools, 2 (1): 21-28, 1997 , described. The method described in the technical article by T. Möller et al. Is based on simple geometrical considerations. It is therefore technically easy to implement. By means of the number of intersections of these straight lines with triangular surfaces of the closed surface, it is thus determined whether, at a point which represents a candidate for a collision of the two spatial objects, this collision actually takes place. If there is no intersection of the straight line with a triangular area, or if the number of points of intersection is straight, the point is outside the volume enclosed by the triangular areas. There is no collision in this case. On the other hand, if the number of intersections is odd, the point lies within that volume and there is a collision of the observed spatial objects at that point. This fact is generally stated in the technical article of S. Krishnan et. al., "Interactive Boundary Computation of Boolean Combinations of Sculptured Solids ", Computer Graphics Forum 16, 3 (August 1997), pp. 67-78, ISSN 1067-7055 pointed.

Advantageous are spaced apart for both lines Defined support points and the support points of the both lines mutually to triangulation, d. H. for the formation of the triangular surfaces, interconnected. Also advantageous for the triangulation of a closed line whose center of gravity determined. Subsequently the formation of the triangular surfaces is done by connecting lines between points of the line and between the center of gravity. On This is the way of the triangular surfaces form enclosed volumes in a simple manner. This one represents Section of the spatial object between the two lines of lines represents.

expedient All points within a closed surface another bounding volume, to a virtual collision area summarized. Following is for the collision area a deformation of the spatially related to the points Calculated objects. For touching points no further expansion is possible. This will be for example achieved by placing an appropriate on the spatial objects acting force is simulated. In this way, a mutual non-reality-based penetration of the spatial Objects avoided. By simulating a reciprocal acting Rather, it comes to a realistic deformation of the spatial Objects during their virtualization. This material data and Consider the geometry of the spatial objects in the calculation.

In an advantageous variant, the closed polylines the geometry of circular rings. By means of such circular rings Tubular or channel-like structures can be made more realistic Mimic manner.

In an advantageous variant is in the by wire mesh models represented spatial objects around the thighs a stent. A stent is generally a medical implant, a lattice framework in metal or tubular form Plastic has and serves as a vascular prosthesis. Thus, a stent serves to dilate a vessel narrowing, a so-called stenosis. Furthermore, a stent is used, a vascular dilation, a so-called aneurysm, to bridge. These are the meshes of the lattice framework lined with a synthetic fabric. The blood is flowing now within the walls of the vascular prosthesis. The Aneurysm is no longer supplied with blood. The danger of a Tear of the aneurysm is therefore no longer.

The medical background for this is in the dissertation of S. Märchen, "The Inflammatory Abdominal Aortic Aneurysm: Comparison of the Endovascular Stent-Graft Implantation with the Conventionally Open Shut-Off", University of Ulm, 2004 , described. The insertion of such a stent is performed endoscopically. This is a so-called endovascular therapy procedure. A method for endoscopic navigation within the body of a person is in the WO 200725081 A2 described.

Further Details of the treatment of an aneurysm using a stent as a vascular prosthesis are the embodiment refer to.

Of the Stent is about a small incision in the groin area in the folded state of a pelvic artery of a patient fed. Subsequently, the stent is using of the endoscopic navigation method to the aneurysm to be treated navigated and expanded there. He nestles from the inside to the vessel walls. It comes thus to one Collision between the stent and the vessel wall as spatial objects. For the therapy of a Aneurysms that are in the area of the branching of the heart Aorta descendens to the two leg arteries, the iliac arteries communes, a stent is inserted in Y-shape. At a such stent also touch when expanding the two Y-thighs of the stent. In this case, the two Y-thighs as spatial objects that collide with each other. Therefore, a virtual collision area of the two is determined Y-thighs of the stent. By considering this Collisions make it possible to unfold or expand realistic replication of the stent inside the body. The Collisions between individual thighs of the stent or between Thigh and vessel wall influence the expansion of the stent. The stent can thus be in virtual form in the body be introduced and its expansion or its subsequent Seat at its destination can be checked. From the determined virtual collision area can thus the expected actual collision area for a stent can be determined with these dimensions. From the determined Collision area of the legs of the stent can now reciprocally force acting on the legs and the resulting deformation the thighs are determined.

Thus, in the run-up to surgery, an exact design and dimensioning of the stent is possible. Due to the exact dimensioning, an optimal fit of the stent in a blood vessel can be achieved. The risk of a patient being subjected to risky surgery to replace the faulty stent due to incorrect sizing can be deducted, can thus be greatly reduced. With a Y stent, which is used as a vascular prosthesis for bridging an aneurysm, the risk of failure of this vascular prosthesis can be significantly reduced by slipping out of its position. Since internal bleeding can occur immediately in such a case, which normally does not lead to the death of the patient only in the case of an immediate intervention, the exact rate of dimensioning of such a stent also allows the mortality rate of patients who suffer from such aneurysm. reduce. If it is further recognized after such an operation by means of an imaging method that the stent was dimensioned incorrectly, then this stent must be removed by means of a conventional operation on the open abdomen. Also, the risk of such reoperation is thus reduced by an improved interpretation of the stent.

In An advantageous development is the size the spatial objects and / or their position to each other changed. This can be done on a virtual level capture exactly when there is a collision of two each by means of a wireframe model represented spatial Objects is coming. In this way, processes can be time-resolved to capture.

So Complex robot-controlled manufacturing processes, such as the insertion of a dashboard through the front opening of a Motor vehicle body by means of a robot, in advance on a play through virtual level and based on the knowledge gained adjust the movement of the robot so that it is at the actual Assembly not to collisions between different components comes. Also the unfolding of a stent at its destination in individual stages play like a cartoon.

The Problem is further solved by a device for Dimensioning of medical stents according to a procedure of the previous one Claims. They are directed to the process Variants and their benefits to transmit the device mutatis mutandis.

following an embodiment of the invention with reference to a Drawing explained in more detail:

1 the position of the descending aorta in a person's body,

2 a detail from 1 .

3 the detail 2 with an aneurysm,

4 a Y-stent with two Y-legs in the final assembled state and in the disassembled state,

5 to 8th insertion of the Y stent into the aorta around the aneurysm 3 .

9 a process scheme for detecting collisions of the Y-limbs of the Y-stent,

10 a wireframe model of the Y stent,

11 a detail of the two y-legs out 10 .

12 schematically the determination of limiting volumes in two steps,

13 the determination of the vertices of the bounding volumes,

14 the wireframe model 10 with bounding volumes in the area of the two Y-legs,

15 the triangulation of two adjacent polylines of the wireframe model in three steps,

16 to 18 checking whether a point lies inside or outside the volume enclosed by the triangular surfaces,

19 the wireframe model 12 in the virtual inserted into the aorta state,

20 in a total of twelve steps, the virtual insertion and expansion of the Y stent into the aorta, as well as

21 the wireframe model of a Y-stent, if no collisions of the Y-legs are considered.

1 shows the hull 1 a person with the heart roughly centered in its upper half 2 , From the heart 2 out extends in the vertical direction 3 down towards the unillustrated legs the descending aorta 4 , In the lower part of the body trunk 1 branches the descending aorta 4 in the two leg arteries, the so-called arteries iliacae communes 5 . 5 ' , The two arteries iliacae communes 5 . 5 ' provide the two legs of the person with oxygen-containing blood.

2 shows the area of branching of the descending aorta 4 in the two arteries iliacae communes 5 . 5 ' ,

3 shows a pathological enlargement of the descending aorta 4 in the area of the branch too the arteria iliacae communes 5 . 5 ' , a so-called aneurysm 6 , The following medical consideration is taken in particular from the dissertation already cited in the description by S. Märchen. Aneurysm comes from a diseased, an innate, or an acquired change in the wall of the descending aorta 4 conditions. In the area of the aneurysm 6 is due to the widening of the descending aorta 4 the flow velocity of the blood greatly reduced. This increases according to the Bernoulli laws the risk of rupture of the aortic wall. The result is a very rapid loss of blood into the abdomen, which requires immediate medical attention. However, even with immediate medical intervention, the patient often dies. Is it due to a specific symptoms to a diagnosis of an aneurysm 6 For example, by an imaging method, such as computed tomography or an ultrasound measurement method, so can be against tearing the vessel wall take operational measures.

Such a measure consists in the insertion of a vascular prosthesis. In the so-called conventional surgical method, the abdominal cavity is first opened. Subsequently, a vascular prosthesis in Y-shape in the area between the descending aorta 4 and the two arteries iliacae communes 5 used. The one end of the vascular prosthesis reaches into the descending aorta 4 and the two Y-thighs engage in the two arteries iliacae communes 5 , The existing plastic vascular prosthesis is fixed in position by sewn. The operation method is described, for example, in the dissertation by S. Märchen, which was already cited in the description. Such surgery on the open abdomen is associated with a high burden on the patient. Therefore, an alternative method of operation has developed in recent years.

Here is a in the 4 Y stent shown on the left in the final assembled state 7 used in place of the conventional vascular prosthesis on the endovascular route. This Y stent 7 has a Y base 8th on, resting in two y-thighs 9 . 9 ' branched. The Y base 8th and the two Y-thighs 9 . 9 ' consist of a lattice framework 10 , whose interstices are filled with a plastic. The lattice framework 10 of the Y-stent 7 is so collapsible that the Y-stent 7 in the collapsed state can be pushed through the blood vessels of the patient. For this purpose, the Y-stent is inserted through a small incision in the groin into one of the two iliac artery arteries 5 . 5 ' used. The insertion of the Y-stent 7 should be described only briefly. For easier assembly, the Y-stent is 7 executed in two parts. The right side of the 4 shows the Y-stent 7 in disassembled condition. The first major part includes the Y base 8th and the first Y-thigh 9 while the second part only the second Y-leg 9 ' includes.

According to 5 becomes a navigational wire 11 through one of the two arteries iliacae communes 5 threaded and in the direction of the descending aorta 4 drawn. On the navigational wire 11 is a balloon catheter 12 attached, over which the larger part of the Y-stent 7 with the Y base 8th and one of the two Y-thighs 9 pushed. By means of a control system, as for example in the WO 2007/025081 A2 is described, is constantly the location of the Y-stent 7 controlled. Once the Y base 8th into the descending aorta 4 and the Y-thigh 9 into the common iliac artery 5 engages, an actuation of the balloon catheter takes place 12 , The lattice framework 10 from Y base 8th and Y-thighs 9 expands. According to 7 is the Y base 8th in the descending aorta 4 and the Y-thigh 9 in the common iliac artery 5 fixed. Already before expanding the balloon catheter 12 became a second navigational wire 11 ' with a second balloon catheter 12 ' from the second common iliac artery 5 ' through the descending aorta 4 drawn. About the balloon catheter 12 ' is the second part of the Y stent 7 that only the second Y-leg 9 ' includes, pushed. Once the Y-thigh 9 ' with its one end to the Y base 8th connects and with its second end in the second common iliac artery 5 ' engages, an actuation of the balloon catheter takes place 12 ' for fixing the second Y-leg 9 in the second common iliac artery 5 ' ,

8th shows the Y-stent 7 in the final assembly state. The Y stent is now by the spring pressure of its lattice framework 10 held in place and fixed. Since the openings of the lattice framework are filled with a plastic, the blood now flows on the inner walls of the Y-stent 7 along. The aneurysm 6 is thus through the Y-stent 7 as it were bridged. Thus, a tight fit of the Y-stent 7 assuming his position, no surgery on the open abdomen necessary. It is vividly clear that the Y stent 7 the better it is adapted to the anatomical conditions of the patient, the better it is held in position. This applies in particular to the diameters of the Y base 8th and the two Y-thighs 9 . 9 ' as well as their length. When designing the Y-stent 7 It should be noted that the two Y-legs 9 . 9 ' at her to the Y base 8th attaching branch point in a collision area 13 touch. This collision area 13 must be taken into account in the interpretation of the Y-stent, since thereby in the final assembly state, the diameter of the two Y-legs 9 . 9 ' is affected.

9 shows a calculation method for virtual insertion of a Y-stent 7 , Core of Ver driving is here the detection of collisions of the two Y-legs 9 . 9 ' of the Y-stent 7 together.

For this purpose, first by means of a medical diagnostic device 14 Image information BI of the aneurysm 6 and the aneurysm 6 adjacent sections of the descending aorta 4 and the two arteries iliacae communes 5 . 5 ' detected. At the medical diagnostic device 14 For example, it may be a computed tomograph, an angiograph or an ultrasound meter. It is important that the captured image information BI rich in contrast and have the highest possible spatial resolution.

The image information BI becomes a processing unit 15 a computer system 16 to hand over. The processing unit 15 prepares the image information BI for display on a designed as a monitor display element 17 on. Furthermore, the processing unit causes 15 the storage of the image information BI in a data memory 18 , The on the display element 17 displayed image information BI are read by a diagnosing physician by means of an image analysis device 19 analyzed. This is in particular a on the computer system 16 installed computer program for image analysis. Access to this image analyzer 19 ge happens by means of a user interface 20 , Under the user interface 20 are all input elements, such as a computer mouse or a computer keyboard, by means of which access to the image analysis device 19 is possible. By means of the image analysis device 19 can be length and maximum diameter of the aneurysm 6 as well as the diameter of the aneurysm 6 adjacent aorta descendens 4 and the two arteries iliacae communes 5 . 5 ' determine. From the measured data can be determined the Y-stent 7 especially with regard to the diameter and the length of its Y-base 8th and his two y-thighs 9 . 9 ' dimension.

Out of those via the user interface 20 given data is for the Y stent 7 in a process step 21 calculated a wireframe model comprising a plurality of closed and interconnected by means of connecting lines. In one process step 22 For each two adjacent lines of the wireframe model, a bounding volume that completely accommodates both lines is calculated. In a further process step 23 From the two lines of lines, a closed surface of triangular surfaces is formed by means of triangulation. Subsequently, in a process step 24 for every closed polyline of a Y-thigh 9 . 9 ' checks whether at least one of its points within a bounding volume of the wireframe model of the other Y-leg 9 . 9 ' lies. In a further process step 25 becomes for every point of a Y-thigh 9 . 9 ' in a bounding volume of the other Y-leg 9 . 9 ' is still checked whether it is also within the closed surface. If both Y-thighs 9 . 9 ' of the Y-stent 7 touch, so will for the collision area 13 the deformation of the two Y-legs 9 . 9 ' in a process step 26 certainly. The image information BI / S of the Y-stent 7 is adapted and processed by the processing unit 15 together with the image information BI of the medical diagnostic device 14 on the display element 17 shown.

This procedure allows unfolding of the Y-stent 7 the two balloon catheters 12 . 12 ' gradually simulate in the manner of a cartoon. For each iteration step t1, t2, ... tN, the size of the wireframe model becomes the simulated expansion of the Y stent 7 customized.

As a result, the process steps 21 to 26 explained in more detail.

10 shows a wireframe model 27 as it is for a Y stent 7 in process step 21 is calculated. The wireframe model 27 includes closed circular lines 28 that by means of connecting lines 29 connected to each other. The lines 28 and the connecting lines 29 are in particular in the 11 recognizable, the middle area of the wireframe model 27 of the Y-stent 7 shows. The wireframe model 27 provides a vectorized image of the Y stent 7 With it, the geometric relationships of the Y-stent can be 7 , in particular the length and the diameter of the Y-base 8th and the two Y-thighs 9 . 9 ' play in a space-saving way. For better orientation, the reference numbers are for the Y base 8th as well as the two Y-thighs 9 . 9 ' in the 10 with brackets.

The 10 shows the wireframe model 27 of the Y-stent 7 im completely by means of the two balloon catheter 12 . 12 ' expanded state. How the particular 11 can be seen, touch the two Y-leg 9 . 9 ' in their mouth area in the Y base 8th on their sides facing each other in a virtual collision area 30 , In this collision area 30 practice the two y-thighs 9 . 9 ' pressure on each other, resulting in a slight deformation of the two Y-legs 9 . 9 ' leads. For the graphical representation of the wireframe model 27 Here is already the sequence of process steps 22 to 26 considered.

12 shows on the left side a section of the wireframe model 27 one of the two Y-thighs 9 . 9 ' , The better overview are half only the lines 28 but not the lines 28 interconnecting connecting lines 29 shown. To two adjacent lines 28 of the wireframe model 27 is now in the process step 22 a two lines 28 fully accommodating bounding volume 31 calculated. 12 shows on the right side again the four lines 28 with each of them receiving three bounding volumes in pairs 31 ,

The bounding volumes 31 are formed according to the so-called AABB method as so-called axis-oriented bounding volumes. In the 13 is a single such bounding volume 31 shown. For clarity, the lines are 28 in the 13 not marked with. For each of the two lines 28 With respect to the Cartesian x, y, and z coordinates, their minimum value for both polylines becomes respectively 28 determined. In this way you get the first corner 32 the limit volume 31 , Analog becomes for both polylines 28 the maximum value with respect to the Cartesian x, y, and z coordinates with respect to both polylines 28 educated. In this way you get the second corner 32 ' the limit volume 31 , Now be through both vertices 32 . 32 ' three straight lines placed parallel to the axes of the Cartesian coordinate system 33 are aligned. This results in a total of six points of intersection 34 that work together with the two vertices 32 . 32 ' the total of eight corners of the bounding volume 31 form. The AABB method is described in detail in the document cited by G. van den Bergen already in the description.

14 shows the wireframe model 27 with stacked boundary volumes 31 on both y-thighs 9 . 9 ' , For the determination of the collision area 30 only the two Y-thighs become 9 . 9 ' considered. The Y base 8th The Y-stent can expand unhindered and is not expanded by another spatial object 9 . 9 ' but only through the wall of the descending aorta 4 limited.

In process step 23 will be out of the two in a bounding volume 31 contained line trains 28 a closed surface formed. For this purpose, as shown in the left diagram of the 15 can be seen, first for both polylines 28 mutually spaced support points 35 established. These interpolation points 35 are then, as the middle schematic of the 15 shows, mutually to the formation of triangular surfaces 36 connected with each other. As it is the closed line trains 28 is circular rings, you get a cylinder jacket.

In the right sketch of the 15 Finally, first for each of the two closed polylines 28 its focus 37 determined. Subsequently, by the connection of the support points 35 with the main focus 37 triangular surfaces 38 educated. Assign, as in the example, the closed polylines 28 the geometry of circular rings forms the totality of all triangular surfaces 38 for a polyline 28 a circular disk. Through the entirety of the triangular surfaces 36 and 38 is a closed surface 36 . 38 formed, which now forms part of the limiting volume 31 encloses. This is schematically the 16 to 18 refer to. This from the triangular areas 36 . 38 enclosed volume approaches that of the Y-leg 9 . 9 ' occupied volumes between the two lines 28 at.

In process step 24 is checked for a point 39 . 39 ' . 39 ' 'one to a Y-thigh 9 . 9 ' appropriate line train 28 in a bounding volume 31 the other Y-leg 9 . 9 ' lies. In the 16 to 18 is each a point 39 . 39 ' . 39 '' shown in the bounding volume 31 lies. Only such points 39 . 39 ' . 39 '' will undergo further investigation. Roughly speaking, such points are 39 . 39 ' 39 '' , near each other's Y-thigh 9 . 9 ' and thus are candidates for a collision of the two Y-thighs 9 . 9 ' , Due to the Cartesian coordinates of the bounding volumes 31 such a test is possible very quickly.

In the process step 25 Now it is checked if such points 39 . 39 ' . 39 '' , the candidates for a touch of the two Y-thighs 9 . 9 ' even inside of the triangular areas 36 . 38 enclosed volume lie. This becomes the point to be examined 39 . 39 ' . 39 '' over a straight line 40 . 40 ' . 40 '' with one outside the limit volume 31 lying auxiliary point 41 connected. As a result, the number of triangular faces 36 . 38 determined by the straight line 40 . 40 ' . 40 '' get cut.

In 16 cuts the straight line 40 none of the triangular surfaces 36 . 38 , It is therefore also clear that the point 39 outside of the triangular areas 36 . 38 enclosed volume and thus there is no collision of the two Y-thighs at this point 9 . 9 ' comes.

In 17 cuts the straight line 40 ' one of the triangular surfaces 36 . 38 namely, the triangular area 36 ' , Thus lies the point 39 ' inside of the triangular areas 36 . 38 enclosed volume. Therefore, there is a collision of the two Y-legs 9 . 9 ' ,

In 18 cuts the straight line 40 '' two of the triangular surfaces 36 . 38 namely the two triangles areas 36 '' , The straight line penetrates 40 one of the two triangular surfaces 36 '' from the outside and then the second of the two triangular surfaces 36 '' from the inside. The point 39 '' is therefore outside of the triangular areas 36 . 38 enclosed volume.

Generally the point lies 39 . 39 '' for an even number of interfaces including the zero outside of the triangle areas 36 . 38 enclosed volume. The point is still there 39 ' for an odd number of intersections inside of the triangle faces 36 . 38 enclosed volume. Only in this last case, touch the two Y-thighs 9 . 9 ' of the Y-stent 7 ,

All determined points 39 in which the two Y-thighs 9 . 9 ' collide with each other, together form the virtual collision area 30 the two Y-thighs 9 . 9 ' , Based on this determined virtual collision area 30 can be in the process step 26 calculate the extent to which the two Y-thighs 9 . 9 ' deform due to mutual contact. For this purpose, first the force is determined that the two Y-stents 9 . 9 ' to each other. Based on the geometric features and the material data of the two Y-legs 9 . 9 ' can be calculated their deformation. At this calculated deformation of the two Y-legs 9 . 9 ' becomes the wireframe model 27 of the Y-stent 7 adjusted accordingly. The corresponding image information BI / S of the Y-stent 7 is sent to the processing unit 15 pass and on the display element 17 displayed.

19 shows the wireframe model 27 of the Y-stent 7 in the not yet fully expanded state.

The wireframe model 27 is here virtually used at its destination in the body of a patient. The anatomy of the destination became as image information BI with the medical diagnostic device 14 detected. The Y-base takes effect 8th of the Y-stent 7 into the descending aorta 4 , The two Y-thighs 9 . 9 ' engage in the two arteries iliacae communes 5 . 5 ' ,

The procedure according to 19 allows viewing the expansion of the Y stent 7 by means of the wireframe model 27 , In doing so, the diameters of the Y base become 8th and the two Y-thighs 9 . 9 ' successively increased. This is in the twelve phase images of the 20 shown. Each of these phase images shows an already inserted Y-stent according to 19 , The individual Y-stents 7 take from left to right and from top to bottom in the diameter of Y-base 8th and Y-leg 9 . 9 ' to. For each of the phase images shown, the process steps become 21 to 26 run through. At the end of each pass, the virtual collision area becomes 30 between the two Y-thighs 9 . 9 ' determined. This results in the instantaneous deformation of the two Y-legs 9 . 9 ' , The phase image at the bottom right in the 20 shows the Y-stent 7 in the fully expanded state. A medical doctor can now on the display element 17 check if the Y stent 7 has an optimum fit and whether the diameter of Y-base 8th the descending aorta 4 and the two Y-thighs 9 . 9 ' the arteria iliacae communes 5 . 5 ' so that allow unimpeded blood flow across the canal structure of the Y stent 7 is possible. Only in this case, the aneurysm can be 6 namely, safely bypass, so the risk of tearing the aneurysm 6 and a bleeding into the abdomen of the patient is safely avoided.

The virtual collision area 30 corresponds to the expected collision area 13 if you have a Y stent 7 inserted with the given properties as a vascular prosthesis in the aneurysm.

Used in the simulated expansion of the Y-stent 7 found that its seat is not optimal, so by means of the user interface 20 slightly changed parameters for the Y-stent 7 specified. This procedure is repeated until an optimal fit of the Y stent on the display element 17 is detectable. The surgery will be followed by a Y-stent 7 used with the determined stent parameters.

21 shows a Y stent 7 ' in which the collision area between his two thighs 9 '' and 9 ''' was not considered. The two Y-thighs 9 '' . 9 ''' grab the Y-base at their starting points 8th' into each other and overlap. The Y stent 7 ' is thus represented with an error. Therefore, there is the danger that a doctor finding a treatment will choose a stent which does not optimally fit the application.

QUOTES INCLUDE IN THE DESCRIPTION

This list The documents listed by the applicant have been automated generated and is solely for better information recorded by the reader. The list is not part of the German Patent or utility model application. The DPMA takes over no liability for any errors or omissions.

Cited patent literature

• - WO 200725081 A2 [0021]
• WO 2007/025081 A2 [0050]

Cited non-patent literature

• - Per-Olof Persson and Gilbert Strang, "A Simple Mesh Generator in Matlab," SIAM Review, Volume 46 (2), pp. 329-345, July 2004 [0003]
• G. van den Bergen, "Efficient Collision Detection of Complex Deformable Models using AABB Trees", Journal of Graphics Tools, 2 (4): 1-14, 1997 [0015]
• T. Möller and B. Trumbore, "Fast, Minimum Storage Ray-Triangle Intersection", Journal of Graphics Tools, 2 (1): 21-28, 1997 [0016]
• - S. Krishnan et. al., "Interactive Boundary Computation of Boolean Combinations of Sculptured Solids", Computer Graphics Forum 16, 3 (August 1997), pp. 67-78, ISSN 1067-7055 [0016]
• - S. Märchen, "The Inflammatory Abdominal Aortic Aneurysm: Comparison of the Endovascular Stent Graft Implantation with the Conventionally Open Shutoff ", University of Ulm, 2004 [0021]

Claims (10)

Calculation method for detecting collisions of at least two by means of a respective wireframe model ( 27 ) represented spatial objects ( 9 . 9 ' ), Wherein the wireframe models ( 27 ) a plurality of closed and by means of connecting lines ( 29 ) interconnected polylines ( 28 ), wherein - in each case two adjacent lines ( 28 ) of a wireframe model ( 27 ) a both lines ( 28 ) completely accommodating limiting volume ( 31 ) is calculated, - whereby from the two line trains ( 28 ) using triangulation a closed surface of triangular surfaces ( 36 . 38 ), where - for each closed polyline ( 28 ) of a wireframe model ( 28 ) checks whether at least one of its points ( 39 . 39 ' . 39 '' ) within a bounding volume ( 31 ) of a wireframe model ( 27 ) of another object ( 9 . 9 ' ), and - for each point ( 39 . 39 ' . 39 '' ) of a wireframe model ( 27 ) contained in a bounding volume ( 31 ) of another wireframe model ( 27 ), it is further checked whether it is also within the closed surface. Method according to claim 1, wherein the two lines ( 27 ) receiving limiting volume ( 31 ) is formed according to the AABB method. Method according to one of claims 1 or 2, - wherein a within a bounding volume ( 31 ) lying point ( 39 . 39 ' . 39 '' ) of another wireframe model ( 27 ) with one outside this limit volume ( 31 ) further point ( 41 ) over a straight line ( 40 . 40 ' . 40 '' ) and - whereby the number of intersection points of the straight line ( 40 . 40 ' . 40 '' ) with triangular surfaces ( 36 ' . 36 '' ) of the closed surface, it is decided whether the point ( 39 . 39 ' . 39 '' ) lies inside or outside the closed surface. Method according to one of the preceding claims, - wherein for both lines ( 27 ) spaced apart support points ( 35 ) and - whereby the reference points ( 35 ) of the two lines ( 27 ) mutually to form the triangular surfaces ( 36 ). Method according to one of the preceding claims, - wherein for the triangulation of a closed polyline ( 27 ) whose focus ( 37 ) and - whereby the formation of the triangular surfaces ( 38 ) by connecting lines to the line ( 27 ) and the focus ( 37 ) he follows. Method according to one of the preceding claims, - wherein all points ( 39 ' ) within a closed surface of another bounding volume ( 31 ), to a virtual collision area ( 30 ) and - whereby the deformation of the points ( 39 ' ) belonging spatial objects ( 9 . 9 ' ) is determined. Method according to one of the preceding claims, wherein the closed polylines ( 27 ) have the geometry of circular rings. Method according to one of claims 1 to 5, wherein the by wire mesh models ( 27 ) represented spatial objects the legs ( 9 . 9 ' ) of a Y-stent ( 7 ) are. Method according to one of the preceding claims, wherein the size and / or the position of the spatial objects ( 9 . 9 ' ) is changed. Device for sizing medical Y-stents ( 7 ) according to a method of the preceding claims.