JP2000503561A - Calibration of X-ray geometry - Google Patents

Abstract

(57) Abstract An X-ray geometry calibration system has a calibration frame that can be mounted near at least a part of a patient's body, such as a patient's head. The X-ray source cooperates with a target located at a given orientation and a given distance from said part of the patient's body to form an image of said part of the patient and at least a corresponding part of the calibration frame. The calibration frame has a coded arrangement for partially determining the correspondence between the corresponding portion of the calibration frame and the calibration frame such that the orientation and the distance are uniquely determined from the image of the corresponding portion of the calibration frame. .

Description

DETAILED DESCRIPTION OF THE INVENTION           X-ray geometry calibration system and method   The present invention is directed to a system and method for calibrating geometry in an x-ray imaging configuration. In particular, a system that can establish a correlation between a three-dimensional object and a two-dimensional image About.   Interest in reconstructing the vasculature from x-ray angiograms has increased, It has a number of uses. Various methods have been proposed and some such methods are , Contrast bolus is digital subtraction angiographic sequence Consider the dynamics of the contrast bolus as it evolves over time Trying to do that. These methods involve imaging systems or X-ray sources and The image intensifier system rotates around the patient and is well calibrated. It is assumed that Therefore, these methods are not suitable for reconstitution of the vascular system. Use the geometry of the easing system.   However, many X-ray systems are not well calibrated in practice, Errors propagate throughout the reconstruction process, reducing the accuracy of the results.   "X-ray geometry calibrator" by the inventor on the same filing date as the present application. A pending application is also referred to herein. To use. The co-pending application is concerned with content that is closely related to the content of the present application. The disclosure of the time-pending application is incorporated herein by reference to the extent it does not conflict with this application. It is rare.   Briefly, the X-ray geometry calibration device disclosed in the co-pending application is: X-ray irradiated part of patient's body, X-ray source, and imaging apparatus for forming image Used in combination with the imaging device to provide a given oriente And a calibration ring, which is located on the part of the patient's body. Tightened and fixed, the calibration ring shows the first degree of transparency to X-rays, The coding device also has a coding arrangement, which is integrated into the calibration ring The coding arrangement has multiple coding units, and the coding units are transparent Exhibit a second degree of transparency, the second degree of transparency being the first degree of transparency Unlike coding units, coding units are arranged in a predetermined manner, i.e. Image of the calibration ring with the position and the essence formed by the imaging device The two-dimensional projected image is required to determine the orientation and distance uniquely. Pattern of contrast intensity that can contribute to at least part of the coding arrangement Are arranged in a predetermined form so as to indicate   The co-pending application further discloses a calibration device, which comprises a three-dimensional object. Between the two-dimensional projected image Used in conjunction with X-ray imaging to establish Has a calibration ring, which is part of the anatomy of the subject for X-ray examination The ring consists of a carrier part attached to the subject and a calibration ring carrier. A coding arrangement embedded in the rear part, the coding arrangement comprising a plurality of cores. The code word form is different from the degree of transparency of the carrier part. The code word form is along the peripheral path along the ring. Are placed in a sequence in a given direction, separated by spaces. That is, such words must be read in this given direction, and every word More words do not appear, and none of the words are useful if read in the opposite direction to this given direction. Does not form a valid word.   The co-pending application also discloses a calibration device, which comprises a three-dimensional object and its Used in conjunction with X-ray imaging to establish correspondence between two-dimensional projection images The calibration device has a calibration ring, which is the anatomical part of the X-ray subject. The calibration ring, which corresponds to the part, A coding arrangement embedded in the carrier portion of the tag. It has a plurality of code word forms, and the code word form is composed of an image of a carrier portion and contrast. Formed from the substance that forms the calibration image in the calibration image. Part or whole smaller It is possible to establish uniquely from the part.   It can be seen that the algorithm of the present invention allows the patient to be Dynamic school of X-ray system by introducing surrounding signs That is what must be possible. One object of the present invention is to provide the most efficient Intelligence on such labels to obtain the most reliable calibration procedure in To achieve the event design.   In one aspect of the invention, the calibration frame is referred to herein as a calibration ring. Having one set of bearing balls disposed on the surface of the cylinder to be formed. Proofreading The structure of the ring is simulated. View from different directions of the calibration ring And especially the calibration ring image in a typical X-ray imaging system, This is shown using the simulation procedure in e.   In one aspect of the invention, an x-ray geometry calibration system comprises at least a patient body. Also has a calibration frame attached to a part of the body, has an X-ray source, and is in front of the patient's body. Forming an image of the portion and at least a corresponding or related portion of the calibration frame A predetermined orientation from said part of the patient's body to cooperate with the x-ray source to And a target located at a distance and the calibration frame is a calibration frame. Uniquely, the correspondence between the image of the corresponding portion or the related portion and the calibration frame is determined. Coding means to determine orientation and distance It can be uniquely obtained from the image of the corresponding part of the normal frame.   In another aspect of the invention, the portion of the patient's body and the corresponding portion of the calibration frame Are contained in a substantially flat plane.   In yet another aspect of the invention, the coded arrangement is adapted for X-rays of the coded arrangement. And a second part relatively opaque to X-rays It has at least a part of a calibration frame indicating minutes.   In yet another aspect of the invention, the coding arrangement comprises the coding of the first part. With the arrangement.   In yet another aspect of the invention, the coding arrangement comprises the coding of the second part. With the arrangement.   In yet another aspect of the invention, the coding arrangement comprises a first part and a second part. Has the coded arrangement of   In yet another aspect of the invention, the first portion of the calibration frame comprises metal spheres. You.   In yet another aspect of the invention, the second part of the calibration frame is the calibration frame. Consisting of a hole in the system.   In yet another aspect of the invention, the calibration frame exhibits a three-dimensional shape.   In yet another aspect of the invention, the calibration frame comprises an image in a substantially flat plane. From any of the three-dimensional shapes Finds which part is proximal to the image and which part is distal to the image Exhibit properties that are possible.   In yet another aspect of the invention, the calibration frame exhibits a three-dimensional shape.   In yet another aspect of the invention, the calibration frame has a flat ribbon-like structure .   In yet another aspect of the invention, the flat ribbon-like structure is formed into an endless loop. Has been established.   In yet another aspect of the invention, the pair of patient body and calibration frame The image with the corresponding part is included in the two-dimensional plane.   In yet another aspect of the invention, the calibration frame can clamp the patient's head. Configuration.   In yet another aspect of the present invention, an X-ray geometry calibration system includes a patient body. Comprising a calibration frame corresponding to the part, comprising an X-ray source, and said part of the patient's body. An apparatus for forming an X-ray image with a corresponding part of the calibration frame is provided. In order to uniquely determine the correspondence from the image of the corresponding part of the normal frame, a pair of calibration frames A coding arrangement for determining the correspondence between the corresponding part and the calibration frame.   In yet another aspect of the invention, a calibration file corresponding to at least a portion of a patient's body. Determining the correspondence between the X-ray image of the frame and the position of the calibration frame with respect to the X-ray device A method comprising: (a) encoding at least a portion of a patient's body with a coded three-dimensional Positioning a calibration frame; and (b) a calibration frame with said portion of the patient's body. A given orientation from the X-ray source to at least a portion of the frame. Exposure at a given distance and projection of said part of the patient and said part of the calibration frame Forming an image; and (c) univocally determining an orientation and a distance. Image of said part of the calibration frame and the three-dimensional calibration frame Is uniquely obtained from the observation of the image.   In yet another aspect of the invention, a calibration file corresponding to at least a portion of a patient's body. Determining the correspondence between the X-ray image of the frame and the position of the calibration frame with respect to the X-ray device A method comprising: (a) encoding at least a portion of a patient's body with a coded three-dimensional Positioning the calibration frame, the calibration frame internalizing the placement of the globules (B) X-rays of said part of the patient's body and at least a part of the calibration frame The source allows exposure at a given orientation and a given distance from the X-ray source, Forming a projected image of the part of the person and the part of the calibration frame. (C) the correspondence between said part of the calibration frame and the three-dimensional calibration frame is represented by an image (D) orientation and distance Compute And steps.   The present invention will be described in detail below with reference to the drawings.   FIG. 1 shows a configuration used in connection with the present invention, showing an X-ray source S and an image intensity. Schematic diagram showing the relative positions of the fire I and the calibration frame F with respect to the patient's head FIG. 2 shows perspective views (a) and (d), a plan view (b) viewed from above, and the application of the present invention. FIG. 3 is a schematic diagram showing the calibration ring of the present invention with a typical view (c), and FIG. Balls or bearing balls (BBs) applicable to one embodiment of the present invention 4) is a schematic diagram showing, in perspective projection, the change in dimensions and the change in spacing between them, FIG. The center of the projection is located outside of the cylinder and between the two bases of the cylinder and built into them Word changes the sign or polarity of the surface normal component in the viewing direction. And FIG. 5 is a schematic view showing a cylinder applicable to the present invention. The blank space depends on the sign of the surface normal component in the viewing direction. Indicates whether it is a 5-digit code 54321 or 12345 (a). Whether this device is excluded to unify the detection of blank spaces FIG. 6B is a schematic diagram showing a configuration applicable to one embodiment of the present invention, and FIG. The table is organized so that the total number of BBs in the next word changes as little as possible. FIG. 7 is a table showing sequential words for one embodiment of the present invention, Indicates that the center of the coordinate system is a cylindrical support At the center of the chart, each point is defined by its three coordinates θ, ρ, and z FIG. 8 is a schematic diagram showing a cylindrical coordinate system applied to one embodiment of the present invention, and FIG. FIG. 6A and FIG. 6B show the parameters of the calibration ring of one embodiment of the present invention. FIG. 9 is a general view (a) of FIG. (B), which is a typical image, shows a typical X-ray imaging system of the present invention. FIG. 7 is a schematic diagram showing a simulation result for an X-ray image of a middle calibration ring.   A calibration system and apparatus according to the present invention will be described using one illustrated embodiment. In the present embodiment, one set of markers placed around the patient's head, It has a “calibration frame”. FIG. 1 shows the relative position of the X-ray source S, Fig. 2 shows a tensifier I, a calibration frame F and a patient's head. Reference coordinate frame xyz is attached to the calibration frame. The three-dimensional shape of the calibration frame is It is exactly known in the target frame. Barry built into the calibration frame Ball is made of steel, which is significantly larger than human bones and tissues. Has a high absorption coefficient. This allows detection and / or segmentation in the image. Is greatly facilitated. Detection, segmentation and calibration procedures Are well known to those skilled in the art and need not be described in detail herein.   One problem in this case is that the calibration frame Find the 3D-2D correspondence between the 3D model and its image, then use the reference coordinate frame The purpose of the present invention is to reproduce the position and orientation of the X-ray source in the xyz.   This problem is called "absolute orientation recovery" in photogrammetry. Similar to what is called, and "like seeking pose" in computer vision ing. One object of the present invention is that 3D-2D correspondence is easily and uniquely required. And that is possible. The calibration ring reconstructs the relevant small area (local reconstruction). Used for Set the image acquisition settings so that the relevant area fills the entire field of view. Only a part of the calibration ring is present in the x-ray image. this thing It turns out that the design of the calibration ring is smart enough, i.e. That is, it is possible to uniquely determine a portion appearing in the image.   The calibration frame of the present invention is formed by one set of bearing balls (BBs). It has a circular ring R formed. The calibration ring is described in detail below . FIG. 2 shows a plan view and a perspective view of the calibration frame of the present invention. What's further in Therefore, it is explained whether this is a smart arrangement of the calibration BBs. The calibration procedure is as follows Having the following steps. Sphere (BBs) to X-ray image based on the variation of X-ray image intensity ) And detecting the 2D between the image and the calibration ring in this way. Steps to find 3D support and related calibration rings Computes the orientation of camera R and its location T in a coordinate system Computer calculation.   The design of the calibration ring is important in the present invention because this design Because it affects all three steps. Therefore, the design is performed next. The present invention simulated the structure of the calibration ring to explain its performance . Different views of the calibration ring and particularly typical X-ray imaging systems The calibration ring image in the system is simulated using Map software. This is indicated by the action procedure.   The point that can be understood from this is that the calibration ring can be easily adapted to 3D-2D and is unique Must be designed so that it can be sought. However In addition to seeking such a response from a single diagram, certain features, such as the patient's brain, When it is necessary to concentrate on a certain part, see the correspondence from the partial view of the calibration frame. It is desirable to attach. In the above embodiment, the calibration frame is a bearing frame. The ball has a circular ring R formed by a set of balls (BBs). A subset of BBs is vertically₀Are arranged on the periphery. Such an arrangement Sequence defines code words to code different sections of the ring. Justify. The following considerations apply when designing this encoding: Used. The relative spacing between BBs varies under perspective projection. Therefore BBs Are not used to code the calibration frame.   But it is possible to use intervals to facilitate this decoding process This will be described later. The dimensions of the BBs change under perspective projection. Therefore The different dimensions of the BBs are not used to code the calibration frame. this It is possible to select two BBs in this way, one smaller and the other smaller. In contrast, a large choice is made so that they can be distinguished under typical perspective transformation. is there. The detection of small BBs in this case is not possible due to their reduced size. It can always be difficult and inaccurate. In order to solve this problem, high absorption of X-rays It is also possible to use modulus materials. But this is a bigger BBs image May lead to saturation, which makes their centroid difficult and inaccurate to detect . Accordingly, BBs of the same dimensions are used in forming the ring.   In one embodiment of the invention, one particular arrangement of BBs is used, Determines the space word (blank space) between two sequential codes You. To define blank spaces and other code words in rings: The following rules apply. In X-ray imaging, surface patches are displayed in the visual direction. Different images depending on the sign of the surface normal component To run. Thus, one word is the relative position of the ring and the imaging system. Depends on two different ways, one being the flipped version of the other ).   FIG. 4 shows one clear example of this effect. A transparent cylinder is visible. projection Is located outside the cylinder between its two bases. How words are visual When the sign of the surface normal component in the direction changes, it can be seen whether the sign is reversed. One A further rule is that a word is a codeword if its inverted form is a codeword That is what is impossible. Especially symmetric words cannot be code words. is there. In the present embodiment, the blank space is the sign of the surface normal component in the viewing direction. Is a 5-digit code 54321 or 12345 depending on the sign of FIG. And FIG. 6). The positioning code is a 3-digit word with 6 bases. word 234 (or 432) is omitted to unambiguously detect blank spaces . W is a collection of all possible codewords in observing all the above rules and precautions. It is. W = [112,445,311,435,122,255,114,321,5 24,213,452,132,542,551,223,415,124,2 53,331,235,241,145,412, 154, 511, 244, 215, 334, 422, 225, 143, 513, 134,531,152,441,233,342,521,423,314, 351]   FIG. 6 shows a diagram of W. All visible BBs in a partial view of the ring (see FIG. 9) The number should change as little as possible. This ensures uniformity in calibration results Accuracy is maintained. The code words in W are the total number of BBs in a sequential word. They are classified so that they change as little as possible. This has a small number of BBs 445 following the code 112 that has only 4BBs means. Therefore, it is preferable to use even code words. Therefore, this embodiment The spacing between successive BBs should be such that they have even codewords if necessary. To be changed. The blank space is 5 digits and has 15BBs (Fig. 5), on the other hand, the code word has an average of 8BBs in only 3 digits. I can't do it. This results in a small window containing one word and one blank space. In the dough, the number of visible BBs within the range of 23 ± 5 The ratio becomes smaller between the fluctuations.   If the blank space has smaller digits or BBs, this ratio Increase, which is desirable Absent. The design of the calibration ring experiences different changes during the test process. Therefore the book The invention incorporates a flexible design, where the radius R of the ring, the radius r of the BBs , Spacing d between code words_Two, And each of the horizontal / vertical BBs in the codeword Mutual spacing d₁And d_ThreeAre all variables. Geo of one embodiment of a calibration ring In addition to the above description of the metrology, the design algorithm A description will be given using one example given.   This information can be obtained simply by changing some of the parameters. It is believed to be helpful in designing additional embodiments for the new ring. This design results in one particular arrangement of BBs. Therefore, with this design, A set of BBs locations in one particular coordinate system given to the ring is obtained You. The cylindrical coordinate system is clearly preferred as a natural choice. The center of the coordinate system is calibrated Centered on the cylindrical support of the ring. Each BB position A point is defined by its three coordinates θ, ρ, z in FIG.   Since all BBs are located on the surface of the ring, ρ = R. All parameters θ, α in the design₁, Α_Two, D₁, D_Two, And d_Three Is a function of the calibration ring dimensions and the BBs dimensions (R and r). This embodiment Then r = 0.6mm BB s is used and BBs are located on the surface of a cylinder with a radius R = 114.54 mm. You. The used tube is injection molded with acrylic resin and has an average outer diameter of 224.7 mm. Have. BBs on a 1.58mm thick Lexan sheet band Once positioned, the band is then wrapped around the outer periphery of the tube. this thing The point understood from this is how the parameters of the calibration ring are determined. One example at each step is given using the values obtained for this embodiment. available.   First, θ is determined as follows (FIG. 8).   θ = arcsin (r / R)   For example,   θ = Arctan (0.6 / 112.37) = 0.0053 radians   Α in FIG.₁Is the center of two successive BBs at the surface of the ring in radians Determine the angle between them. When these BBs are in contact, this angle is 2 * θ equal. Spacing must be provided between BBs to facilitate detection. B The larger the Bs, the greater the spacing. Therefore, the interval is reset between BBs. Around the ring₁* Specified to have an angular distance of θ Is done. In the first embodiment, λ₁= 3.   Therefore,   α₁= 2θ + λ₁θ   For example,   α₁= 0.026 radians   Thus, between the centers of two successive BBs in one codeword in millimeters The length of the arc is d₁= Α₁R.   For example,   d₁= 2.98 mm   Similarly, it is possible to determine the distance between two successive code words. This distance Is also preferably a function of the dimensions of the BB, but λ_Two* Θ is the code It is made larger than the interval between successive BBs.   In the first embodiment, λ_Two= 5.5. In the last BB of one codeword The angle from the center to the center of the first BB of the next code and the arc length (see FIG. 8) In radians and millimeters respectively:   α_Two= 2θ + λ_Twoθ   For example,   α_Two= 0.0397 radians   d_Two= Α_TwoR   For example,   d_Two= 4.46mm   The ring has a series of sequential pairs of blank spaces and code words. blank The space and codeword are M digits and N digits, respectively, and the first In the embodiment, if M = 5 and N = 3, the respective order around the calibration ring The angle occupied by the next pair is given by:   s = (N + M-2) α₁+ 2α_Two   For example,   s = 0.238 radians   Therefore the maximum number of (blank + code) words that can be placed around the ring Is expressed as follows.   k = trunc (2π / s)   For example,   k = 26   Then α_TwoMust be adjusted to ensure regular spacing between words No. If k (blank + code) is used as defined above, That there is an extra space of 2π-ks between the subsequent code and the first space. Please be careful. In this example, this interval is 2π−26 * 0.238 = 6.283− 6.188 = 0.095 radians. So this extra space around the ring The sources are equally distributed in all spaces between successive codewords.   α_Two= Α_Two+ (Π / ks−2)   For example,   α_Two= 0.0414 radians   Therefore d_TwoIs also changed as follows.   d_Two= Α_TwoR   For example,   d_Two= 4.65mm   The last parameter to set is the vertical distance between BBs, ie d in FIG._Three It is. D when two BBs touch each other_Three= 2r. This distance BB radius d_Three= 2 * π + λ_Three* It must be a function of r. Further λ_ThreeIs one The vertically arranged group of BBs that define one digit of the code word of It is selected so that it can be easily separated from the widget. This is λ_Three<Λ₁<Λ_TwoArticle Means that the matter must be imposed.   In the first embodiment, λ_ThreeIs λ_Three= 1 and d_Three= 4.2r = 2.5mm Has been taken.   At this point, a ring can be formed. Of all points on the surface of the ring The second cylindrical coordinates are the same, ie ρ = R. Two separate parameters θ and z Defines the position of each BB on the surface of the ring. The procedure is a coordinate point (θ = 0, ρp = R, z = 0), ie, simply start at the point (0, R, 0). Then Procedure approximates all words and blank spaces around the ring I do. In fact, the procedure is based on the coding table of FIG. W [i], i = 1. . . and s are alternately selected one by one.   BBs too small or ring too large Unless otherwise, W always provides enough code words (in the first embodiment, s = 26). Last regarding whether even 30 code words are used Is confirmed, and in the case of no, λ₁And λ_TwoIs changed, and thus l₁And l_TwoIs strange This places even code words around the calibration ring.   613BBs are used in the first design. The attached table is all 6 13 shows the positions of 13 BBs. Complete calibration ring and imaging geometry The simulation was formed using Maple. FIG. 9 shows a typical X-ray image. Simulate the image that can be obtained by the amazing geometry I have. In this simulation, the X-ray source and the image intensifier The distance between is 98cm and the ring is centered 78cm away from the X-ray source The image size is 17cm^TwoHas been taken.   Obviously, the invention uses a suitably programmed digital computer. It can be implemented using programming techniques obvious to those skilled in the art.   The invention has been described with reference to examples. However, various modifications and changes may be made in the spirit of the present invention. Can be implemented without departing from the above. Eg different coding schemes It is self-evident that this is possible as a variant of the coding scheme described here. is there. In addition, inversion, which interchanges opaque and transmissive parts, can cause any manifestation of the operation. Obviously, this can be done without significant changes. Such changes, etc. It is included in the scope of the invention defined by the claims.

[Procedure of Amendment] Article 184-8, Paragraph 1 of the Patent Act [Submission Date] January 16, 1998 (1998.1.16) [Correction contents]                           Specification                     Calibration of X-ray geometry   The present invention relates to the calibration of geometry in X-ray imaging configurations, and in particular, Method and method for establishing a correlation between an original (3D) object and a two-dimensional (2D) image About the steps.   Interest in reconstructing the vasculature from x-ray angiograms has increased, It has a number of uses. Various methods have been proposed and some such methods are , Contrast bolus is digital subtraction angiograph (DSA ) The dynamics of the contrast bolus as it evolves over the sequence Trying to take into account the costs. These methods are an imaging system The X-ray source and the image intensifier system It is assumed that it rotates around the body and is well calibrated. So these methods Uses the geometry of the imaging system for vasculature reconstruction.   However, many X-ray systems are not well calibrated in practice, Errors propagate throughout the reconstruction process, reducing the accuracy of the results.   U.S. Pat. No. 5,442,674 (and corresponding French Patent No. 2709) No. 09). The calibration device has a plurality of metal beads, and the metal beads are mounted on a plexiglass cylindrical wall. Spiraled inside, this spiral makes a complete turn above and below the central bead The other beads that extend across the center bead are all the same size Greater than.   In this arrangement, as long as the center bead projection is included in the x-ray image, Other beads in the image can be numbered in absolute form, Determining the correspondence between the known relative position in 3D of the code and the 2D-X-ray image Is possible.   Another known alignment device that uses metal beads is International Publication Pamphlet No. No. 9513018. Any ball in this design Can be made into balls of different dimensions or different X-ray absorption or different shapes It is possible to detect individually in the X-ray image. One advantage this has Is the 3D of the ball as long as at least one ball of the calibration device is in the image It is possible to determine the correspondence between the known relative position in the 2D-X-ray image and It is to be.   However, as will be explained later in the detailed description, a relatively small ball is actually detected. Difficult to emerge, relatively large balls with high X-ray absorption are displayed saturated in the image This makes their centroid difficult and inaccurate to detect. More ball drawings The size of the image depends on the projection, This is a potential source of trouble when placing effectiveness on ball dimensions in image interpretation. Become.   In a first aspect of the present invention,   Mounting means for attaching a calibration frame to a subject exposed to X-rays; A plurality of X-ray contrast features distributed around the frame X-ray system calibration frame   Each X-ray contrast feature is encoded in a coded pattern The orientation of the image to the frame and the distance from the frame to the image. In a 2D-X-ray image containing a projection of at least one of the features By identifying at least one such coded pattern of A plurality of X-ray contrast elements arranged so that they can be obtained A calibration frame for X-ray system calibration is provided.   In this way, the dynamic calibration of the X-ray system can be By placing a sign in the form of a trust feature around the patient in that area. Can be achieved.   In one advantageous embodiment of the invention, the calibration frame is cylindrical or ring-shaped. Yes, with one set of bearing balls as X-ray contrast element, bare The ring ball is located on the surface of the cylinder. The structure of the calibration ring is simulated. Has been installed. Calibration phosphorus Views from different directions of the imaging system, especially in a typical X-ray imaging system. Use the simulation procedure in “Maple” for the positive ring image. It will be described later.   In a second aspect of the present invention,   An X-ray source,   A calibration frame according to the first aspect of the invention,   X-ray X-ray image of at least a part of the calibration frame exposed by X-rays from the X-ray source An X-ray imaging device arranged to form the X-ray imaging device; An X-ray calibration system is provided, comprising:   In a third aspect of the present invention,   2D-X of a 3D subject, where the image includes a projection of at least a portion of the x-ray calibration frame A line image processing method, wherein the method comprises the steps of: In a method comprising detecting a position of a contrast element,   From the detected position of the contrast element, at least one code The converted contrast features are stored in the 3D model stored in the calibration frame. From the file,   Image and calibration frame from at least one identified contrast feature 2D-3D correspondence between   From the obtained 2D-3D correspondence, the position of the image in the coordinate system of the calibration frame and the 2D-X-ray image processing characterized by computing the orientation A method is provided.   In a fourth aspect of the present invention,   2D-X of a 3D subject, where the image includes a projection of at least a portion of the x-ray calibration frame An image processing system for line image processing, the system comprising: A system including a detecting means for detecting a position of a contrast element in an image as a foundation. In the system   From the detected position of the contrast element, at least one code Discriminating contrast features from the 3D model of the calibration frame The 3D model includes an identification unit, and the 3D model includes a plurality of X-ray contrast features. These X-ray contrast features contain positional information about the X-ray contrast element. See   Image and calibration frame from at least one identified contrast feature Means for determining a 2D-3D correspondence between   From the obtained 2D-3D correspondence, the position of the image in the coordinate system of the calibration frame and the Image characterized by comprising means for computing the orientation A processing system is provided.   In a fifth aspect of the present invention,   The calibration frame according to the first aspect of the present invention is applied to a subject exposed to X-rays. Position the   In order to form a projected 2D-X-ray image of a subject on an X-ray imaging apparatus Exposing the subject and the calibration frame by X-rays,   X-ray processing an image by the method according to the third aspect of the present invention An imaging method is provided.   To facilitate an understanding of the invention, reference will now be made, by way of example, to the accompanying drawings, in which: I do.   FIG. 1 is a view showing the relationship between the X-ray source S, the image intensifier I, and the calibration frame F. It is a schematic diagram showing a relative position to a person's head.   FIG. 2 shows perspective views (a) and (d), a plan view (b) viewed from above, and the applied reference. FIG. 4 is a schematic view showing the calibration ring of the present invention by a schematic view (c).   FIG. 3 shows the changes in the dimensions of the globules or bearing balls (BB's) and their variations. FIG. 7 is a schematic diagram showing a change in the interval between the images by perspective projection.   FIG. 4 is a schematic diagram showing a cylinder, where the projection center is outside the cylinder and the two The words that are located between and embedded in them are the surface normals in the viewing direction. Indicates that the sign is inverted when the sign or polarity of the component changes.   Figure 5 shows how the blank space looks in the viewing direction 5 digit code 54321 depending on the sign of the surface normal component at Or 12345 (a), showing how the device Excluded to unambiguously (ie, make it not ambiguous) (B) It is a schematic diagram which shows.   FIG. 6 shows that the total number of BB's in successive words varies as little as possible. A sequential word for one embodiment of the invention in which the table elements are arranged FIG.   FIG. 7 shows that the center of the coordinate system is at the center of the cylindrical support and each point is FIG. 3 is a schematic diagram showing a cylindrical coordinate system defined by coordinates θ, ρ, and z.   FIG. 8 shows the parameters of the calibration ring in the plan view (a) and the lateral view (b). FIG.   FIG. 9 shows a general view (a) and a partial view which is a more typical image in local reconstruction. (B) X-ray image of a calibration ring in a typical X-ray imaging system FIG. 4 is a schematic diagram showing a simulation result for an image.   The calibration system and device will be described using one illustrated embodiment. Of this implementation The form has a set of markers located around the patient's head, i.e., The "calibration frame". FIG. 1 shows the relative position of the X-ray source S, Tensifier I, calibration 3 shows a frame F and a patient's head. The reference coordinate frame xyz is taken as the calibration frame. It is attached. The three-dimensional shape of the calibration frame is exactly Is knowledge. Bearing balls built into the calibration frame are made of steel As a result, this steel has a significantly higher absorption coefficient compared to human bones and tissues. This greatly facilitates detection and / or segmentation in the image . Detection, segmentation and calibration procedures are well known to those skilled in the art, Not described in detail herein.   One problem in this case is that the 3D model of the calibration frame and its image Between the 3D-2D correspondences and then the location of the X-ray source in the reference coordinate frame xyz To reproduce the orientation and orientation.   This problem is called "absolute orientation recovery" in photogrammetry. Similar to what is called, and "like seeking pose" in computer vision ing. Ideally, 3D-2D support should be easily and uniquely required . The calibration ring is used for the reconstruction of this small area (local reconstruction). The image acquisition settings are implemented so that the area in question fills the entire field of view, i.e. Only a part of the image is present in the X-ray image. The calibration ring design appears in the image. It should be possible to determine the part to be determined uniquely.   The calibration frame is formed by one set of bearing balls (BB's) With a circular ring R. The calibration ring is described in detail below. FIG. Shows a plan view and a perspective view of the calibration frame. In the following, why this is calibration B It is explained whether the arrangement of B's is appropriate.   The calibration procedure has the following steps:   Detecting the projection of a sphere (BB's) onto an X-ray image based on the variation in X-ray image intensity Steps   Finding a 2D-3D correspondence between the image and the calibration ring;   The orientation of the camera R in the coordinate system associated with the calibration ring and its And computing the location T of the computer.   The design of the calibration ring is important because this design is This is because it affects all three steps. So the design is done next . The present invention simulates the structure of the calibration ring to explain its performance. Was. Different views of the calibration ring and especially typical X-ray imaging systems The calibration ring image in the system is simulated using the "Maple" software. As indicated by the translation procedure.   The calibration ring can easily handle 3D-2D It must be designed so that it can be determined uniquely. Only one figure In addition to seeking such a response from, certain parts of the patient's brain, for example When it is necessary to concentrate on Is desirable. In the above embodiment, the calibration frame is a bearing ball. It has a circular ring R formed by one set of (BB's). BB ' A subset of s is a plurality of vertical lines R₀Are arranged on the periphery. This Such a sequence of line arrays is used to encode different sections of the ring. Define the Hardened language. The following considerations apply when designing this encoding. BB The relative spacing between 's varies under perspective projection. Therefore the interval between BB's Is not used to encode the calibration frame.   But nevertheless, intervals to facilitate this decoding process Can also be used, as will be described later. BB's dimensions are perspective projection Changes below. Thus different dimensions of BB's code the calibration frame Not used for In this way, it is possible to select two BB's. One is small and the other is relatively large, which results in a typical perspective change. Under commutation they are identifiable. In this case, the detection of small BB's Very difficult and inaccurate due to the reduced dimensions of May be. To solve this problem, use materials with high absorption coefficient of X-ray It is also possible. However, this may lead to larger BB's image saturation. This makes their centroid difficult and inaccurate to detect. Therefore the same dimensions BB's are used in forming the ring.   In one embodiment of the invention, one special arrangement of BB's is used, A space word (blank space) between two sequential codes is determined by the arrangement. It is. To define blank spaces and other code words in the ring: The following rules apply. In X-ray imaging, the surface patch is Different imaging is performed depending on the sign of the surface normal component. Therefore one The term is used in two different ways depending on the relative position of the ring and the imaging system. One is a flipped version of the other.   FIG. 4 shows one clear example of this effect. A transparent cylinder is visible. projection Is located outside the cylinder between its two bases. How words are visual When the sign of the surface normal component in the direction changes, it can be seen whether the sign is reversed. One A further rule is that a word is a codeword if its inverted form is a codeword That is what is impossible. Especially symmetric words cannot be code words. is there. In the present embodiment, The rank space is 5 digits depending on the sign of the surface normal component in the viewing direction. The code 54321 or 12345 (see FIGS. 5 and 6). Positioning co The code is a 3-digit word with a base number of 6. The word 234 (or 432) Excluded to unify rank space detection. W is defined by all of the aforementioned rules. The set of all possible code words when observing rules and precautions. W = [112,445,311,435,122,255,114,321,5 24,213,452,132,542,551,223,415,124,2 53,331,235,241,145,412,154,511,244,2 15,334,422,225,143,513,134,531,152,4 41, 233, 342, 521, 423, 314, 351]   FIG. 6 shows a diagram of W. The visible BB's in the partial view of the ring (see FIG. 9) The total number should change as little as possible. This allows the calibration results to be averaged. Uniform accuracy is maintained. The code word in W is BB's in two sequential words Are classified such that the total number, i.e., the sum, changes as little as possible. this is, Code with a small number of BB's, for example only 4BB 'Code with a large number of BB's, i.e., 44 5 is followed. Therefore, it is preferable to use even code words. Therefore, in this embodiment, the interval between successive BB's is set to an even number if necessary. Changed to have a C language. The blank space is 5 digits Has 15BB's (FIG. 5), whereas the code word has an average of only 3 digits Has only 8BB's. Thus, one word and one blanks Within the small window containing the pace, the average number of visible BB's 23 (23 = 15 + 8) ± 5 (4-13) The ratio between the variation of the number of visible BB's Becomes smaller.   If the blank space has smaller digits or BB's The ratio increases, which is undesirable. Calibration ring design may vary during the test process. Experience change. The design is therefore flexible, ie the radius R of the ring , BB's radius r, spacing d between codewords_Two, And horizontal / vertical in code words The mutual spacing d of BB's₁And d_ThreeAre all variables. Calibration ring 1 In addition to the above description of the geometry of one embodiment, the design algorithm The system is described using one example that gives different measurements.   This information simply changes some of the parameters. In designing additional embodiments for the required new ring It is thought to be useful. This design results in one particular arrangement of BB's You. Therefore, this design allows BB in one particular coordinate system to be imparted to the ring. One set of locations is obtained. Cylindrical coordinate systems are clearly preferred as natural choices You. The center of the coordinate system is located at the center of the cylindrical support of the calibration ring. One B Each point that is the position of B is defined by the three coordinates θ, ρ, and z in FIG. It is.   Since all BB's are located on the surface of the ring, for all points ρ = R. All parameters θ, α in the design₁, Α_Two, D₁, D_Two, And d_Three Is a function of the dimension R of the calibration ring and the dimension r of BB's. In this embodiment Uses BB's with r = 0.6 mm, and BB's has a radius R = 114.54 mm Are arranged on the surface of the cylinder. The used tube is injection molded with acrylic resin, It has an average outer diameter of 224.7 mm. BB's is 1.58 mm thick lexan (Lexan) placed on a band of sheets, then the band is rolled up and tubed Where BB's are located on the outer periphery of the tube. From this What you can see is how the parameters of the calibration ring are determined. One example in each of these steps is given using the values obtained for this embodiment. It is.   First, θ is determined as follows (FIG. 8).   θ = arcsin (r / R)   For example,   θ = Arctan (0.6 / 112.37) = 0.0053 radians   FIG. 8a₁Is the center of two successive BB's at the surface of the ring in radians Determine the angle between If these BB's are in contact, this angle is 2 * equal to θ. A spacing must be provided between BB's to facilitate detection. No. The larger the BB's, the larger the spacing must be. Therefore, the interval is BB ' λ around the ring between s₁* Specified to have an angular distance of θ. First In the embodiment, λ₁= 3.   Therefore,   α₁= 2θ + λ₁θ   For example,   α₁= 0.026 radians   Thus between two successive BB's centers in one codeword in millimeters Arc length d₁= Α₁R.   For example,   d₁= 2.98 mm   Similarly, it is possible to determine the distance between two successive code words. This distance Is also preferably a function of the dimensions of BB's, but_Two* Θ is the code It is made larger than the interval between successive BB's.   In the first embodiment, λ_Two= 5.5. In the last BB of one codeword The angle from the center to the center of the first BB of the next code and the arc length (see FIG. 8) In radians and millimeters respectively:   α_Two= 2θ + λ_Twoθ   For example,   α_Two= 0.0397 radians   d_Two= Α_TwoR   For example,   d_Two= 4.46mm   The ring has a series of sequential pairs of blank spaces and code words. blank The space and codeword are M digits and N digits, respectively, and the first In the embodiment, if M = 5 and N = 3, the respective order around the calibration ring The angle occupied by the next pair is given by:   s = (N + M-2) α₁+ 2α_Two   For example,   s = 0.238 radians   Therefore the maximum number of (blank + code) words that can be placed around the ring Is expressed as follows.   k = trunc (2π / s)   For example,   k = 26   Then α_TwoMust be adjusted to ensure regular spacing between words No. If k (blank + code) is used as defined above, That there is an extra space of 2π-ks between the subsequent code and the first space. Please be careful. In this example, this interval is 2π−26 * 0.238 = 6.283− 6.188 = 0.095 radians. So this extra space around the ring The sources are equally distributed in all spaces between successive codewords.   α_Two= Α_Two+ (Π / ks−2)   For example,   α_Two= 0.0414 radians   Therefore d_TwoIs also changed as follows.   d_Two= Α_TwoR   For example,   d_Two= 4.65mm   The last parameter to set is the vertical distance between BB's, ie d in FIG._Three It is. D when two BB's touch each other_Three= 2r. This distance Is also the radius d of BB_Three= 2 * π + λ_Three* It must be a function of r. Further λ_ThreeIs 1 BB's of vertically arranged groups that define one digit of one codeword It is chosen so that it can be easily separated from another digit. This is λ_Three<Λ₁<Λ_Two And that the condition must be imposed.   In the first embodiment, λ_ThreeIs λ_Three= 1 and d_Three= 4.2r = 2.5mm Has been taken.   At this point, a ring can be formed. Of all points on the surface of the ring The second cylindrical coordinates are the same, ie ρ = R. Two separate parameters θ and z Defines the position of each BB on the surface of the ring. The procedure is a coordinate point (θ = 0, ρp = R, z = 0), ie, simply start at the point (0, R, 0). Then Procedure approximates all words and blank spaces around the ring I do. In fact, the procedure is based on the coding table of FIG. W [i], i = 1. . . and s are alternately selected one by one.   W is always sufficient unless BB's is too small or the ring is not too large (In the first embodiment, more than s = 26). Even number A final check is made as to whether 30 code words have been used and a no In case λ₁And λ_TwoIs changed, and thus l₁And l_TwoHas been changed, which has An even number of code words are placed around the ring.   613BB's are used in the first design. The attached table is all 613 position of BB's Show. A complete simulation of the calibration ring and imaging geometry is given by M formed using an apple. FIG. 9 shows a typical X-ray imaging geometry Simulates an image that can be obtained by This simulation The distance between the X-ray source and the image intensifier is 98 cm Where the ring is centered 78 cm away from the x-ray source and the image size is 17 cm^TwoHas been taken.   This embodiment uses an appropriately programmed digital computer and This can be achieved using programming techniques that are obvious to the merchant.   The invention has been described with reference to examples. However, implementing various modifications and changes Is possible. For example, various coding schemes may be used to implement the coding schemes described herein. Obviously, it is possible as a modification of the schema. In addition, opaque and transparent parts Inversion interchangeable with minutes is feasible without any significant changes in operation It is self-evident.                     The scope of the claims 1. Mounting means for mounting a calibration frame on a subject exposed to X-rays, A plurality of X-ray contrast features distributed around a calibration frame; X-ray system calibration frame having     Each x-ray contrast feature is a coded pattern, Orientation of the image with respect to the frame and from the frame to the image And a projection of a feature of at least one of said features. Identify at least one such coded pattern in the 2D-X-ray image. A plurality of X-rays arranged so that they can be uniquely obtained by differentiating A calibration frame for calibrating an X-ray system, comprising a contrast element. 2. Patterns of contrast elements that form each contrast feature Extend in two dimensions and represent one code word. Positive frame. 3. Any code word pattern among the code word patterns seen from the rear The code word pattern has an appearance different from that when viewed from the front, and the code word pattern Any other code in the 3. The proofreading frame according to claim 2, wherein the proofreading frame has an appearance different from the word pattern. Home. 4. Contrast features are distributed in a ring around the calibration frame, The contrast elements of each pattern are arranged in the peripheral direction and the axial direction of the ring. The calibration frame according to claim 2, wherein the calibration frame extends. 5. Multiple codewords distributed around the calibration frame A first set of codewords having a last element pattern, wherein 5. The code word according to claim 2, wherein the code words are different from each other. Calibration frame. 6. Contrasts forming each said code word in a first set of code words The two-dimensional pattern of the contrast elements is adjacent to a first fixed number of Tangent lines, wherein the number of said contrast elements per line is one and the first And each of said code words in the first set is of base m represented by a sequence of n digits, where n is the first fixed number of the lines Where m is equal to the first maximum number of the contrast elements per line. The calibration frame according to claim 5, wherein: 7. Replacing the first set of code words with the first set The sum of the number of contrast elements in the codewords adjacent to each other Characterized in that it is arranged around the calibration frame so that it changes only slightly. A calibration frame according to claim 6. 8. Multiple codewords distributed around the calibration frame 2 and each of the code words is identical to each other and has the same contra Having a strike element pattern and being different from any code word of the first set. The calibration frame according to claim 5, 6 or 7, wherein 9. Code words around a frame, wherein the second set of code words is the first set 9. The calibration file according to claim 8, wherein the calibration word is arranged so as to alternate with the code word. Laem. Ten. 2 of the contrast elements forming a code word in the second set of code words A two-dimensional fixed number of adjacent ones of said contrast elements Lines, wherein the number of said contrast elements per line is between 1 and a second maximum number And thus the code words of each of the second set are represented by a base number m 'By the same sequence of n' digits, where n 'is the second of the lines Equal to a fixed number, m 'is the second of the contrast elements per line Equal to the maximum number The calibration frame according to claim 8, wherein 11． The average number of contrast elements in the first set of codewords is calculated as the second set of codewords. The value obtained by dividing by the number of contrast elements in each code word is 1 / The calibration frame according to claim 8, wherein the number is 2 or less. 12． The calibration frame has the form of a band, said band forming in an endless loop Characterized in that it has a flat ribbon-like structure The calibration frame according to any one of claims 1 to 11. 13. An X-ray source (S),     A calibration frame according to any one of claims 1 to 12. And     At least a part of a calibration frame exposed by X-rays from the X-ray source (S) The X-ray image of the above is arranged in the X-ray imaging apparatus (I) An X-ray calibration system, comprising: an X-ray imaging apparatus (I). 14. 2D- of the 3D subject, where the image includes a projection of at least a portion of the x-ray calibration frame An X-ray image processing method,     The method includes determining the position of a contrast element in an image based on changes in image intensity. To detect the location Including the method,     At least the calibration frame from the detected position of the contrast element One coded contrast feature is stored in the calibration frame From the 3D model     The image from at least one identified contrast feature A 2D-3D correspondence with the calibration frame is obtained,     From the obtained 2D-3D correspondence, the image in the coordinate system of the calibration frame 2D-X, wherein position and orientation are computed Line image processing method. 15． 2D- of the 3D subject, where the image includes a projection of at least a portion of the x-ray calibration frame An image processing system for X-ray image processing,   The system includes a contrast element in the image based on a change in image intensity. In a system including detection means for detecting the position of the child,     From the detected position of the contrast element, less of the calibration frame Both one coded contrast feature and the 3D An identification means for identifying the 3D model from the plurality of X-ray controllers; The X-ray core of each of the X-ray contrast features of the trust feature Contrast element Including location information,     The image from at least one identified contrast feature Means for determining a 2D-3D correspondence with the calibration frame,     The image in the coordinate system of the calibration frame is obtained from the obtained 2D-3D correspondence. Providing means for computing the position and orientation of the image Characteristic image processing system. 16． 13. The method according to claim 1, wherein the subject is exposed to X-rays. Positioning the calibration frame according to any one of the claims,     To form a projected 2D-X-ray image of a subject on an X-ray imaging apparatus Irradiating the subject and the calibration frame with X-rays,     An X-ray image processing an image by the method according to claim 14. Zing method.

Claims (1)

[Claims] 1. A calibration frame mounted near at least a portion of the patient's body,     Equipped with an X-ray source,     Image of the part of the patient's body and at least the corresponding part of the calibration frame A predetermined dose from said part of the patient's body to cooperate with said x-ray source to form X-ray geometry calibration with target located at the orientation and distance In the system,     The calibration frame is configured to store the image of the corresponding portion of the calibration frame and the calibration Coding means for uniquely determining the correspondence with the normal frame, whereby the The orientation and the distance are determined by the image of the corresponding part of the calibration frame. X-ray geometry calibration system characterized in that it can be uniquely determined from an image Stem. 2. An image of the part of the patient's body and the corresponding part of the calibration frame is substantially flat. 2. The X-ray geometry calibration system according to claim 1, wherein the X-ray geometry calibration system is included in a plane. Tem. 3. The coding means being relatively opaque to X-rays of said coding means; Calibration showing one part and a second part that is relatively transparent to X-rays The X-ray geometries according to claim 2, comprising at least a part of a frame. Metric calibration system. 4. Encoding means having an encoded arrangement of the first part. The X-ray geometry calibration system according to claim 3, wherein 5. Encoding means having an encoded arrangement of the second part. The X-ray geometry calibration system according to claim 3. 6. Encoding means having an encoded arrangement of the first part and the second part The X-ray geometry calibration system according to claim 3, wherein: 7. 5. The method according to claim 4, wherein the first part of the calibration frame comprises metal balls. An X-ray geometry calibration system as described. 8. A second portion of the calibration frame comprises a hole in the calibration frame. The X-ray geometry calibration system according to claim 5, wherein 9. The X-ray scanner according to claim 1, wherein the calibration frame has a three-dimensional shape. Ometry calibration system. Ten. From the image in the almost flat plane, the calibration frame is To determine which part is proximal to the image and which part is distal to the image. Characteristics that can be The X-ray geometry calibration system according to claim 9, wherein: 11． The X-ray according to claim 10, wherein the calibration frame has a three-dimensional shape. Geometry calibration system. 12． 2. The method according to claim 1, wherein the calibration frame has a flat ribbon-like structure. X-ray geometry calibration system. 13. Characterized by a flat ribbon-like structure formed in an endless loop An X-ray geometry calibration system according to claim 12. 14. An image of the part of the patient's body and the corresponding part of the calibration frame is a two-dimensional plane. 2. The X-ray geometry calibration system according to claim 1, wherein the X-ray geometry calibration system is included. M 15． The calibration frame has a structure for clamping and fixing the patient's head. The X-ray geometry calibration system according to claim 1. 16． A calibration frame corresponding to a part of the patient's body,     Equipped with an X-ray source,     Forming an X-ray image of the part of the patient's body and the corresponding part of the calibration frame An X-ray geometry calibration system comprising:     The calibration frame, the calibration frame of the Before the calibration frame to uniquely determine the correspondence from the image of the corresponding part Having coding means for obtaining a correspondence between the corresponding portion and the calibration frame. An X-ray geometry calibration system, characterized in that: 17． The calibration frame is relatively opaque to X-rays of the calibration frame and X 17. The X of claim 16 including a portion that is relatively transparent to the line. Line geometry calibration system. 18． The opaque portion forms a coded arrangement. 8. The X-ray geometry calibration system according to 7. 19. 18. The method of claim 17, wherein the permeable portion forms a coded arrangement. 2. The X-ray geometry calibration system according to 1. 20. Make sure that the opaque and permeable parts form a coded arrangement. The X-ray geometry calibration system according to claim 17, characterized in that: twenty one. A calibration frame comprising: a substrate portion of a material that is relatively transparent to X-rays; A plurality of objects embedded in the substrate portion and relatively impermeable to X-rays; The X-ray geometry calibration system according to claim 1, wherein: twenty two. The object embedded in the substrate part is composed of metal globules Claim 21 X-ray geometry calibration system. twenty three. The calibration frame comprises a substrate portion of a material that is relatively impermeable to X-rays, Having a plurality of points relatively transparent to X-rays in the substrate portion. 2. The X-ray geometry calibration system according to claim 1, wherein: twenty four. 24. The X-ray device of claim 23, wherein the point in the substrate portion comprises a hole. Ometry calibration system. twenty five. An image of the portion of the patient's body and the corresponding portion of the calibration frame is substantially flat. 2. The X-ray geometry calibration according to claim 1, wherein the X-ray geometry calibration is included in a plane. system. 26. The coding arrangement allows the position of the calibration frame from the image in a substantially flat plane. Pattern in the image so that the position and orientation can be determined uniquely. 26. The X-ray geometry calibration according to claim 25, wherein a pattern is formed. system. 27. The X-ray scanner according to claim 1, wherein the calibration frame has a three-dimensional shape. Ometry calibration system. 28. 2. The method according to claim 1, wherein the calibration frame has a flat ribbon-like structure. X-ray geometry calibration system. 29. Characterized by a flat ribbon-like structure formed in an endless loop An X-ray geometry calibration system according to claim 28. 30. An image of the part of the patient's body and the corresponding part of the calibration frame is in a ring shape. The X-ray geometry calibration system according to claim 1, wherein: 31. The calibration frame is configured to be able to clamp and fix said part of the patient's body. The X-ray geometry calibration system according to claim 1, wherein: 32. An X-ray image of a calibration frame corresponding to at least a part of a patient's body; In a method for determining the correspondence between the position of the calibration frame with respect to     (A) a coded three-dimensional calibration file for at least a part of the patient's body; Positioning the frame;     (B) X-rays the part of the patient's body and at least a part of the calibration frame A source for exposure at a given orientation and at a given distance from said X-ray source. Forming a projected image of the portion of the patient and the portion of the calibration frame When,     (C) The orientation and the distance can be uniquely obtained. So that the calibration Viewing the correspondence between the image of the portion of the frame and the three-dimensional calibration frame in the image A method comprising the step of uniquely obtaining from insight. 33. An X-ray image of a calibration frame corresponding to at least a part of a patient's body; In a method for determining the correspondence between the position of the calibration frame with respect to     (A) a coded three-dimensional calibration file for at least a part of the patient's body; Positioning the frame, wherein the calibration frame has a globular arrangement therein. Including     (B) X-rays of said part of the patient's body and at least a part of said calibration frame A source for exposure at a given orientation and at a given distance from said X-ray source. Forming a projection image of the part of the patient and the part of the calibration frame. Have     (C) a pair between the part of the calibration frame and the three-dimensional calibration frame Response from the observation of the image,     (D) calculating the orientation and the distance by computer; And a step.