A method, system, device and medium for CBCT image osteosclerotic artifact correction

Abstract

The present application relates to image correction technology, disclose a kind of CBCT image bone hardening artifact correction method, comprising: the high-energy projection image of target object and low-energy projection is collected and is calculated after image reconstruction to obtain preliminary artifact-free image, based on threshold segmentation from preliminary artifact-free image extraction bone image and soft tissue image and calculate skeleton weight, artifact image of target object is used as training input data, bone image and soft tissue image are used as reference data, and dual-channel artifact correction model is trained, and preprocessed bone image and preprocessed soft tissue image are output according to the bone hardening artifact image to be processed according to dual-channel artifact correction model, according to the preprocessed bone image and preprocessed soft tissue image are weighted fusion according to skeleton weight, and artifact-free reconstruction image is obtained.The present application also proposes a kind of CBCT image bone hardening artifact correction device, electronic equipment and storage medium.The present application can improve the effect of image artifact correction.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of image correction, in particular to a CBCT image bone hardening artifact correction method, system, device and medium. BACKGROUND

[0002] CBCT (Cone Beam Computed Tomography) is widely used in oral medicine. The X-rays used by CBCT have polychromaticity, that is, they contain photons of multiple energies. When these photons pass through human tissues, low-energy photons are more easily absorbed, while high-energy photons are relatively difficult to absorb. Due to its high density and high atomic number, bone has a very strong ability to absorb X-rays. When X-rays pass through bone, a large number of low-energy photons are absorbed, while a relatively large number of high-energy photons penetrate the bone. The density of bone is much higher than that of surrounding tissues, so its attenuation coefficient of X-rays is also higher. This means that X-rays will undergo significant attenuation when passing through bone. When X-rays pass through bone, a large number of low-energy photons are absorbed, resulting in mainly high-energy photons reaching the detector. This "hardened" X-ray beam appears as radial or strip-like artifacts around the bone area on the image. Due to the different absorption abilities of bone to X-rays, the energy distribution of the X-ray beam after passing through the bone is uneven. This energy difference is difficult to completely correct during image reconstruction, resulting in artifacts.

[0003] Existing correction methods for bone hardening artifacts usually use polynomial fitting to correct. However, due to the large difference in bone density among different patients, the correction effect of polynomial fitting may be affected by this difference, affecting the correction effect. SUMMARY

[0004] The present application provides a CBCT image bone hardening artifact correction method, system, device and medium, which mainly aims to solve the problem of poor correction effect of existing CBCT image bone hardening artifact correction methods.

[0005] To achieve the above purpose, the present application provides a CBCT image bone hardening artifact correction method, which comprises:

[0006] Optionally, the acquisition of the high-energy projection image and the low-energy projection image of the target object by the preset scanning device comprises:

[0007] acquiring the high-energy projection image of the target object according to the preset first scanning parameter by the scanning device;

[0008] acquiring the low-energy projection image of the target object according to the preset second scanning parameter by the scanning device.

[0009] Optionally, the image reconstruction of the high-energy projection image and the low-energy projection image to obtain a reconstructed high-energy projection image and a reconstructed low-energy projection image comprises:

[0010] The inverse projection path of each pixel point of the high-energy projection image and the low-energy projection image is determined according to a predetermined projection angle;

[0011] The high-energy projection image and the low-energy projection image are integrated according to the inverse projection path, and the integration result is accumulated to a preset reconstruction space to obtain an inverse projection high-energy image and an inverse projection low-energy image;

[0012] The inverse projection high-energy image and the inverse projection low-energy image are subjected to Fourier transform to obtain a frequency domain high-energy image and a frequency domain low-energy image;

[0013] The frequency domain high-energy image and the frequency domain low-energy image are multiplied by a preselected filter function and subjected to inverse Fourier transform to obtain the reconstructed high-energy projection image and the reconstructed low-energy projection image.

[0014] Optionally, the primary artifact-free image is calculated according to the reconstructed high-energy projection image and the reconstructed low-energy projection image, comprising:

[0015] The primary artifact-free image is calculated by using the following formula:

[0016] I t =I H +α(I H ―I L )

[0017] Wherein, I t is the primary artifact-free image, I H is the reconstructed high-energy projection image, I L is the reconstructed low-energy projection image, and α is a predetermined weighting coefficient.

[0018] Optionally, the bone image and the soft tissue image are extracted from the primary artifact-free image based on threshold segmentation, comprising:

[0019] The gray level histogram of the primary artifact-free image is calculated;

[0020] The peak value feature of the gray level histogram is extracted, and the segmentation threshold is determined according to the peak value feature;

[0021] The image part with a gray value greater than the segmentation threshold in the primary artifact-free image is extracted to obtain a primary bone image;

[0022] The image part with a gray value less than or equal to the segmentation threshold in the primary artifact-free image is extracted to obtain a primary soft tissue image.

[0023] The primary bone image and the primary soft tissue image are subjected to image filtering processing to obtain a bone image and a soft tissue image.

[0024] Optionally, the calculation of the bone weight according to the bone image and the primary artifact-free image comprises:

[0025] Area data of the artifact-free image is obtained to obtain an artifact-free image area.

[0026] Area data of the bone image is obtained to obtain a bone image area.

[0027] The ratio of the bone image area to the artifact-free image area is calculated to obtain a bone weight.

[0028] Optionally, the training of the dual-channel artifact correction model according to the training input data and the reference data comprises:

[0029] A pre-established neural network model is used to obtain a predicted output result according to the training input data.

[0030] A pre-defined loss function is used to calculate the difference value between the predicted output result and the reference data.

[0031] A back propagation algorithm is used to calculate the gradient of the loss value function for each parameter according to the difference value to obtain gradient information.

[0032] A gradient descent algorithm is used to adjust the parameters of the neural network model according to the gradient information.

[0033] When the preset number of iterations is reached, it is confirmed that the model training is completed, and the dual-channel artifact correction model is obtained.

[0034] To solve the above problems, the application further provides a CBCT image bone hardening artifact correction device, which comprises:

[0035] An image acquisition module is configured to acquire high-energy projection images and low-energy projection images of a target object by using a preset scanning device.

[0036] An image reconstruction module is configured to reconstruct the high-energy projection images and the low-energy projection images to obtain reconstructed high-energy projection images and reconstructed low-energy projection images.

[0037] A primary artifact-free image is calculated according to the reconstructed high-energy projection images and the reconstructed low-energy projection images.

[0038] an image segmentation module configured to extract a bone image and a soft tissue image from the primary artifact-free image based on threshold segmentation, and to calculate a bone weight according to the bone image and the primary artifact-free image;

[0039] a model training module configured to obtain an artifact image of the target object as training input data, to obtain the bone image and the soft tissue image as reference data of the training input data, and to train a double-channel artifact correction model according to the training input data and the reference data;

[0040] an artifact correction module configured to output a preprocessed bone image and a preprocessed soft tissue image according to the double-channel artifact correction model and a pre-acquired bone-hardening artifact image, and to obtain an artifact-free reconstructed image by weighted fusion of the preprocessed bone image and the preprocessed soft tissue image according to the bone weight.

[0041] To solve the above problems, the present application further provides an electronic device, which comprises:

[0042] at least one processor;

[0043] and a memory in communication connection with the at least one processor;

[0044] 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 to enable the at least one processor to execute the CBCT image bone-hardening artifact correction method described above.

[0045] To solve the above problems, the present application further provides a computer readable storage medium, which stores at least one computer program, and the at least one computer program is executed by a processor in an electronic device to implement the CBCT image bone-hardening artifact correction method described above.

[0046] This invention employs a pre-set scanning device to acquire high-energy and low-energy projection images of a target object. Image reconstruction is then performed on these images to obtain reconstructed high-energy and low-energy projection images. A primary artifact-free image is calculated based on these images. Bone and soft tissue images are extracted from the primary artifact-free image using threshold segmentation. Bone weights are calculated based on the bone and primary artifact-free images. Artifact images of the target object are obtained as training input data, and the bone and soft tissue images are used as reference data. A dual-channel artifact correction model is trained using the training input data and the reference data. This model outputs pre-processed bone and soft tissue images based on pre-acquired bone sclerosis artifact images. The pre-processed bone and soft tissue images are then weighted and fused according to the bone weights to obtain an artifact-free reconstructed image. This method improves the correction effect for bone sclerosis artifacts in CBCT images. Therefore, the CBCT image bone sclerosis artifact correction method, device, electronic device and computer-readable storage medium proposed in this invention can solve the problem of poor correction effect of existing CBCT image bone sclerosis artifact correction methods. Attached Figure Description

[0047] Figure 1 A schematic flowchart of a CBCT image bone sclerosis artifact correction method provided in an embodiment of the present invention;

[0048] Figure 2 This is a schematic diagram of an image reconstruction process provided in an embodiment of the present invention;

[0049] Figure 3 This is a schematic diagram of an image segmentation process provided in an embodiment of the present invention;

[0050] Figure 4 This is a functional block diagram of a CBCT image bone sclerosis artifact correction device provided in an embodiment of the present invention;

[0051] Figure 5 This is a schematic diagram of the structure of an electronic device for implementing the CBCT image bone sclerosis artifact correction method according to an embodiment of the present invention;

[0052] Figure 6 This is a schematic diagram illustrating the principle of obtaining images after bone artifact removal, as provided in an embodiment of the present invention.

[0053] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0054] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0055] This application provides a method for correcting bone sclerosis artifacts in CBCT images. The execution subject of this method includes, but is not limited to, at least one electronic device that can be configured to execute the method provided in this application, such as a server or a terminal. In other words, the method can be executed by software or hardware installed on a terminal device or a server device, and the software may be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cluster of cloud servers. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0056] Reference Figure 1 The diagram shown is a flowchart illustrating a method for correcting bone sclerosis artifacts in CBCT images according to an embodiment of the present invention. In this embodiment, the method for correcting bone sclerosis artifacts in CBCT images includes:

[0057] S1. Use a preset scanning device to acquire high-energy projection images and low-energy projection images of the target object.

[0058] In this embodiment of the invention, the scanning device may be a CBCT device. Specifically, the CBCT device uses an X-ray emitter to generate a cone-shaped X-ray beam, which is composed of multiple beams and can simultaneously emit multiple X-rays at different angles to obtain projected images of the patient.

[0059] In this embodiment of the invention, the target object may be the patient's oral cavity.

[0060] In this embodiment of the invention, the acquisition of high-energy projection images and low-energy projection images of the target object using a preset scanning device is achieved by setting different currents and voltages for the scanning device to perform scanning, thereby obtaining projection images of different energy levels.

[0061] In this embodiment of the invention, the step of acquiring high-energy projection images and low-energy projection images of the target object using a preset scanning device includes:

[0062] The scanning device is used to acquire a high-energy projection image of the target object according to preset first scanning parameters;

[0063] The scanning device is used to acquire a low-energy projection image of the target object according to the preset second scanning parameters.

[0064] Specifically, the preset first scan parameter can be 120kV 4mA (120 kV, 4 mA).

[0065] Specifically, the preset second scan parameter can be 94kV 4mA (94 kilovolts, 4 milliamperes).

[0066] In this embodiment of the invention, by using a preset scanning device to acquire high-energy projection images and low-energy projection images of the target object, the efficiency of subsequent image reconstruction is improved.

[0067] S2. Perform image reconstruction on the high-energy projection image and the low-energy projection image to obtain the reconstructed high-energy projection image and the reconstructed low-energy projection image.

[0068] In this embodiment of the invention, the image reconstruction of the high-energy projection image and the low-energy projection image is performed using the FDK reconstruction algorithm.

[0069] In detail, the FDK reconstruction algorithm is a classic algorithm for CT image reconstruction. This algorithm recovers high-quality CT images by performing inverse projection and filtering operations on X-ray data.

[0070] In this embodiment of the invention, reference is made to Figure 2 The diagram illustrates a process for image reconstruction according to an embodiment of the present invention. The process of reconstructing the high-energy projection image and the low-energy projection image to obtain reconstructed high-energy projection images and reconstructed low-energy projection images includes:

[0071] S21. Determine the inverse projection path of each pixel of the high-energy projection image and the low-energy projection image according to the predetermined projection angle.

[0072] S22. Integrate the high-energy projection image and the low-energy projection image according to the inverse projection path, and accumulate the integration result into the preset reconstruction space to obtain the inverse projection high-energy image and the inverse projection low-energy image.

[0073] S23. Perform Fourier transform on the inverse projection high-energy image and the inverse projection low-energy image to obtain the frequency domain high-energy image and the frequency domain low-energy image.

[0074] S24. Multiply the high-energy image in the frequency domain and the low-energy image in the frequency domain with a pre-selected filter function and then perform an inverse Fourier transform to obtain a reconstructed high-energy projection image and a reconstructed low-energy projection image.

[0075] In detail, the predetermined projection angle refers to the angular relationship between the X-rays of the scanning device and the projected image.

[0076] In detail, the Fourier transform is a method for converting a signal from its original domain to the frequency domain, and the inverse Fourier transform is a process for converting a signal from the frequency domain back to its original time or spatial domain. It is the inverse process of the Fourier transform and is used to reconstruct the original signal from its frequency components.

[0077] Specifically, the pre-selected filtering functions include the Ram-Lak filtering function, the Shepp-Logan filtering function, etc. These filtering functions can enhance the high-frequency components in the image to improve image clarity.

[0078] In detail, the process of multiplying the high-energy image and the low-energy image in the frequency domain with a pre-selected filtering function and then performing an inverse Fourier transform to obtain the reconstructed high-energy projection image and the reconstructed low-energy projection image is a filtering process that can remove noise from the image.

[0079] In this embodiment of the invention, by performing image reconstruction on the high-energy projection image and the low-energy projection image, the reconstructed high-energy projection image and the reconstructed low-energy projection image are obtained, which improves the efficiency of subsequent calculation of the primary artifact-free image.

[0080] S3. Calculate the primary artifact-free image based on the reconstructed high-energy projection image and the reconstructed low-energy projection image.

[0081] In this embodiment of the invention, calculating the primary artifact-free image based on the reconstructed high-energy projection image and the reconstructed low-energy projection image means fusing the reconstructed high-energy projection image and the reconstructed low-energy projection image according to a preset method based on a predetermined weighting coefficient to obtain the primary artifact image.

[0082] In this embodiment of the invention, the step of calculating the primary artifact-free image based on the reconstructed high-energy projection image and the reconstructed low-energy projection image includes:

[0083] The primary artifact-free image is calculated using the following formula:

[0084] I t =I H +α(I H —I L )

[0085] Among them, I t For the primary artifact-free image, I H For the reconstructed high-energy projection image, I LFor the reconstructed low-energy projection image, α is a pre-determined weighting coefficient.

[0086] In detail, the predetermined weighting coefficients need to be determined through extensive experiments and optimization to achieve optimal image quality or meet specific diagnostic needs. By adjusting the values ​​of the weighting coefficients, artifacts caused by differences between images of different energy levels can be reduced or eliminated to a certain extent, thereby obtaining clearer and more accurate images.

[0087] In this embodiment of the invention, a primary artifact-free image is calculated based on the reconstructed high-energy projection image and the reconstructed low-energy projection image, thereby improving the efficiency and accuracy of calculating and extracting bone and soft tissue images.

[0088] S4. Extract bone images and soft tissue images from the primary artifact-free image based on threshold segmentation, and calculate bone weights based on the bone images and the primary artifact-free image.

[0089] In this embodiment of the invention, the extraction of bone and soft tissue images from the primary artifact-free image based on threshold segmentation refers to using the grayscale value of the image as a threshold, determining the threshold based on the grayscale histogram or local features of the primary artifact-free image, and segmenting the primary artifact-free image based on the determined threshold to obtain bone and soft tissue images.

[0090] In this embodiment of the invention, reference is made to Figure 3 The diagram illustrates a process for image segmentation according to an embodiment of the present invention. The step of extracting bone and soft tissue images from the primary artifact-free image based on threshold segmentation includes:

[0091] S31. Calculate the grayscale histogram of the primary artifact-free image;

[0092] S32. Extract the peak features of the grayscale histogram and determine the segmentation threshold based on the peak features;

[0093] S33. Extract the image portion with a gray value greater than the segmentation threshold from the primary artifact-free image to obtain the primary bone image;

[0094] S34. Extract the image portion of the primary artifact-free image whose grayscale value is less than or equal to the segmentation threshold to obtain the primary soft tissue image;

[0095] S35. Perform image filtering processing on the primary bone image and the primary soft tissue image to obtain the bone image and the soft tissue image.

[0096] In this embodiment of the invention, the calculation of bone weight based on the bone image and the primary artifact-free image is based on the area ratio of the bone image in the primary artifact-free image.

[0097] In this embodiment of the invention, calculating bone weights based on the bone image and the primary artifact-free image includes:

[0098] The bone weights are calculated using the following formula:

[0099]

[0100] Where ω is the bone weight, S t S is the area of ​​the primary artifact-free image. bone The area of ​​the bone image is denoted as .

[0101] In this embodiment of the invention, calculating bone weights based on the bone image and the primary artifact-free image includes:

[0102] Obtain the area data of the artifact-free image to obtain the area of ​​the artifact-free image;

[0103] Obtain the area data of the bone image to obtain the bone image area;

[0104] The bone weight is obtained by calculating the ratio of the area of ​​the bone image to the area of ​​the artifact-free image.

[0105] In this embodiment of the invention, bone images and soft tissue images are extracted from the primary artifact-free image based on threshold segmentation, which improves the efficiency of calculating bone weights. By calculating bone weights based on the bone images and the primary artifact-free image, the efficiency and accuracy of subsequent weighted fusion are improved.

[0106] S5. Obtain the artifact image of the target object as training input data, and use the bone image and the soft tissue image as reference data for the training input data. Train a dual-channel artifact correction model based on the training input data and the reference data.

[0107] In this embodiment of the invention, obtaining the artifact image of the target object as training input data means obtaining a CBCT image of the target object with noise band bone sclerosis artifacts.

[0108] In this embodiment of the invention, training a dual-channel artifact correction model based on the training input data and the reference data includes:

[0109] The predicted output is obtained using a pre-established neural network model based on the training input data;

[0110] The difference between the predicted output and the reference data is calculated using a predefined loss function.

[0111] The gradient of the loss function with respect to each parameter is calculated using the backpropagation algorithm based on the difference value, thus obtaining the gradient information;

[0112] The gradient descent algorithm is used based on the gradient information and the parameters of the neural network model;

[0113] Once the preset number of iterations is reached, the model training is confirmed to be complete, and a dual-channel artifact correction model is obtained.

[0114] In detail, the step of obtaining the predicted output result based on the training input data using a pre-established neural network model involves using a forward propagation algorithm to input the training input data into the neural network model and calculating the output of each layer layer by layer to obtain the predicted result.

[0115] In detail, the backpropagation algorithm is a core algorithm used in deep learning to train neural networks. It updates the network weights by calculating the gradient of the loss function with respect to the network parameters, thereby minimizing the loss function.

[0116] In detail, the use of gradient descent algorithm based on the gradient information based on the parameters of the neural network model refers to using the learning rate, weight parameters, and bias term of the neural network model.

[0117] In this embodiment of the invention, by acquiring the artifact image of the target object as training input data, and using the bone image and the soft tissue image as reference data for the training input data, a dual-channel artifact correction model is trained based on the training input data and the reference data, thereby improving the efficiency of subsequent output preprocessed bone images and preprocessed soft tissue images.

[0118] S6. Using the dual-channel artifact correction model, output preprocessed bone images and preprocessed soft tissue images based on the pre-acquired bone sclerosis artifact images to be processed.

[0119] In this embodiment of the invention, the preprocessed bone image and the preprocessed soft tissue image are the output results of the dual-channel artifact correction model after segmenting the bone sclerosis artifact image to be processed.

[0120] In detail, the step of using the dual-channel artifact correction model to output preprocessed bone images and preprocessed soft tissue images based on the pre-acquired bone sclerosis artifact images to be processed involves performing convolution and pooling operations on the bone sclerosis artifact images to be processed using the convolutional and pooling layers of the dual-channel artifact correction model, adding nonlinear factors using activation layers, and finally summarizing the output results of each neuron using fully connected layers based on weight parameters and bias terms to obtain the preprocessed bone images and preprocessed soft tissue images.

[0121] In this embodiment of the invention, by utilizing the dual-channel artifact correction model to output preprocessed bone images and preprocessed soft tissue images based on the pre-acquired bone sclerosis artifact images to be processed, the efficiency of subsequent fusion to obtain artifact-free reconstructed images is improved.

[0122] S7. The preprocessed bone image and the preprocessed soft tissue image are weighted and fused according to the bone weights to obtain an artifact-free reconstructed image.

[0123] In this embodiment of the invention, the weighted fusion is a technique for merging information from multiple images into a single image. It involves assigning a weight to each image and then combining the pixel values ​​of the images based on these weights.

[0124] For details, refer to Figure 6 The diagram shown is a schematic representation of the principle of obtaining images after bone artifact removal according to an embodiment of the present invention.

[0125] In this embodiment of the invention, by weighted fusing the preprocessed bone image and the preprocessed soft tissue image according to the bone weight, an artifact-free reconstructed image is obtained, which improves the efficiency and effect of CBCT image bone sclerosis artifact correction.

[0126] like Figure 4 The diagram shown is a functional block diagram of a CBCT image bone sclerosis artifact correction device provided in an embodiment of the present invention.

[0127] The CBCT image bone sclerosis artifact correction device 100 of the present invention can be installed in an electronic device. Depending on the functions implemented, the CBCT image bone sclerosis artifact correction device 100 may include an image acquisition module 101, an image reconstruction module 102, an image segmentation module 103, a model training module 104, and an artifact correction module 105. The module described in this invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by the processor of an electronic device and can perform a fixed function, and which are stored in the memory of the electronic device.

[0128] In this embodiment, the functions of each module / unit are as follows:

[0129] The image acquisition module 101 is used to acquire high-energy projection images and low-energy projection images of the target object using a preset scanning device.

[0130] The image reconstruction module 102 is used to perform image reconstruction on the high-energy projection image and the low-energy projection image to obtain the reconstructed high-energy projection image and the reconstructed low-energy projection image.

[0131] A primary artifact-free image is calculated based on the reconstructed high-energy projection image and the reconstructed low-energy projection image;

[0132] The image segmentation module 103 is used to extract bone images and soft tissue images from the primary artifact-free image based on threshold segmentation, and to calculate bone weights based on the bone images and the primary artifact-free image.

[0133] The model training module 104 is used to acquire the artifact image of the target object as training input data, and to use the bone image and the soft tissue image as reference data for the training input data, and to train a dual-channel artifact correction model based on the training input data and the reference data.

[0134] The artifact correction module 105 is used to output a preprocessed bone image and a preprocessed soft tissue image based on the pre-acquired bone sclerosis artifact image to be processed using the dual-channel artifact correction model, and to perform weighted fusion of the preprocessed bone image and the preprocessed soft tissue image according to the bone weight to obtain an artifact-free reconstructed image.

[0135] In detail, the modules in the CBCT image bone sclerosis artifact correction device 100 described in this embodiment of the invention employ the same methods as described above during use. Figures 1 to 3 The method used is the same as the CBCT image bone sclerosis artifact correction method described above, and can produce the same technical effect, so it will not be repeated here.

[0136] like Figure 5 The diagram shown is a schematic diagram of the electronic device for the CBCT image bone sclerosis artifact correction method provided in an embodiment of the present invention.

[0137] The electronic device 1 may include a processor 10, a memory 11, a communication bus 12, and a communication interface 13. It may also include a computer program stored in the memory 11 and capable of running on the processor 10, such as a CBCT image bone sclerosis artifact correction program.

[0138] In some embodiments, the processor 10 may be composed of integrated circuits, such as a single packaged integrated circuit or multiple integrated circuits with the same or different functions, including combinations of one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and various control chips. The processor 10 is the control unit of the electronic device, connecting various components of the entire electronic device through various interfaces and lines. It executes programs or modules stored in the memory 11 (e.g., CBCT image bone sclerosis artifact correction programs) and calls data stored in the memory 11 to perform various functions of the electronic device and process data.

[0139] The memory 11 includes at least one type of readable storage medium, including flash memory, portable hard drive, multimedia card, card-type memory (e.g., SD or DX memory), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the memory 11 can be an internal storage unit of an electronic device, such as a portable hard drive. In other embodiments, the memory 11 can be an external storage device of the electronic device, such as a plug-in portable hard drive, Smart Media Card (SMC), Secure Digital (SD) card, Flash Card, etc. Furthermore, the memory 11 can include both internal and external storage units of the electronic device. The memory 11 can be used not only to store application software and various types of data installed on the electronic device, such as the code for CBCT image bone sclerosis artifact correction programs, but also to temporarily store data that has been output or will be output.

[0140] The communication bus 12 can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. This bus can be divided into an address bus, a data bus, a control bus, etc. The bus is configured to enable communication between the memory 11 and at least one processor 10, etc.

[0141] The communication interface 13 is used for communication between the aforementioned 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, Bluetooth interface, etc.), typically used to establish communication connections between the electronic device and other electronic devices. The user interface may be a display, an input unit (such as a keyboard), or, optionally, a standard wired or wireless interface. Optionally, in this embodiment, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen, etc. The display may also be appropriately referred to as a screen or display unit, used to display information processed in the electronic device and to display a visual user interface.

[0142] 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, or combine certain components, or have different component arrangements.

[0143] For example, although not shown, the electronic device may also include a power supply (such as a battery) to power the various components. Preferably, the power supply can be logically connected to the at least one processor 10 through a power management device, thereby enabling functions such as charging management, discharging management, and power consumption management. The power supply may also include one or more DC or AC power supplies, recharging devices, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components. The electronic device may also include various sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be described in detail here.

[0144] It should be understood that the embodiments described are for illustrative purposes only and are not limited to this structure in the scope of the patent application.

[0145] The CBCT image bone sclerosis artifact correction program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions, which, when run in the processor 10, can achieve the following:

[0146] High-energy projection images and low-energy projection images of the target object are acquired using a pre-set scanning device;

[0147] Image reconstruction is performed on the high-energy projection image and the low-energy projection image to obtain the reconstructed high-energy projection image and the reconstructed low-energy projection image;

[0148] A primary artifact-free image is calculated based on the reconstructed high-energy projection image and the reconstructed low-energy projection image;

[0149] Bone images and soft tissue images are extracted from the primary artifact-free images based on threshold segmentation, and bone weights are calculated based on the bone images and the primary artifact-free images.

[0150] The artifact image of the target object is obtained as training input data, and the bone image and the soft tissue image are used as reference data for the training input data. A dual-channel artifact correction model is trained based on the training input data and the reference data.

[0151] The dual-channel artifact correction model is used to output preprocessed bone images and preprocessed soft tissue images based on the pre-acquired bone sclerosis artifact images to be processed.

[0152] The preprocessed bone image and the preprocessed soft tissue image are weighted and fused according to the bone weights to obtain an artifact-free reconstructed image.

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

[0154] Furthermore, if the modules / units integrated in the electronic device 1 are implemented as software functional units and sold or used as independent products, they 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 may include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, or a read-only memory (ROM).

[0155] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor of an electronic device, can perform the following:

[0156] High-energy projection images and low-energy projection images of the target object are acquired using a pre-set scanning device;

[0157] Image reconstruction is performed on the high-energy projection image and the low-energy projection image to obtain the reconstructed high-energy projection image and the reconstructed low-energy projection image;

[0158] A primary artifact-free image is calculated based on the reconstructed high-energy projection image and the reconstructed low-energy projection image;

[0159] Bone images and soft tissue images are extracted from the primary artifact-free images based on threshold segmentation, and bone weights are calculated based on the bone images and the primary artifact-free images.

[0160] The artifact image of the target object is obtained as training input data, and the bone image and the soft tissue image are used as reference data for the training input data. A dual-channel artifact correction model is trained based on the training input data and the reference data.

[0161] The dual-channel artifact correction model is used to output preprocessed bone images and preprocessed soft tissue images based on the pre-acquired bone sclerosis artifact images to be processed.

[0162] The preprocessed bone image and the preprocessed soft tissue image are weighted and fused according to the bone weights to obtain an artifact-free reconstructed image.

[0163] In the embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and other division methods may be used in actual implementation.

[0164] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0165] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0166] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0167] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0168] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0169] Furthermore, 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 recited in a system claim may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method of CBCT image osteosclerotic artifact correction, characterized in that, The method comprises: acquiring high-energy projection images and low-energy projection images of a target object by using a preset scanning device; reconstructing the high-energy projection images and the low-energy projection images to obtain reconstructed high-energy projection images and reconstructed low-energy projection images; calculating a primary artifact-free image according to the reconstructed high-energy projection images and the reconstructed low-energy projection images; extracting a bone image and a soft tissue image from the primary artifact-free image based on threshold segmentation, and calculating a bone weight according to the bone image and the primary artifact-free image; obtaining an artifact image of the target object as training input data, taking the bone image and the soft tissue image as reference data of the training input data, and training a double-channel artifact correction model according to the training input data and the reference data; outputting a preprocessed bone image and a preprocessed soft tissue image according to a pre-acquired to-be-processed bone-hardening artifact image by using the double-channel artifact correction model; weighting and fusing the preprocessed bone image and the preprocessed soft tissue image according to the bone weight to obtain an artifact-free reconstructed image.

2. The CBCT image bone hardening artifact correction method of claim 1, wherein, The method comprises: acquiring high-energy projection images of the target object according to preset first scanning parameters by using the scanning device; acquiring low-energy projection images of the target object according to preset second scanning parameters by using the scanning device.

3. The CBCT image bone hardening artifact correction method of claim 2, wherein, The method comprises: determining an inverse projection path of each pixel point of the high-energy projection images and the low-energy projection images according to a pre-determined projection angle; integrating the high-energy projection images and the low-energy projection images according to the inverse projection path, and accumulating the integration results to a preset reconstruction space to obtain inverse projection high-energy images and inverse projection low-energy images; performing Fourier transform on the inverse projection high-energy images and the inverse projection low-energy images to obtain frequency domain high-energy images and frequency domain low-energy images; multiplying the frequency domain high-energy images and the frequency domain low-energy images by a pre-selected filter function, and then performing inverse Fourier transform to obtain the reconstructed high-energy projection images and the reconstructed low-energy projection images.

4. The CBCT image bone hardening artifact correction method of claim 1, wherein, The method comprises: calculating the primary artifact-free image by using the following formula: I t = I H + α(I H - I L ) where I t is the primary artifact-free image, I H is the reconstructed high-energy projection image, I L is the reconstructed low-energy projection image, and a is a predetermined weighting factor.

5. The CBCT image bone hardening artifact correction method of claim 1, wherein, The method comprises: calculating a gray level histogram of the primary artifact-free image; extracting peak value features of the gray level histogram, and determining a segmentation threshold according to the peak value features; extracting an image part with a gray value greater than the segmentation threshold in the primary artifact-free image to obtain a primary bone image; extracting an image part with a gray value less than or equal to the segmentation threshold in the primary artifact-free image to obtain a primary soft tissue image; performing image filtering processing on the primary bone image and the primary soft tissue image to obtain a bone image and a soft tissue image.

6. The CBCT image bone hardening artifact correction method of claim 1, wherein, The calculating the bone weight according to the bone image and the primary artifact-free image comprises: obtaining area data of the artifact-free image to obtain an artifact-free image area; obtaining area data of the bone image to obtain a bone image area; calculating a ratio of the bone image area to the artifact-free image area to obtain a bone weight.

7. The CBCT image bone hardening artifact correction method of claim 1, wherein, The training of the dual-channel artifact correction model according to the training input data and the reference data comprises: obtaining a predicted output result according to the training input data by using a pre-established neural network model; calculating a difference value between the predicted output result and the reference data by using a pre-defined loss function; calculating a gradient of a loss value function for each parameter according to the difference value by using a back propagation algorithm to obtain gradient information; updating parameters of the neural network model according to the gradient information by using a gradient descent algorithm; when a preset iteration number is reached, confirming that the model training is completed to obtain the dual-channel artifact correction model.

8. A CBCT image bone hardening artifact correction apparatus, characterized by, The device comprises: an image acquisition module configured to acquire high-energy projection images and low-energy projection images of a target object by using a preset scanning device; an image reconstruction module configured to perform image reconstruction on the high-energy projection images and the low-energy projection images to obtain reconstructed high-energy projection images and reconstructed low-energy projection images; a primary artifact-free image is calculated according to the reconstructed high-energy projection images and the reconstructed low-energy projection images; an image segmentation module configured to extract a bone image and a soft tissue image from the primary artifact-free image based on threshold segmentation, and to calculate a bone weight according to the bone image and the primary artifact-free image; a model training module configured to obtain an artifact image of the target object as training input data, to obtain the bone image and the soft tissue image as reference data of the training input data, and to train a dual-channel artifact correction model according to the training input data and the reference data; an artifact correction module configured to output a preprocessed bone image and a preprocessed soft tissue image according to a previously obtained bone-hardening artifact image to be processed by using the dual-channel artifact correction model, and to obtain an artifact-free reconstructed image by performing weighted fusion on the preprocessed bone image and the preprocessed soft tissue image according to the bone weight.

9. An electronic device, comprising: The electronic device comprises: at least one processor; and a memory in communication connection with 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 to enable the at least one processor to execute the CBCT image bone-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, The computer program is executed by the processor to implement the CBCT image bone-hardening artifact correction method according to any one of claims 1 to 7.

Images