Scattering correction method and scattering correction method system for static ct

Abstract

Disclosed are a scattering correction method and a scattering correction system for static CT. The scattering correction method comprises: acquiring scattering data in advance by means of a scattering detector arranged on at least one side of a main detector to obtain a scattering correction table, and obtaining a pre-training model by means of model training, wherein the scattering correction table comprises scattering distribution corresponding to each bed position and angle; obtaining an original image of a user; pre-correcting the original image to obtain a first image; eccentrically cutting the first image to obtain a second image; determining whether a CT scanning mode of the user is spiral scanning or not; if the result is yes, carrying out scattering correction on the second image by using the scattering correction table; and if the result is no, obtaining, on the basis of the pre-training model, scattering distribution corresponding to the second image, and carrying out scattering correction on the second image. According to the scattering correction method, interpolation is used to obtain the scattering correction table, enabling precise scattering correction and improving the scattering correction effect.

Description

A scatter correction method and scatter correction system for static CT Technical Field

[0001] The present invention relates to a scatter correction method for static CT and also to a corresponding scatter correction system, belonging to the technical field of radiation imaging. Background Art

[0002] Currently, there are many methods for CT scatter correction, such as the BSA (beam stop array) method and the SKS (scatter kernel convolution) method. The BSA method uses a designed two-dimensional grid of ray blockers to measure scattering, then interpolates the scatter distribution for scatter correction. This method requires two imaging sessions, with and without BSA, which increases scan time, adds additional dose, and may increase errors caused by object motion. The SKS method does not require any hardware adjustment or additional scans, and can provide a relatively good estimate of scattering. However, this method is complex to adjust parameters and has certain instabilities. The estimation accuracy is poor for scattering from complex objects and cross-scattering from multi-source scanning.

[0003] However, due to the diversity and complexity of scatter suppression, a well-defined standard method currently exists for static CT. Because static CT employs a dual-ring structure, multiple sources are imaged on the same detector, resulting in significant variations in radiation angles. This makes it impossible to employ the anti-scatter grids commonly used in traditional spiral CT to suppress scatter. Consequently, designing an appropriate anti-scatter grid for static CT is currently difficult, resulting in a large amount of scattered radiation being received by the detector. This is particularly true in multi-source exposure modes, where cross-scattering is extremely severe, leading to severe scatter artifacts and significantly degrading image quality. Summary of the Invention

[0004] The primary technical problem to be solved by the present invention is to provide a scatter correction method for static CT.

[0005] Another technical problem to be solved by the present invention is to provide a scatter correction system for static CT.

[0006] In order to achieve the above technical objectives, the present invention adopts the following technical solutions:

[0007] According to a first aspect of an embodiment of the present invention, a scatter correction method for static CT is provided, comprising the following steps:

[0008] Scattering data is collected in advance by a scattering detector provided on at least one side of the main detector to obtain a scattering correction table by interpolation, and the scattering data is used to perform model training to obtain a pre-trained model; wherein the scattering correction table includes a scattering distribution corresponding to each bed position and angle;

[0009] Obtain the original image of the static CT scan performed by the user;

[0010] Pre-correcting the original image to obtain a first image;

[0011] Obtain a second image by performing an eccentric cut on the first image;

[0012] Determining whether the CT scanning mode of the user is a spiral scan;

[0013] If the judgment result is yes, the scatter correction table is used to perform scatter correction on the second image; if the judgment result is no, the scatter distribution corresponding to the second image is obtained based on the pre-trained model, and scatter correction is performed on the second image.

[0014] Preferably, the scatter correction table is obtained by:

[0015] Collecting a scattering image based on a scattering detector disposed on at least one side of the main detector and performing data preprocessing;

[0016] Based on the pre-processed data, the scattering data is extracted and placed into an array, and corresponds to the bed position and angle; wherein the array includes data information for each bed position and angle;

[0017] Based on the scattering data, extract all scattering values ​​of each pixel and the bed position and angle corresponding to each pixel;

[0018] In the bed and angle plane, the scattering value of each grid position in the plane is obtained by interpolation according to the set bed and angle grids;

[0019] Arrange the interpolated scattering values ​​according to the detection direction of the main detector and perform smoothing;

[0020] The scatter distributions of all main detectors in the detection direction and the corresponding bed positions and angles are stored as the scatter correction table.

[0021] Preferably, performing scatter correction based on the scatter correction table specifically includes:

[0022] inputting the second image;

[0023] extracting bed position and angle information of the second image;

[0024] For each row of main detectors, the detection angle is converted to 0-360°, and calculation is performed based on the bed position and row number of the second image to obtain the bed position corresponding to the row of main detectors;

[0025] Finding the scattering value corresponding to each row of detectors in the scatter correction table according to the bed position and angle information corresponding to the row of detectors to obtain a scattering distribution corresponding to the second image;

[0026] performing an antilogarithmic transformation on the second image and obtaining the inverse thereof to obtain an image that has not been scatter-corrected;

[0027] subtracting the scattering distribution corresponding to the second image from the obtained image not subjected to scattering correction to obtain a scattering-corrected image;

[0028] A negative logarithm operation is performed on the scatter-corrected image, thereby completing the scatter correction of the second image.

[0029] Preferably, obtaining the scattering distribution corresponding to the second image based on the pre-trained model specifically includes:

[0030] Acquiring scattering data collected by the scattering detector and projection data corresponding to the second image;

[0031] Inputting the scattering data and the projection image data into a pre-trained model;

[0032] Output a scattering distribution corresponding to the second image based on the pre-trained model.

[0033] Preferably, the interpolation method includes at least cubic interpolation or biharmonic spline interpolation, and CUDA acceleration is performed during the interpolation process.

[0034] Preferably, the pre-correction of the original image specifically includes:

[0035] Performing background correction on the original image to obtain a background corrected image;

[0036] Performing gain correction on the background-corrected image to obtain a gain-corrected image;

[0037] Air correction is performed on the gain-corrected image to obtain a first image.

[0038] Preferably, in the axial scanning state, if scattering detectors are provided on both sides of the main detector, the scattering values ​​measured by the scattering detectors on both sides and the low-frequency characteristics of the scattering distribution are used to obtain the scattering value of each pixel on the main detector by using the surface interpolation method for scattering subtraction;

[0039] If only one side of the main detector is equipped with a scattering detector, the scattering kernel convolution method or AI method is used to estimate the scattering value of each pixel on the main detector for scattering subtraction.

[0040] Preferably, the arrangement interval of the scattering detectors is greater than the arrangement interval of the main detectors; wherein, the larger the arrangement interval, the sparser the arrangement, and the smaller the arrangement interval, the denser the arrangement.

[0041] Preferably, the scattering detector collecting scattering data at least includes:

[0042] Scattering data collected under single-ray source exposure; or, scattering data collected under multiple-ray source alternating exposure; or, scattering data collected under multiple-ray source simultaneous exposure.

[0043] According to a second aspect of an embodiment of the present invention, a scatter correction system for static CT is provided, comprising a processor and a memory, wherein the processor reads a computer program in the memory to perform the following operations:

[0044] Scattering data is collected in advance by a scattering detector provided on at least one side of the main detector to obtain a scattering correction table by interpolation, and the scattering data is used to perform model training to obtain a pre-trained model; wherein the scattering correction table includes a scattering distribution corresponding to each bed position and angle;

[0045] Obtain the original image of the static CT scan performed by the user;

[0046] Pre-correcting the original image to obtain a first image;

[0047] Obtain a second image by performing an eccentric cut on the first image;

[0048] Determining whether the CT scanning mode of the user is a spiral scan;

[0049] If the judgment result is yes, the scatter correction table is used to perform scatter correction on the second image; if the judgment result is no, the scatter distribution corresponding to the second image is obtained based on the pre-trained model, and scatter correction is performed on the second image.

[0050] Compared with the prior art, the present invention has the following technical effects:

[0051] 1. Using a scatter detector to measure scattering in real time during CT scanning, and leveraging the characteristics of static CT, the scattering distribution of all positions and angles of the scanned object can be obtained very accurately through interpolation, allowing for precise scattering correction and improving the effect of scattering correction.

[0052] 2. This scatter correction method is easy to implement and has low algorithm complexity.

[0053] 3. Can be widely used in single-source and multi-source exposure modes. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] FIG1A is a schematic diagram of a scattering detection principle according to an embodiment of the present invention;

[0055] FIG1B is a schematic diagram of an arrangement of a main detector and a scattering detector provided by an embodiment of the present invention;

[0056] FIG1C is a schematic diagram of another arrangement of a main detector and a scattering detector provided by an embodiment of the present invention;

[0057] FIG1D is a schematic diagram of another arrangement of a main detector and a scattering detector provided by an embodiment of the present invention;

[0058] FIG2 is a flow chart of obtaining a scatter correction table according to an embodiment of the present invention;

[0059] FIG3 is a schematic diagram of the entire curved surface obtained by interpolation based on measured values ​​in an embodiment of the present invention;

[0060] FIG4 is a flow chart of a scatter correction method for static CT provided by an embodiment of the present invention;

[0061] FIG5 is a flow chart of performing scatter correction based on a scatter correction table in an embodiment of the present invention;

[0062] FIG6 is a structural diagram of a scatter correction system for static CT provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The technical content of the present invention is described in detail below with reference to the accompanying drawings and specific embodiments.

[0064] In an embodiment of the present invention, the scattered component of the projection image is measured by arranging a scatter detector on the main detector. As shown in Figure 1A, the detector is divided into a main detector 1 and a scatter detector 2, with the main detector 1 located in the middle of the detector and the scatter detectors 2 located on either side of the main detector 1. In actual applications, the outgoing beam of X-ray source 10 is constrained to a pre-designed field of view (FOV) by a beam limiting device 20. In this case, scatter detector 2 is located outside the FOV and is not covered by the main beam. The signal it measures is the X-ray scatter signal generated by the X-rays passing through the object being measured.

[0065] It will be appreciated that, in this embodiment, the scatter detector 2 and the main detector 1 are independent of their specific implementation. For example, the scatter detector 2 can be a separate, independently controlled detection unit from the main detector 1, or an independent photodiode (PD) array; or a detector pixel array (including but not limited to PD+DMS, CMOS+PD, TFT PD, IGZO, photon flux detector, CZT, photon counting detector, etc.). Furthermore, the scatter detector 2 can be provided only on one side of the main detector 1.

[0066] In this embodiment, the pixel array of main detector 1 is uniformly arranged. The pixel array of main detector 1 is represented as I(R*M), i.e., R rows and M columns. The scatter detector 2 can be arranged in the ZX plane in the form of a sparse matrix. The scatter matrix is ​​represented as S(r*m), i.e., r rows and m columns. When m=M, the number of columns of main detector 1 and scatter detector 2 is the same.

[0067] As shown in Figures 1B to 1D, a partial view of a detector unit is used as an example. In Figure 1B, only one scattering detector 2 is set on one side of the main detector 1, and the pixel arrays of the two scattering detectors 2 are sparsely and symmetrically distributed. In Figure 1C, only one scattering detector 2 is set on one side of the main detector 1, and the pixel arrays of the scattering detector 2 are sparsely and asymmetrically distributed. In Figure 1D, a scattering detector 2 is set on both sides of the main detector 1, and the pixel arrays of the two scattering detectors 2 are sparsely and symmetrically distributed, and the pixel arrays of the two scattering detectors 2 on both sides are symmetrical about the center of the main detector 1.

[0068] It can be understood that the above Figures 1B to 1D only show part of the arrangement of the main detector 1 and the scattering detector 2. In other embodiments, the arrangement of the main detector 1 and the scattering detector 2 can be adaptively adjusted as needed, as long as the scattering data can be collected for scattering correction.

[0069] During a static CT scan, scatter detector 2 can collect scatter data in real time. This data can be pre-collected by the scatter detector, and a scatter correction table can be obtained by interpolation. This scatter data can then be used to train a model to obtain a pre-trained model. In practical applications, the patient's original image can be scatter-corrected based on this scatter correction table and the pre-trained model to obtain high-quality detection images.

[0070] As shown in FIG2 , in one embodiment of the present invention, the scatter correction table is obtained through steps S10 to S60:

[0071] First, a scattering image is collected based on a scattering detector arranged on at least one side of the main detector, and data preprocessing is performed;

[0072] The scatter data collected by the scatter detector includes at least: scatter data collected during single-source exposure; scatter data collected during alternating exposure using multiple ray sources; or scatter data collected during simultaneous exposure using multiple ray sources. In other words, the scatter correction method of this embodiment is applicable to both single-source and multi-source exposure modes.

[0073] Here, the scattering detector collects the scattering image and the preprocessing of the scattering image is a prior art and thus will not be described in detail. In an ideal state, no preprocessing is required.

[0074] Based on the preprocessed data, the scattered data is extracted and placed into an array, corresponding to the bed position and angle; wherein the array includes data information for each bed position and angle, specifically including the following steps:

[0075] S30: Extracting the scattering value corresponding to each pixel and the bed position and angle corresponding to each pixel.

[0076] According to the bed control unit in the CT machine, the current bed position information is obtained. The bed position and angle are the most basic information of CT scanning. Each time an image is acquired, the bed position and angle of the data acquisition will be stored. The bed position is the position of the bed, and the angle is the angle information.

[0077] The object to be measured is a water phantom or a body phantom, which is prior art and will not be described in detail here.

[0078] The scatter value detected by each pixel in the scatter detector is stored in correspondence with the bed position and angle.

[0079] S40: In the bed position and angle plane, according to the set bed position and angle grid, obtain the scattering value of each grid position in the plane by interpolation;

[0080] 3 , the x coordinate is the angle; the y coordinate is the bed position; the z coordinate is the scattering value; the gray circled points are the measured scattering values, and the entire gray surface is the entire surface obtained by interpolation based on the measured values.

[0081] The interpolation method includes at least cubic interpolation or biharmonic spline interpolation, and CUDA (Compute Unified Device Architecture) acceleration is performed during the interpolation process.

[0082] S50: Arrange the interpolated scattering values ​​according to the detection direction of the main detector and perform smoothing processing;

[0083] The detection direction of the main detector is the x-direction of the main detector array. In this case, only the interpolated scattering value is used, and the detected scattering value is not considered. This is because if the (x, y) coordinates coincide, the interpolated scattering value and the detected scattering value are exactly the same.

[0084] S60: The scatter distributions of all main detectors in the detection direction and the corresponding bed positions and angles are stored as a scatter correction table.

[0085] Specifically, the scattering distributions of multiple main detectors are summed up and stored to form a scattering correction table.

[0086] On this basis, FIG4 shows a flowchart of a scatter correction method for static CT provided by an embodiment of the present invention. In actual use, the patient is on the bed, the CT machine is started to obtain the original image, and then the scatter correction method provided by the embodiment of the present invention is used. Specifically, steps S1 to S5 are as follows:

[0087] S1: Pre-correct the original image to obtain a first image.

[0088] Specifically, it includes steps S11 to S13:

[0089] S11: performing background correction on the original image to obtain a background corrected image;

[0090] S12: performing gain correction on the background correction image to obtain a gain correction image;

[0091] S13: Perform air correction on the gain-corrected image to obtain a first image.

[0092] The air data is a set of data obtained by performing a series of scans without any objects placed within the scanning range. The actual scan data is then divided by a reference value obtained by scanning only the air, and the negative logarithm (-log) of the reference value is taken. Specifically, in one embodiment of the present invention, the air correction process includes: dividing the scattering data from the scattering detector by the air value to obtain the scattering percentage, removing the air from the main detector, and performing a negative logarithm operation to obtain the first image. Background correction and gain correction are both common knowledge in the art and will not be elaborated on here.

[0093] It should be noted that, unless otherwise specified, the original image, first image, second image, etc. in the present invention all refer to images obtained by detection using the main detector.

[0094] S2: Cut the first image according to the eccentricity to obtain the second image.

[0095] In this embodiment, since the static CT system uses a full-ring detector, the detector illuminated by each radiation source is fixed. Therefore, the eccentricity data obtained from calibration determines the range of the cut image, thereby obtaining the projection image corresponding to the source, i.e., the second image. This is prior art and will not be further described.

[0096] S3: Determine whether the CT scanning mode is spiral scanning.

[0097] Specifically, based on the parameters collected during the static CT scan, the parameter configuration is read to determine whether the scan is a spiral scan. The process of determining whether the scan is a spiral scan based on the system parameter configuration is common knowledge in the art and will not be elaborated on here.

[0098] S4: If the judgment result is yes, perform scatter correction on the second image based on a scatter correction table; wherein the scatter correction table includes a scatter distribution corresponding to each bed position and angle.

[0099] Referring to FIG5 , this step specifically includes steps S41 to S47:

[0100] S41: inputting a second image (i.e., a projection image, the same below);

[0101] S42: Extracting bed position and angle information corresponding to the second image;

[0102] S43: For each row of main detectors, convert the detection angle to 0-360°, and perform calculations based on the bed position and row number in the second image to obtain the bed position corresponding to the row of main detectors;

[0103] For example, if the projection angle of a certain acquisition is 10°, the bed is 20mm, and the bed corresponding to each row of detectors in the z direction of the main detector is 0.165mm, then the bed corresponding to the 30th row of detectors is: 20mm+30*0.165mm, and the angle is still 10°

[0104] It is understandable that in a static CT scanning system, the main detector is a detector ring, which is composed of multiple main detector modules. A row of main detectors here refers to multiple main detector modules arranged along the Z direction.

[0105] The specific calculation process is as follows:

[0106] Assume that the bed position corresponding to the second image is a, the z-width of the main detector is b, the magnification from the rotation center to the main detector is c, and the row number of the main detector is d, then the calculation result of the bed position corresponding to this row of main detectors is: a+b*d / c.

[0107] S44: Finding the scattering value corresponding to each row of detectors in the scattering correction table according to the bed position and angle information corresponding to the row of detectors to obtain a scattering distribution corresponding to the second image;

[0108] S45: performing an antilogarithmic transformation on the second image and obtaining the inverse thereof to obtain an image that has not been subjected to scattering correction;

[0109] S46: subtracting the scattering distribution corresponding to the second image from the obtained image without scattering correction to obtain a scattering corrected image;

[0110] S47: performing a negative logarithm operation on the image after scattering correction, thereby completing scattering correction for the second image.

[0111] S5: If the judgment result is no, obtaining the scattering distribution corresponding to the second image based on the pre-trained model and performing scattering correction.

[0112] Specifically, in this embodiment, the working principle of the pre-training model is as follows:

[0113] Acquiring scattering data collected by the scattering detector and projection data corresponding to the second image;

[0114] Inputting the scattering data and the projection image data into a pre-trained model;

[0115] A scattering distribution corresponding to the second image is output based on the pre-trained model.

[0116] It is understandable that the above steps S4 and S5 are parallel steps and there is no order of precedence. Different data processing processes are performed based on different CT scanning methods to complete the scatter correction.

[0117] Furthermore, in the above embodiment, when the CT scanner is in an axial scan mode, if scatter detectors are installed on both sides of the main detector, the scatter values ​​measured by the scatter detectors on both sides and the low-frequency characteristics of the scatter distribution are used to obtain the scatter value of each pixel on the main detector using areal interpolation for scatter subtraction. If a scatter detector is installed on only one side of the main detector, the scatter value of each pixel on the main detector is estimated using a scatter kernel convolution method or an AI method for scatter subtraction.

[0118] Furthermore, the spacing of the scattered detectors is preferably greater than that of the main detectors; a larger spacing indicates a sparser arrangement, while a smaller spacing indicates a denser arrangement. This allows for a sparse arrangement of scattered detectors by leveraging the low-frequency characteristics of the spatial distribution of scattered scattering intensity, thereby making the arrangement of scattered detectors more flexible. This eliminates the need for shielding with the main detectors and the need for detectors with the same pixel size and density as the main detectors, thereby saving costs.

[0119] Based on the above-described scatter correction method for static CT, the present invention further provides a scatter correction system for static CT. As shown in Figure 6 , the scatter correction system includes one or more processors 21 and a memory 22. The memory 22 is coupled to the processor 21 and stores a program. When the program is executed by the processor 21, the processor 21 implements the scatter correction method for static CT described in the above-described embodiment.

[0120] The processor 21 is used to control the overall operation of the scatter correction system to complete all or part of the steps of the above-described scatter correction method for static CT. The processor 21 may be a central processing unit (CPU), a graphics processing unit (GPU), a field programmable gate array (FPGA), an application-specific integrated circuit (ASIC), a digital signal processing (DSP) chip, etc. The memory 22 is used to store various types of data to support the operation of the scatter correction system. This data may include, for example, instructions for any application or method operating on the scatter correction system, as well as application-related data. The memory 22 may be implemented by any type of volatile or non-volatile storage device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, etc.

[0121] In an exemplary embodiment, the scatter correction system can be implemented as a computer chip or entity, or as a product with certain functions, to perform the aforementioned scatter correction method for static CT and achieve the same technical effects as the aforementioned method. A typical embodiment is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, an in-vehicle human-computer interaction device, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or any combination of these devices.

[0122] In another exemplary embodiment, the present invention further provides a computer-readable storage medium comprising program instructions, which, when executed by a processor, implement the steps of the scatter correction method for static CT described in any of the aforementioned embodiments. For example, the computer-readable storage medium may be the aforementioned memory comprising the program instructions, which may be executed by a processor of a scatter correction system to perform the aforementioned scatter correction method for static CT and achieve the same technical effects as the aforementioned method.

[0123] In summary, the scatter correction method and system for static CT provided by the embodiments of the present invention have the following beneficial effects:

[0124] 1. Using a scatter detector to measure scattering in real time during CT scanning, and utilizing the characteristics of static CT, the scattering distribution of all positions and angles of the scanned object can be obtained very accurately through interpolation, so as to perform precise scattering correction and improve the effect of scattering correction.

[0125] 2. The scatter correction method is easy to implement and has low algorithm complexity.

[0126] 3. It can be widely used in single-source and multi-source exposure modes.

[0127] The scatter correction method and system for static CT provided by the present invention have been described in detail above. Any obvious modification to the present invention without departing from its essence would constitute an infringement of the present invention's patent rights and would incur corresponding legal liability.

Claims

1. A scatter correction method for static CT, characterized in that The steps include: Preliminarily collecting scattering data through a scattering detector disposed on at least one side of the main detector to obtain a scattering correction table by interpolation, and using the scattering data to perform model training to obtain a pre-trained model; wherein the scattering correction table includes a scattering distribution corresponding to each bed position and angle; Obtaining the original image of the static CT scan performed by the user; Pre-correcting the original image to obtain a first image; Obtain a second image by performing an eccentric cut on the first image; Determining whether the CT scanning method of the user is spiral scanning; If the judgment result is yes, the scatter correction table is used to perform scatter correction on the second image; if the judgment result is no, the scatter distribution corresponding to the second image is obtained based on the pre-trained model, and scatter correction is performed on the second image.

2. The scatter correction method according to claim 1, characterized in that The scatter correction table is obtained by: Collecting a scattering image based on a scattering detector disposed on at least one side of the main detector, and performing data preprocessing; Based on the preprocessed data, the scattered data is extracted and put into an array, and corresponds to the bed position and angle; wherein the array includes data information of each bed position and angle; Based on the scattering data, extract all scattering values ​​of each pixel and the bed position and angle corresponding to each pixel; In the bed and angle plane, the scattering value of each grid position in the plane is obtained by interpolation according to the set bed and angle grids; The interpolated scattering values ​​are arranged according to the detection direction of the main detector and smoothed; The scatter distributions of all main detectors in the detection direction and the corresponding bed positions and angles are stored as the scatter correction table.

3. The scatter correction method according to claim 2, characterized in that Performing scatter correction based on the scatter correction table specifically includes: inputting the second image; extracting bed position and angle information of the second image; For each row of main detectors, the detection angle is converted to 0-360 degrees, and a calculation is performed according to the bed position and the row number of the second image to obtain the bed position corresponding to the row of main detectors; According to the bed position and angle information corresponding to the row of detectors, finding the scattering value corresponding to each row of detectors in the scattering correction table to obtain the scattering distribution corresponding to the second image; Performing an antilogarithmic transformation on the second image and obtaining an inverse thereof to obtain an image that has not been subjected to scatter correction; subtracting the scattering distribution corresponding to the second image from the obtained image that has not been scatter-corrected to obtain a scatter-corrected image; A negative logarithm operation is performed on the scatter-corrected image, thereby completing the scatter correction of the second image.

4. The scatter correction method according to claim 1, characterized in that The acquiring the scattering distribution corresponding to the second image based on the pre-trained model specifically includes: Acquiring scattering data collected by the scattering detector and projection data corresponding to the second image; Inputting the scattering data and the projection image data into a pre-trained model; Output a scattering distribution corresponding to the second image based on the pre-trained model.

5. The scatter correction method according to claim 2, wherein: The interpolation method at least includes cubic interpolation or biharmonic spline interpolation, and CUDA acceleration is performed during the interpolation process.

6. The scatter correction method according to claim 1, characterized in that The pre-correction of the original image specifically includes: Performing background correction on the original image to obtain a background corrected image; Performing gain correction on the background correction image to obtain a gain correction image; Air correction is performed on the gain-corrected image to obtain a first image.

7. The scatter correction method according to claim 1, wherein: In the axial scanning state, if scattering detectors are set on both sides of the main detector, the scattering value measured by the scattering detectors on both sides and the low-frequency characteristics of the scattering distribution are used to obtain the scattering value of each pixel on the main detector by using the surface interpolation method for scattering subtraction; If only one side of the main detector is provided with a scattering detector, a scattering kernel convolution method or an AI method is used to estimate the scattering value of each pixel on the main detector for scattering subtraction.

8. The scatter correction method according to claim 1, wherein: The arrangement interval of the scattering detectors is greater than the arrangement interval of the main detectors; wherein, the larger the arrangement interval, the sparser the arrangement, and the smaller the arrangement interval, the denser the arrangement.

9. The scatter correction method according to claim 1, characterized in that The scattering detector collects scattering data, including at least: Scattering data collected under single-ray source exposure; or, scattering data collected under multiple-ray source alternating exposure; or, scattering data collected under multiple-ray source simultaneous exposure.

10. A scatter correction system for static CT, characterized in that The device comprises a processor and a memory, wherein the processor reads a computer program in the memory to perform the following operations: Preliminarily collecting scattering data through a scattering detector disposed on at least one side of the main detector to obtain a scattering correction table by interpolation, and using the scattering data to perform model training to obtain a pre-trained model; wherein the scattering correction table includes a scattering distribution corresponding to each bed position and angle; Obtaining the original image of the static CT scan performed by the user; Pre-correcting the original image to obtain a first image; Obtain a second image by performing an eccentric cut on the first image; Determining whether the CT scanning method of the user is spiral scanning; If the judgment result is yes, the scatter correction table is used to perform scatter correction on the second image; if the judgment result is no, the scatter distribution corresponding to the second image is obtained based on the pre-trained model, and scatter correction is performed on the second image.

Images