A method for calibrating the position of surgical navigation reference markers
By adding verification points on the registration block and collecting their coordinates with a probe, the connection between the verification point coordinate system and the reference array coordinate system is dynamically established, which solves the problem of coordinate error of the center of the metal ball in the registration block, realizes high-precision calibration and simplifies operation.
Patent Information
- Application Number
- CN202211165363.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-09-23
- Publication Date
- 2025-09-23
- Estimated Expiration
- 2042-09-23
AI Technical Summary
In the existing technology, the coordinate value of the center of the metal ball of the registration block in the optical reference array coordinate system in surgical navigation is highly dependent on the hardware processing accuracy, resulting in an error between the actual value and the design value, and the process of collecting the surface points of the metal ball is cumbersome.
By adding verification points on the registration block, using the optical tracking probe to collect the verification point coordinates, dynamically establishing the connection between the verification point coordinate system and the reference array coordinate system, and calculating the exact position of the center of the metal ball in the reference array coordinate system.
The accurate calibration of the coordinate values of the center of the metal ball is achieved, the influence of manufacturing errors is avoided, the operation process is simplified, and the position flexibility of the reference array is improved.
Smart Images

Figure CN115568945B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of image navigation technology, and in particular to a method for calibrating the position of a surgical navigation reference mark point. Background Art
[0002] In image-guided surgery, a registration block is used to register the patient's CT images. Its main body is made of a polymer material that is invisible in CT images and has at least three metal balls attached to it as fiducial markers. The block also has an optical reference array with reflective balls or sheets mounted on it that can be recognized by the positioning camera.
[0003] Disadvantages of existing technology:
[0004] The coordinates of the center of the metal ball in the optical reference array coordinate system are determined in the design drawing of the registration block and used for registration. Disadvantages: Highly dependent on hardware processing accuracy, there may be errors between the geometric design value and the actual geometric value.
[0005] Before using the registration block for CT image registration, an optical probe is used to collect the coordinates of the metal sphere's surface points on the registration block within the block's reference array. The sphere's center is then determined by fitting the sphere's surface, achieving pre-calibration. Disadvantages: In some applications, the metal spheres are only 2-3 mm in diameter and embedded within the registration block, making it difficult to obtain their surface points using a probe. Collecting all the sphere's surface points is also cumbersome. Therefore, we propose a method for calibrating the position of surgical navigation fiducial markers. Summary of the Invention
[0006] Based on the technical problems existing in the background technology, the present invention proposes a position calibration method for surgical navigation reference marking points, which achieves the calibration purpose by obtaining the accurate position of the center of the metal ball serving as the reference marking point in the reference array coordinate system of the registration block.
[0007] The present invention provides the following technical solution: a method for calibrating the position of a surgical navigation reference mark point, comprising the following steps:
[0008] S1. Obtain the coordinate values of the reference mark point and the verification point of the center of the metal ball in the coordinate system of the three-dimensional model of the registration block, establish a coordinate system in the three-dimensional model of the registration block, and obtain the coordinate values of the three verification points on the registration block;
[0009] S2. Calculate the transformation relationship from the verification point coordinate system to the registration block 3D model coordinate system, as follows:
[0010] Using the coordinate values of the three verification points on the registration block in the three-dimensional model coordinate system of the registration block in step S1, the unit direction vectors of the three coordinate axes of the coordinate system formed by the three verification points of the registration block are calculated;
[0011] The unit direction vectors of the three coordinate axes and the coordinate value of one of the verification points form a 4X4 matrix. This matrix is the transformation matrix from the 3D model coordinate system of the registration block to the verification point coordinate system. The transformation matrix from the verification point coordinate system to the 3D model coordinate system of the registration block can be obtained by calculating its inverse matrix.
[0012] S3, using an optical tracking probe to collect the coordinate values of the verification point in the reference array coordinate system of the registration block;
[0013] S4, calculating the conversion relationship between the registration block reference array coordinate system and the registration block three-dimensional model coordinate system;
[0014] S5. Calculate the coordinate value of the reference mark point at the center of the metal ball in the registration block reference array coordinate system.
[0015] Preferably, the specific content of step S1 is as follows:
[0016] S11. Import the registration block 3D model into the engineering drawing software and establish a 3D orthogonal coordinate system F at any position in the software virtual 3D scene. design ;
[0017] S12, obtain the model image coordinate system F design Under the following conditions, the coordinate values of the center of all metal balls used as reference points are:
[0018]
[0019] S13, obtain the model image coordinate system F design The coordinate values of the three verification points on the registration block are:
[0020]
[0021] Preferably, the specific content of step S2 is as follows:
[0022] S21. Use model image coordinate system F design The coordinate values c1, c2, c3 of the three verification points on the lower registration block are calculated in the model image coordinate system F design Next, the unit direction vectors of the three coordinate axes of the coordinate system formed by the three verification points of the registration block;
[0023]
[0024] The values of the three vectors are recorded as follows:
[0025]
[0026] S21, use the above three vectors and the coordinate value of point c1 to form a 4X4 matrix TdesignToCheckPoints as follows:
[0027]
[0028] Matrix T designToCheckPoints is the transformation matrix from the registration block 3D model coordinate system to the verification point coordinate system. By finding its inverse matrix, we can get the transformation matrix T from the verification point coordinate system to the registration block 3D model coordinate system. checkPointsToDesign ;
[0029] T checkPointsToDesign =(T designToCheckPoints ) -1 .
[0030] Preferably, the specific content of step S3 is as follows:
[0031] S31, place the probe tip at the registration block verification point i, i = 1, 2, 3, and collect the 4X4 transformation matrix from the optical positioning camera coordinate system to the probe coordinate system and the registration block reference array coordinate system respectively T cameraToProbe_i With T cameraToBlock_ ;
[0032] S32: The homogeneous coordinate value q of the verification point i in the registration block reference array coordinate system can be obtained by using the following coordinate system transformation. i ':
[0033]
[0034] S33, repeat the above steps to obtain the homogeneous coordinate values (4x1 sequence) of all three verification points in the registration block reference array coordinate system, which are q1', q2', q3', and their corresponding three-dimensional space coordinate values (3x1 sequence) are c1', c2', c3';
[0035] S34. Use the coordinate values obtained by the above calculation to calculate the unit direction vectors of the three coordinate axes of the coordinate system formed by the registration block verification points in the registration block reference array coordinate system:
[0036]
[0037] The above three vectors are written as follows:
[0038]
[0039] S35. Use the above three vectors and the coordinate value of c1' [x1', y1', z1'] to create a 4X4 matrix T blockToCheckPoints , which is the transformation matrix from the registration block reference array coordinate system to the registration block verification point coordinate system:
[0040]
[0041] Preferably, the specific content of step S4 is as follows:
[0042] Use the coordinate system transformation relationship T calculated in steps S2 and S3 blockToCheckPoints With T checkPointsToDesign , calculate the transformation relationship T from the registration block reference array coordinate system to the registration block three-dimensional model coordinate system blockToDesign ;
[0043] T blockToDsign =T blockToCheckPoints ·T checkPointsToDesign .
[0044] Preferably, the specific content of step S5 is as follows:
[0045] In the registration block 3D model coordinate system F design The coordinate value of the center of the metal ball No. i is m i , its coordinate value m in the registration block reference array coordinate system i 'for:
[0046] m′ i =T blockToDesign ·m i .
[0047] The present invention provides a method for calibrating the position of surgical navigation reference markers. By adding verification points on the base plate of the registration block and collecting their coordinates with a probe, the connection between the verification point coordinate system and the reference array coordinate system is dynamically established. The positional relationship between the verification points themselves and the metal balls can directly use the designed values. Since they are all located on the base plate that is processed in one piece, the processing accuracy is high enough, and the accurate position of the center of the metal ball in the reference array coordinate system is obtained. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Figure 1 This is a schematic diagram of the registration block structure of the present invention;
[0049] Figure 2 Schematic diagram of the optical tracking probe structure of the present invention;
[0050] Figure 3 Schematic diagram of registering the block with the patient's target surgical area CT image;
[0051] Figure 4 Schematic diagram for establishing the conversion relationship between the optical reference array coordinate system and the positioning camera coordinate system;
[0052] Figure 5 is a schematic diagram of calibration error;
[0053] Figure 6 This is a schematic diagram of calibration error 2;
[0054] Figure 7 Schematic diagram of calibration error 3. DETAILED DESCRIPTION
[0055] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.
[0056] The present invention provides a technical solution: a method for calibrating the position of a surgical navigation reference marker, the specific steps of which are as follows:
[0057] S1. Obtain the coordinate values of the reference mark point and the verification point of the center of the metal ball in the three-dimensional model coordinate system of the registration block.
[0058] Import the registration block 3D model into the engineering drawing software and establish a 3D orthogonal coordinate system F at any position in the software virtual 3D scene. design .
[0059] Get the model image coordinate system F design Under the following conditions, the coordinate values of the center of all metal balls used as reference points are: if the number of metal balls is n, and n≥3, then the coordinate value of the center of the i-th metal ball is recorded as
[0060]
[0061] Get the model image coordinate system F design The coordinate values of the three verification points on the registration block are
[0062]
[0063] S2. Calculate the transformation relationship from the verification point coordinate system to the registration block three-dimensional model coordinate system.
[0064] Use the model image coordinate system F design The coordinate values c1, c2, and c3 of the three verification points on the lower registration block are calculated as follows:
[0065]
[0066] The values of the three vectors are recorded as follows:
[0067]
[0068] The actual meaning of these three vectors is that in the model image coordinate system F designBelow, the unit direction vectors of the three coordinate axes of the coordinate system formed by the three verification points of the registration block.
[0069] Use the above three vectors and the coordinates of point c1 to form a 4X4 matrix T designToCheckPoints as follows:
[0070]
[0071] Matrix T designToCheckPoints That is, the transformation matrix from the registration block 3D model coordinate system to the verification point coordinate system. The transformation matrix T from the verification point coordinate system to the registration block 3D model coordinate system can be obtained by calculating its inverse matrix. checkPointsToDesign .
[0072] T checkPointsToDesign =(T designToCheckPoints ) -1 .
[0073] S3. Acquire the coordinate values of the verification points in the reference array coordinate system of the registration block.
[0074] The probe tip is placed at the registration block verification point i (i = 1, 2, 3), and the 4X4 transformation matrix (which can be directly obtained from the positioning system) from the optical positioning camera coordinate system to the probe coordinate system and the registration block reference array coordinate system (set based on its reflective sphere) is collected. cameraToProbe_i With T cameraToBlock_i .
[0075] The homogeneous coordinate value q of the verification point i in the registration block reference array coordinate system can be obtained by the following coordinate system transformation: i '
[0076]
[0077] Repeat the above steps to obtain the homogeneous coordinate values (4X1 series) of all three verification points in the registration block reference array coordinate system, which are q1', q2', q3', and their corresponding three-dimensional space coordinate values (3X1 series) are c1', c2', c3'.
[0078] Use the coordinate values calculated above to calculate the following three unit vectors:
[0079]
[0080] The above three vectors are written as follows:
[0081]
[0082] The actual meaning of these three vectors is the unit direction vectors of the three coordinate axes of the coordinate system formed by the registration block verification points in the registration block reference array coordinate system;
[0083] Use the above three vectors and the coordinate values of c1' [x1', y1', z1'] to create a 4X4 matrix T blockToCheckPoints, This matrix is the transformation matrix from the registration block reference array coordinate system to the registration block verification point coordinate system;
[0084]
[0085] S4. Calculate the conversion relationship between the registration block reference array coordinate system and the registration block three-dimensional model coordinate system.
[0086] Use the coordinate system transformation relationship T obtained above blockToCheckPoints With T checkPointsToDesign , calculate the transformation relationship T from the registration block reference array coordinate system to the registration block three-dimensional model coordinate system blockToDesign .
[0087] T blockToDsign =T blockToCheckPoints ·T checkPointsToDesign .
[0088] S5. Calculate the coordinate value of the reference mark point at the center of the metal ball in the registration block reference array coordinate system.
[0089] In the registration block 3D model coordinate system F design The coordinate value of the center of the metal ball No. i is m i , its coordinate value m in the registration block reference array coordinate system i 'for
[0090] m′ i =T blockToDesign ·m i .
[0091] At this point, in the registration block reference array coordinate system, the coordinate values of the centers of the metal balls serving as reference marking points can be calculated using the above formula.
[0092] The apparatus used in the above method includes a registration block and an optical tracking probe.
[0093] like Figure 1 As shown in the figure, the hardware of the original registration block is modified as follows: three probe groove verification points that are not on the same straight line are set on the bottom plate component embedded with metal balls.
[0094] like Figure 2 As shown, the optical tracking probe is equipped with three or four reflective balls to form a tracking array. The origin of the probe coordinate system is located at the tip of the probe, and the direction of the coordinate axis is not specified. The probe is pre-calibrated by shaking the pen to fit the center of the ball so that the origin of its coordinate system completely coincides with the tip.
[0095] In image-guided surgery, the registration principle is briefly described as follows:
[0096] 1. Fix the registration block to the patient's target surgical area (usually bone) in some way (such as with bone screws), and take a CT scan of the registration block and the patient's target surgical area. The resulting CT image is as follows: Figure 3 As shown, the metal ball image on the registration block and the patient's bones are in the same image coordinate system, and the set of sphere centers of all metal ball images is recorded as P1;
[0097] 2. Such as Figure 4 As shown in Figure 1, during image-guided surgery, the positioning camera can capture the position of the optical spheres on the reference array of the registration block, thereby establishing a transformation relationship between the optical reference array coordinate system and the positioning camera coordinate system. Furthermore, the distribution of the individual metal spheres on the registration block in the optical reference array coordinate system can be determined in advance using design parameters. Therefore, the reference array of the registration block used by the positioning camera is the metal spheres on the registration block. The centers of all the metal spheres in the positioning camera coordinate system constitute point set P2.
[0098] 3. By calculating the transformation relationship between point set P2 and point set P1, the transformation relationship between the positioning camera coordinate system and the CT image coordinate system is established, and CT image registration is completed. The bone image in the CT image is mapped to the patient's actual bone, thus enabling image-guided surgery.
[0099] From the brief description of the principle of using the above registration block, it can be seen that the coordinates of the center of each metal ball on the registration block and the relative position of the reflective balls on the reference array (that is, the coordinate values of the center of each metal ball in the reference array coordinate system) must be known, and the accuracy of their values directly affects the accuracy of image navigation.
[0100] The hardware structure of the registration block causes the following three errors in the coordinate values of the metal ball center in the reference array coordinate system:
[0101] Error 1, such as Figure 5 As shown, the registration block reference array and the registration block base plate with metal balls embedded therein are two independent components, which are fixedly connected by a connecting mechanism. The manufacturing error of the connecting mechanism causes the relative position of the metal balls and the reflective balls to deviate from the designer.
[0102] Error 2, such as Figure 6 As shown, the manufacturing error of the registration block reference array causes the position of the reflective ball to deviate from the designed value.
[0103] Error 3, such as Figure 7 As shown in FIG, the manufacturing error of the bottom plate portion of the registration block where the metal balls are embedded causes the position of the metal balls to deviate from the designed value.
[0104] Since the base plate of the registration block with the metal balls embedded in it is an integrally formed part, its own manufacturing accuracy can generally meet the requirements. Therefore, the above three types of manufacturing errors can be ignored in practical applications, while the manufacturing errors of types 1 and 2 must be calibrated.
[0105] The calibration scheme described in this invention dynamically establishes a relationship between the verification point coordinate system and the reference array coordinate system by adding verification points to the registration block baseplate and acquiring their coordinates with a probe. Furthermore, the positional relationship between the verification points and the metal ball can be directly based on the designed values (both located on the integrally machined baseplate with high machining accuracy). This allows the accurate position of the metal ball center in the reference array coordinate system to be determined, resulting in the following advantages:
[0106] Avoid the influence of manufacturing errors in forms 1 and 2.
[0107] The relative position between the reference array and the base plate embedded with the metal balls does not need to be fixed. The reference array can be adjusted as needed (for example, to avoid occlusion). Each time the relative position between the two changes, the probe can be used to collect verification points and recalibrate the registration block, thus increasing the flexibility of the reference array's position. Each calibration only requires the probe to collect three verification points, making the operation relatively simple.
[0108] The above description is only a preferred specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with the technical field, within the technical scope disclosed by the present invention, who makes equivalent replacements or changes based on the technical solution and inventive concept of the present invention, should be covered by the scope of protection of the present invention.
Claims
1. A method for calibrating the position of a surgical navigation reference marker, characterized by: The steps include: S1. Obtain the coordinate values of the reference mark point and the verification point of the center of the metal ball in the coordinate system of the three-dimensional model of the registration block, establish a coordinate system in the three-dimensional model of the registration block, and obtain the coordinate values of the three verification points on the registration block; the registration block is provided with three probe groove verification points that are not on the same straight line on the bottom plate component embedded with the metal balls; the positioning camera captures the position of the optical ball on the reference array of the registration block, thereby establishing a conversion relationship between the optical reference array coordinate system and the positioning camera coordinate system, and at the same time, the distribution position of each metal ball on the registration block in the optical reference array coordinate system is known through the design parameters; the registration block reference array and the bottom plate of the registration block embedded with the metal balls are two independent components, and the two are fixedly connected by a connecting mechanism; S2. Calculate the transformation relationship from the verification point coordinate system to the registration block 3D model coordinate system, as follows: Using the coordinate values of the three verification points on the registration block in the three-dimensional model coordinate system of the registration block in step S1, the unit direction vectors of the three coordinate axes of the coordinate system formed by the three verification points of the registration block are calculated; The unit direction vectors of the three coordinate axes and the coordinate value of one of the verification points form a 4X4 matrix. This matrix is the transformation matrix from the 3D model coordinate system of the registration block to the verification point coordinate system. The transformation matrix from the verification point coordinate system to the 3D model coordinate system of the registration block can be obtained by calculating its inverse matrix. S3, using an optical tracking probe to collect the coordinate values of the verification point in the reference array coordinate system of the registration block; S4, calculating the conversion relationship between the registration block reference array coordinate system and the registration block three-dimensional model coordinate system; S5. Calculate the coordinate value of the reference mark point at the center of the metal ball in the registration block reference array coordinate system.
2. The method for calibrating the position of a surgical navigation reference marker according to claim 1, wherein: The specific content of step S1 is as follows: S11. Import the registration block 3D model into the engineering drawing software and establish a 3D orthogonal coordinate system F at any position in the software virtual 3D scene. design ; S12, obtain the model image coordinate system F design Under the following conditions, the coordinate values of the center of all metal balls used as reference points are: S13, obtain the model image coordinate system F design The coordinate values of the three verification points on the registration block are:
3. The method for calibrating the position of a surgical navigation reference marker according to claim 2, wherein: The specific content of step S2 is as follows: S21. Use model image coordinate system F design The coordinate values c1, c2, c3 of the three verification points on the lower registration block are calculated in the model image coordinate system F design Next, the unit direction vectors of the three coordinate axes of the coordinate system formed by the three verification points of the registration block; The values of the three vectors are recorded as follows: S21, use the above three vectors and the coordinate value of point c1 to form a 4X4 matrix T designToCheckPoints as follows: Matrix T designToCheckPoints is the transformation matrix from the registration block 3D model coordinate system to the verification point coordinate system. By finding its inverse matrix, we can get the transformation matrix T from the verification point coordinate system to the registration block 3D model coordinate system. checkPointsToDesign ; T checkPointsToDesign =(T designToCheckPoints ) -1 。 4. The method for calibrating the position of a surgical navigation reference marker according to claim 3, wherein: The specific content of step S3 is as follows: S31, place the probe tip at the registration block verification point i, i = 1, 2, 3, and collect the 4X4 transformation matrix from the optical positioning camera coordinate system to the probe coordinate system and the registration block reference array coordinate system respectively T cameraToProbe_i With T cameraToBlock_ ; S32: The homogeneous coordinate value q of the verification point i in the registration block reference array coordinate system can be obtained by using the following coordinate system transformation. i ': S33, repeat the above steps to obtain the homogeneous coordinate values of all three verification points in the registration block reference array coordinate system, the homogeneous coordinate values are a 4x1 sequence, namely q1', q2', q3', and the corresponding three-dimensional space coordinate values, the three-dimensional space coordinate values are a 3x1 sequence, namely c1', c2', c3'; S34. Use the coordinate values obtained by the above calculation to calculate the unit direction vectors of the three coordinate axes of the coordinate system formed by the registration block verification points in the registration block reference array coordinate system: The above three vectors are written as follows: S35. Use the above three vectors and the coordinate value of c1' [x1', y1', z1'] to create a 4X4 matrix T blockToCheckPoints , which is the transformation matrix from the registration block reference array coordinate system to the registration block verification point coordinate system:
5. The method for calibrating the position of a surgical navigation reference marker according to claim 1, wherein: The specific content of step S4 is as follows: Use the coordinate system transformation relationship T calculated in steps S2 and S3 blockToCheckPoints With T checkPointsToDesign , calculate the transformation relationship T from the registration block reference array coordinate system to the registration block three-dimensional model coordinate system blockToDesign ; T blockToDesign =T blockToCheckPoints ·T checkPointsToDesign 。 6. The method for calibrating the position of a surgical navigation reference marker according to claim 1, wherein: The specific content of step S5 is as follows: In the registration block 3D model coordinate system F design The coordinate value of the center of the metal ball No. i is m i , its coordinate value m in the registration block reference array coordinate system i 'for: m′ i =T blockToDesign ·m i 。
Citation Information
Patent Citations
Surgical navigation device and system, calibration method, medium and electronic equipment
CN113143463A
Method for reconstructing a 3D image from 2D X-ray images
EP2868277A1