Dual-energy CBCT hardening artifact correction method, device and equipment and storage medium

Abstract

The invention relates to the field of image processing, and discloses a dual-energy CBCT (cone beam computed tomography) hardening artifact correction method which comprises the following steps: acquiring a dual-energy original projection image of a scanned object, performing dual-energy image decomposition on the dual-energy original projection image, combining the decomposed images to obtain a virtual single-energy image, and performing CBCT hardening artifact correction on the virtual single-energy image. And calculating a loss value of a preset correction training model by using the virtual single-energy image, optimizing parameters of the correction training model according to the loss value to obtain a dual-energy CBCT hardening artifact correction model, and carrying out deconvolution on the obtained hardening artifact image by using the dual-energy CBCT hardening artifact correction model to obtain a reconstructed image after hardening artifact suppression. The quality of a dual-energy CBCT image generated through CBCT scanning can be improved, and the accuracy of weakening hardening artifacts of the dual-energy CBCT image can be improved.

Description

Technical Field

[0001] The present application relates to the field of image processing, and in particular to a dual-energy CBCT hardening artifact correction method, device, electronic device and computer-readable storage medium. Background Art

[0002] When X-rays penetrate the human body, the main effects of X-rays on matter are different depending on the energy, resulting in different image quality in the final CBCT imaging. Classical analytical CT reconstruction requires that the projection data is completely linear, which means that the average energy of the X-ray beam must remain constant as it passes through the entire object. However, in reality, since X-rays have stronger penetrating power, the average energy of the beam gradually increases as it penetrates deeper into matter. This phenomenon is called the ray hardening effect. In the image results of CBCT (cone beam computed tomography), this effect often leads to the appearance of strip-like artifacts around high-density areas, namely hardening artifacts, which have an adverse effect on image quality.

[0003] The dual-energy CBCT images generated by CBCT scanning in the prior art have low quality, and the accuracy of reducing hardening artifacts in the dual-energy CBCT images is low. Summary of the invention

[0004] The present application provides a dual-energy CBCT hardening artifact correction method, device, equipment and computer-readable storage medium, the main purpose of which is to solve the problem that the dual-energy CBCT image generated by CBCT scanning has low quality and low accuracy in reducing the hardening artifacts of the dual-energy CBCT image.

[0005] To achieve the above-mentioned purpose, the present application provides a dual-energy CBCT hardening artifact correction method, comprising:

[0006] Acquire a dual-energy original projection image of the scanned object;

[0007] Performing dual-energy image decomposition on the dual-energy original projection image, and combining the decomposed images to obtain a virtual mono-energy image;

[0008] Calculating a loss value of a preset correction training model using the virtual monoenergetic image;

[0009] Optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model;

[0010] The dual-energy CBCT hardening artifact correction model is used to perform deconvolution on the acquired hardening artifact image to obtain a reconstructed image after hardening artifact suppression.

[0011] In some embodiments, performing dual-energy image decomposition on the dual-energy original projection image and combining the decomposed images to obtain a virtual single-energy image includes:

[0012] Performing FDK reconstruction on the dual-energy original projection image to obtain a dual-energy reconstructed image;

[0013] Decomposing the dual-energy reconstructed image into a dual-energy image to obtain a high-energy image and a low-energy image;

[0014] Performing forward projection processing on the high-energy image and the low-energy image to obtain a projection result;

[0015] Recombining the high-energy image and the low-energy image according to the projection results to obtain a single-energy combined image;

[0016] The monoenergetic combined image is enhanced to obtain a virtual monoenergetic image.

[0017] In some embodiments, performing forward projection processing on the high-energy image and the low-energy image to obtain a projection result includes:

[0018] Acquire voxel matrices of the high-energy image and the low-energy image;

[0019] Get the source point position and the detector plane position;

[0020] Calculating a first distance from each voxel in the voxel matrix to the source point position and a second distance from each voxel in the voxel matrix to the detector plane position;

[0021] Calculate the attenuation of the radioactive isotope passing through each voxel in the voxel matrix according to the first distance and the second distance;

[0022] Attenuation correction is performed on each voxel in the voxel matrix according to the attenuation amount, and the projection value after attenuation correction is projected onto the detector plane to form a projection result.

[0023] In some embodiments, the recombining the high-energy image and the low-energy image according to the projection result to obtain a single-energy combined image includes:

[0024] Aligning each voxel in the high-energy image and the low-energy image in spatial position;

[0025] Performing energy correction on the aligned high-energy image and the low-energy image;

[0026] The corrected high-energy image and low-energy image are weighted combined to obtain a single-energy combined image.

[0027] In some embodiments, the step of performing image enhancement on the single-energy combined image to obtain a virtual single-energy image includes:

[0028] Calculating the minimum and maximum pixel values ​​of the monoenergetic combined image;

[0029] Calculate a parameter gain factor and an offset according to the minimum value and the maximum value;

[0030] Calculate and update pixel value according to the gain factor and the offset;

[0031] The updated pixel values ​​are used to replace the pixel values ​​in the monoenergetic combined image to obtain a virtual monoenergetic image.

[0032] In some embodiments, optimizing the correction training model parameters according to the loss value to obtain the dual-energy CBCT hardening artifact correction model includes:

[0033] Calculate the gradient of the loss value to the correction training model parameters using a back propagation algorithm;

[0034] Updating parameters of the correction training model according to the gradient;

[0035] Determining whether the loss value is less than a preset loss value;

[0036] If the loss value is greater than a preset loss value, then continuing to perform hardening artifact correction on the virtual monoenergetic image;

[0037] If the loss value is less than or equal to the preset loss value, the correction training model parameter optimization is completed, and the dual-energy CBCT hardening artifact correction model is obtained.

[0038] In some embodiments, the step of deconvolving the acquired hardening artifact image using the dual-energy CBCT hardening artifact correction model to obtain a reconstructed image after hardening artifact suppression includes:

[0039] acquiring a pre-processed hardening artifact image;

[0040] converting the hardening artifact image into an image pixel matrix;

[0041] The image pixel matrix is ​​convolved using a dual-energy CBCT hardening artifact correction network model, and a feature channel is obtained through a ReLU function;

[0042] Performing maximum pooling on the feature channels to obtain a pooling matrix;

[0043] Filling the pooling matrix with null values ​​according to a preset size to obtain a filled matrix;

[0044] Setting a sliding step size according to a preset deconvolution kernel size, and performing sliding coverage on the filling matrix according to the sliding step size;

[0045] Performing a dot product sum on the pixel points covered by sliding in the filling matrix and a preset deconvolution kernel, and obtaining a deconvolution matrix according to the dot product sum;

[0046] The deconvolution matrix is ​​convolved again to obtain a reconstructed image with hardening artifacts suppressed.

[0047] In order to solve the above problems, the present invention also provides a dual-energy CBCT hardening artifact correction device, the device comprising:

[0048] An image acquisition module, used for acquiring a dual-energy original projection image of a scanned object;

[0049] An image preprocessing module, used for performing dual-energy image decomposition on the dual-energy original projection image, and combining the decomposed images to obtain a virtual mono-energy image;

[0050] A loss value calculation module, used to calculate the loss value of a preset correction training model using the virtual monoenergetic image;

[0051] A model optimization module, used for optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model;

[0052] The image convolution module is used to deconvolve the acquired hardening artifact image using the dual-energy CBCT hardening artifact correction model to obtain a reconstructed image after the hardening artifact is suppressed.

[0053] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0054] at least one processor; and,

[0055] a memory communicatively connected to the at least one processor; wherein,

[0056] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the dual-energy CBCT hardening artifact correction method described above.

[0057] In order to solve the above problem, the present invention also provides a computer-readable storage medium, in which at least one computer program is stored. The at least one computer program is executed by a processor in an electronic device to implement the above-mentioned dual-energy CBCT hardening artifact correction method. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A schematic diagram of a flow chart of a dual-energy CBCT hardening artifact correction method provided in one embodiment of the present application;

[0059] Figure 2 A functional module diagram of a dual-energy CBCT hardening artifact correction device provided in one embodiment of the present application;

[0060] Figure 3 A schematic diagram of the structure of an electronic device for implementing the dual-energy CBCT hardening artifact correction method provided in one embodiment of the present application.

[0061] The realization of the purpose, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0062] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0063] The embodiment of the present application provides a dual-energy CBCT hardening artifact correction method. The execution subject of the dual-energy CBCT hardening artifact correction method includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided in the embodiment of the present application. In other words, the dual-energy CBCT hardening artifact correction method can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc. The server can be an independent server, or it can be a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and big data and artificial intelligence platforms.

[0064] Reference Figure 1 FIG. 1 is a flow chart of a dual-energy CBCT hardening artifact correction method provided in an embodiment of the present application. In this embodiment, a dual-energy CBCT hardening artifact correction method includes:

[0065] S1. Obtaining a dual-energy original projection image of a scanned object.

[0066] In the embodiment of the present invention, the same scan object is scanned by a CBCT device to obtain a dual-energy original projection image, and the dual-energy original projection image includes: a low-energy original projection image and a high-energy original projection image.

[0067] Dual-energy CBCT is a cone-beam CT technology that uses X-ray beams of different energies for scanning and obtains quantitative information on material composition through algorithm processing. This technology scans the target by emitting two X-rays of different energies (usually high energy and low energy). By acquiring low-energy and high-energy original projection images of the same scanned object through CBCT equipment, more comprehensive and detailed information can be obtained. Low-energy projection images help improve the resolution and contrast of soft tissue structures, while high-energy projection images are more suitable for detecting and analyzing bone structure and density changes, providing more accurate data support for subsequent model hardening artifact suppression.

[0068] S2. Perform dual-energy image decomposition on the dual-energy original projection image, and combine the decomposed images to obtain a virtual mono-energy image.

[0069] In an embodiment of the present invention, performing dual-energy image decomposition on the dual-energy original projection image and combining the decomposed images to obtain a virtual mono-energy image includes:

[0070] Performing FDK reconstruction on the dual-energy original projection image to obtain a dual-energy reconstructed image;

[0071] Decomposing the dual-energy reconstructed image into a dual-energy image to obtain a high-energy image and a low-energy image;

[0072] Performing forward projection processing on the high-energy image and the low-energy image to obtain a projection result;

[0073] Recombining the high-energy image and the low-energy image according to the projection results to obtain a single-energy combined image;

[0074] The monoenergetic combined image is enhanced to obtain a virtual monoenergetic image.

[0075] In detail, the FDK algorithm is mainly divided into two steps: back projection and filtering. Back projection is to project the X-ray data back to the original image space, and filtering is to filter the back projection results to remove artifacts and noise, and finally obtain a dual-energy reconstructed image. The dual-energy reconstructed image is scanned by two X-ray sources with different energies to obtain the X-ray imaging data of the dual-energy reconstructed image, and the material decomposition model is used to calculate the X-ray imaging data to decompose the high-energy image and the low-energy image.

[0076] In the embodiment of the present invention, the forward projection processing is performed on the high-energy image and the low-energy image to obtain a projection result, including:

[0077] Acquire voxel matrices of the high-energy image and the low-energy image;

[0078] Get the source point position and the detector plane position;

[0079] Calculating a first distance from each voxel in the voxel matrix to the source point position and a second distance from each voxel in the voxel matrix to the detector plane position;

[0080] Calculate the attenuation of the radioactive isotope passing through each voxel in the voxel matrix according to the first distance and the second distance;

[0081] Attenuation correction is performed on each voxel in the voxel matrix according to the attenuation amount, and the projection value after attenuation correction is projected onto the detector plane to form a projection result.

[0082] In detail, each voxel in the voxel matrix has a grayscale value, which reflects the density or absorption characteristics of the voxel point. Forward projection processing can convert the high-energy image and the low-energy image into corresponding projection results. Projection processing can help correct artifacts and errors in the imaging process, provide more comprehensive material density information, reduce common problems such as metal artifacts, and projections at different energies can provide more accurate functional and metabolic information through combination.

[0083] In an embodiment of the present invention, the step of recombining the high-energy image and the low-energy image according to the projection result to obtain a single-energy combined image includes:

[0084] Aligning each voxel in the high-energy image and the low-energy image in spatial position;

[0085] Performing energy correction on the aligned high-energy image and the low-energy image;

[0086] The corrected high-energy image and low-energy image are weighted combined to obtain a single-energy combined image.

[0087] In detail, the weighted combination assigns different weight coefficients to the high-energy image and the low-energy image based on the quality and required features of the corrected high-energy image and the low-energy image. The formula of the weighted combination is as follows:

[0088] P=w1×P h +w2×P l

[0089] Where P represents the pixel value in the single-energy combined image, w1 represents the weight coefficient of the high-energy image, w2 represents the weight coefficient of the low-energy image, and P h Represents the pixel value in the high-energy image, P l Represents the pixel value in the low-energy image.

[0090] Single-energy combined images integrate the advantages of high-energy and low-energy image information, which can optimize imaging quality, enhance contrast, reduce artifacts, and provide richer diagnostic information, more comprehensive and accurate anatomical and functional information in medical imaging.

[0091] In the embodiment of the present invention, the step of performing image enhancement on the monoenergetic combined image to obtain a virtual monoenergetic image includes:

[0092] Calculating the minimum and maximum pixel values ​​of the monoenergetic combined image;

[0093] Calculate a parameter gain factor and an offset according to the minimum value and the maximum value;

[0094] Calculate and update pixel value according to the gain factor and the offset;

[0095] The updated pixel values ​​are used to replace the pixel values ​​in the monoenergetic combined image to obtain a virtual monoenergetic image.

[0096] In detail, the parameter gain factor α and offset β are calculated by the following formulas:

[0097]

[0098] β=min new ―α*min old

[0099] Among them, α represents the contrast gain factor, β represents the brightness offset, and max new Indicates the preset maximum pixel value threshold, min new Indicates the preset minimum pixel value threshold, max old Indicates the maximum pixel value of the single energy combined image, min old Indicates the minimum pixel value of the monoenergetic combined image.

[0100] Substitute the calculated gain factor α and offset β values ​​into the following linear transformation formula, and use the obtained updated pixel values ​​to generate a virtual monoenergetic image.

[0101] I new =αI old +β

[0102] Among them, I new Indicates updating pixel value, I old represents the original pixel value in the monoenergetic combined image, α represents the contrast gain factor, and β represents the brightness offset.

[0103] Linear contrast adjustment is a simple and effective image enhancement technology used to improve the contrast of single-energy combined images. It adjusts the brightness range of the image through linear transformation and expands the darkest and brightest parts of the image to the entire grayscale range. This adjustment expands the contrast of the image within a linear range, enhances the visual effect and detail visibility, and obtains a virtual single-energy image with high image quality after image enhancement.

[0104] The virtual monoenergetic image obtained after preprocessing the dual-energy original projection image improves the signal-to-noise ratio of the image by merging the information of the high-energy and low-energy images, thereby enhancing the clarity and resolution of the image. This preprocessing method not only improves the image quality, but also helps to make subsequent analysis more accurate and reliable.

[0105] S3. Calculate the loss value of a preset correction training model using the virtual monoenergetic image.

[0106] In an embodiment of the present invention, the step of calculating the loss value of a preset correction training model using the virtual monoenergetic image includes:

[0107] Acquire an image label of the virtual monoenergetic image;

[0108] Performing hardening artifact correction on the virtual monoenergetic image using a preset correction training model to obtain a predicted label of the virtual monoenergetic image;

[0109] A loss value of the corrected training model is calculated based on the image label and the predicted label.

[0110] In detail, the loss value of the correction training model is calculated by the following formula:

[0111]

[0112] in, represents the loss value of the correction training model, n represents the total number of virtual monoenergetic image labels, i represents the i-th label of the virtual monoenergetic image, y(i) represents the image label of the i-th virtual monoenergetic image, represents the predicted label of the i-th virtual monoenergetic image.

[0113] By using virtual monoenergetic images to calculate the loss value of the preset correction training model, the performance of the model can be evaluated and optimized during the training process. The error of the correction training model in the hardening artifact correction task can be accurately measured. By analyzing the loss value, the model parameters can be further adjusted to improve the accuracy and stability of the correction training model.

[0114] S4. Optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model.

[0115] In an embodiment of the present invention, the step of optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model includes:

[0116] Calculate the gradient of the loss value to the correction training model parameters using a back propagation algorithm;

[0117] Updating parameters of the correction training model according to the gradient;

[0118] Determining whether the loss value is less than a preset loss value;

[0119] If the loss value is greater than a preset loss value, then continuing to perform hardening artifact correction on the virtual monoenergetic image;

[0120] If the loss value is less than or equal to the preset loss value, the correction training model parameter optimization is completed, and the dual-energy CBCT hardening artifact correction model is obtained.

[0121] In detail, back propagation is used to calculate the gradient of the loss function with respect to the correction training model parameters. The stochastic gradient descent method uses partial image labels in the virtual monoenergetic image to update the parameters of the correction training model in each iteration. The stochastic gradient descent method is usually faster and more efficient than the standard gradient descent.

[0122] The main benefit of using virtual monoenergetic images to construct a dual-energy CBCT hardening artifact correction model is that it can effectively solve the common hardening artifact problem in dual-energy CBCT images. By establishing the model, it can be used to process high-energy and low-energy images with artifacts, thereby improving the quality and accuracy of dual-energy CBCT images.

[0123] S5. Deconvolute the acquired hardening artifact image using the dual-energy CBCT hardening artifact correction model to obtain a reconstructed image after hardening artifacts are suppressed.

[0124] In an embodiment of the present invention, the step of deconvolving the acquired hardening artifact image using the dual-energy CBCT hardening artifact correction model to obtain a reconstructed image after hardening artifact suppression includes:

[0125] obtaining a pre-processed hardening artifact image;

[0126] converting the hardening artifact image into an image pixel matrix;

[0127] The image pixel matrix is ​​convolved using a dual-energy CBCT hardening artifact correction network model, and a feature channel is obtained through a ReLU function;

[0128] Performing maximum pooling on the feature channels to obtain a pooling matrix;

[0129] Filling the pooling matrix with null values ​​according to a preset size to obtain a filled matrix;

[0130] Setting a sliding step size according to a preset deconvolution kernel size, and performing sliding coverage on the filling matrix according to the sliding step size;

[0131] Performing a dot product sum on the pixel points covered by sliding in the filling matrix and a preset deconvolution kernel, and obtaining a deconvolution matrix according to the dot product sum;

[0132] The deconvolution matrix is ​​convolved again to obtain a reconstructed image with hardening artifacts suppressed.

[0133] In detail, a convolution operation is performed on the image pixel matrix, and the feature channel is extracted using the ReLU activation function. The feature channel is subjected to a maximum pooling operation to reduce the dimension of the feature map and retain the main features. The pooled matrix is ​​then subjected to a deconvolution operation to increase the size of the feature map, thereby performing upsampling. Finally, a convolution operation is performed on the deconvolved feature map using a preset convolution kernel to further extract and adjust the features, thereby obtaining a reconstructed image after hardening artifacts are suppressed.

[0134] like Figure 2 , which is a functional module diagram of a dual-energy CBCT hardening artifact correction device provided in one embodiment of the present application.

[0135] The dual-energy CBCT hardening artifact correction device 100 of the present invention can be installed in an electronic device. According to the functions to be implemented, the dual-energy CBCT hardening artifact correction device 100 may include an image acquisition module 101, an image preprocessing module 102, a loss value calculation module 103, a model optimization module 104, and an image convolution module 105. The module described in this application may also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0136] In this embodiment, the functions of each module / unit are as follows:

[0137] An image acquisition module 101 is used to acquire a dual-energy original projection image of a scanned object;

[0138] An image preprocessing module 102 is used to perform dual-energy image decomposition on the dual-energy original projection image, and combine the decomposed images to obtain a virtual mono-energy image;

[0139] A loss value calculation module 103, used to calculate the loss value of a preset correction training model using the virtual monoenergetic image;

[0140] A model optimization module 104, configured to optimize the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model;

[0141] The image convolution module 105 is used to perform deconvolution on the acquired hardening artifact image using the dual-energy CBCT hardening artifact correction model to obtain a reconstructed image after hardening artifacts are suppressed.

[0142] In detail, each module described in the dual-energy CBCT hardening artifact correction device 100 in the embodiment of the present invention is used in the same manner as described above. Figure 1 The dual-energy CBCT hardening artifact correction method described in the present invention is the same technical means and can produce the same technical effect, so it will not be repeated here.

[0143] like Figure 3 FIG. 1 is a schematic diagram of the structure of an electronic device for implementing a dual-energy CBCT hardening artifact correction method provided by an embodiment of the present invention.

[0144] The electronic device may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a dual-energy CBCT hardening artifact correction program.

[0145] In some embodiments, the processor 10 may be composed of an integrated circuit, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (ControlUnit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes various functions and processes data of the electronic device by running or executing programs or modules stored in the memory 11 (for example, executing a dual-energy CBCT hardening artifact correction program, etc.), and calling data stored in the memory 11.

[0146] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (for example, SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of an electronic device, such as a mobile hard disk of the electronic device. In other embodiments, the memory 11 may also be an external storage device of an electronic device, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device. Further, the memory 11 may also include both an internal storage unit of the electronic device and an external storage device. The memory 11 can be used not only to store application software and various types of data installed in the electronic device, such as the code of the dual-energy CBCT hardening artifact correction program, but also to temporarily store data that has been output or is to be output.

[0147] The communication bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. The bus is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0148] The communication interface 13 is used for communication between the above-mentioned electronic device and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a display (Display), an input unit (such as a keyboard (Keyboard)), and optionally, the user interface may also be a standard wired interface, a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode, organic light-emitting diode) touch device, etc. Among them, the display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device and to display a visual user interface.

[0149] The figure only shows an electronic device with components. Those skilled in the art will understand that the structure shown in the figure does not constitute a limitation on the electronic device, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0150] For example, although not shown, the electronic device may also include a power source (such as a battery) for supplying power to each component. Preferably, the power source may be logically connected to the at least one processor 10 through a power management device, so that the power management device can realize functions such as charging management, discharging management, and power consumption management. The power source may also include one or more DC or AC power sources, recharging devices, power failure detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0151] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0152] The dual-energy CBCT hardening artifact correction program stored in the memory 11 in the electronic device is a combination of multiple instructions. When running in the processor 10, it can achieve: obtaining a dual-energy original projection image of the scanned object; performing dual-energy image decomposition on the dual-energy original projection image, and combining the decomposed images to obtain a virtual mono-energy image; using the virtual mono-energy image to calculate the loss value of a preset correction training model; optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model; and using the dual-energy CBCT hardening artifact correction model to deconvolve the acquired hardening artifact image to obtain a reconstructed image after hardening artifacts are suppressed.

[0153] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to the description of the relevant steps in the corresponding embodiment of the accompanying drawings, which will not be repeated here.

[0154] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0155] The present invention also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor of an electronic device, it can achieve: obtaining a dual-energy original projection image of a scanned object; performing dual-energy image decomposition on the dual-energy original projection image, and combining the decomposed images to obtain a virtual mono-energy image; using the virtual mono-energy image to calculate a loss value of a preset correction training model; optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model; and using the dual-energy CBCT hardening artifact correction model to deconvolve the acquired hardening artifact image to obtain a reconstructed image after hardening artifacts are suppressed.

[0156] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation.

[0157] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0158] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0159] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0160] Therefore, no matter from which point of view, the embodiments should be regarded as illustrative and non-restrictive, and the scope of the present invention is limited by the appended claims rather than the above description, so it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any attached figure mark in the claims should not be regarded as limiting the claims involved.

[0161] The blockchain referred to in this invention is a new application mode of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanism, encryption algorithm, etc. Blockchain is essentially a decentralized database, a string of data blocks generated by cryptographic methods. Each data block contains a batch of network transaction information, which is used to verify the validity of its information (anti-counterfeiting) and generate the next block. Blockchain can include the underlying blockchain platform, platform product service layer, and application service layer.

[0162] The embodiments of the present application can acquire and process relevant data based on neural networks. Among them, neural networks are machine learning models that imitate the structure and working principle of the human brain's nervous system. They can automatically learn and extract patterns from data, and are used to identify information such as images, voices, and texts, predict trends, generate new data, solve control problems, and even optimize complex systems. It is widely used in computer vision, natural language processing, medical diagnosis, financial forecasting, and other fields to improve the intelligence level of data analysis and decision-making processes.

[0163] In addition, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices stated in the system claim can also be implemented by one unit or device through software or hardware. The words first, second, etc. are used to indicate names, and do not indicate any particular order.

[0164] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A dual-energy CBCT hardening artifact correction method, characterized in that: The method comprises: Acquire a dual-energy original projection image of the scanned object; Performing dual-energy image decomposition on the dual-energy original projection image, and combining the decomposed images to obtain a virtual mono-energy image; Calculating a loss value of a preset correction training model using the virtual monoenergetic image; Optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model; The dual-energy CBCT hardening artifact correction model is used to perform deconvolution on the acquired hardening artifact image to obtain a reconstructed image after hardening artifact suppression.

2. The dual-energy CBCT hardening artifact correction method according to claim 1, characterized in that: The step of performing dual-energy image decomposition on the dual-energy original projection image and combining the decomposed images to obtain a virtual mono-energy image includes: Performing FDK reconstruction on the dual-energy original projection image to obtain a dual-energy reconstructed image; Decomposing the dual-energy reconstructed image into a dual-energy image to obtain a high-energy image and a low-energy image; Performing forward projection processing on the high-energy image and the low-energy image to obtain a projection result; Recombining the high-energy image and the low-energy image according to the projection results to obtain a single-energy combined image; The monoenergetic combined image is enhanced to obtain a virtual monoenergetic image.

3. The dual-energy CBCT hardening artifact correction method according to claim 2, characterized in that: The forward projection processing is performed on the high-energy image and the low-energy image to obtain a projection result, including: Acquire voxel matrices of the high-energy image and the low-energy image; Get the source point position and the detector plane position; Calculating a first distance from each voxel in the voxel matrix to the source point position and a second distance from each voxel in the voxel matrix to the detector plane position; Calculate the attenuation of the radioactive isotope passing through each voxel in the voxel matrix according to the first distance and the second distance; Attenuation correction is performed on each voxel in the voxel matrix according to the attenuation amount, and the projection value after attenuation correction is projected onto the detector plane to form a projection result.

4. The dual-energy CBCT hardening artifact correction method according to claim 2, characterized in that: The step of recombining the high-energy image and the low-energy image according to the projection result to obtain a single-energy combined image comprises: Aligning each voxel in the high-energy image and the low-energy image in spatial position; Performing energy correction on the aligned high-energy image and the low-energy image; The corrected high-energy image and low-energy image are weighted combined to obtain a single-energy combined image.

5. The dual-energy CBCT hardening artifact correction method according to claim 2, characterized in that: The step of performing image enhancement on the monoenergetic combined image to obtain a virtual monoenergetic image comprises: Calculating the minimum and maximum pixel values ​​of the monoenergetic combined image; Calculate a parameter gain factor and an offset according to the minimum value and the maximum value; Calculate and update pixel value according to the gain factor and the offset; The updated pixel values ​​are used to replace the pixel values ​​of the monoenergetic combined image to obtain a virtual monoenergetic image.

6. The dual-energy CBCT hardening artifact correction method according to claim 5, characterized in that: The step of optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model includes: Calculate the gradient of the loss value to the correction training model parameters using a back propagation algorithm; Updating parameters of the correction training model according to the gradient; Determining whether the loss value is less than a preset loss value; If the loss value is greater than a preset loss value, then continuing to perform hardening artifact correction on the virtual monoenergetic image; If the loss value is less than or equal to the preset loss value, the correction training model parameter optimization is completed, and the dual-energy CBCT hardening artifact correction model is obtained.

7. The dual-energy CBCT hardening artifact correction method according to claim 1, characterized in that: The method of using the dual-energy CBCT hardening artifact correction model to deconvolve the acquired hardening artifact image to obtain a reconstructed image after hardening artifact suppression includes: acquiring a pre-processed hardening artifact image; converting the hardening artifact image into an image pixel matrix; The image pixel matrix is ​​convolved using a dual-energy CBCT hardening artifact correction network model, and a feature channel is obtained through a ReLU function; Performing maximum pooling on the feature channels to obtain a pooling matrix; Filling the pooling matrix with null values ​​according to a preset size to obtain a filled matrix; Setting a sliding step size according to a preset deconvolution kernel size, and performing sliding coverage on the filling matrix according to the sliding step size; Performing a dot product sum on the pixel points covered by sliding in the filling matrix and a preset deconvolution kernel, and obtaining a deconvolution matrix according to the dot product sum; The deconvolution matrix is ​​convolved again to obtain a reconstructed image with hardening artifacts suppressed.

8. A dual-energy CBCT hardening artifact correction device, characterized in that: The device comprises: An image acquisition module, used for acquiring a dual-energy original projection image of a scanned object; An image preprocessing module, used for performing dual-energy image decomposition on the dual-energy original projection image, and combining the decomposed images to obtain a virtual mono-energy image; A loss value calculation module, used to calculate the loss value of a preset correction training model using the virtual monoenergetic image; A model optimization module, used for optimizing the correction training model parameters according to the loss value to obtain a dual-energy CBCT hardening artifact correction model; The image convolution module is used to deconvolve the acquired hardening artifact image using the dual-energy CBCT hardening artifact correction model to obtain a reconstructed image after the hardening artifact is suppressed.

9. An electronic device, characterized in that: The electronic device comprises: at least one processor; and, a memory communicatively connected to the at least one processor; wherein, The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can perform the dual-energy CBCT hardening artifact correction method according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the dual-energy CBCT hardening artifact correction method according to any one of claims 1 to 7 is implemented.

Images