Method, device and medium suitable for multi-energy cone beam CT projection interpolation completion

Abstract

The application relates to a method, device and medium suitable for multi-energy cone beam CT projection interpolation completion. The method comprises the following steps: acquiring low-energy projection and high-energy projection of a multi-energy cone beam CT energy spectrum imaging system; performing preliminary iterative sparse reconstruction and material decomposition on the low-energy projection and the high-energy projection respectively to obtain a base material one reconstructed image and a base material two reconstructed image; performing an image enhancement operation on the base material one reconstructed image and the base material two reconstructed image respectively, then performing forward projection, and performing weighted processing by using pre-calibrated high-low energy spectrum information to obtain low-energy complete projection data and high-energy complete projection data under full viewing angle. Therefore, the image domain sparse reconstructed image is used to guide the projection domain high-low energy projection data to perform energy spectrum interpolation completion, the problem of inconsistent high-low energy projection angles of a fast kilovoltage switching energy spectrum CBCT is solved, and the image quality and material decomposition accuracy are greatly improved.

Description

Technical Field

[0001] This application relates to the field of imaging technology, and in particular to a method, device and medium suitable for projection interpolation and completion in multi-energy cone-beam CT. Background Technology

[0002] Dual-energy or multi-energy CT (Computed Tomography) imaging is currently one of the hottest research topics in the field of CT. Multi-energy CT utilizes the correlation between the material attenuation coefficient and X-ray energy to provide more information than conventional CT. Therefore, compared to traditional CT, multi-energy CT can suppress hardening artifacts, improve tissue contrast, enhance image quality, and reduce radiation dose, offering many advantages. Currently, mainstream dual-energy CT systems include technologies such as dual-source dual-detector, fast kilovolt switching, and dual-layer detectors. Each of these technologies has its own advantages and disadvantages, and all have been commercially implemented.

[0003] In the past two decades, with the rise of large-area flat-panel detectors, cone-beam computed tomography (CBCT) has become a major academic hotspot in cutting-edge imaging theory and application research due to its advantages such as high integration, high spatial resolution, and convenience and flexibility, and has become a new and important direction for the development of X-ray imaging. CBCT has broad application prospects in many fields such as industry, agriculture, and medicine, and has already played an indispensable role in areas such as human oral (dental) examinations, image-guided interventional treatments, and radiotherapy. However, CBCT still faces limitations such as radiation scattering contamination and radiation hardening, and image quality urgently needs improvement. Fast kV-switching (FKS) technology can provide multiple sets of projection data with different energies without changing the physical structure of the CT source probe, and is expected to improve the image quality and quantitative imaging performance of CBCT. However, extending this technology to CBCT (Cone-Beam Computed Tomography) still faces a series of challenges, such as radiation scattering, sparse sampling, inconsistent high and low energy projection angles, and limited energy separation.

[0004] The fast kilovolt switching technology is limited by the electronic readout speed of the flat panel detector and the response speed of the scintillator crystal, which cannot support high-frequency switching. At the same time, in order to avoid motion artifacts caused by excessively long scanning time in clinical applications, CBCT has a limited number of projection angles that can be acquired within a fixed scanning time. Sparse sampling further aggravates the problem of inconsistent high and low energy projection angles, which can easily cause bright and dark artifacts in the reconstructed images, affecting image quality and the accuracy of material decomposition. This problem urgently needs to be solved. Summary of the Invention

[0005] This application provides a method for projection interpolation and completion applicable to multi-energy cone-beam CT, which solves the problem of inconsistent high and low energy projection angles in rapid kilovolt switching spectral CBCT, greatly improving image quality and material decomposition accuracy.

[0006] To achieve the above objectives, the first aspect of this application proposes a method for projection interpolation completion in multi-energy cone-beam CT, comprising the following steps: Acquire low-energy and high-energy projections of a multi-energy cone-beam CT spectral imaging system; Preliminary iterative sparse reconstruction and matter decomposition were performed on the low-energy projection and the high-energy projection respectively to obtain the reconstructed image of base material one and the reconstructed image of base material two. Image enhancement operations are performed on the reconstructed images of base material one and base material two respectively, followed by forward projection. Weighting is then performed using pre-calibrated high and low energy spectrum information to obtain low-energy complete projection data and high-energy complete projection data under full viewing angle.

[0007] According to one embodiment of this application, after obtaining the low-energy completed projection data and the high-energy completed projection data under full viewing angle, the method further includes: Based on a preset projection domain material decomposition strategy, projection domain material decomposition is performed on the low-energy completed projection data and the high-energy completed projection data respectively, and energy spectrum image reconstruction is performed based on the decomposition results to obtain the final base material one image and the final base material two image.

[0008] According to one embodiment of this application, the preset projection domain matter decomposition strategy is as follows:

[0009] in, For the final base material image, The final image of the base material 2. To complete the low-energy projection data from all angles, To complete the projection data with high energy from all angles, Perform a matter decomposition operation on the projection domain. For image reconstruction operations.

[0010] According to one embodiment of this application, the preliminary iterative sparse reconstruction and matter decomposition of the low-energy projection and the high-energy projection respectively to obtain a reconstructed image of base material one and a reconstructed image of base material two includes: Based on a preset image domain sparse reconstruction strategy, preliminary iterative sparse reconstruction is performed on the low-energy projection and the high-energy projection respectively to obtain low-energy reconstructed images and high-energy reconstructed images. Based on a preset image domain material decomposition strategy, material decomposition operations are performed on the low-energy reconstructed image and the high-energy reconstructed image in the image domain to obtain the base material one reconstructed image and the base material two reconstructed image.

[0011] According to one embodiment of this application, the preset image domain sparse reconstruction strategy is as follows:

[0012] in, It is a 2-norm. For TV regular expression, For low-energy reconstructed images, For high-energy reconstructed images, This is the system matrix corresponding to the low-energy reconstructed image. is the system matrix corresponding to the high-energy reconstructed image; SSIM is the structural similarity regularization term, with values ​​ranging from [-1, 1]. For the corresponding regularization parameter.

[0013] According to one embodiment of this application, the preset image domain material decomposition strategy is as follows:

[0014] in, For the low-energy reconstructed image, For the high-energy reconstructed image, Perform a matter decomposition operation for the image domain. Reconstructed image of the base material. The image is reconstructed for the base material 2.

[0015] The projection interpolation completion method for multi-energy cone-beam CT proposed in this application firstly involves iterative sparse reconstruction of projection values ​​for different energies, followed by image-domain material decomposition and forward projection. Thus, the image reconstructed from the base material obtained in the image domain guides multi-energy projection for interpolation completion in the projection domain, solving the problem of inconsistent high and low energy projection angles in rapid kilovolt switching spectral CBCT, and significantly improving image quality and material decomposition accuracy.

[0016] To achieve the above objectives, a second aspect of this application provides an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the multi-energy cone-beam CT projection interpolation completion method as described in the above embodiments.

[0017] To achieve the above objectives, a third aspect of this application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the multi-energy cone-beam CT projection interpolation completion method as described in the above embodiments.

[0018] To achieve the above objectives, a fourth aspect of this application provides a computer program product, including a computer program that is executed by a processor to implement the multi-energy cone-beam CT projection interpolation completion method as described in the above embodiments.

[0019] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0020] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart illustrating a projection interpolation completion method for multi-energy cone-beam CT according to an embodiment of this application; Figure 2 This is a schematic diagram of a multi-energy cone-beam CT spectral imaging system based on fast kilovolt switching according to an embodiment of this application; Figure 3 This is a schematic diagram illustrating the inconsistency between high and low energy angles according to an embodiment of this application; Figure 4 This is a schematic diagram of image artifacts related to the decomposition and reconstruction of materials with inconsistent angles, provided according to an embodiment of this application. Figure 5 This is a schematic diagram of the energy spectrum image reconstruction results under different sampling frame numbers in a simulation experiment according to an embodiment of this application, both uncorrected and corrected by energy spectrum interpolation. Figure 6 This is a schematic diagram of the profile lines at the edge of uncorrected and corrected energy spectrum reconstructed images after 200 frames of sampling, according to an embodiment of this application; Figure 7 This is a schematic diagram showing the decomposition and energy spectrum reconstruction results of a multi-energy motif under different sampling frame numbers in an actual experiment according to an embodiment of this application; Figure 8 This is a schematic diagram of the profile lines at the edge of an energy spectrum reconstructed image after uncorrection, spline interpolation correction, and correction by the method of this application, according to an embodiment of this application; Figure 9 This is a flowchart of a method for joint processing of the image domain and projection domain in multi-energy cone-beam CT imaging according to an embodiment of this application; Figure 10This is a flowchart of a projection interpolation completion method for multi-energy cone-beam CT according to an embodiment of this application; Figure 11 This is a schematic diagram of the structure of an electronic device provided according to an embodiment of this application. Detailed Implementation

[0021] The embodiments of this application are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this application, and should not be construed as limiting this application.

[0022] The following description, with reference to the accompanying drawings, describes a method for projection interpolation and completion of multi-energy cone-beam CT (CBCT) according to embodiments of this application. Addressing the problems mentioned in the background art regarding X-ray scattering, sparse sampling, inconsistent high- and low-energy projection angles, and limited energy separation in CBCT, this application proposes a method for projection interpolation and completion of multi-energy cone-beam CT. This method guides high- and low-energy projection data in the projection domain to perform energy spectrum interpolation and completion by using sparsely reconstructed images in the image domain. This solves the problem of inconsistent high- and low-energy projection angles in rapidly switching kilovolt energy spectrum CBCT, significantly improving image quality and material decomposition accuracy.

[0023] Figure 1 This is a flowchart of a method for projection interpolation completion in multi-energy cone-beam CT according to an embodiment of this application.

[0024] In this embodiment, the multi-energy cone-beam CT spectral imaging system based on fast kilovolt switching of this application can be as follows: Figure 2 As shown, Figure 2 This is a schematic diagram of a multi-energy cone-beam CT spectral imaging system based on rapid kilovolt switching according to an embodiment of this application.

[0025] The multi-energy cone-beam CT spectral imaging system based on rapid kilovolt switching includes: an X-ray source module, a flat panel detector module, and mechanical / electrical control and data transmission / processing units.

[0026] Specifically, the system's multi-energy projection is achieved by an X-ray source with rapid voltage switching capability. This multi-energy CT spectral imaging system can acquire projection data of different energies by alternating the X-ray tube voltage at different viewpoints during a single scan. Therefore, the projection data of different energies exhibit inconsistencies in angular direction. Furthermore, in cone-beam CT systems based on flat-panel detectors, limited by the frame rate of the flat-panel detector, a 360° scan in rapid kilovolt switching mode during a conventional CT scan (approximately 0.5 seconds) can only acquire a few hundred projection frames, facing a significant sampling sparsity problem. Sparse sampling further amplifies the problem of inconsistencies in the angles of multi-energy projection data.

[0027] For example, under conditions of a total of 400 frames of sampling and alternating switching between two voltages, the angular interval between different energy projection data increases to approximately 0.9°. Due to the inherent geometric characteristics of cone-beam imaging, each projection angle covers a wide range of phantom structures. This degree of angular interval will cause high- and low-energy projections to fail to obtain strictly corresponding path penetration information, resulting in an angle mismatch problem. The projection domain material decomposition method of energy-spectral CT relies on the physical relationship between different energy projections under the same ray direction. Its theoretical basis is the difference in energy response of each base material in the energy-spectral attenuation model. When the projection angle matching accuracy decreases, the high- and low-energy data that should originally come from the same penetration path are spatially offset, resulting in inconsistencies in the anatomical structures penetrated by the ray path, such as... Figure 3 This severely undermines the fundamental assumptions of the matter decomposition model, particularly at the edges of the scanned object's structure and in areas where different components intersect. In the reconstructed image, this angular mismatch manifests as stripe artifacts, bright and dark bands at the edges, and uneven grayscale and structural distortion in the virtual monoenergetic image, such as… Figure 4 This severely impacts the quantitative performance of energy-spectral cone-beam CT. Therefore, interpolation and completion of multi-energy projections are crucial for multi-energy cone-beam CT imaging systems based on fast kilovolt switching.

[0028] like Figure 1 As shown, the method for projection interpolation completion applicable to multi-energy cone-beam CT includes the following steps: In step S101, the low-energy projection and high-energy projection of the multi-energy cone-beam CT spectral imaging system are obtained.

[0029] Multi-energy cone-beam computed tomography (MCCT) is a type of computed tomography that utilizes a cone-shaped X-ray beam and can operate at multiple energies. A spectral imaging system refers to a device / system that can acquire and process X-ray projection data of different energies to achieve spectral imaging. Low-energy projection refers to the low-energy projection data corresponding to different energy X-rays acquired by a multi-energy cone-beam CT spectral imaging system, while high-energy projection refers to the high-energy projection data corresponding to different energy X-rays acquired by the same system.

[0030] Specifically, the high-energy projection and low-energy projection of the multi-energy cone-beam CT spectral imaging system can be acquired in stages by controlling the X-ray tube to output rays of different energies, or by using a special detector to directly distinguish photons of different energies, and multiple energy projections can be acquired simultaneously in a single scan. No specific limitations are made here.

[0031] In step S102, preliminary iterative sparse reconstruction and material decomposition are performed on the low-energy projection and high-energy projection respectively to obtain the reconstructed image of base material one and the reconstructed image of base material two. Among them, preliminary iterative sparse reconstruction is a CT image generation method that combines iterative optimization and sparsity constraints. Material decomposition utilizes the attenuation differences between X-rays of different energies and matter to decompose the signal of each pixel in the image into two base materials. The base materials refer to pre-selected reference materials (such as water, iodine, bone, and soft tissue commonly used in medicine), whose attenuation patterns at different energies are known, serving as the basic units for material decomposition.

[0032] Specifically, preliminary images are obtained through iterative sparse reconstruction, and then the matter is decomposed by combining multi-energy characteristics, ultimately obtaining pure distribution images of the two base materials, laying the analytical foundation at the material level for subsequent processing.

[0033] Optionally, in some embodiments, preliminary iterative sparse reconstruction and matter decomposition are performed on the low-energy projection and high-energy projection respectively to obtain a reconstructed image of base material one and a reconstructed image of base material two, including: performing preliminary iterative sparse reconstruction on the low-energy projection and high-energy projection respectively based on a preset image domain sparse reconstruction strategy to obtain a low-energy reconstructed image and a high-energy reconstructed image; and performing matter decomposition operation on the low-energy reconstructed image and the high-energy reconstructed image respectively in the image domain based on a preset image domain matter decomposition strategy to obtain a reconstructed image of base material one and a reconstructed image of base material two.

[0034] The preset image domain sparse reconstruction strategy can be a user-defined image domain sparse reconstruction strategy, an image domain sparse reconstruction strategy obtained through a finite number of experiments, or an image domain sparse reconstruction strategy obtained through a finite number of computer simulations; no specific limitation is made here. Similarly, the preset matter decomposition strategy can be a user-defined matter decomposition strategy, a matter decomposition strategy obtained through a finite number of experiments, or a matter decomposition strategy obtained through a finite number of computer simulations; no specific limitation is made here.

[0035] Specifically, the projection data of different energies are first reconstructed into corresponding energy images (preserving energy-specific decay information), and then the difference between the two energy images is used at the image level to obtain a pure base material image through a material decomposition algorithm.

[0036] Furthermore, in some embodiments, the preset image domain sparse reconstruction strategy is as follows:

[0037]

[0038] in, It is a 2-norm. For TV regular expression, For low-energy reconstructed images, For high-energy reconstructed images, is the system matrix corresponding to the high-energy reconstructed image; SSIM is the structural similarity regularization term, with values ​​ranging from [-1, 1]. For the corresponding regularization parameter.

[0039] The calculation method for the structural similarity regularization term is as follows:

[0040] in, These represent the variance of the low-energy image, the variance of the high-energy image, and the covariance between the high- and low-energy images, respectively. This is a constant term to prevent the denominator from being 0.

[0041] Preliminary low-energy and high-energy reconstructed images were obtained. Finally, considering that image-domain material decomposition cannot completely eliminate hardening artifacts and may amplify noise, image enhancement operations, such as multi-material hardening correction, can be used. Further reduce hardening artifacts in high- and low-energy images, and perform image smoothing and noise reduction. Further reduce image noise to obtain an updated high-energy reconstructed image. and the updated low-energy reconstructed image .

[0042]

[0043] By optimizing the high- and low-energy reconstructed images of multi-energy CT in three dimensions—fitting the original projection, ensuring consistency between high- and low-energy image structures, and suppressing artifacts—a more reliable base image is provided for subsequent material decomposition and projection completion.

[0044] Furthermore, in some embodiments, the preset image domain material decomposition strategy is as follows:

[0045] in, For low-energy reconstructed images, For high-energy reconstructed images, Perform a matter decomposition operation for the image domain. Reconstructing images based on base materials, Reconstructed image based on material 2.

[0046] Specifically, updated low-energy and high-energy reconstructed images are obtained. Then, a matter decomposition operation is performed in the image domain. Obtain base material image .

[0047] In step S103, image enhancement operations are performed on the reconstructed images of base material one and base material two respectively, followed by forward projection. Weighting is then performed using pre-calibrated high and low energy spectrum information to obtain the complete projection data of base material one and base material two under full viewing angle.

[0048] Forward projection refers to the process in CT imaging that is the opposite of image reconstruction. The pre-calibrated high and low energy spectrum information can be user-defined, obtained through a limited number of experiments, or obtained through a limited number of computer simulations; no specific limitations are imposed here.

[0049] Specifically, low-energy and high-energy reconstructed images are obtained. Then, a matter decomposition operation is performed in the image domain. Obtain the base material image, and then... Perform image enhancement operations followed by forward projection operations. Using pre-calibrated high and low energy spectrum information Weighting Obtain high and low energy complete projection data from all viewing angles:

[0050] in, To complete the low-energy projection data from all angles (i.e., complete the projection data of the base material). The high-energy projection data is completed from all angles (i.e., the projection data of the second base material).

[0051] For angles missing in the original sampling, the above synthesis method is used to fill in the missing high or low energy projections, and finally a fully registered high and low energy projection pair is obtained.

[0052] Furthermore, in some embodiments, after obtaining high-energy complete projection data and low-energy complete projection data from all angles, the method further includes: performing projection domain material decomposition on the low-energy complete projection data and the high-energy complete projection data respectively based on a preset projection domain material decomposition strategy, and performing energy spectrum image reconstruction based on the decomposition results to obtain the final base material one image and the final base material two image.

[0053] The high-energy completed projection data under full-view conditions is low-energy projection data covering all angles of CT scans after completion processing. The low-energy completed projection data under full-view conditions is high-energy projection data covering all angles of CT scans after completion processing. Projection domain material decomposition, at the projection data level, utilizes different energy projection data and, based on the differences in the attenuation characteristics of materials to different energy rays, decomposes the projection signal into projection components corresponding to various base materials, reducing errors introduced by image reconstruction. Energy spectrum image reconstruction, based on the base material projection components obtained from projection domain material decomposition, reconstructs images reflecting different energy characteristics or base material distributions.

[0054] Specifically, in this application embodiment, the projection data of base material one and base material two, after completion and angle matching, will be supplemented. Performing projection domain material decomposition With image reconstruction The final base material image is obtained, where the preset projection domain material decomposition strategy is:

[0055] in, For the final base material image, The final image of the base material 2. To complete the low-energy projection data from all angles, To complete the projection data with high energy from all angles, Perform a matter decomposition operation on the projection domain. For image reconstruction operations Therefore, for multi-energy spectral cone-beam CT imaging based on rapid kilovolt switching, the interpolated high and low energy projection data provide a good guarantee of data consistency for the projection domain-based material decomposition method.

[0056] To further understand the image domain and projection domain joint processing method applicable to multi-energy cone-beam CT imaging, this application embodiment analyzes the effects of material decomposition and energy spectrum reconstruction after performing 80 / 130kVp energy spectrum imaging on a water-iodine solution phantom at different sampling frame numbers, after energy spectrum interpolation and correction.

[0057] The simulation results show the reconstructed energy spectrum images under different sampling frame numbers, both uncorrected and corrected after energy spectrum interpolation. Figure 5 As shown, Figure 5The results demonstrate the reconstruction performance of energy spectrum images at 200 and 800 frames, both before and after correction using energy spectrum interpolation. Before correction, significant bright and dark artifacts were present at the edges; after correction, these artifacts were significantly reduced, resulting in clearer images. Furthermore, different substances achieved better imaging quality after correction at various sampling frame numbers, validating the effectiveness of the energy spectrum interpolation correction method in improving the reconstruction performance of energy spectrum CT images.

[0058] Specifically, the profile lines at the edges of the uncorrected and corrected energy spectrum reconstructed images at 200 frames of sampling are as follows: Figure 6 As shown, the horizontal axis represents position, and the vertical axis represents the average measurement value. Figure 6 In the image, 'a' represents the water reconstruction image. The blue reference line represents the ideal distribution of reconstructed values. The red uncorrected line deviates significantly from the reference line, exhibiting drastic fluctuations and obvious anomalies. The orange corrected line is closer to the reference line, with significantly reduced fluctuations, indicating that the edge signals of the corrected water reconstruction image are closer to the ideal situation, thus improving accuracy. Figure 6 In the diagram, 'b' represents the iodine reconstruction image. The blue reference line is the ideal reconstruction value; the red uncorrected line deviates from the reference line, exhibiting significant oscillations and abnormal fluctuations; the orange corrected line is smoother and has a higher degree of fit with the reference line, indicating that the correction effectively improves the signal accuracy at the edges of the iodine reconstruction image. Therefore, through correction, the signal performance in the edge regions of both water and iodine energy spectrum reconstruction images is closer to the reference standard, verifying the effectiveness of the correction method in improving the edge accuracy of energy spectrum reconstruction images.

[0059] Actual experiments under 80 / 130 kVp rapid kilovolt switching scanning and different sampling frame numbers yielded the following results: (The rest of the text appears to be a continuation of the previous sentence and can be omitted.) Figure 7 As shown, the horizontal axis represents the number of sampling frames, and the vertical axis represents different processing methods. Figure 7 In the figure, 'a' represents the water reconstruction result. The uncorrected image has obvious bright and dark artifacts and undersampling artifacts, and the water reconstruction image is blurry. Using spline interpolation, the artifacts are reduced to some extent, but the image is still not uniform enough. The image domain and projection domain joint processing method for multi-energy cone-beam CT imaging provided in this application provides the most uniform image, the fewest artifacts, and the clearest water reconstruction image. This shows that the image domain and projection domain joint processing method for multi-energy cone-beam CT imaging used in this application has the best effect on the material decomposition and energy spectrum reconstruction of water. Figure 7In the diagram, 'b' represents the reconstructed iodine result. Uncorrected high-concentration iodine areas exhibit blurred edges and background noise. While spline interpolation improves the bright spot edges, noise remains. Using the image domain and projection domain combined processing method for multi-energy cone-beam CT imaging provided in this application, the high-concentration iodine bright spot edges are sharp, the background is clean, and the iodine distribution quantification is more accurate. Therefore, in actual experiments, this application outperforms uncorrected and spline interpolation methods in both material decomposition and energy spectrum reconstruction at different sampling frame numbers, more accurately displaying the distribution and concentration information of water and iodine, thus verifying the effectiveness of the method.

[0060] Among them, the profile lines at the edges of the energy spectrum reconstructed images after correction (uncorrected, spline interpolation corrected, and the image domain and projection domain joint processing method for multi-energy cone-beam CT imaging provided in this application embodiment) are as follows: Figure 8 As shown, Figure 8 This paper demonstrates the differences in energy spectrum reconstruction results of water and iodine within a specified rectangular region after processing with three methods: uncorrected, spline interpolation corrected, and corrected using the method adopted in this application embodiment, at sampling rates of 200 frames and 1000 frames. The core objective is to compare the impact of different correction methods on image edge accuracy. Figure 8 Figure 'a' shows a profile line drawing of the water edge at 200 frames. The blue line represents the uncorrected image, with drastic curve fluctuations and a large deviation from the ideal signal. The orange line represents the image corrected by spline interpolation, which reduces fluctuations but is still not smooth enough. The image domain and projection domain joint processing method for multi-energy cone-beam CT imaging provided in this application has the smoothest curve, which is closer to the distribution trend of the true reconstructed values. This shows that the method significantly improves the accuracy of water edge reconstruction at low sampling frame numbers. Figure 8 Figure b shows a cross-sectional line graph of the edge of iodine at 200 frames. The blue line represents the uncorrected image, with obvious curve oscillation and peak / valley values ​​deviating from the reasonable range. The orange line represents the image corrected by spline interpolation, which alleviates the oscillation, but the details are still deviated. The image domain and projection domain joint processing method for multi-energy cone-beam CT imaging provided in this application embodiment has a smoother curve, a higher degree of fit with the reference trend, and better improvement in the edge signal accuracy of iodine. Figure 8 Figure 'c' shows a cross-sectional view of the water edge at 1000 frames. The blue line represents the uncorrected image with obvious abnormal fluctuations, while the orange line represents the image corrected by spline interpolation, which reduces fluctuations but still leaves residual deviations. The image domain and projection domain joint processing method for multi-energy cone-beam CT imaging provided in this application provides the smoothest curve, with abnormal fluctuations basically eliminated, and can further optimize the water edge reconstruction even at high sampling frame rates. Figure 8The diagram 'd' shows the edge profile of iodine at 1000 frames. The blue line represents the uncorrected image, with a significant drop near position 80. The orange line indicates that the anomaly has been alleviated, but the transition is still unnatural. The image domain and projection domain joint processing method for multi-energy cone-beam CT imaging provided in this application provides a smooth curve and a natural transition, accurately restoring the edge energy spectrum signal of iodine. Therefore, whether it is low sampling (200 frames) or high sampling (1000 frames), after correction using the image domain and projection domain joint processing method for multi-energy cone-beam CT imaging provided in this application, the edge profiles of the reconstructed energy spectrum images of water and iodine are smoother and closer to the real signal. Compared with uncorrected and spline interpolation correction, the improvement effect on image edge accuracy is better, verifying the effectiveness of this application in energy spectrum reconstruction edge optimization.

[0061] In summary, taking the projected data of both high and low energies as examples, the entire interpolation completion method includes the following main stages, as shown in the flowchart below. Figure 9 As shown, Figure 9 This is a flowchart of a method for joint processing of the image domain and projection domain for multi-energy cone-beam CT imaging according to an embodiment of this application.

[0062] Step 1: Input low-energy projection data and high-energy projection data, and perform image reconstruction on the low-energy projection data and high-energy projection data respectively to obtain low-energy reconstruction results and high-energy reconstruction results. Step 2: Perform image enhancement on the low-energy reconstruction results and high-energy reconstruction results, and update the reconstruction results based on the enhancement effect; Step 3: Using the enhanced low-energy reconstruction results and high-energy reconstruction results, perform material decomposition in the image domain to obtain base material image 1 and base material image 2; Step 4: Perform image enhancement on base material image 1 and base material image 2, and then perform forward projection to obtain low-energy complete projection and high-energy complete projection. Step 5: Using low-energy completion projection and high-energy completion projection, perform material decomposition again in the projection domain to output the final base material image 1 and the final base material image 2.

[0063] Thus, through the bidirectional collaboration of the image domain and the projection domain, the intuitiveness of the image domain is used for preliminary material decomposition, while the completion and energy spectrum weighting of the projection domain are used to repair data gaps and improve the accuracy of material analysis, ultimately achieving accurate identification and imaging optimization of material components by multi-energy CT.

[0064] To facilitate those skilled in the art to further understand the multi-energy cone-beam CT projection interpolation completion method of the embodiments of this application, the following description is provided in conjunction with specific embodiments.

[0065] like Figure 10 As shown, Figure 10 This is a flowchart of a multi-energy cone-beam CT projection interpolation completion method according to an embodiment of this application.

[0066] This multi-energy cone-beam CT projection interpolation completion method includes the following steps: S1001, experimental calibration of the system's equivalent energy spectrum under different voltage parameters.

[0067] S1002 is based on a multi-energy cone-beam CT spectral imaging system, which acquires multi-energy projection data according to predetermined switching parameters such as voltage and current.

[0068] S1003, using the proposed projection interpolation completion method, the missing multi-energy projection data at the corresponding angle is obtained.

[0069] S1004 uses the interpolated and completed multi-energy projection data to perform projection domain material decomposition, and then performs energy spectrum image reconstruction.

[0070] The projection interpolation completion method for multi-energy cone-beam CT proposed in this application firstly involves iterative sparse reconstruction of projection values ​​for different energies, followed by image-domain material decomposition and forward projection. Thus, the image reconstructed from the base material obtained in the image domain guides multi-energy projection for interpolation completion in the projection domain, solving the problem of inconsistent high and low energy projection angles in rapid kilovolt switching spectral CBCT, and significantly improving image quality and material decomposition accuracy.

[0071] Figure 11 This is a schematic diagram of an electronic device provided in an embodiment of the present invention. The electronic device may include: The memory 1101, the processor 1102, and the computer program stored on the memory 1101 and executable on the processor 1102.

[0072] When the processor 1102 executes the program, it implements the method for interpolation completion of multi-energy cone-beam CT projection provided in the above embodiments.

[0073] Furthermore, electronic devices also include: Communication interface 1103 is used for communication between memory 1101 and processor 1102.

[0074] The memory 1101 is used to store computer programs that can run on the processor 1102.

[0075] The memory 1101 may include high-speed RAM (Random Access Memory) memory, and may also include non-volatile memory, such as at least one disk storage.

[0076] If the memory 1101, processor 1102, and communication interface 1103 are implemented independently, then the communication interface 1103, memory 1101, and processor 1102 can be interconnected via a bus to complete communication between them. The bus can be an ISA (Industry Standard Architecture) bus, a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 11 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0077] Optionally, in a specific implementation, if the memory 1101, processor 1102, and communication interface 1103 are integrated on a single chip, then the memory 1101, processor 1102, and communication interface 1103 can communicate with each other through an internal interface.

[0078] The processor 1102 may be a CPU (Central Processing Unit), an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement embodiments of the present invention.

[0079] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for projection interpolation completion in multi-energy cone-beam CT.

[0080] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the above-described method for projection interpolation completion in multi-energy cone-beam CT.

[0081] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0082] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0083] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for projection interpolation and completion applicable to multi-energy cone-beam CT, characterized in that, Includes the following steps: Acquire low-energy and high-energy projections of a multi-energy cone-beam CT spectral imaging system; Preliminary iterative sparse reconstruction and matter decomposition were performed on the low-energy projection and the high-energy projection respectively to obtain the reconstructed image of base material one and the reconstructed image of base material two. Image enhancement operations are performed on the reconstructed images of the first base material and the reconstructed images of the second base material, respectively, and then forward projection is performed. Weighting is performed using pre-calibrated high and low energy spectrum information to obtain low-energy complete projection data and high-energy complete projection data under the full view. The process includes, after obtaining the low-energy complete projection data and the high-energy complete projection data from all angles, the following steps are also included: based on a preset projection domain material decomposition strategy, performing projection domain material decomposition on the low-energy complete projection data and the high-energy complete projection data respectively, and performing energy spectrum image reconstruction based on the decomposition results to obtain the final base material one image and the final base material two image.

2. The method according to claim 1, characterized in that, The preset projection domain matter decomposition strategy is as follows: in, For the final base material image, The final image of the base material 2. To complete the low-energy projection data from all angles, To complete the projection data with high energy from all angles, Perform a matter decomposition operation on the projection domain. For image reconstruction operations.

3. The method according to claim 1, characterized in that, The preliminary iterative sparse reconstruction and matter decomposition of the low-energy projection and the high-energy projection, respectively, to obtain the reconstructed image of base material one and the reconstructed image of base material two, includes: Based on a preset image domain sparse reconstruction strategy, preliminary iterative sparse reconstruction is performed on the low-energy projection and the high-energy projection respectively to obtain low-energy reconstructed images and high-energy reconstructed images. Based on a preset image domain material decomposition strategy, material decomposition operations are performed on the low-energy reconstructed image and the high-energy reconstructed image in the image domain to obtain the base material one reconstructed image and the base material two reconstructed image.

4. The method according to claim 3, characterized in that, The preset image domain sparse reconstruction strategy is as follows: in, It is a 2-norm. For TV regular expression, For low-energy reconstructed images, For high-energy reconstructed images, is the system matrix corresponding to the high-energy reconstructed image; SSIM is the structural similarity regularization term, with values ​​ranging from [-1, 1]. For the corresponding regularization parameter.

5. The method according to claim 3, characterized in that, The preset image domain material decomposition strategy is as follows: in, For the low-energy reconstructed image, For the high-energy reconstructed image, Perform a matter decomposition operation for the image domain. Reconstructed image of the base material. The image is reconstructed for the base material 2.

6. An electronic device, characterized in that, include: A memory, a processor, and a computer program stored in the memory and executable on the processor, the processor executing the program to implement the method for multi-energy cone-beam CT projection interpolation completion as described in any one of claims 1-5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, The program is executed by the processor to implement the method for multi-energy cone-beam CT projection interpolation completion as described in any one of claims 1-5.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the method for projection interpolation completion of multi-energy cone-beam CT as described in any one of claims 1-5.

Images