A method and device for globally optimizing CT geometry parameters

Abstract

The application discloses a CT geometry parameter global optimization method, relates to the CT imaging technical field, and combines a constraint optimization algorithm to obtain initial optimization parameters through iteration, so that a plurality of horizontal offset candidate values are generated; a global optimal horizontal offset correction parameter is screened out through image reconstruction and target function value calculation, and finally, a CT system global optimal geometry correction parameter is generated, so that the problem of insufficient traditional CT geometry correction precision is solved, the CT image reconstruction quality can be effectively improved, and the method is suitable for the geometry parameter correction scene of various medical CT, industrial CT imaging systems.

Description

Technical Field

[0001] This application relates to the field of CT imaging technology, specifically to a method and apparatus for global optimization of CT geometric parameters. Background Technology

[0002] Computed tomography (CT) technology acquires X-ray projection data of an object from different angles and combines it with image reconstruction algorithms to obtain the object's three-dimensional internal structure. It is a core non-destructive testing technology in fields such as medical diagnosis, industrial non-destructive testing, security inspection, and cultural relic archaeology. The accuracy of the geometric parameters of a CT imaging system directly determines the quality of the reconstructed image. If there are deviations in the geometric parameters, the reconstructed image will exhibit serious artifacts, geometric distortion, and reduced resolution. Therefore, geometric correction is an indispensable and crucial step in the CT imaging process.

[0003] Existing CT geometric correction methods are mainly divided into two categories: one is based on specially designed standard phantoms. This type of method uses high-precision calibrated phantoms to extract the projection information of phantom markers and solve for geometric parameters. It achieves high correction accuracy but requires custom-made phantoms and separate scanning for each correction, making the process complex and unsuitable for scenarios such as rapid on-site correction or online correction. The other category is phantomless geometric correction methods, mainly including correction methods based on projection symmetry and correction methods based on reconstructed image quality. These methods do not require scanning a standard phantom; correction can be completed using only the projection data of the object being examined, making them more adaptable and applicable to a wider range of engineering scenarios. Among these, geometric correction methods based on reconstructed image quality directly use the quality evaluation indicators of the reconstructed image as the optimization objective, transforming geometric correction into a multi-parameter constrained optimization problem. The correction results are directly linked to the final CT image quality, and the obtained geometric parameter accuracy is generally higher than that of projection symmetry methods, making it the mainstream research direction for phantomless geometric correction. These methods typically use the gradient, information entropy, high-frequency energy, and sharpness of the reconstructed image as objective functions, and solve for the optimal geometric parameters through iterative optimization algorithms. However, in practical engineering applications, this type of method has an unavoidable core flaw: the key parameter to be corrected—the detector horizontal offset dx—is affected by image interpolation, and its objective function curve has multiple local extrema. Conventional nonlinear optimization algorithms (such as the simplex algorithm, gradient descent, and particle swarm optimization) are prone to getting trapped in local extrema rather than the global optimum, ultimately leading to systematic deviations in the solved geometric parameters. This results in artifacts, blurred edges, and geometric distortions in CT reconstructed images, severely affecting imaging accuracy. To solve the local extrema problem, existing solutions often use global optimization algorithms (such as genetic algorithms and simulated annealing algorithms) to replace conventional optimization algorithms. However, the computational load of these algorithms increases exponentially, requiring numerous iterations and lengthy correction times, making them difficult to implement in engineering. Some solutions improve the probability of global optimization by expanding the parameter search range, but this further increases the computational burden and still cannot completely avoid the risk of getting trapped in local extrema. Summary of the Invention

[0004] The purpose of this application is to provide a global optimization method and apparatus for CT geometric parameters, which solves the problems of insufficient correction accuracy, inability to balance correction accuracy and correction efficiency, and / or CT image reconstruction errors caused by parameter optimization errors in the prior art.

[0005] This application is achieved through the following technical solution:

[0006] A global optimization method for CT geometric parameters, comprising:

[0007] The geometric parameters to be corrected of the CT imaging system are obtained, and an objective function based on the quality of the reconstructed image is constructed; the geometric parameters to be corrected include at least the detector horizontal offset.

[0008] A geometric correction method based on reconstructed image quality is adopted, combined with a conventional constraint optimization algorithm, to iteratively optimize the geometric parameters to be corrected and determine the initial optimized geometric parameters;

[0009] Based on the initialized and optimized geometric parameters, multiple candidate horizontal offset values ​​to be verified are obtained;

[0010] For each candidate horizontal offset value to be verified, CT image reconstruction is performed, and the objective function value corresponding to the reconstructed CT image is obtained according to the objective function based on the quality of the reconstructed image.

[0011] Based on the objective function value corresponding to the reconstructed CT image, the candidate horizontal offset values ​​to be verified are filtered to determine the globally optimal horizontal offset correction parameters;

[0012] Based on the aforementioned globally optimal horizontal offset correction parameters, the final globally optimal geometric correction parameters for the CT system are generated.

[0013] In one possible implementation, an objective function based on the reconstructed image quality is constructed, including:

[0014] An objective function based on the quality of reconstructed images is constructed by using one or more of the following evaluation metrics: gradient of the reconstructed image, high-frequency energy, and image sharpness. When there is more than one quality evaluation metric, the multiple quality evaluation metrics are fused to obtain the objective function based on the quality of the reconstructed image.

[0015] In one possible implementation, the initial optimization set parameters include at least the detector horizontal offset;

[0016] Alternatively, the initial set of optimized parameters may include at least the detector horizontal offset, the distance from the X-ray source to the sample, and the detector rotation angle.

[0017] In one possible implementation, based on the initial optimized geometric parameters, a plurality of candidate horizontal offset values ​​to be verified are obtained, including:

[0018] Based on the pixel size of the CT system detector, multiple candidate values ​​for horizontal offset are generated within the neighborhood of the initial optimized geometric parameters corresponding to the horizontal offset of the detector.

[0019] In one possible implementation, the range of the candidate horizontal offset value to be verified is [dx0-n×p, dx0+n×p]; where dx0 represents the initial optimized geometric parameters corresponding to the horizontal offset of the detector, n represents a positive integer, and p represents the pixel size of the CT system detector.

[0020] In one possible implementation, for each candidate horizontal offset value to be verified, CT image reconstruction is performed, and the objective function value corresponding to the reconstructed CT image is obtained according to the objective function based on the reconstructed image quality, including:

[0021] For each candidate horizontal offset value to be verified, a conventional reconstruction algorithm is used to reconstruct the CT image and obtain the reconstructed CT image; the conventional reconstruction algorithm is set as either a filtered back projection algorithm or an algebraic iteration algorithm.

[0022] The reconstructed CT image is evaluated using the objective function based on the quality of the reconstructed image to determine the quality evaluation index value and obtain the corresponding objective function value.

[0023] In one possible implementation, the candidate horizontal offset values ​​to be verified are filtered based on the objective function value corresponding to the reconstructed CT image to determine the globally optimal horizontal offset correction parameters, including:

[0024] Based on the objective function value corresponding to the reconstructed CT image, the value with the largest objective function value among all candidate horizontal offset values ​​to be verified is taken as the globally optimal horizontal offset correction parameter.

[0025] In one possible implementation, based on the globally optimal horizontal offset correction parameters, the final globally optimal geometric correction parameters for the CT system are generated, including:

[0026] If the geometric parameters to be corrected only include detector horizontal offset, then the globally optimal horizontal offset correction parameter is directly used as the final globally optimal geometric correction parameter of the CT system.

[0027] When the geometric parameters to be corrected include detector horizontal offset and other parameters, the global optimal horizontal offset correction parameter corresponding to the detector horizontal offset and the initial optimized geometric parameters corresponding to the other parameters are used together as the final global optimal geometric correction parameters of the CT system.

[0028] A second aspect of this application provides a global optimization device for CT geometric parameters, comprising:

[0029] An initialization module is used to acquire the geometric parameters to be corrected from the CT imaging system and to construct an objective function based on the quality of the reconstructed image; the geometric parameters to be corrected include at least the detector horizontal offset.

[0030] The conventional iterative module is used to iteratively optimize the geometric parameters to be corrected by employing a geometric correction method based on the quality of the reconstructed image, combined with a conventional constraint optimization algorithm, to determine the initial optimized geometric parameters;

[0031] The candidate value generation module is used to obtain multiple horizontal offset candidate values ​​to be verified based on the initial optimized geometric parameters.

[0032] The reconstruction verification module is used to reconstruct the CT image for each candidate horizontal offset value to be verified, and to obtain the target function value corresponding to the reconstructed CT image according to the target function based on the quality of the reconstructed image.

[0033] The parameter filtering module filters the candidate horizontal offset values ​​to be verified based on the objective function value corresponding to the reconstructed CT image, and determines the globally optimal horizontal offset correction parameter.

[0034] The result output module is used to generate the final global optimal geometric correction parameters of the CT system based on the global optimal horizontal offset correction parameters.

[0035] Compared with the prior art, this application has the following advantages and beneficial effects:

[0036] This application provides a global optimization method for CT geometric parameters. This method combines a constrained optimization algorithm to iteratively obtain initial optimization parameters, thereby generating multiple horizontal offset candidate values. The globally optimal horizontal offset correction parameters are then selected through image reconstruction and objective function value calculation, ultimately generating the globally optimal geometric correction parameters for the CT system. This method solves the problem of insufficient accuracy in traditional CT geometric correction, effectively improves the quality of CT image reconstruction, and is applicable to geometric parameter correction scenarios in various medical CT and industrial CT imaging systems. Attached Figure Description

[0037] To more clearly illustrate the technical solutions of the exemplary embodiments of this application, the accompanying drawings used in the embodiments will be briefly described below. It should be understood that the following drawings only show some embodiments of this application and should not be considered as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort. In the drawings:

[0038] Figure 1 A schematic diagram illustrating a global optimization method for CT geometric parameters provided in an embodiment of this application;

[0039] Figure 2 A graph showing the correspondence between detector horizontal offset and high-frequency energy of reconstructed images provided in the embodiments of this application;

[0040] Figure 3 A graph showing the correspondence between the detector's horizontal offset and the gradient of the reconstructed image provided in the embodiments of this application;

[0041] Figure 4A graph showing the relationship between detector horizontal offset and reconstructed image sharpness provided in the embodiments of this application;

[0042] Figure 5 A schematic diagram of a global optimization device for CT geometric parameters provided in an embodiment of this application;

[0043] The attached diagram shows the markings and corresponding component names:

[0044] 501 Initialization module, 502 Regular iteration module, 503 Candidate value generation module, 504 Reconstruction and verification module, 505 Parameter filtering module, 506 Result output module. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the embodiments and accompanying drawings. The illustrative embodiments and descriptions of this application are only for explaining this application and are not intended to limit this application.

[0046] To address the core shortcomings of existing CT geometric correction techniques based on reconstructed image quality, this application provides a global optimization method for CT geometric parameters, aiming to solve one or more of the following technical problems:

[0047] (1) In the existing geometric correction method based on the quality of reconstructed images, due to the influence of image interpolation calculation, the objective function of detector horizontal offset dx has multiple local extrema. Conventional optimization algorithms are prone to getting trapped in local extrema and cannot obtain the globally optimal geometric parameters, resulting in insufficient correction accuracy.

[0048] (2) Existing global optimization schemes for avoiding local extrema have problems such as large computational load, long correction time, and high difficulty in engineering implementation, and cannot balance correction accuracy and correction efficiency.

[0049] (3) Errors in parameter optimization lead to artifacts, geometric distortions, and reduced resolution in CT reconstructed images, which cannot meet the imaging requirements of scenarios such as medical diagnosis and high-precision industrial inspection.

[0050] like Figure 1 As shown in the figure, this application provides a global optimization method for CT geometric parameters, including:

[0051] S101. Obtain the geometric parameters to be corrected of the CT imaging system, and simultaneously construct an objective function based on the quality of the reconstructed image; the geometric parameters to be corrected include at least the detector horizontal offset.

[0052] S102. Using a geometric correction method based on the quality of the reconstructed image, combined with a conventional constraint optimization algorithm, the geometric parameters to be corrected are iteratively optimized to determine the initial optimized geometric parameters.

[0053] S103. Based on the initialized and optimized geometric parameters, obtain multiple candidate horizontal offset values ​​to be verified;

[0054] S104. For each candidate horizontal offset value to be verified, perform CT image reconstruction, and obtain the target function value corresponding to the reconstructed CT image according to the target function based on the quality of the reconstructed image.

[0055] S105. Based on the objective function value corresponding to the reconstructed CT image, the candidate horizontal offset values ​​to be verified are filtered to determine the globally optimal horizontal offset correction parameters.

[0056] S106. Based on the global optimal horizontal offset correction parameters, generate the final global optimal geometric correction parameters for the CT system.

[0057] In one possible implementation, an objective function based on the reconstructed image quality is constructed, including:

[0058] An objective function based on the quality of reconstructed images is constructed by using one or more of the following evaluation metrics: gradient of the reconstructed image, high-frequency energy, and image sharpness. When there is more than one quality evaluation metric, the multiple quality evaluation metrics are fused to obtain the objective function based on the quality of the reconstructed image.

[0059] In one possible implementation, the initial optimization set parameters include at least the detector horizontal offset;

[0060] Alternatively, the initial set of optimized parameters may include at least the detector horizontal offset, the distance from the X-ray source to the sample, and the detector rotation angle.

[0061] A geometric correction method based on reconstructed image quality can be employed, combined with conventional constrained optimization algorithms (such as the simplex optimization algorithm). The objective function for reconstructed image quality (image sharpness, gradient, high-frequency energy) is maximized, and the geometric parameters to be corrected are iteratively optimized to determine the initial optimized geometric parameters.

[0062] In one possible implementation, based on the initial optimized geometric parameters, a plurality of candidate horizontal offset values ​​to be verified are obtained, including:

[0063] Based on the pixel size of the CT system detector, multiple candidate values ​​for horizontal offset are generated within the neighborhood of the initial optimized geometric parameters corresponding to the horizontal offset of the detector.

[0064] In one possible implementation, the range of the candidate horizontal offset value to be verified is [dx0-n×p, dx0+n×p]; where dx0 represents the initial optimized geometric parameters corresponding to the horizontal offset of the detector, n represents a positive integer, and p represents the pixel size of the CT system detector.

[0065] In one possible implementation, for each candidate horizontal offset value to be verified, CT image reconstruction is performed, and the objective function value corresponding to the reconstructed CT image is obtained according to the objective function based on the reconstructed image quality, including:

[0066] For each candidate horizontal offset value to be verified, a conventional reconstruction algorithm is used to reconstruct the CT image and obtain the reconstructed CT image; the conventional reconstruction algorithm is set as either a filtered back projection algorithm or an algebraic iteration algorithm.

[0067] The reconstructed CT image is evaluated using the objective function based on the quality of the reconstructed image to determine the quality evaluation index value and obtain the corresponding objective function value.

[0068] In one possible implementation, the candidate horizontal offset values ​​to be verified are filtered based on the objective function value corresponding to the reconstructed CT image to determine the globally optimal horizontal offset correction parameters, including:

[0069] Based on the objective function value corresponding to the reconstructed CT image, the value with the largest objective function value among all candidate horizontal offset values ​​to be verified is taken as the globally optimal horizontal offset correction parameter.

[0070] In one possible implementation, based on the globally optimal horizontal offset correction parameters, the final globally optimal geometric correction parameters for the CT system are generated, including:

[0071] If the geometric parameters to be corrected only include detector horizontal offset, then the globally optimal horizontal offset correction parameter is directly used as the final globally optimal geometric correction parameter of the CT system.

[0072] When the geometric parameters to be corrected include detector horizontal offset and other parameters, the global optimal horizontal offset correction parameter corresponding to the detector horizontal offset and the initial optimized geometric parameters corresponding to the other parameters are used together as the final global optimal geometric correction parameters of the CT system.

[0073] For example, this application provides an illustration based on the above-described method and technical solution (i.e., the technical solution described in steps S101 to S106) so that those skilled in the art can more clearly understand the technical solution described in this application. The specific example is as follows.

[0074] like Figures 2-4 As shown, based on the principle that the distance between adjacent local extrema of the horizontal offset dx is one detector pixel size, we propose an improved method for global optimization of geometric parameters. The core of this method is a two-step approach that uses initial optimization followed by neighborhood pixel-level traversal verification to achieve a fast and accurate solution for the globally optimal geometric parameters. The specific technical solution includes the following steps:

[0075] Step 1: Determine the basic calibration parameters and optimization objectives. Identify the set of geometric parameters to be calibrated for the CT imaging system.

[0076] The parameters to be corrected include at least the detector horizontal offset dx; at the same time, an objective function based on the quality of the reconstructed image is selected, which includes, but is not limited to, one or more combinations of the gradient of the reconstructed image, high-frequency energy, and image sharpness evaluation index, and the optimization objective (maximizing / minimizing the objective function value) is determined.

[0077] In this example, the core parameters of the cone-beam CT system used are as follows: the X-ray source is a 9MeV microfocus X-ray source, the source sample is a section of cable, the object distance s1=1000mm, and the source-image distance s2=3940mm; the physical pixel size of the area array detector is 0.139mm, and the pixel merging is 2×2, that is, the sampling pixel of the projected image is p=0.278mm; the geometric parameters to be corrected include the detector horizontal offset dx and the detector deflection angle θ around the optical axis; the optimization objective function is to select the high-frequency energy of the reconstructed slice image, and the optimization objective is to maximize the image sharpness (the higher the image sharpness value, the clearer the image edge and the better the imaging quality).

[0078] The set of geometric parameters to be corrected can be determined as [dx, θ], the objective function is the image sharpness of the CT reconstructed slice image, and the optimization direction is to maximize the value of the objective function.

[0079] Step 2: Optimize initial geometric parameters.

[0080] The traditional geometric correction method based on reconstructed image quality is adopted, and conventional constraint optimization algorithm is combined to iteratively optimize the set of geometric parameters to be corrected, so as to obtain an initial optimized geometric parameter set, wherein the detector horizontal offset parameter obtained by the initial optimization is denoted as dx0.

[0081] A traditional geometric correction method based on image sharpness can be used, combined with a simplex nonlinear optimization algorithm, to iteratively optimize the geometric parameters to be corrected and obtain an initial set of optimized geometric parameters, where the initial optimized horizontal offset dx0=2.070mm and θ=0.15°.

[0082] Step 3: Set candidate values ​​for the horizontal offset parameter.

[0083] Using the pixel size p of the CT system detector as the basic step size, multiple horizontal offset candidate values ​​to be verified are set within the neighborhood of dx0. The range of the candidate values ​​is dx0±n×p, where n is a positive integer from 1 to 5, which can be flexibly adjusted according to the detector pixel specifications and system accuracy requirements.

[0084] For example, based on the detector pixel size p=0.278mm, the traversal range is set to dx0±5×p, with a step size of 0.278mm, generating 11 horizontal offset candidate values, which are: 0.680mm, 0.958mm, 1.236mm, 1.514mm, 1.792mm, 2.070mm, 2.348mm, 2.626mm, 2.904mm, 3.182mm, and 3.460mm.

[0085] Step 4: Calculate the objective function value of the candidate values.

[0086] With all geometric parameters except dx fixed in the initial optimized geometric parameter set, CT image reconstruction is performed for each horizontal offset candidate value, and the objective function value of the corresponding reconstructed image is calculated to obtain the correspondence between all candidate values ​​and objective function values.

[0087] For example, θ can be fixed at 0.15°, and the FDK algorithm can be used to reconstruct CT images for each of the 11 candidate dx values. The image sharpness of the reconstructed CT slice images can be calculated to obtain the objective function value corresponding to each candidate value, as shown in Table 1.

[0088] Table 1. Sharpness of reconstructed CT images corresponding to candidate values ​​of detector horizontal offset.

[0089]

[0090] Step 5: Filter the globally optimal horizontal offset parameters.

[0091] Based on the preset optimization objective, the objective function values ​​corresponding to all candidate values ​​are compared, and the horizontal offset parameter corresponding to the optimal objective function value is selected as the globally optimal horizontal offset correction parameter, denoted as dx_opt.

[0092] By comparing the objective function values ​​of all candidate values, when dx_opt=2.348mm, the image sharpness reaches the maximum value of 1.296, which is the global optimum; while the initial optimization value of dx0=2.070mm is only a local extreme point, and its corresponding objective function value is significantly lower than the global optimum.

[0093] Step 6: Generate the final global optimal geometric correction parameter set.

[0094] The globally optimal horizontal offset correction parameter dx_opt is used to replace dx0 in the initial optimized geometric parameter set. Combined with the remaining optimized geometric parameters, the final globally optimal geometric correction parameter set of the CT system is generated, thus completing the system geometric correction.

[0095] The final horizontal offset correction parameter can be dx_opt=2.348mm. Combined with θ=0.15°, the globally optimal set of geometric correction parameters can be generated to complete the geometric correction of the CT system.

[0096] CT reconstruction using the globally optimal geometric parameters obtained in this invention produces images free from geometric distortion and artifacts, with sharp and clear object edges, high internal detail resolution, and uniform grayscale distribution. In contrast, images reconstructed using local extremum parameters obtained through traditional methods exhibit significant edge blurring and stripe artifacts, and severely insufficient detail resolution. This invention, through only 11 simple reconstructions and target value calculations, can accurately escape local extrema and obtain globally optimal correction parameters, reducing correction time by more than 90% compared to traditional global optimization algorithms, while significantly improving correction accuracy.

[0097] Compared with the prior art, the technical solution provided in this application has the following beneficial effects:

[0098] (1) It fundamentally solves the problem of local extrema and greatly improves the correction accuracy. Based on the core feature that the distance between adjacent local extrema of dx is 1 pixel, this invention performs pixel-level traversal verification in the neighborhood of the initial optimization parameters, which can accurately jump out of local extrema and lock the globally optimal geometric parameters, thus completely avoiding the defect of traditional optimization algorithms that are prone to getting trapped in local extrema, and significantly improving the correction accuracy of geometric parameters.

[0099] (2) Balancing correction accuracy and computational efficiency, this invention is easy to implement in engineering. It only requires small-scale verification within ±5 pixels of the initial optimization parameters, without changing the core framework of the original optimization algorithm or adopting a highly complex global optimization algorithm. The computational load is extremely low and the correction time is short. It can be seamlessly embedded into the geometric correction process of existing CT systems, making it very easy to implement in engineering.

[0100] (3) Significantly improves the quality of CT reconstructed images. Based on the globally optimal geometric parameters obtained by this invention, artifacts, geometric distortions, and edge blurring in reconstructed images can be effectively eliminated, improving the spatial resolution and grayscale uniformity of the images, and ensuring the imaging reliability of scenarios such as medical diagnosis and high-precision non-destructive testing in industry.

[0101] (4) High versatility and wide applicability. This invention does not rely on a specific CT scanning track and can be adapted to various CT scanning methods such as circular track, spiral track, offset scanning track, and linear track; it is not limited to a specific objective function type and can be compatible with various evaluation indicators based on the quality of reconstructed images; at the same time, it is not only applicable to X-ray CT systems, but can also be extended to the geometric correction scenarios of various tomographic imaging systems such as neutron CT, proton CT, and electron beam CT.

[0102] like Figure 5 As shown in the figure, this application provides a global optimization device for CT geometric parameters, including:

[0103] Initialization module 501 is used to acquire the geometric parameters to be corrected of the CT imaging system and to construct an objective function based on the quality of the reconstructed image; the geometric parameters to be corrected include at least the detector horizontal offset.

[0104] The conventional iteration module 502 is used to iteratively optimize the geometric parameters to be corrected by employing a geometric correction method based on the quality of the reconstructed image, combined with a conventional constraint optimization algorithm, to determine the initial optimized geometric parameters;

[0105] The candidate value generation module 503 is used to obtain multiple horizontal offset candidate values ​​to be verified based on the initial optimized geometric parameters.

[0106] The reconstruction verification module 504 is used to reconstruct the CT image for each candidate horizontal offset value to be verified, and to obtain the target function value corresponding to the reconstructed CT image according to the target function based on the quality of the reconstructed image.

[0107] The parameter filtering module 505 filters the candidate horizontal offset values ​​to be verified based on the objective function value corresponding to the reconstructed CT image, and determines the globally optimal horizontal offset correction parameter.

[0108] The result output module 506 is used to generate the final global optimal geometric correction parameters of the CT system based on the global optimal horizontal offset correction parameters.

[0109] The global optimization device for CT geometric parameters provided in this application embodiment can perform the above-described method and technical solution. The principle and beneficial effects are similar, and will not be described again in this application embodiment.

[0110] Based on the same inventive concept, embodiments of this application also provide an electronic device, including a processor and a memory; the memory and the processor are interconnected via a bus.

[0111] The memory stores computer-executed instructions;

[0112] The processor executes computer execution instructions stored in the memory, causing the processor to perform a global optimization method for CT geometric parameters as described in any embodiment of this application.

[0113] For specific examples, memory may include, but is not limited to, random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out (FIFO) memory, and / or first-in-last-out (FILO) memory, etc.; specifically, processor may include one or more processing cores, such as a 4-core processor, an 8-core processor, etc. The processor may be implemented using at least one hardware form of DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), PLA (Programmable Logic Array). Furthermore, the processor may include a main processor and coprocessors. The main processor, also known as the CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state.

[0114] This application also provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the global optimization method for CT geometric parameters described in any of the above embodiments.

[0115] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the global optimization method for CT geometric parameters described in any of the above embodiments.

[0116] All or part of the steps in the above method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a readable memory. When the program is executed, it performs the steps of the above method embodiments; and the aforementioned memory (storage medium) includes: read-only memory (ROM), RAM, flash memory, hard disk, solid-state drive, magnetic tape, floppy disk, optical disk, and any combination thereof.

[0117] This application describes embodiments with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processing unit of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processing unit of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0118] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0119] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0120] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this application are indicated by the claims.

[0121] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this application is limited only by the appended claims.

Claims

1. A global optimization method for CT geometric parameters, characterized in that, include: The geometric parameters to be corrected of the CT imaging system are obtained, and an objective function based on the quality of the reconstructed image is constructed; the geometric parameters to be corrected include at least the detector horizontal offset. A geometric correction method based on reconstructed image quality is adopted, combined with a conventional constraint optimization algorithm, to iteratively optimize the geometric parameters to be corrected and determine the initial optimized geometric parameters; Based on the initialized and optimized geometric parameters, multiple candidate horizontal offset values ​​to be verified are obtained; For each candidate horizontal offset value to be verified, CT image reconstruction is performed, and the objective function value corresponding to the reconstructed CT image is obtained according to the objective function based on the quality of the reconstructed image. Based on the objective function value corresponding to the reconstructed CT image, the candidate horizontal offset values ​​to be verified are filtered to determine the globally optimal horizontal offset correction parameters; Based on the aforementioned globally optimal horizontal offset correction parameters, the final globally optimal geometric correction parameters for the CT system are generated.

2. The global optimization method for CT geometric parameters according to claim 1, characterized in that, Construct an objective function based on the quality of the reconstructed image, including: An objective function based on the quality of reconstructed images is constructed by using one or more of the following evaluation metrics: gradient of the reconstructed image, high-frequency energy, and image sharpness. When there is more than one quality evaluation metric, the multiple quality evaluation metrics are fused to obtain the objective function based on the quality of the reconstructed image.

3. The global optimization method for CT geometric parameters according to claim 1, characterized in that, The initial optimized set of parameters includes at least the detector horizontal offset; Alternatively, the initial set of optimized parameters may include at least the detector horizontal offset, the distance from the X-ray source to the sample, and the detector rotation angle.

4. The global optimization method for CT geometric parameters according to claim 1, characterized in that, Based on the initialized and optimized geometric parameters, multiple candidate horizontal offset values ​​to be verified are obtained, including: Based on the pixel size of the CT system detector, multiple candidate values ​​for horizontal offset are generated within the neighborhood of the initial optimized geometric parameters corresponding to the horizontal offset of the detector.

5. The global optimization method for CT geometric parameters according to claim 4, characterized in that, The range of the candidate horizontal offset values ​​to be verified is [dx0-n×p, dx0+n×p]; where dx0 represents the initial optimized geometric parameters corresponding to the horizontal offset of the detector, n represents a positive integer, and p represents the pixel size of the CT system detector.

6. The global optimization method for CT geometric parameters according to claim 2, characterized in that, For each candidate horizontal offset value to be verified, CT image reconstruction is performed, and the objective function value corresponding to the reconstructed CT image is obtained according to the objective function based on the reconstructed image quality, including: For each candidate horizontal offset value to be verified, a conventional reconstruction algorithm is used to reconstruct the CT image and obtain the reconstructed CT image; the conventional reconstruction algorithm is set as either a filtered back projection algorithm or an algebraic iteration algorithm. The reconstructed CT image is evaluated using the objective function based on the quality of the reconstructed image to determine the quality evaluation index value and obtain the corresponding objective function value.

7. The global optimization method for CT geometric parameters according to claim 1, characterized in that, Based on the objective function value corresponding to the reconstructed CT image, the candidate horizontal offset values ​​to be verified are filtered to determine the globally optimal horizontal offset correction parameters, including: Based on the objective function value corresponding to the reconstructed CT image, the value with the largest objective function value among all candidate horizontal offset values ​​to be verified is taken as the globally optimal horizontal offset correction parameter.

8. The global optimization method for CT geometric parameters according to claim 7, characterized in that, Based on the aforementioned globally optimal horizontal offset correction parameters, the final globally optimal geometric correction parameters for the CT system are generated, including: If the geometric parameters to be corrected only include detector horizontal offset, then the globally optimal horizontal offset correction parameter is directly used as the final globally optimal geometric correction parameter of the CT system. When the geometric parameters to be corrected include detector horizontal offset and other parameters, the global optimal horizontal offset correction parameter corresponding to the detector horizontal offset and the initial optimized geometric parameters corresponding to the other parameters are used together as the final global optimal geometric correction parameters of the CT system.

9. A global optimization device for CT geometric parameters, characterized in that, include: An initialization module is used to acquire the geometric parameters to be corrected from the CT imaging system and to construct an objective function based on the quality of the reconstructed image; the geometric parameters to be corrected include at least the detector horizontal offset. The conventional iterative module is used to iteratively optimize the geometric parameters to be corrected by employing a geometric correction method based on the quality of the reconstructed image, combined with a conventional constraint optimization algorithm, to determine the initial optimized geometric parameters; The candidate value generation module is used to obtain multiple horizontal offset candidate values ​​to be verified based on the initial optimized geometric parameters. The reconstruction verification module is used to reconstruct the CT image for each candidate horizontal offset value to be verified, and to obtain the target function value corresponding to the reconstructed CT image according to the target function based on the quality of the reconstructed image. The parameter filtering module filters the candidate horizontal offset values ​​to be verified based on the objective function value corresponding to the reconstructed CT image, and determines the globally optimal horizontal offset correction parameter. The result output module is used to generate the final global optimal geometric correction parameters of the CT system based on the global optimal horizontal offset correction parameters.

Images