CBCT calibration method for non-ideal circular trajectory and poor motion repeatability

Abstract

The application provides a CBCT calibration method for a non-ideal circular track and poor motion repeatability, comprising the following steps: embedding steel balls on a phantom surface, recording geometric positions of the steel balls, placing the phantom in the center of CBCT, irradiating the phantom, collecting images, and obtaining two-dimensional flat films; processing the flat films through an image processing algorithm to automatically mark the coordinates of the metal points; obtaining the coordinates of each metal mark point in the phantom coordinate system according to the coordinate values of the known steel ball mark points in the phantom and the serial numbers of the steel ball mark points automatically marked in the second step, and finding the corresponding coordinate values in the detector flat plate coordinate system; calculating a matrix; calculating a correction matrix for each Gantry angle θ; correcting the collected projection images of each angle by using the correction matrix, and then performing three-dimensional reconstruction. The method provided by the application can calibrate the CBCT for a non-ideal circular track and poor motion repeatability, and realizes accurate reconstruction of CT images.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of CBCT calibration, in particular to a CBCT calibration method for non-ideal circular trajectory and poor motion repeatability. BACKGROUND

[0002] Computer tomography technology is an imaging technology for obtaining cross-sectional information of an object by measuring the object at different angles of ray projection. According to the ray beam acquisition mode, it can be divided into two categories of fan beam CT and cone beam CT. Among them, the fan beam spiral CT has been widely used in the medical field, and the cone beam CT (CBCT) has become an industry hotspot due to its low dose, high spatial resolution and fast scanning speed. In clinical application, the CBCT image reconstruction usually adopts FDK algorithm to reconstruct the image. FDK algorithm has two prerequisite conditions: first, the line connecting the ray source focus and the detector center must pass through the rotation center line and be perpendicular to the detector plane; second, the rotation axis should be parallel to the center column direction of the detector. However, in the actual construction of the CBCT system, due to the existence of mechanical errors, the CBCT system cannot strictly meet the above two prerequisite conditions, so there is a certain geometric deviation. If the deviation is not corrected, the reconstructed image will produce serious artifacts, affecting the image quality. Therefore, the accurate solution of geometric parameters (i.e. geometric correction) is the premise of realizing the accurate reconstruction of CT image.

[0003] The geometric correction method currently used in the industry needs to design a correction phantom, obtain the projection data of the correction phantom at multiple angles or one angle, use an analytical method to obtain the geometric parameters of the CT system, and then perform subsequent image reconstruction according to the parameters. Various CBCT correction methods require the system motion to be a standard circular motion, and the system motion has a high repeatability (i.e. the motion is completely stable). Under this premise, the correction algorithm corrects the rigid displacement of the system. However, in some actual application scenarios, limited by mechanical precision, the system motion is irregular and has poor repeatability, i.e. the system motion has random deviation, such as image scanning acquisition in a bumpy motion state. For this kind of special application scenario, there is no mature solution in the industry at present. SUMMARY

[0004] The technical problem solved by the present application is to provide a CBCT calibration method for non-ideal circular trajectory and poor motion repeatability to solve the problems mentioned in the technical background. In order to achieve the above purpose, the present application provides the following technical scheme: a CBCT calibration method for non-ideal circular trajectory and poor motion repeatability, specifically including the following contents:

[0005] (1) Steel balls are embedded on the surface of a cylindrical phantom, the geometric position of each steel ball is recorded, the phantom is placed in the center of CBCT, the cylindrical phantom is irradiated, and images are acquired to obtain a two-dimensional flat film.

[0006] (2) The flat plate is processed by an image processing algorithm to automatically mark the coordinates of the metal points, and the measurement result is the coordinate value in the detector flat plate coordinate system.

[0007] (3) Based on the known coordinates of the steel ball markers in the mold and the serial numbers of the steel ball markers automatically marked in the second step, obtain the coordinates p of each metal marker in the mold coordinate system in the second step. i , and find p i The corresponding coordinate value q in the detector flat plate coordinate system i ;

[0008] (4) Calculate matrix M t : Set coordinates The projection onto the flat plate is given by the flat plate coordinate system as follows: but

[0009]

[0010] in M is the transformation matrix from the phantom coordinate system to the CBCT coordinate system. p This is the projection matrix of a point in the CBCT coordinate system onto the flat plate coordinate system, i.e.

[0011]

[0012] set up Extracted from flat slice images The solution is to find the projected coordinates on the flat plate.

[0013]

[0014] achievable For any point p in the peg coordinate system i M t ·p i These are the coordinates of the point in the CBCT coordinate system;

[0015] (5) Calculate M(θ) for each Gantry angle θ;

[0016] (6) The projected images acquired from each angle are corrected using the correction matrix M(θ) and then three-dimensional reconstruction is performed.

[0017] Preferably, in (1), the cylindrical mold surface is embedded with 16 steel balls, which are distributed in a spiral shape.

[0018] Preferably, the method for acquiring images in (1) is as follows: the cylindrical model is irradiated at each angle from 0° to 359° to acquire images and obtain 360 two-dimensional flat images.

[0019] Preferably, the image processing algorithm in (2) is the Hough transform.

[0020] Preferably, the calculation method for (5) is as follows: then under the Gantry angle θ, the equation can be obtained:

[0021] q i =M p ·M(θ)·M t ·p i

[0022] set up Extracted from flat slice images The solution is to find the projected coordinates on the flat plate.

[0023]

[0024] M(θ) is the correction matrix when Gantry is at angle θ.

[0025] Compared with existing technologies, the method provided by this invention can calibrate CBCT with non-ideal circular trajectories and poor motion repeatability, thereby achieving accurate reconstruction of CT images. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of steel balls embedded on the surface of the cylindrical mold in the embodiment;

[0027] Figure 2 This is a schematic diagram of image acquisition in the embodiment.

[0028] To more clearly illustrate the technical solution of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. Detailed Implementation

[0029] To make the technical means, creative features, workflow, and usage methods of this invention readily understandable and effective, the technical solutions in the embodiments of this invention will be clearly and completely described below in conjunction with the embodiments of this invention. Obviously, the described embodiments are merely a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort, and other conclusions derived from non-creative extensions, are within the scope of protection of this invention.

[0030] Example

[0031] For CBCT calibration methods that address non-ideal circular trajectories and poor motion repeatability, the specific methods include the following:

[0032] (1) Sixteen steel balls are embedded on the surface of the cylindrical mold body, and they are arranged in a spiral shape, such as... Figure 1 As shown, the geometric position of each steel ball was recorded. The phantom was placed at the center of the CBCT, and images were acquired by irradiating the cylindrical phantom at each angle from 0° to 359°. Figure 2 As shown, 360 two-dimensional flat images were obtained;

[0033] (2) The flat plate is processed by an image processing algorithm (Hough transform) to automatically mark the coordinates of the metal points, and the measurement result is the coordinate value in the detector flat plate coordinate system.

[0034] (3) Based on the known coordinates of the steel ball markers in the mold and the serial numbers of the steel ball markers automatically marked in the second step, obtain the coordinates p of each metal marker in the mold coordinate system in the second step. i , and find p i The corresponding coordinate value q in the detector flat plate coordinate system i ;

[0035] (4) Calculate matrix M t Let the projection of the coordinates onto the flat plate be in the flat plate coordinate system. but

[0036]

[0037] in M is the transformation matrix from the phantom coordinate system to the CBCT coordinate system. p This is the projection matrix of a point in the CBCT coordinate system onto the flat plate coordinate system, i.e.

[0038]

[0039] set up Extracted from flat slice images The solution is to find the projected coordinates on the flat plate.

[0040]

[0041] achievable For any point p in the peg coordinate system i M t ·p i These are the coordinates of the point in the CBCT coordinate system;

[0042] (5) Calculate M(θ) for each Gantry angle θ:

[0043] Then, at the Gantry angle θ, we can obtain the equation:

[0044] q i =M p ·M(θ)·M t ·p i

[0045] set up Extracted from flat slice images The solution is to find the projected coordinates on the flat plate.

[0046]

[0047] M(θ) is the correction matrix when Gantry is at angle θ.

[0048] (6) The projected images acquired from each angle are corrected using the correction matrix M(θ) and then three-dimensional reconstruction is performed.

[0049] The foregoing description illustrates the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A CBCT calibration method for non-ideal circular trajectories and poor motion repeatability, characterized in that, Specifically, it includes the following: (1) Steel balls are embedded on the surface of a cylindrical phantom. There are 16 steel balls distributed in a spiral shape. The geometric position of each steel ball is recorded. The phantom is placed in the center of CBCT and the cylindrical phantom is irradiated to acquire images and obtain two-dimensional flat films. (2) The flat plate is processed by an image processing algorithm to automatically mark the coordinates of the steel ball markers, and the measurement result is the coordinate value in the detector flat plate coordinate system. ; (3) Based on the known coordinates of the steel ball markers in the mold and the serial numbers of the steel ball markers automatically marked in step (2), obtain the coordinates p of each steel ball marker in the mold coordinate system in step (2). i And find p i The corresponding coordinate value q in the detector flat plate coordinate system i ; (4) Calculate the matrix Let the set of coordinate vectors be... The projection onto the flat plate is given by the flat plate coordinate system as follows: ,but in This is the transformation matrix from the phantom coordinate system to the CBCT coordinate system. This is the projection matrix of a point in the CBCT coordinate system onto the flat plate coordinate system, i.e. set up Extracted from flat slice images The solution is to find the projected coordinates on the flat plate. achievable For any point p in the peg coordinate system i , These are the coordinates of the point in the CBCT coordinate system; (5) Calculate the angle θ for each Gantry angle. ; (6) Correct the projected images acquired from each angle using a correction matrix. Make corrections, and then perform 3D reconstruction.

2. The CBCT calibration method according to claim 1 for non-ideal circular trajectories and poor motion repeatability is characterized in that: The method for acquiring images in (1) is as follows: the cylindrical model is irradiated at each angle from 0° to 359° to acquire images, resulting in 360 two-dimensional flat images.

3. The CBCT calibration method according to claim 1 for non-ideal circular trajectories and poor motion repeatability is characterized in that: The image processing algorithm in (2) is the Hough transform.

4. The CBCT calibration method according to claim 1 for non-ideal circular trajectories and poor motion repeatability is characterized in that: The calculation method for (5) is as follows: then under the Gantry angle θ, the equation can be obtained: set up Extracted from flat slice images The solution is to find the projected coordinates on the flat plate. have to , which is the correction matrix when Gantry is at angle θ.

Images