Dual-energy cone beam CT metal artifact suppression reconstruction method, device, equipment and medium

Abstract

The invention relates to a deep learning technology, and provides a dual-energy cone beam CT metal artifact suppression reconstruction method, device and equipment and a storage medium, and the method comprises the steps: carrying out the attenuation processing of high-energy original data and low-energy original data which are obtained in advance, and obtaining a dual-energy initial projection image of a scanned object; sequentially performing three-dimensional reconstruction, threshold segmentation and forward projection on the dual-energy initial projection image to obtain a metal mask projection image; carrying out metal artifact suppression processing on the high-energy initial reconstruction image and then carrying out forward projection to obtain a metal prior projection image; performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal prior projection image to obtain a dual-energy correction projection image; and generating a dual-energy correction reconstructed image of the dual-energy correction projection image, and performing energy spectrum synthesis on the dual-energy correction reconstructed image to obtain a virtual single-energy image with metal artifacts eliminated. According to the invention, the quality of reconstructed images can be improved.

Description

Technical Field

[0001] The present invention relates to the field of deep learning technology, and in particular to a dual-energy cone-beam CT metal artifact suppression and reconstruction method, device, equipment and medium. Background Art

[0002] Cone-Beam Computed Tomography (CBCT) has been widely used in oral and maxillofacial imaging examinations due to its advantages of high resolution and low radiation dose. However, in actual dental scanning examinations, the patient's mouth usually contains metal objects such as implants or crowns. These metal objects will strongly absorb low-energy photons in the X-ray beam, resulting in increased X-ray beam hardening and even photon starvation effects. This will cause serious metal artifacts in the reconstructed image, interfering with the doctor's reading and diagnosis process.

[0003] In response to this problem, although traditional metal artifact correction methods such as projection interpolation correction can suppress metal artifacts to a certain extent, secondary artifacts are often generated during the processing process, resulting in incomplete artifact elimination or blurred tooth structures near the metal. Dual-energy cone-beam CT can provide energy spectrum images with relatively single energy, which has a better suppression effect on metal artifacts than the mixed energy images of conventional cone-beam CT. However, energy spectrum images at lower energies still show obvious metal artifacts. Therefore, for dental dual-energy cone-beam CT, it is still necessary to develop effective metal artifact suppression and reconstruction methods to obtain higher quality reconstructed images. Summary of the invention

[0004] The present invention provides a dual-energy cone-beam CT metal artifact suppression reconstruction method, device, equipment and computer-readable storage medium, the main purpose of which is to solve the problem of low quality during image reconstruction.

[0005] To achieve the above object, the present invention provides a dual-energy cone-beam CT metal artifact suppression and reconstruction method, comprising:

[0006] Performing attenuation processing on the high-energy raw data and the low-energy raw data acquired in advance to obtain a dual-energy initial projection image of the scanned object;

[0007] Performing three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object;

[0008] Performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain a metal mask image of the scanned object, and performing forward projection on the metal mask image to obtain a metal mask projection image of the scanned object;

[0009] Performing metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward-projecting the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object;

[0010] Performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain a dual-energy corrected projection image of the scanned object;

[0011] A dual-energy corrected reconstructed image of the dual-energy corrected projection image is generated, and energy spectrum synthesis is performed on the dual-energy corrected reconstructed image to obtain a virtual monoenergetic image with metal artifacts eliminated.

[0012] Optionally, performing attenuation processing on the pre-acquired high-energy raw data and low-energy raw data to obtain a dual-energy initial projection image of the scanned object includes:

[0013] The high-energy raw data and the low-energy raw data acquired in advance are attenuated according to the Lambert-Beer law to obtain a dual-energy initial projection image of the scanned object, wherein the Lambert-Beer law is:

[0014]

[0015] Among them, p h is the high-energy initial projection image in the dual-energy initial projection image, p l is the low-energy initial projection image in the dual-energy initial projection image, s h0 is the high-energy incident photon intensity data, s h is the high-energy raw data acquired in advance, s l0 is the low-energy incident photon intensity data, s l It is the low-energy raw data acquired in advance.

[0016] Optionally, performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain the metal mask image of the scanned object includes:

[0017] The high-energy initial reconstructed image in the dual-energy initial reconstructed image is threshold segmented using a preset segmentation algorithm to obtain a metal mask image of the scanned object, wherein the preset segmentation algorithm is:

[0018]

[0019] Among them, M I (i) is the i-th pixel in the metal mask image, I h (i) is the i-th pixel in the high-energy initial reconstructed image, T mis the preset segmentation threshold, and i is the coordinate of the i-th pixel in the image.

[0020] Optionally, the performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain the dual-energy corrected projection image of the scanned object includes:

[0021] The dual-energy initial projection image is corrected according to the metal mask projection image, the metal priori projection image and a preset image correction algorithm to obtain a dual-energy corrected projection image of the scanned object, wherein the preset image correction algorithm is:

[0022]

[0023] Among them, p l_correct is the low-energy corrected projection image in the dual-energy corrected projection image, p h_correct is the high-energy corrected projection image in the dual-energy corrected projection image, M p is the metal mask projection image, p prior is the metal prior projection image, p h is the high-energy initial projection image in the dual-energy initial projection image, p l It is the low-energy initial projection image in the dual-energy initial projection image.

[0024] Optionally, generating the dual-energy corrected reconstructed image of the dual-energy corrected projection image includes:

[0025] The dual-energy corrected projection image is three-dimensionally reconstructed, and a metal backfilling process is performed on the three-dimensionally reconstructed dual-energy corrected projection image according to the metal mask image to obtain a dual-energy corrected reconstructed image of the scanned object.

[0026] Optionally, performing energy spectrum synthesis on the dual-energy corrected reconstructed image to obtain a virtual single-energy image with metal artifacts removed includes:

[0027] Generate energy spectrum dual-material decomposition coefficients of the dual-energy corrected reconstructed image;

[0028] The dual-energy corrected reconstructed image is subjected to energy spectrum synthesis according to the energy spectrum dual-material decomposition coefficient and a preset virtual monoenergetic image synthesis algorithm to obtain a virtual monoenergetic image with metal artifacts eliminated, wherein the preset virtual monoenergetic image synthesis algorithm is:

[0029] I E =b1μ 1_E +b2μ 2_E

[0030] Among them, I Eis a virtual monoenergetic image with metal artifacts eliminated under the preset single-energy level E condition, b1 is the first decomposition coefficient of the energy spectrum dual material, μ 1_E is the attenuation coefficient of the first basis material under single-level E conditions, b2 is the second decomposition coefficient of the energy spectrum double material, μ 2_E is the attenuation coefficient of the second base material under single-level E conditions.

[0031] Optionally, the energy spectrum dual-material decomposition coefficients for generating the dual-energy corrected reconstructed image include:

[0032] The energy spectrum dual-material decomposition coefficients of the dual-energy corrected reconstructed image are generated according to the following algorithm:

[0033]

[0034] Where b1 is the first decomposition coefficient of the energy spectrum double material, I h_correct is the high energy corrected reconstructed image in the dual energy corrected reconstructed image, b2 is the second decomposition coefficient of the energy spectrum dual material, I l_correct is the low energy corrected reconstructed image in the dual energy corrected reconstructed image, μ 1_h is the attenuation coefficient of the first basis material under high energy conditions, μ 2_h is the attenuation coefficient of the second basis material under high energy conditions, μ 2_l is the attenuation coefficient of the first basis material under low energy conditions, μ 1_l is the attenuation coefficient of the second base material under low energy conditions.

[0035] In order to solve the above problems, the present invention also provides a dual-energy cone-beam CT metal artifact suppression and reconstruction device, the device comprising:

[0036] A data attenuation processing module is used to perform attenuation processing on the pre-acquired high-energy raw data and low-energy raw data to obtain a dual-energy initial projection image of the scanned object;

[0037] An image three-dimensional reconstruction module, used for performing three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object;

[0038] An image threshold segmentation module, used for performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain a metal mask image of the scanned object, and performing forward projection on the metal mask image to obtain a metal mask projection image of the scanned object;

[0039] A metal artifact suppression module, used to perform metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward project the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object;

[0040] An image information correction module, used for performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain a dual-energy corrected projection image of the scanned object;

[0041] The image energy spectrum synthesis module is used to generate a dual-energy corrected reconstructed image of the dual-energy corrected projection image, and perform energy spectrum synthesis on the dual-energy corrected reconstructed image to obtain a virtual single-energy image with metal artifacts eliminated.

[0042] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising:

[0043] at least one processor; and,

[0044] a memory communicatively connected to the at least one processor; wherein,

[0045] 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 cone-beam CT metal artifact suppression and reconstruction method described above.

[0046] 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 cone-beam CT metal artifact suppression and reconstruction method.

[0047] The embodiment of the present invention is based on a priori image generation method based on deep learning, combined with dual-energy cone-beam CT technology, to achieve effective suppression of metal artifacts and high-quality image reconstruction. By performing threshold segmentation and forward projection on the high-energy initial reconstructed image, a metal mask image and a metal mask projection image are generated to accurately identify the location of the metal artifact. The pre-trained metal artifact suppression network model performs deep processing on these images to effectively reduce the impact of metal artifacts. In addition, the dual-energy initial projection image is combined with the metal mask projection image and the metal priori projection image to perform information correction, and energy spectrum synthesis is performed to finally generate a virtual monoenergetic image without metal artifacts. This series of steps significantly improves the quality of the reconstructed image. Therefore, the dual-energy cone-beam CT metal artifact suppression reconstruction method, device, equipment and storage medium proposed in the present invention can solve the problem of low quality of the reconstructed image. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 A schematic diagram of a process for metal artifact suppression and reconstruction of dual-energy cone-beam CT provided by an embodiment of the present invention;

[0049] Figure 2 A functional module diagram of a dual-energy cone-beam CT metal artifact suppression and reconstruction device provided by an embodiment of the present invention;

[0050] Figure 3 A schematic structural diagram of an electronic device for implementing the dual-energy cone-beam CT metal artifact suppression and reconstruction method provided by an embodiment of the present invention.

[0051] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0052] 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.

[0053] The embodiment of the present application provides a dual-energy cone-beam CT metal artifact suppression and reconstruction method. The execution subject of the dual-energy cone-beam CT metal artifact suppression and reconstruction method includes but is not limited to at least one of the electronic devices such as a server, a terminal, etc. that can be configured to execute the method provided in the embodiment of the present application. In other words, the dual-energy cone-beam CT metal artifact suppression and reconstruction 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.

[0054] Reference Figure 1 FIG. 1 is a flow chart of a dual-energy cone-beam CT metal artifact suppression and reconstruction method provided by an embodiment of the present invention. In this embodiment, the dual-energy cone-beam CT metal artifact suppression and reconstruction method includes:

[0055] S1. Perform attenuation processing on the high-energy raw data and the low-energy raw data acquired in advance to obtain a dual-energy initial projection image of the scanned object.

[0056] In an embodiment of the present invention, the attenuation processing is performed on the pre-acquired high-energy raw data and the low-energy raw data to obtain a dual-energy initial projection image of the scanned object, including:

[0057] The high-energy raw data and the low-energy raw data acquired in advance are attenuated according to the Lambert-Beer law to obtain a dual-energy initial projection image of the scanned object, wherein the Lambert-Beer law is:

[0058]

[0059] Among them, p h is the high-energy initial projection image in the dual-energy initial projection image, p l is the low-energy initial projection image in the dual-energy initial projection image, s h0 is the high-energy incident photon intensity data, s h is the high-energy raw data acquired in advance, s l0 is the low-energy incident photon intensity data, s l It is the low-energy raw data acquired in advance.

[0060] In detail, in a dual-energy cone-beam CT system, the process of acquiring high- and low-energy raw projection data can be achieved through different technologies, such as two-round repeated dual-energy scanning, high-voltage rapid switching, or dual-tube schemes, all of which can obtain the response data of the scanned object at different energy levels.

[0061] In detail, the two-round repeated dual-energy scanning refers to performing two independent scans using high-energy and low-energy X-rays respectively, thereby obtaining two sets of projection data with different energies.

[0062] In detail, high-voltage fast switching refers to the rapid alternation of the voltage of the X-ray source during a single scan, thereby achieving the acquisition of high-energy and low-energy data in the same scan.

[0063] In detail, the dual tube system refers to the use of two X-ray sources, one set to high energy output and the other set to low energy output, to scan an object simultaneously.

[0064] Furthermore, after obtaining the high-energy and low-energy raw data, it is necessary to apply the Lambert-Beer law to process these data to obtain attenuation-related projection images. The Lambert-Beer law describes the attenuation of light when it passes through an absorbing medium.

[0065] In detail, the high-energy and low-energy incident photon intensity data refer to the number of photons received by the detector when there is no object; the high-energy and low-energy raw data refer to the number of outgoing photons after the ray passes through the scanned object.

[0066] In detail, through logarithmic operations, the measured ratio of the number of outgoing photons to the number of incident photons can be converted into the degree of light attenuation, thereby obtaining attenuation images for each energy level. These images reflect the absorption and scattering characteristics of the object to photons of different energies, providing the necessary data for subsequent image reconstruction and metal artifact suppression.

[0067] S2. Perform three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object.

[0068] In an embodiment of the present invention, the three-dimensional reconstruction of the dual-energy initial projection image to obtain the dual-energy initial reconstructed image of the scanned object includes: performing three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image, that is, a high-energy initial reconstructed image and a low-energy initial reconstructed image. Here, the three-dimensional filtered back projection FDK (Feldkamp-Davis-Kress) reconstruction method is mainly used to convert the dual-energy original projection image to obtain a high-energy initial reconstructed image and a low-energy initial reconstructed image respectively. The FDK reconstruction method mainly performs ramp filtering operations and weighted back projection operations on the projection data to realize the conversion of the projection image to the spatial image.

[0069] Specifically, the image reconstruction process of dual-energy cone-beam CT involves the conversion from initial projection data to three-dimensional space images, which mainly relies on the FDK (Feldkamp-Davis-Kress) reconstruction algorithm.

[0070] In detail, the original dual-energy projection data (high energy and low energy) are first subjected to necessary preprocessing, such as logarithmic transformation and calibration, to prepare the data for subsequent filtering and back-projection operations, and then ramp filtering is performed on the preprocessed projection data, wherein ramp filtering is a frequency domain filtering method used to enhance the spatial resolution of the image and reduce noise, and the filtering process typically involves converting the projection data to the frequency domain, applying a filter, and then converting back to the time domain.

[0071] Furthermore, the filtered projection data is transformed into three-dimensional space through a weighted back-projection operation. In the FDK algorithm, this weighting takes into account the relative angle of the ray path to the reconstruction plane. Back-projection is to "reverse" the filtered data line through the entire three-dimensional volume and assign it to the corresponding voxel according to the geometric path of the original ray.

[0072] Furthermore, since cone-beam scanning usually involves circular or other forms of scanning trajectories, the FDK algorithm reconstructs the value of each voxel by integrating projections at different angles. In dual-energy cone-beam CT, this process is performed separately for high-energy and low-energy projection data to generate two independent sets of reconstructed images.

[0073] S3. Perform threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain a metal mask image of the scanned object, and perform forward projection on the metal mask image to obtain a metal mask projection image of the scanned object.

[0074] In an embodiment of the present invention, performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain the metal mask image of the scanned object includes:

[0075] The high-energy initial reconstructed image in the dual-energy initial reconstructed image is threshold segmented using a preset segmentation algorithm to obtain a metal mask image of the scanned object, wherein the preset segmentation algorithm is:

[0076]

[0077] Among them, M I (i) is the i-th pixel in the metal mask image, I h (i) is the i-th pixel in the high-energy initial reconstructed image, T m is the preset segmentation threshold, and i is the coordinate of the i-th pixel in the image.

[0078] In detail, threshold segmentation is a basic and effective image processing technique used to classify pixels in an image into different categories, usually based on pixel intensity values. In the present invention, this method is used to distinguish metal and non-metal pixels.

[0079] In detail, the preset segmentation threshold is a pre-set threshold used to determine whether each pixel in the high-energy initial reconstructed image belongs to metal. The selection of this threshold is very critical because it directly affects the accuracy of the metal mask.

[0080] In detail, the selection of the threshold is usually based on experimental data or prior knowledge to ensure accurate discrimination of metals from other materials.

[0081] Furthermore, M I (i) is the pixel value of the metal mask image at coordinate i. If the pixel value of the high-energy initial reconstructed image at coordinate i is greater than or equal to the threshold, the pixel is considered to belong to the metal area and is marked as 1 in the metal mask image; otherwise, it is marked as 0, indicating a non-metal pixel.

[0082] In an embodiment of the present invention, the metal mask image is forward projected to obtain the metal mask projection image of the scanned object, wherein forward projection is a process of simulating X-rays passing through an object from a ray source to a detector. It is a mapping process from a three-dimensional object space to a two-dimensional detector plane. Contrary to the back-projection operation used in the FDK reconstruction algorithm, forward projection is a method of converting a volume image into a projection image, which is used to simulate an actual physical projection process.

[0083] S4. Performing metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward-projecting the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object.

[0084] In an embodiment of the present invention, metal artifact suppression processing is a key step, which utilizes a pre-trained deep learning model, usually based on a Unet network architecture.

[0085] In detail, the metal artifact suppression network model is mainly composed of an encoder and a decoder, in which the encoder gradually reduces the resolution of the input image through the maximum pooling layer to generate multi-scale feature layers, and the decoder gradually increases the resolution of these feature layers through the deconvolution layer, and jumps with the encoder feature layers of the corresponding resolution to integrate information of different scales.

[0086] In detail, the feature layer of each scale includes convolutional layer, activation layer and channel attention layer, which enhances the network's ability to learn image features.

[0087] In detail, the channel attention module improves the feature focusing ability of the network by learning the importance of each channel. It first performs global average pooling on each channel of the input feature image, then adjusts the channel weights through a 1x1 convolutional layer, and finally multiplies these weights with the original input feature image to obtain an output image that strengthens the features of specific channels.

[0088] In detail, the training samples include reconstructed images without metal artifacts (as labels) and reconstructed images with metal artifacts (as input). The reconstructed images without metal artifacts are obtained by directly performing 3D reconstruction on the metal-free projection data. The reconstructed images with metal artifacts are obtained by adding metal information to the metal-free projection data and then performing 3D reconstruction.

[0089] Furthermore, these paired samples (images with metal artifacts and images without metal artifacts) are used to train a metal artifact suppression network. The training goal is to minimize the difference between the network output and the image without metal artifacts, so that the network can learn how to remove these artifacts from images with metal artifacts.

[0090] In an embodiment of the present invention, by forward-projecting the image after metal artifact suppression processing again, it can be ensured that in the final CT image, the artifacts caused by metal are effectively reduced, which not only improves the overall quality of the image, but also improves the diagnostic value of the CT scan.

[0091] S5. Correct the information of the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain a dual-energy corrected projection image of the scanned object.

[0092] In an embodiment of the present invention, the information correction of the dual-energy initial projection image according to the metal mask projection image and the metal prior projection image to obtain the dual-energy corrected projection image of the scanned object includes:

[0093] The dual-energy initial projection image is corrected according to the metal mask projection image, the metal priori projection image and a preset image correction algorithm to obtain a dual-energy corrected projection image of the scanned object, wherein the preset image correction algorithm is:

[0094]

[0095] Among them, p l_correct is the low-energy corrected projection image in the dual-energy corrected projection image, p h_correct is the high-energy corrected projection image in the dual-energy corrected projection image, M p is the metal mask projection image, p prior is the metal prior projection image, p h is the high-energy initial projection image in the dual-energy initial projection image, p l It is the low-energy initial projection image in the dual-energy initial projection image.

[0096] In an embodiment of the present invention, a method for correcting a dual-energy initial projection image using a metal mask projection image and a metal priori projection image is described, aiming to reduce or eliminate artifacts caused by metal objects, thereby improving the quality of dual-energy CT images.

[0097] In detail, the correction algorithm is based on a weighted fusion strategy, in which the metal mask projection image is used to determine which parts of the projection data need to be corrected by the metal prior projection image.

[0098] In detail, at the corresponding pixel position of each projection image, if the value in the metal mask projection image is 1 (i.e., the position is identified as a metal area), the projection value of the position will be mainly determined by the metal prior projection image, which represents the correction information generated by the metal artifact suppression network model. If the value in the metal mask projection image is 0 (i.e., the position is not a metal area), the projection value of the position will remain the original p h or p l value.

[0099] In detail, through this correction method, the metal prior projection image can be effectively used to correct those areas affected by metal while keeping the information of non-metal areas unchanged.

[0100] S6. Generate a dual-energy corrected reconstructed image of the dual-energy corrected projection image, perform energy spectrum synthesis on the dual-energy corrected reconstructed image, and obtain a virtual monoenergetic image with metal artifacts eliminated.

[0101] In an embodiment of the present invention, the step of generating the dual-energy corrected reconstructed image of the dual-energy corrected projection image includes:

[0102] The dual-energy corrected projection image is three-dimensionally reconstructed, and a metal backfilling process is performed on the three-dimensionally reconstructed dual-energy corrected projection image according to the metal mask image to obtain a dual-energy corrected reconstructed image of the scanned object.

[0103] In detail, the two corrected projection images are three-dimensionally reconstructed respectively using a standard CT image reconstruction algorithm (such as the FDK algorithm), which will obtain a dual-energy corrected reconstructed image.

[0104] In detail, the previously generated metal mask image is used to determine which areas need to be corrected or "backfilled" during the reconstruction process. On the basis of the three-dimensional reconstruction, metal backfill is performed on these areas according to the metal mask image. This means that in the metal area, appropriate physical or statistical models will be used to estimate and fill in the artifacts or missing information caused by the metal. After the metal backfilling process, the final high-energy corrected reconstructed images and low-energy corrected reconstructed images are obtained. These images should have fewer metal artifacts and more accurately reflect the structure of the scanned object.

[0105] In detail, the following algorithm can be used to perform metal backfilling processing on the dual-energy corrected projection image after 3D reconstruction:

[0106]

[0107] Among them, I h_correct is the high energy corrected reconstructed image, I l_correct is the low energy corrected reconstructed image, M I is the metal mask image, I h is the high-energy initial reconstructed image, I l is the low-energy initial reconstructed image.

[0108] In an embodiment of the present invention, the step of performing energy spectrum synthesis on the dual-energy corrected reconstructed image to obtain a virtual single-energy image with metal artifacts eliminated includes:

[0109] Generate energy spectrum dual-material decomposition coefficients of the dual-energy corrected reconstructed image;

[0110] The dual-energy corrected reconstructed image is subjected to energy spectrum synthesis according to the energy spectrum dual-material decomposition coefficient and a preset virtual monoenergetic image synthesis algorithm to obtain a virtual monoenergetic image with metal artifacts eliminated, wherein the preset virtual monoenergetic image synthesis algorithm is:

[0111] I E =b1μ 1_E +b2μ 2_E

[0112] Among them, I E is a virtual monoenergetic image with metal artifacts eliminated under the preset single-energy level E condition, b1 is the first decomposition coefficient of the energy spectrum dual material, μ1_E is the attenuation coefficient of the first basis material under single-level E conditions, b2 is the second decomposition coefficient of the energy spectrum double material, μ 2_E is the attenuation coefficient of the second base material under single-level E conditions.

[0113] In detail, the energy spectrum dual-material decomposition coefficients for generating the dual-energy corrected reconstructed image include:

[0114] The energy spectrum dual-material decomposition coefficients of the dual-energy corrected reconstructed image are generated according to the following algorithm:

[0115]

[0116] Where b1 is the first decomposition coefficient of the energy spectrum double material, I h_correct is the high energy corrected reconstructed image in the dual energy corrected reconstructed image, b2 is the second decomposition coefficient of the energy spectrum dual material, I l_correct is the low energy corrected reconstructed image in the dual energy corrected reconstructed image, μ 1_h is the attenuation coefficient of the first basis material under high energy conditions, μ 2_h is the attenuation coefficient of the second basis material under high energy conditions, μ 2_l is the attenuation coefficient of the first basis material under low energy conditions, μ 1_l is the attenuation coefficient of the second base material under low energy conditions.

[0117] In detail, water and iodine are often chosen as base substances because water represents soft tissues in the human body, while iodine is a commonly used contrast agent, both of which are representative in medical imaging.

[0118] First, the attenuation coefficients of two base substances (e.g., water and iodine) under high and low energy conditions need to be determined.

[0119] The embodiment of the present invention is based on a priori image generation method based on deep learning, combined with dual-energy cone-beam CT technology, to achieve effective suppression of metal artifacts and high-quality image reconstruction. By performing threshold segmentation and forward projection on the high-energy initial reconstructed image, a metal mask image and a metal mask projection image are generated, and the location of the metal artifact is accurately identified. The pre-trained metal artifact suppression network model performs deep processing on these images to effectively reduce the impact of metal artifacts. In addition, the dual-energy initial projection image is combined with the metal mask projection image and the metal priori projection image to perform information correction, and energy spectrum synthesis is performed to finally generate a virtual monoenergetic image without metal artifacts. This series of steps significantly improves the quality of the reconstructed image. Therefore, the dual-energy cone-beam CT metal artifact suppression reconstruction method proposed in the present invention can solve the problem of low quality of the reconstructed image.

[0120] like Figure 2, which is a functional module diagram of a dual-energy cone-beam CT metal artifact suppression and reconstruction device provided by an embodiment of the present invention.

[0121] The dual-energy cone-beam CT metal artifact suppression and reconstruction device 100 of the present invention can be installed in an electronic device. According to the functions to be implemented, the dual-energy cone-beam CT metal artifact suppression and reconstruction device 100 can include a data attenuation processing module 101, an image three-dimensional reconstruction module 102, an image threshold segmentation module 103, a metal artifact suppression module 104, an image information correction module 105 and an image energy spectrum synthesis module 106. The module of the present invention can also be called 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.

[0122] In this embodiment, the functions of each module / unit are as follows:

[0123] The data attenuation processing module 101 is used to perform attenuation processing on the pre-acquired high-energy original data and low-energy original data to obtain a dual-energy initial projection image of the scanned object;

[0124] The image 3D reconstruction module 102 is used to perform 3D reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object;

[0125] The image threshold segmentation module 103 is used to perform threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain the metal mask image of the scanned object, and forward project the metal mask image to obtain the metal mask projection image of the scanned object;

[0126] The metal artifact suppression module 104 is used to perform metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward project the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object;

[0127] The image information correction module 105 is used to correct the information of the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain the dual-energy corrected projection image of the scanned object;

[0128] The image energy spectrum synthesis module 106 is used to generate a dual-energy corrected reconstructed image of the dual-energy corrected projection image, and perform energy spectrum synthesis on the dual-energy corrected reconstructed image to obtain a virtual mono-energy image with metal artifacts eliminated.

[0129] In detail, each module described in the dual-energy cone-beam CT metal artifact suppression and reconstruction device 100 in the embodiment of the present invention is used in the same manner as described above. Figure 1 The dual-energy cone-beam CT metal artifact suppression and reconstruction 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.

[0130] like Figure 3 , which is a schematic diagram of the structure of an electronic device for implementing a dual-energy cone-beam CT metal artifact suppression and reconstruction method provided by an embodiment of the present invention.

[0131] The electronic device 1 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 cone-beam CT metal artifact suppression reconstruction program.

[0132] 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. 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 or executes programs or modules stored in the memory 11 (for example, executing a dual-energy cone-beam CT metal artifact suppression reconstruction program, etc.), and calls data stored in the memory 11 to execute various functions of the electronic device and process data.

[0133] 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, an 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 may not only be used to store application software and various types of data installed in the electronic device, such as the code of the dual-energy cone-beam CT metal artifact suppression reconstruction program, but also be used to temporarily store data that has been output or is to be output.

[0134] 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.

[0135] 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.

[0136] 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.

[0137] 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.

[0138] It should be understood that the embodiment is for illustration only and the scope of the patent application is not limited to this structure.

[0139] The dual-energy cone-beam CT metal artifact suppression reconstruction program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve:

[0140] Performing attenuation processing on the high-energy raw data and the low-energy raw data acquired in advance to obtain a dual-energy initial projection image of the scanned object;

[0141] Performing three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object;

[0142] Performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain a metal mask image of the scanned object, and performing forward projection on the metal mask image to obtain a metal mask projection image of the scanned object;

[0143] Performing metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward-projecting the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object;

[0144] Performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain a dual-energy corrected projection image of the scanned object;

[0145] A dual-energy corrected reconstructed image of the dual-energy corrected projection image is generated, and energy spectrum synthesis is performed on the dual-energy corrected reconstructed image to obtain a virtual monoenergetic image with metal artifacts eliminated.

[0146] 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.

[0147] 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).

[0148] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement:

[0149] Performing attenuation processing on the high-energy raw data and the low-energy raw data acquired in advance to obtain a dual-energy initial projection image of the scanned object;

[0150] Performing three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object;

[0151] Performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain a metal mask image of the scanned object, and performing forward projection on the metal mask image to obtain a metal mask projection image of the scanned object;

[0152] Performing metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward-projecting the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object;

[0153] Performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain a dual-energy corrected projection image of the scanned object;

[0154] A dual-energy corrected reconstructed image of the dual-energy corrected projection image is generated, and energy spectrum synthesis is performed on the dual-energy corrected reconstructed image to obtain a virtual monoenergetic image with metal artifacts eliminated.

[0155] 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.

[0156] 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.

[0157] 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.

[0158] 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.

[0159] 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.

[0160] The embodiments of the present application can acquire and process relevant data based on deep learning technology. Among them, artificial intelligence (AI) is the theory, method, technology and application system that uses digital computers or machines controlled by digital computers to simulate, extend and expand human intelligence, perceive the environment, acquire knowledge and use knowledge to obtain the best results.

[0161] 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 cone-beam CT metal artifact suppression and reconstruction method, characterized in that: The method comprises: Performing attenuation processing on the high-energy raw data and the low-energy raw data acquired in advance to obtain a dual-energy initial projection image of the scanned object; Performing three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object; Performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain a metal mask image of the scanned object, and performing forward projection on the metal mask image to obtain a metal mask projection image of the scanned object; Performing metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward-projecting the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object; Performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain a dual-energy corrected projection image of the scanned object; A dual-energy corrected reconstructed image of the dual-energy corrected projection image is generated, and energy spectrum synthesis is performed on the dual-energy corrected reconstructed image to obtain a virtual monoenergetic image with metal artifacts eliminated.

2. The dual-energy cone-beam CT metal artifact suppression and reconstruction method according to claim 1, characterized in that: The attenuation processing is performed on the pre-acquired high-energy raw data and the low-energy raw data to obtain a dual-energy initial projection image of the scanned object, including: The high-energy raw data and the low-energy raw data acquired in advance are attenuated according to the Lambert-Beer law to obtain a dual-energy initial projection image of the scanned object, wherein the Lambert-Beer law is: Among them, p h is the high-energy initial projection image in the dual-energy initial projection image, p l is the low-energy initial projection image in the dual-energy initial projection image, s h0 is the high-energy incident photon intensity data, s h is the high-energy raw data acquired in advance, s l0 is the low-energy incident photon intensity data, s l It is the low-energy raw data acquired in advance.

3. The dual-energy cone-beam CT metal artifact suppression and reconstruction method according to claim 1, characterized in that: The step of performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain the metal mask image of the scanned object includes: The high-energy initial reconstructed image in the dual-energy initial reconstructed image is threshold segmented using a preset segmentation algorithm to obtain a metal mask image of the scanned object, wherein the preset segmentation algorithm is: Among them, M I (i) is the i-th pixel in the metal mask image, I h (i) is the i-th pixel in the high-energy initial reconstructed image, T m is the preset segmentation threshold, and i is the coordinate of the i-th pixel in the image.

4. The dual-energy cone-beam CT metal artifact suppression and reconstruction method according to claim 1, characterized in that: The step of performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain the dual-energy corrected projection image of the scanned object includes: The dual-energy initial projection image is corrected according to the metal mask projection image, the metal priori projection image and a preset image correction algorithm to obtain a dual-energy corrected projection image of the scanned object, wherein the preset image correction algorithm is: Among them, p l_correct is the low-energy corrected projection image in the dual-energy corrected projection image, p h_correct is the high-energy corrected projection image in the dual-energy corrected projection image, M p is the metal mask projection image, p prior is the metal prior projection image, p h is the high-energy initial projection image in the dual-energy initial projection image, p l It is the low-energy initial projection image in the dual-energy initial projection image.

5. The dual-energy cone-beam CT metal artifact suppression and reconstruction method according to claim 1, characterized in that: The step of generating the dual-energy corrected reconstructed image of the dual-energy corrected projection image comprises: The dual-energy corrected projection image is three-dimensionally reconstructed, and a metal backfilling process is performed on the three-dimensionally reconstructed dual-energy corrected projection image according to the metal mask image to obtain a dual-energy corrected reconstructed image of the scanned object.

6. The dual-energy cone-beam CT metal artifact suppression and reconstruction method according to claim 1, characterized in that: The step of performing energy spectrum synthesis on the dual-energy corrected and reconstructed image to obtain a virtual single-energy image with metal artifacts eliminated includes: Generate energy spectrum dual-material decomposition coefficients of the dual-energy corrected reconstructed image; The dual-energy corrected reconstructed image is subjected to energy spectrum synthesis according to the energy spectrum dual-material decomposition coefficient and a preset virtual monoenergetic image synthesis algorithm to obtain a virtual monoenergetic image with metal artifacts eliminated, wherein the preset virtual monoenergetic image synthesis algorithm is: I E =b1μ 1_E +b2μ 2_E Among them, I E is a virtual monoenergetic image with metal artifacts eliminated under the preset single-energy level E condition, b1 is the first decomposition coefficient of the energy spectrum dual material, μ 1_E is the attenuation coefficient of the first basis material under single-level E conditions, b2 is the second decomposition coefficient of the energy spectrum double material, μ 2_E is the attenuation coefficient of the second base material under single-level E conditions.

7. The dual-energy cone-beam CT metal artifact suppression and reconstruction method according to claim 6, characterized in that: The energy spectrum dual-material decomposition coefficients for generating the dual-energy corrected reconstructed image include: The energy spectrum dual-material decomposition coefficients of the dual-energy corrected reconstructed image are generated according to the following algorithm: Where b1 is the first decomposition coefficient of the energy spectrum double material, I h_correct is the high energy corrected reconstructed image in the dual energy corrected reconstructed image, b2 is the second decomposition coefficient of the energy spectrum dual material, I l_correct is the low energy corrected reconstructed image in the dual energy corrected reconstructed image, μ 1_h is the attenuation coefficient of the first basis material under high energy conditions, μ 2_h is the attenuation coefficient of the second basis material under high energy conditions, μ 2_l is the attenuation coefficient of the first basis material under low energy conditions, μ 1_l is the attenuation coefficient of the second base material under low energy conditions.

8. A dual-energy cone-beam CT metal artifact suppression and reconstruction device, characterized in that: The device comprises: A data attenuation processing module is used to perform attenuation processing on the pre-acquired high-energy raw data and low-energy raw data to obtain a dual-energy initial projection image of the scanned object; An image three-dimensional reconstruction module, used for performing three-dimensional reconstruction on the dual-energy initial projection image to obtain a dual-energy initial reconstructed image of the scanned object; An image threshold segmentation module, used for performing threshold segmentation on the high-energy initial reconstructed image in the dual-energy initial reconstructed image to obtain a metal mask image of the scanned object, and performing forward projection on the metal mask image to obtain a metal mask projection image of the scanned object; A metal artifact suppression module, used to perform metal artifact suppression processing on the high-energy initial reconstructed image using a pre-trained metal artifact suppression network model, and forward project the high-energy initial reconstructed image after the metal artifact suppression processing to obtain a metal prior projection image of the scanned object; An image information correction module, used for performing information correction on the dual-energy initial projection image according to the metal mask projection image and the metal priori projection image to obtain a dual-energy corrected projection image of the scanned object; The image energy spectrum synthesis module is used to generate a dual-energy corrected reconstructed image of the dual-energy corrected projection image, and perform energy spectrum synthesis on the dual-energy corrected reconstructed image to obtain a virtual single-energy image with metal artifacts eliminated.

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 cone-beam CT metal artifact suppression and reconstruction 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 cone-beam CT metal artifact suppression and reconstruction method according to any one of claims 1 to 7 is implemented.

Images