Dual-energy CT image generation method, apparatus, electronic device and readable storage medium

Abstract

This invention discloses a method, apparatus, electronic device, and readable storage medium for generating dual-energy CT images, belonging to the field of CT image processing technology. The dual-energy CT image generation method includes: acquiring a first energy scan image and a second energy scan image, wherein the first energy and the second energy are different; acquiring a second energy prediction image based on the first energy scan image and a prediction model, wherein the prediction model is pre-trained based on historical first energy scan images and historical second energy scan images; acquiring a second energy composite image based on the second energy prediction image and the second energy scan image; and determining the second energy composite image and the first energy scan image as a dual-energy CT image.

Description

Technical Field

[0001] This invention relates to the field of CT image processing technology, and in particular to a dual-energy CT image generation method, apparatus, electronic device, and readable storage medium. Background Technology

[0002] Computed tomography (CT) is a technique that uses X-rays to irradiate a target and then processes the images using a computer to obtain a tomographic image of the target. Dual-energy CT (DCT) acquires an image of the target by scanning it twice with different energies. Based on the specific attenuation coefficients of different materials for X-rays of different energies, DCT scans can accurately calculate the composition of different objects, thus generating accurate images of the target.

[0003] However, in dual-energy CT, there is an unavoidable time interval between two scans. During this time interval, the contrast agent flows with the blood and the normal movement of organs cause differences in the images acquired by the two scans, resulting in poor matching of images from different energy scans, which affects subsequent treatment and diagnosis. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to overcome the defect of low matching degree of dual-energy CT scan images in the prior art, and to provide a dual-energy CT image generation method, device, electronic device and readable storage medium.

[0005] The present invention solves the above-mentioned technical problems through the following technical solution:

[0006] In a first aspect, embodiments of the present invention provide a method for generating dual-energy CT images, the method comprising:

[0007] Acquire a first energy scan image and a second energy scan image, wherein the first energy and the second energy are different;

[0008] A second energy prediction image is obtained based on the first energy scan image and the prediction model, wherein the prediction model is pre-trained based on the first energy historical scan image and the second energy historical scan image;

[0009] A second energy composite image is obtained based on the second energy prediction image and the second energy scan image, and the second energy composite image and the first energy scan image are determined as dual-energy CT images.

[0010] In one embodiment, obtaining the second energy composite image based on the second energy prediction image and the second energy scan image includes:

[0011] Compare the second energy prediction image and the second energy scan image to obtain inconsistent pixels;

[0012] A fusion mask is obtained based on the non-consistent pixels;

[0013] The second energy scan image and the second energy prediction image are fused using the fusion mask to obtain the second energy composite image.

[0014] In one embodiment, the step of comparing the second energy prediction image and the second energy scan image to obtain non-consistent pixels includes:

[0015] Pixels whose absolute difference between the CT values ​​in the second energy prediction image and the second energy scan image is greater than or equal to a set threshold are defined as non-consistent pixels.

[0016] In one embodiment, determining the fusion mask based on the non-consistent pixels includes:

[0017] Based on the pixels in the first energy scan image that are at the same position as the non-consistent pixels, the non-consistent pixels are divided into prediction error pixels and scan error pixels;

[0018] The mask is generated based on the prediction error pixels and the scanning error pixels.

[0019] In one embodiment, the method further includes:

[0020] Training data was selected from the first energy history scan image and the second energy history scan image;

[0021] The model is trained based on the training data to obtain the prediction model.

[0022] In one embodiment, the step of filtering training data from the first energy history scan image and the second energy history scan image includes:

[0023] Image matching is performed on the first energy history scan image and the second energy history scan image;

[0024] The first energy history scan image and the second energy history scan image that meet the matching criteria are used as the training data.

[0025] In one embodiment, the step of filtering training data from the first energy history scan image and the second energy history scan image includes:

[0026] Image matching is performed on the first energy history scan image and the second energy history scan image;

[0027] The first energy historical scan image and the second energy historical scan image that do not meet the matching criteria are registered, and the registered images are then matched a second time.

[0028] The first energy history image and the second energy history image that meet the secondary image matching criteria are used as the training data.

[0029] Secondly, embodiments of the present invention provide a dual-energy CT image generation device, the device comprising:

[0030] The first acquisition module is used to acquire a first energy scan image and a second energy scan image, wherein the first energy and the second energy are different.

[0031] The second acquisition module is used to acquire a second energy prediction image based on the first energy scan image and the prediction model, wherein the prediction model is pre-trained based on the first energy historical scan image and the second energy historical scan image.

[0032] The determination module is used to obtain a second energy composite image based on the second energy prediction image and the second energy scan image, and to determine the second energy composite image and the first energy scan image as dual-energy CT images.

[0033] Thirdly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the dual-energy CT image generation method as described in the first aspect above.

[0034] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the dual-energy CT image generation method provided in the first aspect.

[0035] The positive and progressive effects of this invention are as follows:

[0036] The dual-energy CT image generation method provided in this invention predicts a second energy prediction image from a first energy scan image to avoid structural differences caused by the time difference between the two scans. The second energy prediction image is then corrected using the second energy scan image to obtain a second energy composite image, ensuring its accuracy. Furthermore, a dual-energy image is formed from the first energy scan image and the second energy composite image, providing an accurate reference for subsequent treatment. Attached Figure Description

[0037] Figure 1 This is a flowchart illustrating a dual-energy CT image generation method according to an exemplary embodiment;

[0038] Figure 2 This is a flowchart illustrating step S103 according to an exemplary embodiment;

[0039] Figure 3 This is a flowchart illustrating step S1032 according to an exemplary embodiment;

[0040] Figure 4 This is a flowchart illustrating a dual-energy CT image generation method according to another exemplary embodiment;

[0041] Figure 5 This is a flowchart illustrating step S104 according to an exemplary embodiment;

[0042] Figure 6 This is a block diagram of a dual-energy CT image generation apparatus according to an exemplary embodiment;

[0043] Figure 7 This is a block diagram illustrating a determining module according to an exemplary embodiment;

[0044] Figure 8 This is a block diagram illustrating an acquisition unit according to an exemplary embodiment;

[0045] Figure 9 This is a block diagram of a dual-energy CT image generation apparatus according to another exemplary embodiment;

[0046] Figure 10 This is a block diagram illustrating a filtering module according to an exemplary embodiment;

[0047] Figure 11 This is a schematic diagram illustrating the structure of an electronic device according to an exemplary embodiment. Detailed Implementation

[0048] The present invention will be further illustrated by way of embodiments below, but the present invention is not limited to the scope of the embodiments described herein.

[0049] The terminology used in this disclosure is for the purpose of describing particular embodiments only and is not intended to be limiting of the disclosure. Unless otherwise defined, the technical or scientific terms used in this disclosure should be understood in their ordinary sense by one of ordinary skill in the art to which this disclosure pertains. The words “a” or “one” and similar terms used in this disclosure and the claims do not indicate a limitation of quantity, but rather indicate the presence of at least one. Unless otherwise stated, the words “comprising” or “including” and similar terms mean that the elements or objects preceding “comprising” or “including” encompass the elements or objects listed following “comprising” or “including” and their equivalents, and do not exclude other elements or objects. The words “connected” or “linked” and similar terms are not limited to physical or mechanical connections and can include electrical connections, whether direct or indirect.

[0050] The singular forms “a,” “the,” and “the” used in this disclosure and the appended claims are also intended to include the plural forms, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0051] Among related technologies, dual-source scanning, rapid kV switching scanning, and dual-layer detectors have limited the clinical application of dual-energy CT imaging due to their high cost. Slow kV switching scanning, which requires two scans, not only imposes additional doses on patients, but also results in low image matching between the two scans due to unavoidable contrast agent flow and organ movement during the time difference between the two acquisitions, significantly impacting post-imaging clinical diagnosis.

[0052] To address the aforementioned problems, embodiments of the present invention provide a dual-energy CT image generation method, achieving a balance between high matching accuracy and low hardware cost in dual-energy CT image generation. The technical solution provided by these embodiments is applicable to various CT geometries, including but not limited to parallel-beam and cone-beam scanning. Furthermore, the technical solution provided by these embodiments is applicable to various CT scanning modes, including but not limited to plain tomography and spiral scanning. The embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0053] Example 1

[0054] Figure 1 This is a flowchart illustrating a dual-energy CT image generation method according to an exemplary embodiment. Figure 1 As shown, the method includes:

[0055] Step S101: Obtain a first energy scan image and a second energy scan image, wherein the first energy and the second energy are different.

[0056] The first and second energy scan images are scanned images of the same object under different energy spectrum X-rays. Different energy spectrum X-rays are achieved by adjusting the tube voltage of the X-ray tube. Due to the time difference between the two scans, the flow of contrast agent and organ movement during this time difference inevitably cause structural differences between the first and second energy scan images.

[0057] Optionally, the first energy is lower than the second energy. In this case, the target region in the first energy scan image obtained using a relatively low-energy X-ray scan exhibits greater attenuation, making it easier to identify different tissues or sites and providing accurate input for image prediction in subsequent steps.

[0058] Optionally, a second energy scan image is acquired using sparse acquisition and low-dose acquisition methods to reduce the radiation dose received by the scanned object. Sparse acquisition refers to projective scanning at larger intervals, increasing the acquisition rate and thus reducing the radiation dose received by the scanned object. Furthermore, the scan dose is also related to the tube current of the X-ray tube; the scan dose can be changed by adjusting the tube current. In step S101, low-dose acquisition is achieved by reducing the tube current of the X-ray tube.

[0059] Step S102: Obtain a second energy prediction image based on the first energy scan image and the prediction model, wherein the prediction model is pre-trained based on the first energy historical scan image and the second energy historical scan image.

[0060] In conjunction with step S101, when the first energy is low, the prediction accuracy of the second energy prediction image obtained from the first energy scan image is higher. Optionally, in step S102, the prediction model is trained using deep learning based on the first energy historical scan image and the second energy historical scan image. The specific method for obtaining this prediction model will be described in detail below.

[0061] Step S103: Obtain a second energy composite image based on the second energy prediction image and the second energy scan image, and determine the second energy composite image and the first energy scan image as a dual-energy CT image.

[0062] Due to physical limitations, the first energy image and the second energy image do not satisfy a perfect mapping relationship. Therefore, the second energy prediction image obtained solely from the first energy scan image contains uncertainties and error risks. In step S103, the second energy scan image is used to assist in correcting the second energy prediction image to obtain a second energy composite image, thereby optimizing the accuracy and effectiveness of the finally acquired dual-energy CT image.

[0063] In summary, the dual-energy CT image generation method provided in this invention predicts a second energy prediction image using a first energy scan image to avoid structural differences caused by the time difference between the two scans. The second energy prediction image is then corrected using the second energy scan image to obtain a second energy composite image, ensuring its accuracy. Furthermore, a dual-energy image is formed from the first energy scan image and the second energy composite image, providing an accurate reference for subsequent treatment.

[0064] In one example, step S103 is implemented in the following way. Figure 2 This is a flowchart illustrating step S103 according to an exemplary embodiment. For example... Figure 2 As shown, step S103 includes:

[0065] Step S1031: Compare the second energy prediction image and the second energy scan image to obtain non-consistent pixels.

[0066] Optionally, step S1031 specifically identifies pixels whose absolute difference between the CT values ​​in the second energy prediction image and the second energy scan image is greater than or equal to a set threshold as non-consistent pixels. Since the CT values ​​of the same substance in the second energy prediction image and the second energy scan image are consistent or not significantly different, the CT value is used as the criterion to determine the different pixels in the two images.

[0067] Step S1032: Obtain a fusion mask based on non-consistent pixels.

[0068] Further, a fusion mask is determined based on the non-uniform pixels identified in step S1032. Optionally, Figure 3 This is a flowchart illustrating step S1032 according to an exemplary embodiment. For example... Figure 3 As shown, step S1032 includes:

[0069] Step S301: Based on the pixels in the first energy scan image that are at the same position as the non-consistent pixels, the non-consistent pixels are divided into prediction error pixels and scan error pixels.

[0070] Among them, prediction error pixels refer to non-consistent pixels caused by prediction errors, while scan error pixels refer to non-consistent pixels caused by organ movement and contrast agent flow. Scan error pixels are the unavoidable pixel differences between two scans.

[0071] The first energy scan image and the second energy scan image are two actual image scans, and scanning errors are inevitable. The second energy prediction image is predicted based on the first energy scan image. Therefore, the tissue structure in the second energy prediction image and the first energy scan image is the same, but the CT values ​​of the same pixels are different. In this case, in step S301, the scanning error pixels caused by structural differences can be determined based on the first energy scan image and the second energy scan image, and then the non-consistent pixels are divided into scanning error pixels and prediction error pixels.

[0072] Optionally, when determining the scanning error pixels, the geometric and anatomical structure information in the first energy scan image, the second energy scan image, and the second energy prediction image are used as references, and the scanning error pixels are determined by methods such as image matching (e.g., image matching based on grayscale gradient).

[0073] Step S302: Generate a mask based on the prediction error pixels and the scanning error pixels.

[0074] The mask is a matrix of the same size as the first and second energy scan images. Regarding the mask implementation, for example, the values ​​in the mask are either 0 or 1. Specifically, for consistent pixels and inconsistent pixels classified as scan error pixels, the corresponding value in the mask is 1. In this case, the final second energy composite image uses the pixel values ​​of the second energy prediction image. For inconsistent pixels not classified as prediction errors, the corresponding value in the mask is 0. In this case, the final second energy composite image uses the pixel values ​​of the second energy scan image. For example, the values ​​in the mask are greater than or equal to 0 and less than or equal to 1. The image quality of the second energy composite image obtained in this way is better, especially for the more realistic image synthesis effect in areas with inconsistent pixels.

[0075] Continuing with step 2, after step S1032, step S1033 is executed, as follows:

[0076] Step S1033: Perform image fusion on the second energy scan image and the second energy prediction image according to the fusion mask to obtain the second energy composite image.

[0077] Based on the above explanation of step S302, the second energy synthesis image is obtained in the following way:

[0078] Image_synthetic=Image_predict×mask+Image_scan×(1–mask)

[0079] Wherein, Image_synthetic is the second energy synthesized image, Image_predict is the second energy predicted image, Image_scan is the second energy scan image, and mask is the mask.

[0080] In summary, dual-energy CT images were generated through steps S101 to S103. Combining this invention's complete dual-energy CT image generation method, the reason for reducing radiation dose in step S101 by decreasing the X-ray tube current is explained again.

[0081] Typically, radiation dose is reduced during CT scans by decreasing the tube current of the X-ray tube. However, reducing the tube current can increase image noise, affecting image quality. Therefore, in dual-energy CT image generation methods provided by related technologies, the tube voltage of dual-energy scanning needs to be increased to ensure image quality, which in turn leads to an increase in radiation dose.

[0082] In this embodiment of the invention, a second energy prediction image is predicted using a first energy scan image, and the second energy prediction image is optimized using the second energy scan image to improve the accuracy of the generated dual-energy CT image. Therefore, the requirements for image noise are reduced for the first and second energy scan images obtained in step S101. Under such circumstances, a lower tube current can be used when scanning the image in step S101, thereby achieving the goal of reducing radiation dose.

[0083] Figure 4 This is a flowchart illustrating a dual-energy CT image generation method according to another exemplary embodiment. Figure 4 As shown, the method further includes:

[0084] Step S104: Select training data from the first energy history scan image and the second energy history scan image.

[0085] By filtering and optimizing the training data, the prediction model trained based on the training data can more accurately predict the second energy scan image based on the first energy scan image.

[0086] As an example, Figure 5 This is a flowchart illustrating step S104 according to an exemplary embodiment. For example... Figure 5 As shown, step S104 includes:

[0087] Step S1041: Perform image matching on the first energy history scan image and the second energy history scan image.

[0088] Step S1042: Use the first energy history scan image and the second energy history scan image that meet the matching criteria as training data.

[0089] Step S1043: Perform registration processing on the first energy historical scan image and the second energy historical scan image that do not meet the matching criteria, and perform secondary image matching on the registered images.

[0090] Step S1044: Use the first energy history image and the second energy history image that meet the secondary image matching criteria as training data.

[0091] In step S1041, the first energy historical scan image and the second energy historical scan image are clinical data or phantom data. The acquisition methods include, but are not limited to, plain tomography and spiral scanning. Furthermore, the specific method for image matching is not limited; for example, image matching can be performed using structural similarity, cross-correlation coefficients as references, and / or based on grayscale gradients.

[0092] In steps S1042 and S1043, the criteria for successful matching are determined based on the specific image matching method used. For example, a structural similarity greater than or equal to a set threshold is considered a successful match, while a structural similarity less than the set threshold is considered a failed match.

[0093] In step S1042, the first and second energy historical scan images that meet the matching criteria are used as training data, which optimizes the effectiveness of the training data and makes the prediction model accurately reflect the relationship between the first and second energy historical scan images that match.

[0094] Steps S1043 and S1044 perform registration processing on the first and second energy historical scan images that do not meet the matching criteria, thereby improving the matching degree between the first and second energy historical scan images. By using this method, the first and second energy historical scan images that meet the matching criteria after registration processing are also used as training data. This lowers the selection criteria for the training dataset, making it easier to collect a sufficient amount of training data and preventing the trained prediction model from being too idealistic, thus more closely reflecting the data obtained in real-world applications.

[0095] Continue to refer to Figure 4 After step S104, step S105 is executed, as follows.

[0096] Step S105: Train the model based on the training data to obtain the prediction model.

[0097] Optionally, a deep learning model is used to obtain the prediction model. For example, the first energy historical scan image in the training data is used as the input data for model training, and the second energy historical scan image in the training data is used as the gold standard for model training. A loss function is constructed using this gold standard and the image output by the model. During model training, the prediction model is determined by the loss function converging to a preset threshold. The deep learning network used in step S105 includes, but is not limited to, Transformer, CNN, and GAN. Furthermore, other algorithms can also be used in step S105 to obtain the prediction model based on the training data; this embodiment of the invention does not impose specific limitations.

[0098] In summary, the dual-energy CT image generation method provided by this invention predicts a second energy prediction image from a first energy scan image to avoid structural differences caused by the time difference between the two scans. The second energy prediction image is corrected using the second energy scan image to obtain a second energy composite image, ensuring the accuracy and reliability of the second energy composite image. Furthermore, a dual-energy image is formed from the first energy scan image and the second energy composite image, providing an accurate reference for subsequent treatment. Both scans can be performed with radiation doses lower than those in related technologies, reducing the radiation dose received by the scanned object. Moreover, the dual-energy CT image generation method provided by this invention has low hardware costs, facilitating the implementation and promotion of dual-energy CT scanning.

[0099] Example 2

[0100] Figure 6 This is a block diagram illustrating a dual-energy CT image generation apparatus according to an exemplary embodiment. Figure 6 As shown, the device includes: a first acquisition module 610, a second acquisition module 620, and a determination module 630.

[0101] The first acquisition module 610 is used to acquire a first energy scan image and a second energy scan image, wherein the first energy and the second energy are different.

[0102] The second acquisition module 620 is used to acquire a second energy prediction image based on the first energy scan image and the prediction model. The prediction model is pre-trained based on the first energy historical scan image and the second energy historical scan image.

[0103] The determination module 630 is used to obtain a second energy composite image based on the second energy prediction image and the second energy scan image, and to determine the second energy composite image and the first energy scan image as a dual-energy CT image.

[0104] In one embodiment, Figure 7 This is a block diagram illustrating a determining module according to an exemplary embodiment. Figure 7As shown, the determining module 630 includes: a comparison unit 631, an acquisition unit 632, and a fusion unit 633.

[0105] The comparison unit 631 is used to compare the second energy prediction image and the second energy scan image to obtain inconsistent pixels.

[0106] The acquisition unit 632 is used to acquire a fusion mask based on non-uniform pixels.

[0107] The fusion unit 633 is used to perform image fusion on the second energy scan image and the second energy prediction image according to the fusion mask to obtain the second energy composite image.

[0108] In one embodiment, the comparison unit 631 is specifically used to identify pixels whose absolute difference between the CT values ​​in the second energy prediction image and the second energy scan image is greater than or equal to a set threshold as non-consistent pixels.

[0109] In one embodiment, Figure 8 This is a block diagram illustrating an acquisition unit according to an exemplary embodiment. For example... Figure 8 As shown, the acquisition unit 632 includes: a division subunit 6321 and a generation subunit 6322.

[0110] The dividing subunit 6321 is used to divide the non-consistent pixels into prediction error pixels and scanning error pixels based on the pixels in the first energy scan image that are at the same position as the non-consistent pixels.

[0111] The generation subunit 6322 is used to generate a mask based on the prediction error pixels and the scanning error pixels.

[0112] In one embodiment, Figure 9 This is a block diagram of a dual-energy CT image generation apparatus according to another exemplary embodiment. Figure 9 As shown, the device also includes a screening module 640 and a training module 650.

[0113] The filtering module 640 is used to filter training data from the first energy history scan image and the second energy history scan image.

[0114] The training module 650 is used to train the model based on the training data to obtain the prediction model.

[0115] In one embodiment, Figure 10 This is a block diagram illustrating a filtering module according to an exemplary embodiment. For example... Figure 10 As shown, the filtering module 640 includes an image matching unit 641 and a first determining unit 642.

[0116] The image matching unit 641 is used to perform image matching on the first energy history scan image and the second energy history scan image.

[0117] The first determining unit 642 is used to use the first energy history scan image and the second energy history scan image that meet the matching criteria as training data.

[0118] In one embodiment, the screening module 640 further includes: a registration processing unit 643 and a second determination unit 644.

[0119] The registration processing unit 643 is used to perform registration processing on the first energy historical scan image and the second energy historical scan image that do not meet the matching standards, and to perform secondary image matching on the registered image.

[0120] The second determining unit 644 is used to use the first energy history image and the second energy history image that meet the secondary image matching criteria as training data.

[0121] In summary, the dual-energy CT image generation apparatus provided in this invention predicts a second energy prediction image from a first energy scan image to avoid structural differences caused by the time difference between the two scans. The second energy prediction image is corrected using the second energy scan image to obtain a second energy composite image, ensuring the accuracy and reliability of the second energy composite image. Furthermore, a dual-energy image is formed from the first energy scan image and the second energy composite image, providing an accurate reference for subsequent treatment. Both scans can be performed with radiation doses lower than those in related technologies, reducing the radiation dose received by the scanned object. Moreover, the dual-energy CT image generation method provided in this invention has low hardware costs, facilitating the implementation and promotion of dual-energy CT scanning.

[0122] Example 3

[0123] This invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the dual-energy CT image generation method as provided in Embodiment 1.

[0124] Optionally, the electronic device is an electronic device that works in conjunction with a CT scanning device to generate dual-energy CT images based on images acquired by the CT scanning device. Optionally, the electronic device is a CT scanning device.

[0125] Figure 11 This is a schematic diagram of the structure of an electronic device provided in this embodiment. The electronic device includes at least one processor and a memory communicatively connected to the at least one processor. The memory stores a computer program that can be executed by the at least one processor, and the computer program is executed by the at least one processor to enable the at least one processor to perform the dual-energy CT image generation method of Embodiment 1. Figure 11The electronic device 3 shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of the present invention.

[0126] The components of the electronic device 3 may include, but are not limited to: at least one processor 4, at least one memory 5, and a bus 6 connecting different system components (including memory 5 and processor 4).

[0127] Bus 6 includes a data bus, an address bus, and a control bus.

[0128] The memory 5 may include volatile memory, such as random access memory (RAM) 51 and / or cache memory 52, and may further include read-only memory (ROM) 53.

[0129] The memory 5 may also include a program / utility 55 having a set (at least one) of program modules 54, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0130] The processor 4 executes various functional applications and data processing by running computer programs stored in the memory 5, such as the dual-energy CT image generation method described above.

[0131] Electronic device 3 can also communicate with one or more external devices 7 (e.g., keyboard, pointing device, etc.). This communication can be performed through input / output (I / O) interface 8. Furthermore, electronic device 3 can also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN), and / or public networks, such as the Internet) via network adapter 9. Figure 11 As shown, network adapter 9 communicates with other modules of electronic device 3 via bus 6. It should be understood that, although... Figure 11 Not shown, it can be combined with electronic device 3 to use other hardware and / or software modules, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0132] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

[0133] Example 4

[0134] This invention provides a computer-readable storage medium storing a computer program thereon, characterized in that the computer program, when executed by a processor, implements the dual-energy CT image generation method as described in Embodiment 1.

[0135] The readable storage medium may be more specifically adopted, including but not limited to: portable disk, hard disk, random access memory, read-only memory, erasable programmable read-only memory, optical storage device, magnetic storage device, or any suitable combination thereof.

[0136] In a possible implementation, the present invention can also be implemented as a program product comprising program code, which, when the program product is run on a terminal device, causes the terminal device to perform the steps in the pulse ablation region prediction method provided in the second aspect above.

[0137] The program code for executing the present invention can be written in any combination of one or more programming languages. The program code can be executed entirely on the user device, partially on the user device, as a standalone software package, partially on the user device and partially on a remote device, or entirely on a remote device.

[0138] While specific embodiments of the present invention have been described above, those skilled in the art should understand that these are merely illustrative examples, and the scope of protection of the present invention is defined by the appended claims. Those skilled in the art can make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but all such changes and modifications fall within the scope of protection of the present invention.

Claims

1. A method for generating dual-energy CT images, characterized in that, The method includes: Acquire a first energy scan image and a second energy scan image, wherein the first energy and the second energy are different; A second energy prediction image is obtained based on the first energy scan image and the prediction model, wherein the prediction model is pre-trained based on the first energy historical scan image and the second energy historical scan image; Compare the second energy prediction image and the second energy scan image to obtain inconsistent pixels; Based on the pixels in the first energy scan image that are at the same position as the non-consistent pixels, the non-consistent pixels are divided into prediction error pixels and scan error pixels; A fusion mask is generated based on the predicted error pixels and the scan error pixels; The second energy scan image and the second energy prediction image are fused according to the fusion mask to obtain the second energy composite image; The second energy composite image and the first energy scan image are identified as dual-energy CT images.

2. The method according to claim 1, characterized in that, The step of comparing the second energy prediction image and the second energy scan image to obtain non-consistent pixels includes: Pixels whose absolute difference between the CT values ​​in the second energy prediction image and the second energy scan image is greater than or equal to a set threshold are defined as non-consistent pixels.

3. The method according to claim 1, characterized in that, The method further includes: Training data was selected from the first energy history scan image and the second energy history scan image; The model is trained based on the training data to obtain the prediction model.

4. The method according to claim 3, characterized in that, The process of selecting training data from the first energy history scan image and the second energy history scan image includes: Image matching is performed on the first energy history scan image and the second energy history scan image; The first energy history scan image and the second energy history scan image that meet the matching criteria are used as the training data.

5. The method according to claim 3, characterized in that, The process of selecting training data from the first energy history scan image and the second energy history scan image includes: Image matching is performed on the first energy history scan image and the second energy history scan image; The first energy historical scan image and the second energy historical scan image that do not meet the matching criteria are registered, and the registered images are then matched a second time. The first energy history scan image and the second energy history scan image that meet the secondary image matching criteria are used as the... Training data.

6. A dual-energy CT image generation device, characterized in that, The device includes: The first acquisition module is used to acquire a first energy scan image and a second energy scan image, wherein the first energy and the second energy are different. The second acquisition module is used to acquire a second energy prediction image based on the first energy scan image and the prediction model, wherein the prediction model is pre-trained based on the first energy historical scan image and the second energy historical scan image. The determination module includes a comparison unit, an acquisition unit, and a fusion unit. The acquisition unit includes a segmentation subunit and a generation subunit. The comparison unit is used to compare a second energy prediction image and a second energy scan image to obtain non-consistent pixels. The segmentation subunit is used to divide non-consistent pixels into prediction error pixels and scan error pixels based on pixels in the first energy scan image that are at the same position as the non-consistent pixels. The generation subunit is used to generate a fusion mask based on the prediction error pixels and scan error pixels. The fusion unit is used to perform image fusion on the second energy scan image and the second energy prediction image based on the fusion mask to obtain a second energy composite image. The determining module is further configured to determine the second energy composite image and the first energy scan image as dual-energy CT images.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the dual-energy CT image generation method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the dual-energy CT image generation method as described in any one of claims 1-5.

Images