Scatter correction method and device

Abstract

This application relates to a scatter correction method and apparatus. The scatter correction method includes: obtaining a scout image of a scanned object; determining estimated scatter distribution information of the scanned object based on the scout image based on a trained scatter correction model; obtaining scan information of the scanned object based on the scout image; and performing scatter correction on the scan information based on the estimated scatter distribution information. This application addresses the problem in related technologies of high time costs and low efficiency of the entire scatter correction process due to the time-consuming scatter estimation process. This method saves time for scatter estimation and improves the efficiency of the scatter correction process.

Description

Technical Field

[0001] The present application relates to the field of medical imaging technology, and in particular to a scatter correction method and device. Background Art

[0002] Computed tomography (CT) equipment uses precisely collimated X-rays, gamma rays, ultrasound waves, and other technologies, along with highly sensitive detectors, to perform sequential cross-sectional scans around a specific part of the scanned object. CT devices offer fast scan times and clear images, making them suitable for diagnosing a wide range of diseases. During the CT scan, scattered signals generated by the interaction between the radiation and the scanned object are received by the detector. This can cause artifacts such as dark strips, bands, or uneven cupping in the CT-based scanned image, reducing soft tissue contrast and CT value accuracy, thus impacting image quality. Therefore, before acquiring the final scanned image, scatter correction is required to mitigate the effects of scattered signals on the image.

[0003] In related art, scatter estimation is performed based on the original scanned image of the subject, and scatter correction is performed on the original scanned image based on the scatter estimation results. Since scatter estimation is time-consuming, the entire scatter correction process is time-consuming and inefficient.

[0004] Currently, no effective solution has been proposed to address the problem in related technologies that scatter estimation takes a long time, resulting in high time cost and low efficiency in the entire scatter correction process. Summary of the Invention

[0005] The embodiments of the present application provide a scatter correction method and device to at least solve the problem in the related art that scatter estimation takes a long time, resulting in high time cost and low efficiency of the entire scatter correction process.

[0006] In a first aspect, an embodiment of the present application provides a scatter correction method, comprising:

[0007] obtaining a scout image of the scanned object;

[0008] Determining scatter distribution estimation information of the scanned object based on the trained scatter correction model and the scout image;

[0009] acquiring scanning information of the scanned object based on the scout image;

[0010] Scatter correction is performed on the scan information according to the scatter distribution estimation information.

[0011] In some embodiments,

[0012] The scatter correction model uses a sample positioning image and sample scatter distribution estimation information as a training set, wherein the sample positioning image and the sample scatter distribution estimation information correspond to each other; or

[0013] The scatter correction model uses a sample positioning image and sample scanning information as a training set, wherein the sample positioning image and the sample scanning information correspond to each other.

[0014] In some embodiments, when the scatter correction model uses a sample locator image and sample scanning information as a training set, determining the scatter distribution estimation information of the scanned object includes:

[0015] A scanning information estimation result is determined according to the scatter correction model, and a scatter distribution estimation algorithm is used to simulate and calculate the scanning information estimation result to determine the scatter distribution estimation information, wherein the scatter distribution estimation algorithm includes at least one of the following: a Monte Carlo simulation algorithm, a scatter kernel superposition algorithm, and a deep learning algorithm.

[0016] In some embodiments,

[0017] The scanning information includes projection data and / or scanned images;

[0018] The scatter distribution estimation information includes scatter distribution data corresponding to the projection data and / or a scatter distribution image corresponding to the scan image;

[0019] The scatter correction model includes a first scatter correction model corresponding to the projection data and / or a second scatter correction model corresponding to the scanned image.

[0020] In some embodiments, determining the estimated scatter distribution information of the scanned object according to the scout image based on the trained scatter correction model includes:

[0021] determining initial scatter distribution data corresponding to a preset number of projections based on the first scatter correction model;

[0022] Actual scattering distribution data corresponding to the scanned object is determined according to the preset projection number, the initial scattering distribution data, and the actual projection number corresponding to the scanned object.

[0023] In some embodiments, determining actual scattering distribution data corresponding to the scanned object based on the preset number of projections, the initial scattering distribution data, and the actual number of projections corresponding to the scanned object includes:

[0024] When the actual number of projections is greater than the preset number of projections, interpolating and expanding the initial scattering distribution data to obtain the actual scattering distribution data;

[0025] When the actual number of projections is less than the preset number of projections, the initial scattering distribution data is selected at intervals to obtain the actual scattering distribution data.

[0026] In some embodiments, determining the estimated scatter distribution information of the scanned object according to the scout image based on the trained scatter correction model includes:

[0027] Based on the second scatter correction model, a scatter distribution image of the scanned object is acquired according to the scout image, wherein the scatter distribution image is used to perform scatter correction on the scanned image.

[0028] In some embodiments, before performing scatter correction on the scan information based on the scatter distribution estimation information, the method includes:

[0029] Perform environmental parameter correction on the scan information.

[0030] In some embodiments, performing scatter correction on the scan information according to the scatter distribution estimation information includes:

[0031] An iterative calculation is performed on a difference between the scanning information and the scatter distribution estimation information to obtain scatter-corrected scanning information, wherein a convergence condition for the iterative calculation is that a difference between corresponding iteration parameters in two adjacent iterations is less than or equal to a preset convergence threshold, and the iteration parameters include the scanning information and / or the scatter distribution estimation information.

[0032] In a second aspect, an embodiment of the present application provides a scatter correction device, including an acquisition module, a determination module, and a correction module:

[0033] The acquisition module is configured to acquire a positioning image of the scanned object; and acquire scanning information of the scanned object based on the positioning image;

[0034] The determination module is configured to determine the scatter distribution estimation information of the scanned object according to the scout image based on the trained scatter correction model;

[0035] The correction module is configured to perform scatter correction on the scanning information according to the scatter distribution estimation information.

[0036] In a third aspect, an embodiment of the present application provides an electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the computer program, the scatter correction method as described in the first aspect above is implemented.

[0037] In a fourth aspect, an embodiment of the present application provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the scatter correction method as described in the first aspect above.

[0038] Compared with the related art, the scatter correction method provided in the embodiments of the present application obtains a positioning image of the scanned object; determines scatter distribution estimation information of the scanned object based on the positioning image based on a trained scatter correction model; obtains scanning information of the scanned object based on the positioning image; and performs scatter correction on the scanning information based on the scatter distribution estimation information. This solves the problem in the related art that the scatter estimation process is time-consuming, resulting in high time cost and low efficiency of the entire scatter correction process. Since the acquisition of the scatter distribution estimation information and the acquisition of the scanning information can be performed simultaneously in the present application, the formal scanning information can be immediately subjected to scatter correction after the formal scan is completed, thereby saving the time of scatter estimation and improving the efficiency of the scatter correction process.

[0039] The details of one or more embodiments of the present application are set forth in the following drawings and description to make other features, objects, and advantages of the present application more readily apparent. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0041] Figure 1 This is a schematic diagram of an application environment of a scatter correction method according to an embodiment of the present application;

[0042] Figure 2 is a flow chart of a scatter correction method according to an embodiment of the present application;

[0043] Figure 3 This is a flow chart of a method for acquiring projection threshold scattering distribution data according to one embodiment of the present application;

[0044] Figure 4 is a flow chart of a projection threshold scatter correction method according to another embodiment of the present application;

[0045] Figure 5 is a flow chart of an image threshold scatter correction method according to another embodiment of the present application;

[0046] Figure 6This is a hardware structure block diagram of a terminal for a scatter correction method according to an embodiment of the present application;

[0047] Figure 7 This is a structural block diagram of a scattering correction device according to an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is described and illustrated below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. Based on the embodiments provided in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application. In addition, it can also be understood that although the efforts made in this development process may be complex and lengthy, for ordinary technicians in the field related to the contents disclosed in the present application, some changes such as design, manufacturing or production based on the technical contents disclosed in the present application are only conventional technical means and should not be understood as the contents disclosed in the present application being insufficient.

[0049] References to "embodiments" in this application mean that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearance of this phrase in various places in the specification does not necessarily refer to the same embodiment, nor does it refer to independent or alternative embodiments that are mutually exclusive of other embodiments. It is understood, both explicitly and implicitly, by those skilled in the art that the embodiments described in this application may be combined with other embodiments unless there is a conflict.

[0050] Unless otherwise defined, the technical or scientific terms used in this application should have the ordinary meaning understood by a person of ordinary skill in the technical field to which this application belongs. The words "one", "a", "the" and the like used in this application do not indicate a limit on quantity and may indicate the singular or plural. The terms "include", "comprise", "have" and any variations thereof used in this application are intended to cover non-exclusive inclusions; for example, a process, method, system, product or device that includes a series of steps or modules (units) is not limited to the listed steps or units, but may also include steps or units that are not listed, or may also include other steps or units that are inherent to these processes, methods, products or devices. The words "connect", "connected", "coupled" and the like used in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The word "multiple" used in this application means greater than or equal to two. "And / or" describes the association relationship of associated objects, indicating that three relationships can exist. For example, "A and / or B" can mean: A exists alone, A and B exist at the same time, and B exists alone. The terms "first", "second", "third" and the like involved in this application are merely used to distinguish similar objects and do not represent a specific ordering of the objects.

[0051] The scatter correction method provided in this application can be applied to Figure 1 In the application environment shown, Figure 1 FIG. 1 is a schematic diagram of an application environment of a scatter correction method according to an embodiment of the present application. Figure 1 As shown in the figure, a CT system includes a scanning device 101, a scanning bed 102, a host computer 103, and a reconstruction unit 104. The user controls the scanning device 101 through the host computer 103 to scan the object on the scanning bed 102, obtaining scan information of the object. The host computer 103 sends the acquired scan information to the reconstruction unit 104 for image reconstruction, ultimately obtaining a scanned image. However, scattering events within the CT system can affect the quality of the scanned image. Therefore, scatter correction is required for the scan information to improve the quality of the scanned image.

[0052] This embodiment provides a scattering correction method. Figure 2 FIG. 1 is a flow chart of a scatter correction method according to an embodiment of the present invention. Figure 2 As shown, the method includes the following steps:

[0053] Step S210: Acquire a positioning image of the scanned object.

[0054] In this embodiment, a scout image can be acquired using a medical imaging scanning system, such as a CT system. Specifically, the scanned subject can be a human or an animal. Before the formal scan is performed using the CT system, a rapid scan of the subject is typically performed using the medical imaging scanning system to obtain a planar image of the examination area. This image is called a topogram. This topogram is used to determine the scanning range for the subsequent formal scan, and then the formal scan of the subject is performed. Typically, the scout image is scanned after the scanning bed is positioned. The scout image includes an anteroposterior and / or lateral view.

[0055] Step S220 : Determine scatter distribution estimation information of the scanned object based on the trained scatter correction model and the scout image.

[0056] The scatter correction model in this embodiment can be trained by a deep learning algorithm. The trained scatter correction model can directly obtain scatter distribution estimation information based on the positioning image, or it can first estimate the scanning information based on the positioning image and then determine the scatter distribution estimation information based on the scanning information.

[0057] The scatter distribution estimation information in this embodiment refers to the impact of scatter signals on detector pixels. For example, during a CT system scan, the scatter signals generated by the interaction between X-rays and the scanned object, when received by the detector, can cause darkened strips, bands, or uneven cup-shaped artifacts to appear in the scanned image. This reduces soft tissue contrast and CT value accuracy, impacting scanned image quality. Therefore, CT systems require scatter correction to mitigate the impact of scatter signals on the image.

[0058] Step S230: acquiring scanning information of the scanned object based on the scout image.

[0059] After the scanning range is confirmed based on the scout image, the object can be scanned within the scanning range by the medical scanning imaging system. It should be noted that since the scatter distribution estimation information can be directly obtained based on the scout image, step S230 and step S220 in this embodiment can be performed simultaneously.

[0060] Step S240 : performing scatter correction on the scan information according to the scatter distribution estimation information.

[0061] Specifically, the scatter distribution estimation information may be subtracted from the scanning information, thereby achieving scatter correction.

[0062] Through the above steps S210 to S240, this embodiment can predict scatter distribution estimation information based on the trained scatter correction model according to the scout image. In the related art, scatter distribution estimation usually needs to be started after the scanning information is obtained, and the scatter distribution estimation also takes a long time. Therefore, in this embodiment, since the scatter distribution estimation information can be obtained based on the scout image, the acquisition of the scatter distribution estimation information and the acquisition of the scanning information can be performed simultaneously without waiting for the scanning of the medical scanning imaging system. This optimizes the scatter correction process and reduces time consumption. This solves the problem in the related art that the long scatter estimation process leads to high time cost and low efficiency of the entire scatter correction process, saves time for scatter estimation, and improves the efficiency of the scatter correction process.

[0063] In some embodiments, the scatter correction model includes two categories. The first category of scatter correction models can directly obtain scatter distribution estimation information, and the second category of scatter correction models needs to first estimate scanning information based on the positioning image and then determine the scatter distribution estimation information based on the estimated scanning information.

[0064] Specifically, the first type of scatter correction model is trained using a sample scout image and sample scatter distribution estimation information as a training set. The sample scout image and the sample scatter distribution estimation information correspond to each other. In actual use, after inputting a scout image of the scanned object into the first type of scatter correction model, the scatter distribution estimation information can be obtained. The second type of scatter correction model is trained using a sample scout image and sample scan information as a training set. The sample scout image and the sample scan information correspond to each other. In actual use, after inputting a scout image of the scanned object into the second type of scatter correction model, an estimation result of the scan information corresponding to the scanned object can be obtained. Scatter distribution estimation information is then calculated based on the scan information estimation result.

[0065] In this embodiment, two different types of training methods for scatter correction models are provided to adapt to different scenario requirements and improve the flexibility of the scatter correction process.

[0066] Furthermore, when the scatter correction model uses the sample positioning image and sample scanning information as a training set, determining the scatter distribution estimation information of the scanned object is achieved by the following method: determining the scanning information estimation result through the scatter correction model, and performing simulation calculation on the scanning information estimation result using a scatter distribution estimation algorithm to determine the scatter distribution estimation information, wherein the scatter distribution estimation algorithm includes at least one of the following: a Monte Carlo simulation algorithm, a scatter kernel superposition algorithm, and a deep learning algorithm.

[0067] Specifically, taking the CT system as an example, the Monte Carlo simulation algorithm is a calculation method based on the theoretical methods of probability and statistics. In this embodiment, the system parameters of the medical imaging scanning system that need to be used as algorithm input include: tube energy spectrum and intensity distribution, system geometric parameters, detector position and response, etc. Among them, the system geometric parameters such as CT scan width are generally fixed in the CT system. After obtaining the above system parameters, the current scattering distribution estimation information can be obtained using the fast Monte Carlo simulation algorithm based on the scanning information.

[0068] The Scatter Kernel Superposition (SKS) algorithm is a physical correction algorithm that can estimate scatter distributions using a simplified medical scanning imaging system and simulated scanned objects of varying sizes. During the estimation process, the estimated scatter distribution information is determined through a transformation and convolution calculation based on the estimated scan information. The multiple convolution kernels used in the convolution calculation are determined based on the energy spectrum, scan width, and body part size at different voltages.

[0069] Deep learning algorithms require scan data containing scattering signals as training input, and the scattering distribution estimation results corresponding to the scan data serve as the gold standard for training the deep learning network. During use, the scattering distribution estimation results are determined based on the trained deep learning network and the scan data estimation results corresponding to the scanned object.

[0070] In this embodiment, when the scatter correction model uses the sample positioning image and sample scanning information as a training set, different scatter distribution estimation algorithms can be used to calculate the scatter distribution estimation information, further improving the scene adaptability of the scatter correction method.

[0071] In some embodiments, because the scatter correction process includes projection threshold scatter correction and image threshold scatter correction, the scanning information includes projection data and / or a scanned image. Accordingly, the scatter distribution estimation information includes scatter distribution data corresponding to the projection data and / or a scatter distribution image corresponding to the scanned image. The scatter correction model includes a first scatter correction model corresponding to the projection data and / or a second scatter correction model corresponding to the scanned image. It should be noted that different types of scatter correction models include a first scatter correction model for the projection threshold and a second scatter correction model for the image threshold to accommodate different scenario requirements.

[0072] Specifically, for the first scatter correction model for the projection threshold, which can directly obtain scatter distribution data, the network needs to be trained based on the scout image, using the projection data containing scatter information in the projection threshold as training input, and the actual scatter distribution data of the projection threshold as the gold standard. For the second scatter correction model for the image threshold, which can directly obtain the scatter distribution image, the network needs to be trained using the sample scout image and the corresponding sample scatter distribution image. During the actual scatter correction process, a scatter distribution image of the scanned object is obtained based on the scout image, and this scatter distribution image is used to perform scatter correction on the scanned image. For the first scatter correction model for the projection threshold, which obtains the estimated scan information, the network needs to be trained based on the scout image, using the projection data containing scatter information in the projection threshold as training input, and the scanned data of the projection threshold without scattering as the gold standard. For the second scatter correction model for the image threshold, which obtains the estimated scan information, the network needs to be trained using the scout image as input, and the scanned image without scattering artifacts as the gold standard. In this embodiment, the training process of the scatter correction model for different scenarios is provided to improve the scenario adaptability of the scatter correction process.

[0073] In some embodiments, when performing scatter correction at the projection threshold, it is necessary to calculate the number of projections based on the preset number of projections. Figure 3 FIG. 1 is a flow chart of a method for acquiring projection threshold scattering distribution data according to an embodiment of the present invention. Figure 3 As shown, the method includes the following steps:

[0074] Step S310 : determining initial scatter distribution data corresponding to a preset number of projections based on a first scatter correction model.

[0075] During the training of the first scatter correction model, a quantitative number of projections is obtained using the scout image as input, and the scatter distribution data is trained based on the quantitative number of projections. Accordingly, the amount of scatter distribution data is also fixed. Therefore, based on the preset number of projections in the first scatter model, a corresponding amount of initial scatter distribution data can be obtained.

[0076] Step S320 : determining actual scattering distribution data corresponding to the scanned object according to the preset number of projections, the initial scattering distribution data, and the actual number of projections corresponding to the scanned object.

[0077] After obtaining the actual number of projections of the scanned object, the initial scattering distribution data needs to be adjusted according to a comparison result between the actual number of projections and the preset number of projections, thereby obtaining the actual scattering distribution data.

[0078] Through the above steps S310 and S320, the initial scattering distribution data is adjusted according to the comparison result between the actual number of projections and the preset number of projections, thereby obtaining more accurate actual scattering distribution data and improving the accuracy of the scattering correction process.

[0079] Furthermore, when the actual number of projections is greater than the preset number of projections, the initial scattering distribution data is interpolated and expanded to obtain actual scattering distribution data. When the actual number of projections is less than the preset number of projections, the initial scattering distribution data is selected at intervals to obtain actual scattering distribution data. When the actual number of projections is equal to the preset number of projections, the initial scattering distribution data is used as the final actual scattering distribution data. For example, when the preset number of projections is 1200, the initial scattering distribution data is recorded as Is0. If the actual number of projections is less than 1200, scattering distributions corresponding to the actual number of projections are selected at equal intervals within Is0 as the actual scattering distribution data. If the actual number of projections is greater than 1200, Is0 is interpolated and expanded to a scattering distribution corresponding to the actual number of projections as the actual scattering distribution data. Interpolation methods include, but are not limited to, linear interpolation and nearest neighbor interpolation. In this embodiment, the actual scattering distribution data corresponding to the actual number of projections is obtained through interpolation and expansion or interval selection, further improving the accuracy of the actual scattering distribution data in the projection threshold.

[0080] In some embodiments, before performing scatter correction on the scan information based on the scatter distribution estimation information, the scan information needs to be pre-processed, that is, environmental parameter correction, such as air correction, bad channel correction, etc. The process of air correction is to perform a series of scans on the air without placing any objects in the scanning range to obtain a set of air scanning data, and then subtract the air scanning data from the actual scanning data. The bad channel correction is because the CT system usually has tens of thousands to hundreds of thousands of channels, and the performance of each channel is different. Among them, the channel that cannot respond to the incident radiation is called a bad channel. The presence of the bad channel causes the reconstructed image to have ring artifacts, resulting in a decrease in the quality of clinical diagnosis. In order to ensure the quality of clinical diagnosis, it is necessary to remove the influence of the bad channel on the reconstructed image without introducing additional artifacts. Therefore, it is necessary to calculate the actual ray attenuation value at the location of the bad channel to eliminate the influence of the bad channel. In this embodiment, the accuracy of the scan information can be improved by performing environmental parameter correction on the scan information.

[0081] It should be noted that the environmental parameter correction in this embodiment can also be performed while calculating the scatter distribution estimation information. For example, while determining the scatter distribution estimation information through the positioning image, the scanning object is scanned, and then the environmental parameter correction is performed on the scan information, so as to save scatter correction time and improve the efficiency of scatter correction.

[0082] In some embodiments, scatter correction of scan information based on scatter distribution estimation information includes iteratively calculating the difference between the scan information and the scatter distribution estimation information to obtain scatter-corrected scan information. A convergence condition for the iterative calculation is that the difference between corresponding iteration parameters between two adjacent iterations is less than or equal to a preset convergence threshold. The iteration parameters include the scan information and / or the scatter distribution estimation information. Specifically, the scan information is represented by It, the scatter distribution estimation information is represented by Is, and the scatter-corrected scan information is represented by Ip. For each iteration, after estimating the current Is, scatter correction is performed using Ip = It - Is. A determination is made based on the convergence condition, and whether to proceed to the next iteration is determined based on the determination result. During the iteration, It used in the current calculation is the Ip used in the previous calculation. Typically, the convergence condition can be set as the number of iterations being less than or equal to a preset convergence number, or as the difference between the current iteration parameter and the same iteration parameter used in the previous iteration being less than or equal to a preset convergence threshold. The parameter can be either Is or Ip. Furthermore, a regularization term may be added to perform iterations on the basis of It in each cycle. In this embodiment, the scatter correction process of the scanning information is implemented through iterative calculation, which can further improve the accuracy of the scatter correction.

[0083] The embodiments of the present application are described and illustrated below through preferred embodiments.

[0084] Existing scatter correction methods are mainly divided into hardware, software or a combination of the two. Among them, scatter correction based on software methods requires completing the CT scan, performing scatter estimation based on uncorrected projection data or CT scan images, and then performing scatter correction.

[0085] The following describes the fast Monte Carlo simulation of the projection threshold, the SKS algorithm, and the deep learning algorithm respectively.

[0086] First, the scatter correction of the fast Monte Carlo simulation simulates the scatter distribution based on the CT scan image and the system parameters of the CT system, and then subtracts the scatter distribution from the projection data. The specific process includes:

[0087] (1) After the subject is positioned on the scanning bed, the positioning image scan begins. The positioning image includes the frontal image and / or lateral image;

[0088] (2) Select the scanning range on the scout image and, after confirmation, complete the scan using the CT system to obtain projection data;

[0089] (3) Using a small amount of downsampled projection data, a small amount of correction is performed and then quickly reconstructed to obtain an unscattered CT tomogram. The current scatter distribution data Is is then simulated using the fast Monte Carlo method based on the system parameters. At the same time, scatter correction preprocessing of all projection data is completed to obtain It. The preprocessing specifically involves environmental parameter correction, generally including air correction and bad channel correction. It should be noted that the estimation of the current scatter distribution data Is takes a long time, and after the scatter correction preprocessing is completed, it is necessary to wait for the result of Is before proceeding to the next step.

[0090] (4) Subtract Is from It to obtain the scatter correction result Ip = It - Is, completing the scatter correction;

[0091] (5) Continue to perform other correction and reconstruction processes to obtain the CT tomographic image after scatter correction.

[0092] The specific process of the SKS algorithm includes:

[0093] (1) After the subject is positioned on the scanning bed, the positioning image scan begins. The positioning image includes the frontal image and / or lateral image;

[0094] (2) Select the scanning range on the scout image and, after confirmation, complete the scan using the CT system to obtain projection data;

[0095] (3) Complete the scatter correction preprocessing of all projection data to obtain It, where the preprocessing specifically includes environmental parameter correction, generally including air correction, bad channel correction, etc.

[0096] (4) Using the SKS algorithm, the corresponding scattering convolution kernel is selected according to the system parameters and other factors to perform convolution to obtain the current scattering distribution data Is;

[0097] (5) Subtract Is from It to obtain the scatter correction result Ip = It - Is, completing the scatter correction;

[0098] (6) Continue to perform other correction and reconstruction processes to obtain the CT tomographic image after scatter correction.

[0099] When calibrating the deep learning algorithm, a deep learning network is needed to replace the scattering convolution kernel to obtain the scattering distribution data. The specific process includes:

[0100] (1) After the subject is positioned on the scanning bed, the positioning image scan begins. The positioning image includes the frontal image and / or lateral image;

[0101] (2) Select the scanning range on the scout image and, after confirmation, complete the scan using the CT system to obtain projection data;

[0102] (3) Complete the scatter correction preprocessing of all projection data to obtain It, where the preprocessing specifically includes environmental parameter correction, generally including air correction, bad channel correction, etc.

[0103] (4) Input It into the trained deep learning network to obtain the current scattered signal distribution Is;

[0104] (5) Subtract Is from It to obtain the scatter correction result Ip = It - Is, completing the scatter correction;

[0105] (6) Continue to perform other correction and reconstruction processes to obtain the CT tomographic image after scatter correction.

[0106] Compared with related technologies, Figure 4 FIG. 1 is a flow chart of a projection threshold scatter correction method according to another embodiment of the present invention. Figure 4 As shown, the method includes the following steps:

[0107] Step S410: After the scan subject is positioned on the scanning bed, a scout image scan is started. The scout image includes an anteroposterior image and / or a lateral image.

[0108] Step S420, using the deep learning network to obtain the scattering distribution data Is;

[0109] Step S430, selecting a scanning range on the scout image, performing a CT scan within the scanning range and obtaining original projection data;

[0110] Step S440: Perform scatter correction preprocessing on all projection data to obtain It. It should be noted that steps S430 and S440 can be performed simultaneously with step S420 to reduce the overall scatter correction time.

[0111] Step S450, performing scatter correction on the projection threshold, deducting Is from It to obtain a scatter correction result Ip=It-Is, thus completing the scatter correction;

[0112] Step S460: Continue to perform other correction and reconstruction processes to obtain a scatter-corrected CT tomographic image.

[0113] Through the above steps S410 to S460, CT scatter distribution data can be directly obtained based on the trained deep learning network, thereby optimizing the scatter correction processes of the Monte Carlo simulation algorithm, the SKS algorithm, and the deep learning algorithm, reducing time consumption, and solving the problem in the related art that the scatter estimation is time-consuming, resulting in high time cost and low efficiency of the entire scatter correction process. This saves time for scatter estimation and improves the efficiency of the scatter correction process.

[0114] A deep learning network can use CT images with scattering artifacts to learn the scattering distribution on the corresponding image. This network can then be used to deduct the scattering distribution output by the network from the image. The image threshold deep learning method uses CT images containing scattering artifacts as training input and CT images without scattering artifacts as the gold standard to train the network. The following is an illustration of the image threshold deep learning algorithm:

[0115] (1) After the subject is positioned on the scanning bed, the positioning image scan begins. The positioning image includes the frontal image and / or lateral image;

[0116] (2) Select the scanning range on the scout image and, after confirmation, complete the scan using the CT system to obtain projection data;

[0117] (3) Correction and reconstruction to obtain the unscattered CT tomogram IMGt;

[0118] (4) Input into the trained deep learning network to obtain the image threshold scattering estimate IMGs;

[0119] (5) The scatter distribution image IMGs is deducted from the unscatter corrected CT tomogram IMGt to obtain the scatter corrected CT tomogram IMGp = IMGt - IMGs.

[0120] Compared with related technologies, Figure 5 FIG. 1 is a flow chart of an image threshold scatter correction method according to an embodiment of the present invention. Figure 5 As shown, the method includes the following steps:

[0121] Step S510: After the scan subject is positioned on the scanning bed, a scout image scan is started. The scout image includes an anteroposterior image and / or a lateral image.

[0122] Step S520, inputting the positioning image into the trained deep learning network to obtain the image threshold scattering distribution image IMGs;

[0123] Step S530, completing CT scanning, correction, and reconstruction within the scanning range to obtain an unscattered CT tomogram IMGt; it should be noted that step S520 and step S530 can be performed simultaneously to reduce the overall scattering correction time;

[0124] Step S540 , subtracting the scatter distribution image IMGs from the unscatter corrected CT tomogram IMGt to obtain a scatter corrected CT tomogram IMGp=IMGt−IMGs;

[0125] Step S550: Continue to perform other correction and reconstruction processes to obtain a scatter-corrected CT tomographic image.

[0126] Through steps S510 to S550, a scatter distribution image can be directly obtained based on the trained scatter correction model without waiting for scanning by the medical scanning imaging system. This optimizes the scatter correction process and reduces time consumption. This solves the problem in related technologies of high time cost and low efficiency of the entire scatter correction process due to the long scatter estimation process. This saves time for scatter estimation and improves the efficiency of the scatter correction process.

[0127] It should be noted that the steps shown in the above process or the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0128] The method embodiments provided in this application can be executed in a terminal, a computer or a similar computing device. Taking running on a terminal as an example, Figure 6 This is a hardware structure diagram of a terminal of a scatter correction method according to an embodiment of the present application. Figure 6 As shown, the terminal 60 may include one or more ( Figure 6 Only one is shown) a processor 602 (the processor 602 may include but is not limited to a microprocessor MCU or a programmable logic device FPGA and other processing devices) and a memory 604 for storing data. Optionally, the terminal may also include a transmission device 606 and an input / output device 608 for communication functions. It will be understood by those skilled in the art that Figure 6 The structure shown is only for illustration and does not limit the structure of the above terminal. Figure 6 More or fewer components than shown, or with Figure 6 Different configurations shown.

[0129] Memory 604 can be used to store control programs, such as software programs and modules of application software, such as the control program corresponding to the scatter correction method in the embodiments of the present application. Processor 602 executes the control program stored in memory 604 to execute various functional applications and data processing, thereby implementing the aforementioned method. Memory 604 may include high-speed random access memory (RAM) and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, memory 604 may further include memory remotely located from processor 602, and such remote memory may be connected to terminal 60 via a network. Examples of such networks include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0130] Transmission device 606 is used to receive or transmit data via a network. A specific example of the aforementioned network may include a wireless network provided by the communications provider of terminal 60. In one embodiment, transmission device 606 includes a network interface controller (NIC), which can be connected to other network devices via a base station to enable communication with the Internet. In another embodiment, transmission device 606 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0131] This embodiment also provides a scatter correction device for implementing the above-described embodiments and preferred embodiments. Details already described will not be repeated. As used below, terms such as "module," "unit," and "subunit" may refer to a combination of software and / or hardware that implements a predetermined function. While the devices described in the following embodiments are preferably implemented using software, implementation using hardware, or a combination of software and hardware, is also possible and contemplated.

[0132] Figure 7 This is a structural block diagram of a scatter correction device according to an embodiment of the present application. Figure 7 As shown, the device includes an acquisition module 71, a determination module 72 and a correction module 73:

[0133] The acquisition module 71 is configured to acquire a positioning image of the scanned object and acquire scanning information of the scanned object based on the positioning image.

[0134] The determination module 72 is configured to determine the estimated scatter distribution information of the scanned object based on the trained scatter correction model and the locator image. Specifically, the scatter correction model uses the sample locator image and the estimated scatter distribution information of the sample as a training set, wherein the sample locator image and the estimated scatter distribution information of the sample correspond to each other. Alternatively, the scatter correction model uses the sample locator image and the sample scanning information as a training set, wherein the sample locator image and the sample scanning information correspond to each other.

[0135] The correction module 73 is configured to perform scatter correction on the scan information according to the scatter distribution estimation information.

[0136] This embodiment is based on a trained scatter correction model and can predict scatter distribution estimation information based on the scout image through the determination module 72. In related technologies, scatter distribution estimation usually needs to be started after scanning information is obtained, and scatter distribution estimation also takes a long time. Therefore, in this embodiment, since scatter distribution estimation information can be obtained based on the scout image, the acquisition of scatter distribution estimation information and scanning information can be performed simultaneously without waiting for the scanning of the medical scanning imaging system. This optimizes the scatter correction process and reduces time consumption. This solves the problem in related technologies that the long scatter estimation process leads to high time cost and low efficiency of the entire scatter correction process, saves time for scatter estimation, and improves the efficiency of the scatter correction process.

[0137] In some embodiments, the determination module 72 is further configured to determine a scanning information estimation result according to the scatter correction model when the scatter correction model uses the sample positioning image and the sample scanning information as a training set, and perform simulation calculations on the scanning information estimation result using a scatter distribution estimation algorithm to determine the scatter distribution estimation information, wherein the scatter distribution estimation algorithm includes at least one of the following: a Monte Carlo simulation algorithm, a scatter kernel superposition algorithm, and a deep learning algorithm.

[0138] In some embodiments, the scanning information includes projection data and / or a scanned image; the scatter distribution estimation information includes scatter distribution data corresponding to the projection data and / or a scatter distribution image corresponding to the scanned image; and the scatter correction model includes a first scatter correction model corresponding to the projection data and / or a second scatter correction model corresponding to the scanned image.

[0139] In some embodiments, the determination module 72 is further configured to determine initial scatter distribution data corresponding to a preset number of projections based on the first scatter correction model; and determine actual scatter distribution data corresponding to the scanned object based on the preset number of projections, the initial scatter distribution data, and the actual number of projections corresponding to the scanned object. Furthermore, if the actual number of projections is greater than the preset number of projections, the initial scatter distribution data is interpolated and expanded to obtain actual scatter distribution data; if the actual number of projections is less than the preset number of projections, the initial scatter distribution data is interpolated and expanded to obtain actual scatter distribution data.

[0140] In some embodiments, the determination module 72 is further configured to obtain a scatter distribution image of the scanned object according to the scout image based on the second scatter correction model, wherein the scatter distribution image is used to perform scatter correction on the scanned image.

[0141] In some embodiments, the scatter correction device further includes a pre-processing module, which is configured to perform environmental parameter correction on the scanning information.

[0142] In some embodiments, the correction module 73 is configured to iteratively calculate the difference between the scanning information and the scattering distribution estimation information to obtain scattering-corrected scanning information, wherein a convergence condition for the iterative calculation is that the difference between corresponding iteration parameters in two adjacent iterations is less than or equal to a preset convergence threshold, and the iteration parameters include the scanning information and / or the scattering distribution estimation information.

[0143] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can be located in the same processor; or the above modules can be located in different processors in any combination.

[0144] This embodiment further provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0145] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0146] Optionally, in this embodiment, the processor may be configured to execute the following steps through a computer program:

[0147] S1, obtaining a positioning image of the scanned object.

[0148] S2, based on the trained scatter correction model and the positioning image, determining the scatter distribution estimation information of the scanned object.

[0149] S3, acquiring scanning information of the scanned object based on the positioning image.

[0150] S4, performing scatter correction on the scan information according to the scatter distribution estimation information.

[0151] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementation modes, and this embodiment will not be repeated here.

[0152] In addition, in conjunction with the scatter correction method in the above embodiments, embodiments of the present application may provide a storage medium for implementation. The storage medium stores a computer program; when the computer program is executed by a processor, any of the scatter correction methods in the above embodiments is implemented.

[0153] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned 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.

[0154] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A scatter correction method, characterized in that: include: Obtaining a scout image of the scanned subject; the scout image is a planar image of the examination area obtained by quickly scanning the subject using a medical imaging scanning system before the formal scan using a CT system; Based on the trained scatter correction model, the scatter distribution estimation information of the scanned object is determined according to the positioning image; at the same time, Acquiring scanning information of the scanned object based on the scout image; the scanning information is an image obtained by formally scanning the scanned object within the scanning range through a medical scanning imaging system after confirming the scanning range based on the scout image; Scatter correction is performed on the scan information according to the scatter distribution estimation information.

2. The scatter correction method according to claim 1, wherein: The scatter correction model uses a sample positioning image and sample scatter distribution estimation information as a training set, wherein the sample positioning image and the sample scatter distribution estimation information correspond to each other; or The scatter correction model uses a sample positioning image and sample scanning information as a training set, wherein the sample positioning image and the sample scanning information correspond to each other.

3. The scatter correction method according to claim 2, wherein: Determining the scatter distribution estimation information of the scanned object includes: A scanning information estimation result is determined according to the scatter correction model, and a scatter distribution estimation algorithm is used to simulate and calculate the scanning information estimation result to determine the scatter distribution estimation information, wherein the scatter distribution estimation algorithm includes at least one of the following: a Monte Carlo simulation algorithm, a scatter kernel superposition algorithm, and a deep learning algorithm.

4. The scatter correction method according to claim 1, wherein: The scanning information includes projection data and / or scanned images; The scatter distribution estimation information includes scatter distribution data corresponding to the projection data and / or a scatter distribution image corresponding to the scan image; The scatter correction model includes a first scatter correction model corresponding to the projection data and / or a second scatter correction model corresponding to the scanned image.

5. The scatter correction method according to claim 4, wherein: The determining, based on the trained scatter correction model and according to the scout image, scatter distribution estimation information of the scanned object includes: determining initial scatter distribution data corresponding to a preset number of projections based on the first scatter correction model; Actual scattering distribution data corresponding to the scanned object is determined according to the preset projection number, the initial scattering distribution data, and the actual projection number corresponding to the scanned object.

6. The scatter correction method according to claim 5, characterized in that: The determining, based on the preset number of projections, the initial scattering distribution data, and the actual number of projections corresponding to the scanned object, actual scattering distribution data corresponding to the scanned object comprises: When the actual number of projections is greater than the preset number of projections, interpolating and expanding the initial scattering distribution data to obtain the actual scattering distribution data; When the actual number of projections is less than the preset number of projections, the initial scattering distribution data is selected at intervals to obtain the actual scattering distribution data.

7. The scatter correction method according to claim 4, wherein: The determining, based on the trained scatter correction model and according to the scout image, scatter distribution estimation information of the scanned object includes: Based on the second scatter correction model, a scatter distribution image of the scanned object is acquired according to the scout image, wherein the scatter distribution image is used to perform scatter correction on the scanned image.

8. The scatter correction method according to claim 1, wherein: Before performing scatter correction on the scan information according to the scatter distribution estimation information, the method includes: Perform environmental parameter correction on the scan information.

9. The scatter correction method according to claim 1, wherein: The performing scatter correction on the scanning information according to the scatter distribution estimation information includes: An iterative calculation is performed on a difference between the scanning information and the scatter distribution estimation information to obtain scatter-corrected scanning information, wherein a convergence condition for the iterative calculation is that a difference between corresponding iteration parameters in two adjacent iterations is less than or equal to a preset convergence threshold, and the iteration parameters include the scanning information and / or the scatter distribution estimation information.

10. A scatter correction device, characterized in that: Including acquisition module, determination module and correction module: The acquisition module is configured to acquire a scout image of the scanned subject; and to acquire scanning information of the scanned subject based on the scout image. The scout image is a planar image of the examination area obtained by performing a quick scan of the subject by a medical imaging scanning system before a formal scan is performed by a CT system. The scanning information is an image obtained by performing a formal scan of the scanned subject within the scanning range by the medical scanning imaging system after the scanning range is determined based on the scout image. The determination module is configured to determine scatter distribution estimation information of the scanned object based on the trained scatter correction model and the scout image; the scatter distribution estimation information and the scan information are acquired simultaneously; The correction module is configured to perform scatter correction on the scanning information according to the scatter distribution estimation information.

Images