CBCT image quality enhancement method and system, storage medium and computer

Abstract

The invention provides a CBCT image quality enhancement method and system, a storage medium and a computer. The method comprises the following steps: randomly selecting one CBCT image data from CBCT image data as target image data; determining a deflection angle curve of each CBCT image data based on the deflection angle curve of the target image data; performing data processing on the deflection angle curve of each CBCT image data to obtain panoramic image data; and inputting the CBCT image data and the corresponding panoramic image data into an image processing model for data fusion to obtain enhanced image data. According to the method, the panorama data converted from the CBCT image data is used as the input of the network and the output of auxiliary diagnosis, so that artifacts and noise in the CBCT image are remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the field of image processing technology, and in particular to a CBCT image quality enhancement method, system, storage medium and computer. Background Art

[0002] Radiotherapy: is a method of using radiation to treat malignant tumors. Its principle is to destroy the DNA of cancer cells through high-energy radiation, thereby inhibiting the growth and division of cancer cells, and even causing the death of cancer cells. However, during the treatment process, the patient's body shape and tumor tissue may change significantly, resulting in a large deviation between the original planned dose and the actual dose, which may reduce the radiation dose received by the tumor and increase the damage to the surrounding normal tissues, thereby affecting the treatment effect and increasing side effects.

[0003] Adaptive Radiation Therapy (ART): is an advanced form of radiotherapy that can solve the above problems. It obtains real-time imaging information of patients during treatment and adjusts the treatment plan to adapt to changes in tumors and surrounding normal tissues. This technology can locate tumors more accurately, improve tumor control rates, reduce damage to surrounding normal tissues, and reduce the incidence of complications related to radiotherapy. However, traditional CT has problems with positioning, delineating the target area, and long radiotherapy time.

[0004] Cone Beam CT (CBCT): It provides strong support for accurate tumor positioning, formulation and adjustment of treatment plans. CBCT uses a cone-shaped X-ray beam to rotate around the patient, while the detector captures the X-rays passing through the patient. The data is processed by a computer to reconstruct a three-dimensional image of the patient. This technology can provide more detailed images than traditional CT, and is particularly suitable for small-scale imaging, such as oral and maxillofacial areas, cranial brain, etc.

[0005] However, the current CBCT technology still has the following problems: 1. CBCT images often have considerable artifacts, such as streaks, shadows, cupping, and scattered contamination, resulting in severe inaccuracies in Hounsfield unit (HU) values. Current ART practice uses deformed fan-beam CT (pCT) images for planning to match the anatomical structures of CBCT for dose calculation. However, due to the variations in anatomical structures and artifacts on CBCT images, the quality of image registration usually depends on the operator's experience and intuition, limiting the accuracy of ART implementation.

[0006] 2. Narrow scanning area: CBCT has the problem of narrow scanning area, which limits the observation range of the tumor target area and surrounding normal organs.

[0007] 3. Image quality evaluation standards: There is currently a lack of evaluation standards and quality control methods for CBCT image quality, which limits the design and evaluation of radiotherapy plans based on CBCT images. Summary of the invention

[0008] Based on this, the purpose of the present invention is to provide a CBCT image quality enhancement method, system, storage medium and computer to at least solve the deficiencies in the above-mentioned technology.

[0009] The present invention provides a CBCT image quality enhancement method, comprising: Acquire a plurality of CBCT image data, and randomly select one CBCT image data from each of the CBCT image data as target image data; Identifying a deflection curve of the target image data, and determining a deflection curve of each of the CBCT image data based on the deflection curve of the target image data; Performing data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data; An image processing model is constructed, and each of the CBCT image data and its corresponding panoramic image data are input into the image processing model for data fusion to obtain corresponding enhanced image data.

[0010] Furthermore, the step of performing data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data includes: Performing equidistant sampling on the deflection curve of each of the CBCT image data to obtain a plurality of sampling points of each of the CBCT image data; Determine the sampling points with the same position in each of the CBCT image data as a group of sampling points, obtain a sampling matrix corresponding to each group of sampling points, and perform weighted forward projection processing on each of the sampling matrices to obtain a plurality of weighted matrices; The weighted matrices are spliced, and the panoramic image data corresponding to the CBCT image data are obtained based on the splicing result.

[0011] Furthermore, the steps of constructing the image processing model include: Constructing a generative adversarial network model, and reconstructing the backbone network structure of the generative adversarial network model to obtain a reconstructed generative adversarial network model; A diffusion model is constructed, and the diffusion model is fused with the reconstructed generative adversarial network model to obtain the image processing model.

[0012] Furthermore, the step of inputting each of the CBCT image data and its corresponding panoramic image data into the image processing model for data fusion to obtain corresponding enhanced image data includes: Performing a convolution and pooling operation on each of the CBCT image data and the corresponding panoramic image data to obtain a corresponding convolution and pooling result; The features extracted from the convolution pooling result are used as the A field, and the CT scan with the same Hu value domain is used as the B field, and the image processing model is trained according to the mapping between the A field and the B field to obtain corresponding enhanced image data.

[0013] The present invention also provides a CBCT image quality enhancement system, comprising: A data selection module, used for acquiring a plurality of CBCT image data, and randomly selecting a CBCT image data from each of the CBCT image data as target image data; A curve identification module, used to identify the deflection curve of the target image data, and determine the deflection curve of each of the CBCT image data based on the deflection curve of the target image data; A data processing module, used for performing data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data; The data fusion module is used to construct an image processing model, and input each of the CBCT image data and its corresponding panoramic image data into the image processing model for data fusion to obtain corresponding enhanced image data.

[0014] Furthermore, the data processing module includes: An equidistant sampling unit, used for performing equidistant sampling on the deflection curve of each of the CBCT image data to obtain a plurality of sampling points of each of the CBCT image data; A projection processing unit, used to determine the sampling points with the same position in each of the CBCT image data as a group of sampling points, obtain a sampling matrix corresponding to each group of sampling points, and perform weighted forward projection processing on each of the sampling matrices to obtain a plurality of weighted matrices; The matrix stitching unit is used to stitch the weighted matrices together and obtain the panoramic image data corresponding to the CBCT image data based on the stitching result.

[0015] Furthermore, the data fusion module includes: A model reconstruction unit, used to construct a generative adversarial network model and reconstruct the backbone network structure of the generative adversarial network model to obtain a reconstructed generative adversarial network model; The model fusion unit is used to construct a diffusion model and fuse the diffusion model with the reconstructed generative adversarial network model to obtain the image processing model.

[0016] Furthermore, the data fusion module also includes: A convolution pooling unit, used for performing a convolution pooling operation on each of the CBCT image data and the corresponding panoramic image data to obtain a corresponding convolution pooling result; An image processing unit is used to use the features extracted from the convolution pooling result as the A field and the CT scan with the same Hu value domain as the B field, and train the image processing model according to the mapping between the A field and the B field to obtain corresponding enhanced image data.

[0017] The present invention also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the above-mentioned CBCT image quality enhancement method is implemented.

[0018] The present invention also proposes a computer, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned CBCT image quality enhancement method when executing the computer program.

[0019] The CBCT image quality enhancement method, system, storage medium and computer of the present invention process a number of CBCT image data to obtain corresponding panoramic image data, use the panoramic image as the input of the network and the output of the auxiliary diagnosis, so as to increase the ability of the network to extract information, greatly reduce the dependence of the network on image registration, and expand the ability to extract global information of the scan; and use the constructed image processing model to fuse the panoramic image data to obtain synthetic CT image data, optimize the synthetic CT image, and use the quality evaluation function to score the optimized synthetic CT image to obtain enhanced image data, significantly reduce artifacts and noise in the CBCT image, and improve the accuracy of the Hounsfield unit (HU) value, thereby providing clearer image data for radiotherapy. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 is a flow chart of a CBCT image quality enhancement method in a first embodiment of the present invention; Figure 2 for Figure 1 Detailed flow chart of step S103; Figure 3 for Figure 1 Detailed flow chart of step S104; Figure 4is a structural block diagram of a CBCT image quality enhancement system in a second embodiment of the present invention; Figure 5 FIG. 4 is a structural block diagram of a computer in a third embodiment of the present invention.

[0021] The following specific implementation manner will further illustrate the present invention in conjunction with the above-mentioned drawings. DETAILED DESCRIPTION

[0022] In order to facilitate the understanding of the present invention, the present invention will be described more fully below with reference to the relevant drawings. Several embodiments of the present invention are given in the drawings. However, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. On the contrary, the purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. The terms used herein in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0024] Embodiment 1 See also Figure 1 , which shows a CBCT image quality enhancement method in a first embodiment of the present invention, and the method specifically includes steps S101 to S104: S101, acquiring a plurality of CBCT image data, and randomly selecting a CBCT image data from each of the CBCT image data as target image data; In a specific implementation, a plurality of continuous CBCT image data are acquired, and one CBCT image data is randomly selected from the continuous CBCT image data as the target image data. In this embodiment, the CBCT image data is a two-dimensional image, and the transmitted image data contains DICOM data.

[0025] S102, identifying a deflection curve of the target image data, and determining a deflection curve of each of the CBCT image data based on the deflection curve of the target image data; In specific implementation, the calculation process of the deflection curve is as follows: 1. Edge detection: Use edge detection algorithms, such as the Canny edge detector, to identify edges in the image.

[0026] 2. Contour tracking: After edge detection, the contour tracking algorithm can be applied to track the contour of the object.

[0027] 3. Angle calculation: For each contour point, calculate the angle between it and the reference line, which can be achieved by calculating the angle of the tangent.

[0028] Curve fitting: Connect the calculated angle points to form a deflection curve. Polynomial fitting or other curve fitting techniques can be used to smooth the curve.

[0029] S103, performing data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data; For further information, see Figure 2 , the step S103 specifically includes steps S1031 to S1033: S1031, performing equidistant sampling on the deflection curve of each of the CBCT image data to obtain a plurality of sampling points of each of the CBCT image data; S1032, determining the sampling points with the same position in each of the CBCT image data as a group of sampling points, and obtaining a sampling matrix corresponding to each group of sampling points, and performing weighted forward projection processing on each of the sampling matrices to obtain a plurality of weighted matrices; S1033: stitching the weighted matrices together, and obtaining panoramic image data corresponding to the CBCT image data based on the stitching result.

[0030] In a specific implementation, equidistant sampling is performed on the deflection curve of each CBCT image data to obtain multiple sampling points on each CBCT image data, multiple sampling points with the same position in each CBCT image data are determined as a group of sampling points, and a sampling matrix corresponding to each group of sampling points is obtained. A weighted forward projection operation is performed on each sampling matrix to obtain multiple weighted matrices, and weights are assigned to each matrix as an arrangement order according to the position code extracted from the feature, and each weighted matrix is ​​spliced ​​in the arrangement order, and panoramic image data of the CBCT image data is obtained according to the splicing result.

[0031] It can be understood that the panoramic image data converted from CBCT image data is used as the input of the network and the output of auxiliary diagnosis. At the same time, it solves the problem that CBCT images often have considerable artifacts, such as stripes, shadows, cupping and scattered contamination, which lead to serious inaccuracy in the Hounsfield unit (HU) value and the narrow scanning area of ​​CBCT images. The introduction of panoramic image data not only increases the network's ability to extract information, but also greatly reduces the network's dependence on image registration, and expands the ability to extract global scanning information.

[0032] S104, constructing an image processing model, and inputting each of the CBCT image data and its corresponding panoramic image data into the image processing model for data fusion to obtain corresponding enhanced image data.

[0033] For further information, see Figure 3 , the step S104 specifically includes steps S1041-S1042: S1041, constructing a generative adversarial network model, and reconstructing a backbone network structure of the generative adversarial network model to obtain a reconstructed generative adversarial network model; S1042, constructing a diffusion model, and fusing the diffusion model with the reconstructed generative adversarial network model to obtain the image processing model.

[0034] In the specific implementation, a generative adversarial network model is constructed. In this embodiment, the generative adversarial network model is a CycleGAN network model. The backbone network structure of the CycleGAN network model is reconstructed, and the pre-constructed diffusion model is fused. A fusion network is used, and the inputs are added and multiplied as inputs. The results of the two models are added and multiplied as a standard for training, thereby obtaining an image processing model. It can be understood that the CycleGAN network model has the characteristics of unmatched pair training. By improving the network structure of CycleGAN, the stability of training can be improved.

[0035] Specifically, the step S104 further includes steps S1141-S1142: S1141, performing a convolution pooling operation on each of the CBCT image data and the corresponding panoramic image data to obtain a corresponding convolution pooling result; S1142, using the features extracted from the convolution pooling result as the A domain, and the CT scan with the same Hu value range as the B domain, and training the image processing model according to the mapping between the A domain and the B domain to obtain corresponding enhanced image data.

[0036] In this embodiment, the training of the image processing model uses the features extracted after convolution pooling of the original CBCT image and the generated panoramic image as the A field, and the CT scan with the same hu value range as the B field, learns the mapping from the A field to the B field, and uses the trained generator to infer the available simulated CT image through the original CBCT and the panoramic image. In this embodiment, the ResCNN convolutional network is used to embed the CBCT data, the matrix after image preprocessing is used as input, and the panoramic image is used as output. The Unet structure is used for up and down sampling, with the purpose of training a network that can do CBCT to panoramic CBCT, and then the CBCT and panoramic CBCT are input into CycleGAN as input to train higher quality output.

[0037] It can be understood that the CycleGAN network model of the fusion diffusion model is used to fuse CBCT and panoramic images, simulate CT field features to generate synthetic CT (sCT), and solve several key problems existing in the current cone beam computed tomography (CBCT) technology in the field of radiotherapy. Through advanced image processing technology, the artifacts and noise in CBCT images are significantly reduced, and the accuracy of the Hounsfield unit (HU) value is improved, thereby providing clearer image data for radiotherapy.

[0038] Furthermore, the generated enhanced image data are optimized by metal artifact removal and super-resolution, and the processed sCT images are scored by a quality assessment function and compared with those before processing to provide a comprehensive evaluation of image quality.

[0039] In this embodiment, the evaluation index is a general index for judging whether all synthetic CT images are like CT images. The specific image quality estimation function is: a (PSNR) +b (SSIM) +c(MSE); Among them, abc are all obtained by neural network fitting. A linear fitting curve is constructed, and the function is obtained by training using the input and the output of known standard values.

[0040] It can be understood that this embodiment provides a more effective CBCT image quality evaluation standard, which comprehensively considers information such as the hu value and similarity of the output sCT, comprehensively evaluates the image quality enhancement effect, and formulates a comprehensive CBCT image quality evaluation system to provide a scientific basis for the design and evaluation of radiotherapy plans, ensure the quality and safety of treatment plans, and achieve the purpose of improving the accuracy and efficiency of radiotherapy, reducing patients' treatment time and discomfort, and improving patients' overall quality of life.

[0041] In summary, the CBCT image quality enhancement method in the above-mentioned embodiment of the present invention processes a number of CBCT image data to obtain corresponding panoramic image data, uses the panoramic image as the input of the network and the output of the auxiliary diagnosis, thereby increasing the ability of the network to extract information, greatly reducing the dependence of the network on image registration, and expanding the ability to extract global scanning information; and uses the constructed image processing model to perform data fusion on the panoramic image data to obtain synthetic CT image data, optimizes the synthetic CT image, and uses the quality evaluation function to score the optimized synthetic CT image to obtain enhanced image data, significantly reducing artifacts and noise in the CBCT image, and improving the accuracy of the Hounsfield unit (HU) value, thereby providing clearer image data for radiotherapy.

[0042] Embodiment 2 Another aspect of the present invention is to provide a CBCT image quality enhancement system. Figure 4 , shown is a CBCT image quality enhancement system in a second embodiment of the present invention, the system comprising: A data selection module 11 is used to obtain a plurality of CBCT image data and randomly select a CBCT image data from each of the CBCT image data as target image data; A curve identification module 12, used to identify the deflection curve of the target image data, and determine the deflection curve of each of the CBCT image data based on the deflection curve of the target image data; A data processing module 13 is used to perform data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data; Furthermore, the data processing module 13 includes: An equidistant sampling unit, used for performing equidistant sampling on the deflection curve of each of the CBCT image data to obtain a plurality of sampling points of each of the CBCT image data; A projection processing unit, used to determine the sampling points with the same position in each of the CBCT image data as a group of sampling points, obtain a sampling matrix corresponding to each group of sampling points, and perform weighted forward projection processing on each of the sampling matrices to obtain a plurality of weighted matrices; The matrix stitching unit is used to stitch the weighted matrices together and obtain the panoramic image data corresponding to the CBCT image data based on the stitching result.

[0043] The data fusion module 14 is used to construct an image processing model, and input each of the CBCT image data and its corresponding panoramic image data into the image processing model for data fusion to obtain corresponding enhanced image data.

[0044] Furthermore, the data fusion module 14 includes: A model reconstruction unit, used to construct a generative adversarial network model and reconstruct the backbone network structure of the generative adversarial network model to obtain a reconstructed generative adversarial network model; The model fusion unit is used to construct a diffusion model and fuse the diffusion model with the reconstructed generative adversarial network model to obtain the image processing model.

[0045] Furthermore, the data fusion module 14 also includes: A convolution pooling unit, used for performing a convolution pooling operation on each of the CBCT image data and the corresponding panoramic image data to obtain a corresponding convolution pooling result; An image processing unit is used to use the features extracted from the convolution pooling result as the A field and the CT scan with the same Hu value domain as the B field, and train the image processing model according to the mapping between the A field and the B field to obtain corresponding enhanced image data.

[0046] The functions or operation steps implemented when the above modules and units are executed are generally the same as those in the above method embodiments, and will not be repeated here.

[0047] The CBCT image quality enhancement system provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference may be made to the corresponding contents in the aforementioned method embodiment.

[0048] Embodiment 3 The present invention also provides a computer, see Figure 5 , shown is a computer in the third embodiment of the present invention, including a memory 10, a processor 20, and a computer program 30 stored in the memory 10 and executable on the processor 20. When the processor 20 executes the computer program 30, the above-mentioned CBCT image quality enhancement method is implemented.

[0049] The memory 10 includes at least one type of storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory, etc.), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 10 may be an internal storage unit of a computer, such as a hard disk of the computer. In other embodiments, the memory 10 may also be an external storage device, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card, etc. Further, the memory 10 may also include both an internal storage unit of the computer and an external storage device. The memory 10 may be used not only to store application software and various types of data installed in the computer, but also to temporarily store data that has been output or is to be output.

[0050] Among them, in some embodiments, the processor 20 can be an electronic control unit (Electronic Control Unit, abbreviated as ECU, also known as a vehicle computer), a central processing unit (Central Processing Unit, CPU), a controller, a microcontroller, a microprocessor or other data processing chip, used to run the program code stored in the memory 10 or process data, such as executing access restriction programs, etc.

[0051] It should be pointed out that Figure 5 The structure shown does not constitute a limitation on the computer. In other embodiments, the computer may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0052] The embodiment of the present invention further provides a storage medium on which a computer program is stored. When the program is executed by a processor, the CBCT image quality enhancement method as described above is implemented.

[0053] Those skilled in the art will appreciate that the logic and / or steps represented in the flowchart or otherwise described herein, for example, may be considered as an ordered list of executable instructions for implementing logical functions, and may be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For purposes of this specification, "computer-readable medium" may be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0054] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0055] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or a combination thereof: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0056] The technical features of the above-described embodiments may be arbitrarily combined. To make the description concise, not all possible combinations of the technical features in the above-described embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0057] The above-mentioned embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the attached claims.

Claims

1. A CBCT image quality enhancement method, characterized in that: include: Acquire a plurality of CBCT image data, and randomly select one CBCT image data from each of the CBCT image data as target image data; Identifying a deflection curve of the target image data, and determining a deflection curve of each of the CBCT image data based on the deflection curve of the target image data; Performing data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data; An image processing model is constructed, and each of the CBCT image data and its corresponding panoramic image data are input into the image processing model for data fusion to obtain corresponding enhanced image data.

2. The CBCT image quality enhancement method according to claim 1, characterized in that: The step of performing data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data comprises: Performing equidistant sampling on the deflection curve of each of the CBCT image data to obtain a plurality of sampling points of each of the CBCT image data; Determine the sampling points with the same position in each of the CBCT image data as a group of sampling points, obtain a sampling matrix corresponding to each group of sampling points, and perform weighted forward projection processing on each of the sampling matrices to obtain a plurality of weighted matrices; The weighted matrices are spliced, and the panoramic image data corresponding to the CBCT image data are obtained based on the splicing result.

3. The CBCT image quality enhancement method according to claim 1, characterized in that: The steps to build an image processing model include: Constructing a generative adversarial network model, and reconstructing the backbone network structure of the generative adversarial network model to obtain a reconstructed generative adversarial network model; A diffusion model is constructed, and the diffusion model is fused with the reconstructed generative adversarial network model to obtain the image processing model.

4. The CBCT image quality enhancement method according to claim 1, characterized in that: The step of inputting each of the CBCT image data and the corresponding panoramic image data into the image processing model for data fusion to obtain the corresponding enhanced image data comprises: Performing a convolution and pooling operation on each of the CBCT image data and the corresponding panoramic image data to obtain a corresponding convolution and pooling result; The features extracted from the convolution pooling result are used as the A field, and the CT scan with the same Hu value domain is used as the B field, and the image processing model is trained according to the mapping between the A field and the B field to obtain corresponding enhanced image data.

5. A CBCT image quality enhancement system, characterized in that: include: A data selection module, used for acquiring a plurality of CBCT image data, and randomly selecting a CBCT image data from each of the CBCT image data as target image data; A curve identification module, used to identify the deflection curve of the target image data, and determine the deflection curve of each of the CBCT image data based on the deflection curve of the target image data; A data processing module, used for performing data processing on the deflection curves of each of the CBCT image data to obtain panoramic image data corresponding to each of the CBCT image data; The data fusion module is used to construct an image processing model, and input each of the CBCT image data and its corresponding panoramic image data into the image processing model for data fusion to obtain corresponding enhanced image data.

6. The CBCT image quality enhancement system according to claim 5, characterized in that: The data processing module comprises: An equidistant sampling unit, used for performing equidistant sampling on the deflection curve of each of the CBCT image data to obtain a plurality of sampling points of each of the CBCT image data; A projection processing unit, used to determine the sampling points with the same position in each of the CBCT image data as a group of sampling points, obtain a sampling matrix corresponding to each group of sampling points, and perform weighted forward projection processing on each of the sampling matrices to obtain a plurality of weighted matrices; The matrix stitching unit is used to stitch the weighted matrices together and obtain the panoramic image data corresponding to the CBCT image data based on the stitching result.

7. The CBCT image quality enhancement system according to claim 5, characterized in that: The data fusion module comprises: A model reconstruction unit, used to construct a generative adversarial network model and reconstruct the backbone network structure of the generative adversarial network model to obtain a reconstructed generative adversarial network model; The model fusion unit is used to construct a diffusion model and fuse the diffusion model with the reconstructed generative adversarial network model to obtain the image processing model.

8. The CBCT image quality enhancement system according to claim 5, characterized in that: The data fusion module also includes: A convolution pooling unit, used for performing a convolution pooling operation on each of the CBCT image data and the corresponding panoramic image data to obtain a corresponding convolution pooling result; An image processing unit is used to use the features extracted from the convolution pooling result as the A field and the CT scan with the same Hu value domain as the B field, and train the image processing model according to the mapping between the A field and the B field to obtain corresponding enhanced image data.

9. A storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the CBCT image quality enhancement method according to any one of claims 1 to 4 is implemented.

10. A computer comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the CBCT image quality enhancement method according to any one of claims 1 to 4 is implemented.

Images