Dual-energy cone beam CT two-dimensional modulation device, imaging system and imaging method

Abstract

The invention discloses a dual-energy cone beam CT two-dimensional modulation device, an imaging system and an imaging method. A body of the two-dimensional modulation device of the dual-energy cone-beam CT imaging detector is in a square grating plate shape, and the size of the body is matched with that of the dual-energy cone-beam CT imaging detector. The dual-energy cone beam CT imaging system comprises an X-ray source, a detector, a dual-energy cone beam CT two-dimensional modulation device, an equipment rack, a rotary rack, a rotary driving motor and a data processing system. The dual-energy cone beam CT imaging method comprises the following steps: S1, acquiring dual-energy empty scanning data and dual-energy cone beam CT projection data; s2, obtaining line integral data by using logarithm transformation; s3, constructing a high-low energy data extraction template; s4, calculating high-energy line integral data and low-energy line integral data; s5, constructing an objective function; s6, performing iteration to obtain a CT image; compared with an existing interpolation method, an iteration method is adopted in the method, interpolation errors do not exist, and the image reconstruction precision is improved.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, and in particular to a dual-energy cone-beam CT two-dimensional modulation device, an imaging system and an imaging method. Background Art

[0002] Cone-beam CT has a simple and flexible structure. As an important means of structural imaging, it is widely used in clinical practice. For example, it is used to detect pre-treatment positioning errors in radiotherapy and to diagnose oral diseases in dental clinics. However, the lack of quantitative function of cone-beam CT limits the high-end application of cone-beam CT in clinical practice. It is limited to high-contrast structural imaging and is difficult to meet clinical needs. As the development direction of the next generation of tomographic imaging, dual-energy CT transforms the energy spectrum and adds one data acquisition based on the traditional CT single-energy spectrum imaging method, so that conventional CT structural imaging introduces energy information. It has received widespread attention in clinical applications due to its excellent energy resolution characteristics. The combination of dual-energy and cone-beam CT is expected to enhance the quantitative capabilities of cone-beam CT and expand the further high-end clinical applications of cone-beam CT.

[0003] However, it is difficult to integrate efficient dual-energy CT acquisition solutions on existing cone-beam CT platforms. The traditional method of changing the voltage to acquire two cycles doubles the radiation dose to the patient. At the same time, during the two acquisition processes, the patient's internal organs and tissues move or shift to varying degrees, resulting in a mismatch between high-energy and low-energy images in the reconstructed images. Diagnostic dual-energy CT usually uses tube voltage switching, dual-ball tubes, dual-energy plate detectors, and photon technology detection, all of which require complex hardware support and are expensive, making them particularly difficult to implement on existing cone-beam CT platforms.

[0004] Chinese invention patent publication number CN111366982A discloses a detector device, a dual-energy CT system and an image reconstruction method for a dual-energy CT system, which uses low-energy detection strips and high-energy detection strips arranged alternately. This method performs unidirectional sampling on the projection, and the vertical bar modulator is limited to unidirectional modulation. The modulation effect is significant in fan-beam two-dimensional CT dual-energy imaging. However, cone-beam CT imaging belongs to large cone-angle three-dimensional tomographic imaging. Only the imaging geometry of the middle layer of the three-dimensional image is completely consistent with the spatial geometry of fan-beam two-dimensional CT imaging. Due to the existence of the cone angle, the image reconstruction of the non-middle layer cannot meet this imaging geometry, resulting in a significant decrease in the one-dimensional unidirectional modulation effect, and ultimately resulting in a decrease in image quality. At the same time, the reconstruction method uses the interpolation-before-reconstruction method, which has interpolation errors and limited reconstructed image accuracy.

[0005] Chinese invention patent publication number CN111134710A discloses a multi-energy CT imaging system, which uses two one-dimensional grids with different X-ray attenuation effects to overlap to form an energy spectrum modulator, and places the energy spectrum modulator between the ray source and the object to be imaged. Although the subunits of the modulator with the same attenuation proposed by the system are distributed alternately in two directions, their arrangement is a single arrangement, that is, the alternating arrangement of the subunits in each row arrangement composed of subunits with the same attenuation is exactly the same. This arrangement is a simple superposition of vertical stripe modulators in two directions, that is, the superposition of the vertical stripe modulators in Chinese invention patent publication number CN 111366982 A. This method cannot utilize the three-dimensional large cone angle imaging characteristics of cone beam CT, especially in the non-coplanar layer imaging process, the data integrity of the same energy spectrum captured is poor, and the imaging quality is reduced. At the same time, due to the thickness of the energy spectrum modulator, the X-rays attenuate unevenly when passing through the edge of the modulator, and the signals reaching the detector cannot be used directly, resulting in a serious penumbra effect. The system places the energy spectrum modulator between the ray source and the object to be imaged, which amplifies the number of signals with uneven attenuation, causing serious information loss and waste, and reducing imaging quality. Summary of the invention

[0006] In view of the shortcomings of the prior art, one object of the present invention is to provide a dual-energy cone-beam CT two-dimensional modulation device capable of bidirectional sampling and having a stronger modulation performance.

[0007] Another object of the present invention is to provide an imaging system having the above dual-energy cone-beam CT two-dimensional modulation device, which can obtain dual-energy cone-beam CT data through single-circle scanning at low cost.

[0008] Another object of the present invention is to provide a method for dual-energy cone-beam CT imaging using the above imaging system, thereby achieving fast and accurate imaging of single-circle scanning dual-energy cone-beam CT projection data at a low cost.

[0009] The objective of the present invention is achieved through the following technical solutions:

[0010] A two-dimensional modulation device of a dual-energy cone-beam CT imaging detector, the two-dimensional modulation device is an integrally formed square plate, comprising an outer frame portion on the outside and a grid portion on the inside; the grid portion is composed of A×B grids, wherein the grid comprises through holes and modulation units arranged at intervals in each row, and the modulation units have the same area as the through holes; in each row of the two-dimensional modulation device, the arrangement order of the modulation units and the through holes is opposite to that of the adjacent rows;

[0011] The modulation unit is used to modulate the X-rays passing through the object to be measured into high-energy X-rays; the X-rays passing through the object to be measured then pass through the through hole to become low-energy X-rays;

[0012] The four corners of each through hole are chamfered.

[0013] Preferably, the thickness of the two-dimensional modulation device is ≤2 mm, and A=B.

[0014] Preferably, the material of the two-dimensional modulation device is a metal, a metal mixture or a metal oxide with an attenuation effect, wherein the metal includes molybdenum, copper, tin, tungsten or gold; the metal mixture is a metal mixture of two or more metals among molybdenum, copper, tin, tungsten or gold; and the metal oxide is Fe2O3.

[0015] A dual-energy cone-beam CT imaging system equipped with a two-dimensional modulation device comprises an X-ray source, a detector, a two-dimensional modulation device, an equipment frame, a rotating frame, a rotating drive motor and a data processing system; the X-ray source is arranged on the upper part of the rotating frame, and the detector is arranged on the lower part of the rotating frame; the middle part of the rotating frame is hinged with the middle part of the equipment frame, the rotating frame is rotated by the rotating drive motor, and the two-dimensional modulation device is installed on the detector; when the detector rotates with the rotating frame, the two-dimensional modulation device and the rotating frame are rotated synchronously; the size of the two-dimensional modulation device is adapted to the detector.

[0016] Preferably, the minimum value of the side length of the modulation unit is the pixel size of the detector; and the maximum value of the side length of the modulation unit is one quarter of the length of the shortest side of the detector.

[0017] Preferably, the data processing system includes a data transmission module, a data reconstruction module, a data display module and a data storage module;

[0018] The data transmission module is used to transmit the projection data from the detector to the data reconstruction module, and then use the data reconstruction module to perform three-dimensional data reconstruction, display the reconstruction result on the data display module, and store the result in the data storage module.

[0019] A method for performing dual-energy cone-beam CT imaging using a dual-energy cone-beam CT imaging system comprises the following steps:

[0020] S1, using the detector to obtain dual-energy air scan data I 0 and dual-energy cone-beam CT projection data I;

[0021] S2, use logarithmic transformation to get line integral data q:

[0022]

[0023] Where, ln is the natural logarithm;

[0024] S3, build high and low energy data extraction template:

[0025] Construct a high-energy template matrix M according to the dimension of the detection matrix of the detector H and the low energy template matrix M L ; The M H and the M L The dimension of is the same as the pixel dimension of the detection matrix;

[0026] S4, calculate the high energy line integral data y H and low energy line integral data y L :

[0027] y H =M H ⊙q,

[0028] y L =M L ⊙q;

[0029] Among them, ⊙ represents the Hadamard product;

[0030] S5, construct the objective function:

[0031]

[0032] in: express The optimal solution of represents the high-energy CT image to be reconstructed, represents the low-energy CT image to be reconstructed; Represents the vectorized high-energy line integral data y H , Represents the vectorized low-energy line integral data y L ; T represents the transposition operator; F is the forward projection operator, M is the correction operator related to the modulation frequency, M = [M H ,M L ]; ‖●‖2 is the bi-norm; δ is used to control the strength of regularization, δ∈[0,1]; represents a total variation regularization term, which is used to reduce noise and maintain boundaries of a high-energy CT image to be reconstructed or a low-energy CT image to be reconstructed;

[0033] is the characteristic function:

[0034]

[0035] in, p represents the pth pixel of the CT image to be reconstructed, x Hp With x Lp represent the CT value of the pth pixel in the high-energy CT image and the low-energy CT image to be reconstructed respectively;

[0036] S6, iterate to get the CT image:

[0037] Use the Chambolle-Pock algorithm to iterate until the curve of the objective function in S5 converges to obtain the optimal solution

[0038] The high-energy CT image to be reconstructed in the optimal solution at this time is output as the final high-energy CT image, and the low-energy CT image to be reconstructed in the optimal solution is output as the final low-energy CT image.

[0039] The detection matrix in S3 is divided into a modulation unit covering part and a through-hole part;

[0040] The high energy template matrix M H , the value of the position corresponding to the covered part of the modulation unit is set to 1, and the value of the position corresponding to the through hole part is set to 0;

[0041] The low energy template matrix M L In the embodiment, the value of the position corresponding to the covering part of the modulation unit is set to 0, and the value of the position corresponding to the through hole part is set to 1.

[0042] The modulation unit covering portion is a portion where the modulation unit and the detector are attached, corresponding to a position in the detection matrix;

[0043] The through hole portion is such that after the two-dimensional modulation device is attached to the detector, the position of the through hole corresponds to the position in the detection matrix.

[0044] The specific method of iterative solution using Chambolle-Pock algorithm in S6 is:

[0045] Set the first intermediate variable and the second intermediate variable k is the number of iterations, and the following three formulas are obtained:

[0046]

[0047] Wherein, formula (I) represents the objective function of S5 The proximal mapping of For the kth iteration Formula (II) represents the objective function of S5 The proximal mapping is used to update the first intermediate variable Formula (III) represents the objective function of S5 The proximal mapping is used to update the second intermediate variable

[0048] k∈[1,K], K is the maximum number of iterations; D represents the gradient operator, s and t are both iteration steps, s∈[0.1,10]; t∈[0.1,10];

[0049] Iterate using the Chambolle-Pock algorithm and Until the curve of the objective function in S5 converges, the optimal solution is obtained

[0050] Compared with the prior art, the present invention has the following beneficial effects:

[0051] 1. The modulation units of the present invention are alternately distributed in two directions, realizing a leap from one-dimensional modulation to two-dimensional modulation, and progressing from unidirectional sampling to bidirectional sampling, with stronger modulation performance.

[0052] 2. The two-dimensional modulation method of the present invention is more in line with the large cone-angle imaging geometric characteristics of cone-beam CT imaging, with higher imaging quality and significantly suppressed imaging artifacts.

[0053] 3. The imaging method of the present invention adopts an iterative method. Compared with the existing interpolation method, there is no interpolation error, which improves the image reconstruction accuracy.

[0054] 4. The imaging method of the present invention can realize the rapid acquisition and accurate reconstruction of single-circle scanning dual-energy cone-beam CT data at low cost.

[0055] 5. In the present invention, the two-dimensional modulation device is close to the detector, which reduces the penumbra effect, increases the number of effective signals, and ultimately improves the image quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 is a schematic diagram of the top view of the two-dimensional modulation device of the present invention;

[0057] Figure 2 It is a structural schematic diagram of the dual-energy cone-beam CT imaging system of the present invention;

[0058] Figure 3 It is a schematic diagram of the composition of the dual-energy cone-beam CT data processing system of the present invention;

[0059] Figure 4 is a flow chart of the dual-energy cone-beam CT imaging method of the present invention;

[0060] Figure 5 is a schematic diagram comparing the results obtained by different imaging methods in the comparative examples;

[0061] Among them, 1. X-ray source, 2. detector, 3. two-dimensional modulation device of dual-energy cone beam CT, 4. equipment frame, 5. rotating frame, 6. rotating drive motor. DETAILED DESCRIPTION

[0062] The technical solution of the present invention is described in detail below in conjunction with the accompanying drawings and embodiments.

[0063] Embodiment 1

[0064] A two-dimensional modulation device for dual-energy cone-beam CT imaging detector, the structure of which is as follows Figure 1 The size of the two-dimensional modulation device of the dual-energy cone-beam CT imaging detector is adapted to the size of the dual-energy cone-beam CT detector.

[0065] The two-dimensional modulation device is an integrally formed square plate, comprising an outer frame portion on the outside and a grid portion on the inside; the grid portion is composed of A×B grids, wherein the grid comprises through holes and modulation units alternately arranged in each row, and the modulation units have the same area as the through holes; the arrangement order of the modulation units and the through holes in each row of the two-dimensional modulation device is opposite to that of the adjacent rows;

[0066] The modulation unit is used to modulate the X-rays passing through the object to be measured into high-energy X-rays; the X-rays passing through the object to be measured then pass through the through hole to become low-energy X-rays;

[0067] The frame width of the outer frame is 5mm-15mm, and A=B.

[0068] The four corners of each through hole must be chamfered, and the chamfer radius is larger than the pixel size of a single detector but not more than 5 mm.

[0069] like Figure 1 As shown, the two-dimensional modulation device is surrounded by an outer frame, and a plurality of square modulation units are arranged on the grid; Figure 1 In the figure, the black blocks represent modulation units composed of modulation materials, and the white blocks represent through holes, that is, there is no modulation material at the white block position. The modulation units are arranged alternately in directions parallel to the adjacent two sides of the two-dimensional modulation device body. The arrangement order of black blocks and white blocks in each row is opposite to that of the adjacent two rows, and the arrangement order of black blocks and white blocks in each column is opposite to that of the adjacent two columns, presenting a chessboard-like arrangement. The spacing between any two square modulation units is equal to the side length of the modulation unit. Figure 1 This is only one embodiment of the present invention. In other embodiments, the upper left corner point of the grid portion of the two-dimensional modulation device may be a modulation unit.

[0070] The material of the two-dimensional modulation device is made of metal, metal mixture or metal oxide with obvious attenuation effect, including but not limited to one of molybdenum, copper, tin, tungsten and gold, or a metal mixture of two or more of these metals. The metal oxide is Fe2O3.

[0071] The minimum side length of the square modulation unit is the pixel size of the detector;

[0072] Since the detector may be rectangular, the maximum side length of the square modulation unit is one quarter of the shortest side length of the detector.

[0073] The thickness of the modulation device comprehensively considers the attenuation effect of the material on X-rays and the signal-to-noise ratio of the detection signal. The thicker the modulation device, the stronger the attenuation effect on X-rays and the smaller the signal-to-noise ratio of the detection signal.

[0074] When the material of the modulation device is copper, the thickness of the modulation device is 0.2-0.8 mm;

[0075] When the material of the modulation device is molybdenum, the thickness of the modulation device is 0.1-0.5 mm;

[0076] When the material of the modulation device is tin, the thickness of the modulation device is 0.1-0.5 mm;

[0077] When the material of the modulation device is gold, the thickness of the modulation device is 0.1-0.5 mm;

[0078] When the material of the modulation device is tungsten, the thickness of the modulation device is 0.1-0.4 mm.

[0079] When the material of the modulation device is other materials, the thickness thereof shall not exceed 2 mm, and the effective energy attenuation deviation shall be ≥10 keV.

[0080] like Figure 2 As shown, when in use, the two-dimensional modulation device of the dual-energy cone-beam CT is placed closely against the surface of the detector, and the center of the two-dimensional modulation device of the dual-energy cone-beam CT is placed on the same vertical line as the X-ray source.

[0081] In the above structure, the modulation units are arranged bidirectionally to sample X-rays in both horizontal and vertical directions, satisfying the non-mid-plane imaging characteristics of cone-beam CT and realizing two-dimensional modulation of data, thus achieving better modulation performance.

[0082] Embodiment 2

[0083] A dual-energy cone-beam CT imaging system, the structure of which is as follows Figure 2 As shown, it includes an X-ray source, a detector, a two-dimensional modulation device of a dual-energy cone-beam CT, an equipment frame, a rotating frame, a rotating drive motor and a data processing system (not shown in the figure).

[0084] Among them, the X-ray source is arranged on the upper part of the rotating frame, and the detector is arranged on the lower part of the rotating frame; the two-dimensional modulation device of the dual-energy cone-beam CT is installed on the surface of the detector, the middle part of the rotating frame is hinged with the middle part of the equipment frame, and the rotating frame is rotated by a rotating drive motor, thereby driving the X-ray source, detector, and two-dimensional modulation device thereon to rotate synchronously.

[0085] like Figure 3 As shown, the data processing system includes a data transmission module, a data reconstruction module, a data display module and a data storage module; the data transmission module is used to transmit the projection data from the detector to the data reconstruction module, use the reconstruction module to perform three-dimensional data reconstruction, display the reconstruction results on the data display module, and store the results in the data storage module. The data reconstruction module generally uses a PCI-E data acquisition card.

[0086] When the dual-energy cone-beam CT imaging system is in use, the X-ray source forward projects the X-rays onto the object to be imaged. After the X-rays pass through the object to be imaged, the modulation unit on the two-dimensional modulation device modulates the X-rays reaching the detector into high-energy X-rays. The X-rays that are not modulated by the modulation unit and directly pass through the through hole to reach the detector are low-energy X-rays.

[0087] As the super-rotating gantry rotates, the two-dimensional modulation device of the dual-energy cone-beam CT rotates one circle with the detector, and the detector can synchronously collect high-energy X-rays and low-energy X-rays, realizing single-circle scanning dual-energy cone-beam CT data acquisition.

[0088] Data scanning modes include: Half-fan mode, Short-scan mode, and Full-fan full-scan mode.

[0089] Embodiment 3

[0090] A dual-energy cone-beam CT imaging method, such as Figure 4 As shown, the following steps are included:

[0091] S1, using the detector to obtain dual-energy cone-beam CT empty scan data I 0 And dual-energy cone-beam CT projection data I. Since the present invention uses a two-dimensional modulation device, the dual-energy cone-beam CT projection data I consisting of high-energy cone-beam CT projection data and low-energy cone-beam CT projection data can be obtained in one scan.

[0092] S2, use logarithmic transformation to convert the projection data into line integral data q, as shown in formula (1):

[0093]

[0094] Here, ln is the natural logarithm.

[0095] S3, build high and low energy data extraction template:

[0096] A high-energy template matrix M for high-energy data extraction is constructed based on the structure of the grid portion of the two-dimensional modulation device. H And the low energy template matrix M extracted from low energy data L , where H represents high energy and L represents low energy. High energy template matrix M H With the low energy template matrix M L The dimension of is exactly the same as the pixel dimension of the detection matrix of the detector (e.g., pixel 1024*768). Assume that the dimension of the detection matrix of the detector is [N a ,N b ], then the template matrix M H With M L The dimensions are [N a ,N b ]. High energy template matrix M H In the low-energy template matrix M, the value of the corresponding position of the part covered by the modulation unit is set to 1, and the value of the part of the through hole is set to 0. L In the example, the value of the portion covered by the modulation unit is set to 0, and the value of the portion through the hole is set to 1.

[0097] S4, calculate the high energy line integral data y H and low energy line integral data y L ,have:

[0098] y H =M H ⊙q, (2)

[0099] y L =M L ⊙q; (3)

[0100] Where ⊙ represents the Hadamard product of the matrix.

[0101] S5, construct the objective function:

[0102] The objective function consists of a data fidelity term and a regularization term. The data fidelity term is in the form of a binary norm. The regularization term is used to reduce noise and maintain the image signal-to-noise ratio. Its form includes but is not limited to image total variation, deep denoising operator, etc.

[0103] The noise reduction part in the objective function constructed by the present invention uses a variational regularization term, and the objective function is constructed specifically as follows:

[0104]

[0105] in: for The optimal solution of

[0106] represents the high-energy CT image to be reconstructed, represents the low-energy CT image to be reconstructed;

[0107] Represents the vectorized high-energy line integral data y H , Represents the vectorized low-energy line integral data y L .

[0108] T represents the transposition operator. F is the forward projection operator, M is the correction operator related to the modulation frequency, M = [M H ,M L ].

[0109] ||·||2 is the binary norm; δ is used to control the strength of regularization, ranging from 0 to 1. The larger the δ, the higher the noise suppression level and the more blurred the image; the smaller the δ, the lower the noise suppression level and the sharper the image edge;

[0110] represents the total variation regularization term, which is used to reduce noise and preserve boundaries of the reconstructed image;

[0111] is the characteristic function, which is defined as:

[0112]

[0113] in, p represents the pth pixel of the CT image to be reconstructed, x Hp With x Lp They represent the CT value (linear attenuation coefficient) of the pth pixel in the high-energy CT image and low-energy CT image to be reconstructed, respectively.

[0114] S6, iteratively obtain CT image: use Chambolle-Pock iterative method to solve the objective function, set the first intermediate variable and the second intermediate variable Let D represent the gradient operator, s and t are the iteration steps, k is the number of iterations, and the following three equations are obtained:

[0115]

[0116] Among them, formula (6) represents The proximal mapping of For the kth iteration Formula (7) indicates The proximal mapping is used to update the first intermediate variable Formula (8) indicates The proximal mapping is used to update the second intermediate variable k∈[1,K], K is the maximum number of iterations; s∈[0.1,10]; t∈[0.1,10].

[0117] Iterate using the Chambolle-Pock iterative method and Until the curve of the objective function (4) converges, the optimal solution is obtained.

[0118] The high-energy CT image to be reconstructed in the optimal solution at this time is output as the final high-energy CT image, and the low-energy CT image to be reconstructed in the optimal solution is output as the final low-energy CT image.

[0119] For the Chambolle-Pock iterative method used in S6, in (J.Huang, X.Liu and Y.Lei, "X-Ray Phase-Contrast Image Reconstruction Based on the Chambolle-PockAlgorithm," in IEEE Access, vol.9, pp.23120-23126, 2021, doi:10.1109 / ACCESS.2021.3056489.keywords: {X-ray imaging; Image reconstruction; Imaging; TV; Optimization; Measurement; Convergence; X-ray; image reconstruction; algorithms; phase detection}) and (C.Luo et al. al.,"Cone-Beam Computed Tomography DeblurringUsing an Overrelaxed Chambolle-Pock Algorithm,"in IEEE Access, vol.7, pp.1247-1259, 2019, doi:10.1109 / ACCESS.2018.2883297. keywords:{Computed tomography;Indexes;Computed tomography;deconvolution;spatial resolution}) both describe this iterative method in detail.

[0120] In order to obtain fair comparison results, the control variable method is used for comparison. The present invention and the methods of CN111366982A and CN111134710A use the same modulation unit material, the same modulation unit thickness and ensure that other imaging-related parameters such as voltage, current, and number of projections are exactly the same. The difference is that the present invention uses bidirectional modulation, while CN111366982A uses unidirectional modulation, and CN111134710A uses a one-dimensional grid bidirectional overlapping modulation method for dual-energy CT imaging and reconstruction. The three methods all ignore the penumbra effect in this experiment. The reconstruction results are shown in Figure 2. Figure 5 shown.

[0121] Among them: H-a1 and H-b1 are the true values ​​of the mid-plane and non-mid-plane of the high-energy CT image, respectively; L-a1 and L-b1 are the true values ​​of the mid-plane and non-mid-plane of the low-energy CT image, respectively; H-a2 and L-a2 are the mid-plane high-energy and low-energy CT images reconstructed using the method disclosed in CN 111366982 A; H-a3 and L-a3 are the mid-plane high-energy and low-energy CT images reconstructed using the method disclosed in CN111134710A; H-a4 and L-a4 are the mid-plane high-energy and low-energy CT images reconstructed using the method of the present invention; H-b2 and L-b2 are the non-mid-plane high-energy and low-energy CT images reconstructed using the method disclosed in CN 111366982 A; H-b3 and L-b3 are the non-mid-plane high-energy and low-energy CT images reconstructed using the method disclosed in CN111134710A; H-b4 and L-b4 ​​are the non-mid-plane high-energy and low-energy CT images reconstructed using the method of the present invention.

[0122] from Figure 5 It can be seen that the unidirectional modulation method and the one-dimensional grid bidirectional overlapping modulation method disclosed in CN 111366982 A and CN111134710A are more limited to mid-plane imaging, and there are obvious artifacts in cone beam CT non-mid-plane imaging, as shown in Figures H-b2, H-b3, L-b2, and L-b3. In contrast, the present invention is closer to the true value and has no artifacts, as shown in Figures H-b4 and L-b4.

[0123] The above reconstruction results prove that the imaging effect of dual-energy cone-beam CT using the two-dimensional modulation device of the present invention is significantly better than that of the modulation methods disclosed in CN 111366982A and CN111134710A.

Claims

1. A two-dimensional modulation device for a dual-energy cone-beam CT imaging detector, characterized in that: The two-dimensional modulation device is an integrally formed square plate, comprising an outer frame portion on the outside and a grid portion on the inside; the grid portion is composed of A×B grids, wherein the grid comprises through holes and modulation units alternately arranged in each row, and the modulation units have the same area as the through holes; the arrangement order of the modulation units and the through holes in each row of the two-dimensional modulation device is opposite to that of the adjacent rows; The modulation unit is used to modulate the X-rays passing through the object to be measured into high-energy X-rays; the X-rays passing through the object to be measured then pass through the through hole to become low-energy X-rays; The four corners of each through hole are chamfered.

2. The two-dimensional modulation device of the dual-energy cone-beam CT imaging detector according to claim 1, characterized in that: The thickness of the two-dimensional modulation device is ≤2mm, and A=B.

3. The two-dimensional modulation device of the dual-energy cone-beam CT imaging detector according to claim 1, characterized in that: The material of the two-dimensional modulation device is a metal, a metal mixture or a metal oxide with an attenuation effect, wherein the metal includes molybdenum, copper, tin, tungsten or gold; the metal mixture is a metal mixture of two or more metals among molybdenum, copper, tin, tungsten or gold; and the metal oxide is Fe2O3.

4. A dual-energy cone-beam CT imaging system equipped with a two-dimensional modulation device according to any one of claims 1 to 3, comprising an X-ray source, a detector, a two-dimensional modulation device, an equipment frame, a rotating frame, a rotating drive motor and a data processing system; the X-ray source is arranged on the upper part of the rotating frame, and the detector is arranged on the lower part of the rotating frame; the middle part of the rotating frame is hinged with the middle part of the equipment frame, and the rotating frame is rotated by the rotating drive motor, characterized in that: The two-dimensional modulation device is installed on the detector; when the detector rotates with the rotating frame, the two-dimensional modulation device and the rotating frame rotate synchronously; the size of the two-dimensional modulation device is adapted to the detector.

5. The dual-energy cone-beam CT imaging system according to claim 2, characterized in that: The minimum value of the side length of the modulation unit is the pixel size of the detector; the maximum value of the side length of the modulation unit is one quarter of the length of the shortest side of the detector.

6. The dual-energy cone-beam CT imaging system according to claim 5, characterized in that: The data processing system includes a data transmission module, a data reconstruction module, a data display module and a data storage module; The data transmission module is used to transmit the projection data from the detector to the data reconstruction module, and then use the data reconstruction module to perform three-dimensional data reconstruction, display the reconstruction result on the data display module, and store the result in the data storage module.

7. A method for performing dual-energy cone-beam CT imaging using the dual-energy cone-beam CT imaging system described in any one of claims 4 to 6, characterized in that: The following steps are involved: S1, using the detector to obtain dual-energy air scan data I 0 and dual-energy cone-beam CT projection data I; S2, use logarithmic transformation to get line integral data q: Where, ln is the natural logarithm; S3, build high and low energy data extraction template: Construct a high-energy template matrix M according to the dimension of the detection matrix of the detector H and the low energy template matrix M L ; The M H and the M L The dimension of is the same as the pixel dimension of the detection matrix; S4, calculate the high energy line integral data y H and low energy line integral data y L : y H =M H ⊙q, y L =M L ⊙q; Among them, ⊙ represents the Hadamard product; S5, construct the objective function: in: express The optimal solution of represents the high-energy CT image to be reconstructed, represents the low-energy CT image to be reconstructed; Represents the vectorized high-energy line integral data y H , Represents the vectorized low-energy line integral data y L ; T represents the transposition operator; F is the forward projection operator, M is the correction operator related to the modulation frequency, M = [M H ,M L ]; ||●||2 is the two-norm; δ is used to control the strength of regularization, δ∈[0,1]; represents a total variation regularization term, which is used to reduce noise and maintain boundaries of a high-energy CT image to be reconstructed or a low-energy CT image to be reconstructed; is the characteristic function: in, p represents the pth pixel of the CT image to be reconstructed, x Hp With x Lp represent the CT value of the pth pixel in the high-energy CT image and the low-energy CT image to be reconstructed respectively; S6, iterate to get the CT image: Use the Chambolle-Pock algorithm to iterate until the curve of the objective function in S5 converges to obtain the optimal solution The high-energy CT image to be reconstructed in the optimal solution at this time is output as the final high-energy CT image, and the low-energy CT image to be reconstructed in the optimal solution is output as the final low-energy CT image.

8. The dual-energy cone-beam CT imaging method according to claim 7, characterized in that: The detection matrix in S3 is divided into a modulation unit covering part and a through-hole part; The high energy template matrix M H , the value of the position corresponding to the covered part of the modulation unit is set to 1, and the value of the position corresponding to the through hole part is set to 0; The low energy template matrix M L In the embodiment, the value of the position corresponding to the covering part of the modulation unit is set to 0, and the value of the position corresponding to the through hole part is set to 1.

9. The dual-energy cone-beam CT imaging method according to claim 8, characterized in that: The modulation unit covering portion is a portion where the modulation unit and the detector are attached, corresponding to a position in the detection matrix; The through hole portion is such that after the two-dimensional modulation device is attached to the detector, the position of the through hole corresponds to the position in the detection matrix.

10. The dual-energy cone-beam CT imaging method according to claim 9, characterized in that: The specific method of iterative solution using Chambolle-Pock algorithm in S6 is: Set the first intermediate variable and the second intermediate variable k is the number of iterations, and the following three formulas are obtained: Wherein, formula (I) represents the objective function of S5 The proximal mapping of For the kth iteration Formula (II) represents the objective function of S5 The proximal mapping is used to update the first intermediate variable Formula (III) represents the objective function of S5 The proximal mapping is used to update the second intermediate variable k∈[1,K], K is the maximum number of iterations; D represents the gradient operator, s and t are both iteration steps, s∈[0.1,10]; t∈[0.1,10]; Iterate using the Chambolle-Pock algorithm and Until the curve of the objective function in S5 converges, the optimal solution is obtained

Images