Tomographic reconstruction method and device based on dual-plane long-stroke scanning

Abstract

The present invention discloses a tomographic reconstruction method and device based on dual-plane long-stroke scanning, which is applied to a dual-plane X-ray machine system. The method comprises the following steps: setting a dual-plane X-ray scanning mode and a geometric environment to obtain original images of an object to be reconstructed in different relative position relationships; preprocessing the original image to obtain grayscale values ​​of the original image, performing logarithmic transformation to convert the grayscale values ​​into corresponding X-ray absorptivity, and determining a tomographic reconstruction range according to the long-stroke scanning geometric environment; selecting a reconstruction width, traversing all reconstruction range points with the reconstruction width, and completing reconstruction of coronal and sagittal tomographic images by filtered back projection; obtaining overlapping parts of the reconstructed tomographic images, and splicing the overlapping parts using a weighted superposition method to form a long-stroke image, thereby realizing a tomographic reconstruction algorithm for dual-plane long-stroke scanning, realizing reconstruction of coronal and sagittal weighted-position tomographic images, expanding the tomographic image reconstruction range, and improving the image inter-layer resolution.

Description

Technical Field

[0001] The present invention relates to the field of medical imaging technology, and in particular to a tomographic reconstruction method and device based on dual-plane long-stroke scanning. Background Art

[0002] Orthopedics is a clinical medical discipline that mainly studies the anatomy, physiology, pathology and treatment of the skeletal muscle system.

[0003] Currently, clinical challenges remain to be overcome across various sub-disciplines of orthopedics. Adolescent idiopathic scoliosis is a common spinal condition with an overall prevalence of 0.47-5.2%. The male-to-female ratio ranges from 1:1.5 to 1:3, and the prevalence is significantly higher in women than in men among those with larger Cobb angles. The combined incidence of adjacent segment degeneration after spinal fusion is 29.3%, with 32.8% in the cervical spine and 26.6% in the lumbar spine. The revision rate after unicompartmental knee arthroplasty is higher than that after traditional total knee arthroplasty, with 36% attributed to postoperative non-replacement compartment and progressive patellofemoral arthritis. Anterior cruciate ligament rupture accounts for 20% of sports knee injuries, and the incidence of progressive knee arthritis exceeds 50% 10 years after ACL rupture. However, even with ACL reconstruction, 40-90% of patients will still experience arthritis for the rest of their lives. These objective clinical challenges in the fields of spinal surgery, joint surgery, and sports medicine require targeted, innovative medical equipment to assist orthopedic surgeons in addressing these challenges.

[0004] Assessing the anatomical morphology of a patient's diseased bones and joints through medical imaging technology can help orthopedic clinicians diagnose disease types, plan surgeries, and track postoperative outcomes, thereby improving the diagnosis and treatment of orthopedic diseases. Imaging techniques used in orthopedic clinical diagnosis mainly include: digital X-ray imaging, computed tomography (CT), magnetic resonance imaging (MRI), ultrasound imaging, and tomosynthesis imaging. Digital X-ray imaging technology is easy to store, has a low radiation dose, and offers high imaging quality. However, because it is based on cone-beam projection, it cannot distinguish and identify lesions at different depths and cannot perform three-dimensional measurements. CT can quickly and clearly achieve three-dimensional imaging of bones, muscles, and soft tissues, but due to its high radiation dose, it is not suitable for one-time full-body examinations or regular follow-up. MRI uses different sequences to achieve radiation-free and non-invasive imaging of key areas such as bones, muscles, and the spinal cord. However, due to the long imaging time and the small number of domestically installed MRI machines, it is difficult to ensure widespread orthopedic clinical application. Ultrasound imaging can more clearly diagnose fractures and muscle and ligament injuries, but currently ultrasound signals are difficult to diagnose important joint surfaces. Tomosynthesis is a technology that uses digital X-rays to acquire images at a narrow projection angle and reconstructs them using a back-projection algorithm. It offers low dose, short scan times, the ability to remove tissue structure aliasing, and enhance local tissue contrast. It has been shown to significantly improve orthopedic clinical resolution of complex anatomical structures, microfractures, and joint surface morphology. Therefore, tomosynthesis is a promising clinical orthopedic imaging technology.

[0005] In existing technologies, tomosynthesis reconstruction technology can be divided into three categories based on the scanning method: flat panel detector and radiation source in-plane reverse scanning, flat panel detector and radiation source parallel circumferential reverse scanning, and flat panel detector and radiation source in-plane and in-direction scanning. The key technical features of these three categories of tomosynthesis reconstruction technology are as follows:

[0006] (1) Flat panel detector-radiation source coplanar reverse scanning includes four common scanning modes: ① The flat panel detector and the radiation source move synchronously in opposite directions along a linear trajectory (e.g. Figure 1 A); ② The flat panel detector and the radiation source move synchronously in opposite directions along a circular trajectory (as shown in Figure 1 B); ③ The flat panel detector moves in a linear trajectory, and the radiation source moves in a circular trajectory in synchronous opposite directions (as shown in Figure 1 C); ④ The flat panel detector remains stationary, and the radiation source moves in a circular trajectory (as shown in Figure 1D). Because the flat-panel detector and radiation source always move in the same plane, geometric parameters such as the relative position of the flat-panel detector and radiation source during scanning, the source opening angle, rotation angle or movement distance, and target region depth can be used to determine the relationship between voxels in the target region, X-ray position, and absorptivity. Using back-projection, a tomographic image of the target region can be reconstructed. Because current clinical equipment consists of a flat-panel detector and radiation source combination, this scanning mode primarily reconstructs coronal tomographic images.

[0007] (2) Flat panel detector-radiation source parallel circular reverse scanning mode Figure 2 As shown. The center of the flat panel detector and the radiation source are in two parallel planes, around the same axis ( Figure 2 The Z axis in A is scanned in a circular orbit. During the scanning process, the central ray of the radiation source is kept perpendicular to the flat panel detector and passes through the center of the flat panel detector. When the opening angle of the radiation source is small, the actual cone beam X-ray can be approximately equivalent to a set of parallel beams. For any flat panel detector-radiation source position to shoot the target area, it can be approximately equivalent to sampling the target area at a specific angle section (such as Figure 2 (As shown in Figure A). In the corresponding three-dimensional Fourier space, the spatial resolution information of the target region can be approximately sampled at a specific angle. After frequency filtering and performing an inverse Fourier transform, a tomographic image of the target region can be reconstructed. Because current clinical equipment consists of a flat-panel detector-radiation source, this scanning mode primarily reconstructs coronal tomographic images.

[0008] (3) Flat panel detector-radiation source co-planar co-directional scanning mode is a long-stroke tomosynthesis imaging technology first proposed by Koichi Shibata et al. Figure 3 As shown. Different from the traditional scanning mode of synchronous reverse motion of flat-panel detector and radiation source, this scanning mode sets the center of the flat-panel detector and the radiation source to scan synchronously and in the same direction in the same plane with parallel lines as the trajectory. The scanning and imaging space range depends on the radiation source opening angle, the linear motion distance and interval of the flat-panel detector and radiation source, and the depth information of the reconstructed target area. Take a strip space parallel to the plane of the flat-panel detector in any imaging space, imitate the local detector-radiation source reverse linear scanning mode through reasonable data rearrangement, and realize the reconstruction of the strip tomographic image with the back projection algorithm, traverse all possible strip space positions within the target imaging space, and realize multi-depth long-stroke tomographic synthesis reconstruction. Since the current clinical equipment used is a set of flat-panel detectors and radiation sources, this type of scanning mode mainly scans and reconstructs coronal tomographic images.

[0009] The shortcomings of the existing technology are as follows: (1) The traditional tomosynthesis reconstruction space area is small, and the image aliasing in the non-central area is serious. The traditional tomosynthesis reconstruction method relies on five scanning modes: flat panel detector-radiation source reverse linear motion, flat panel detector-radiation source reverse circular motion, flat panel detector linear-radiation source circular motion, flat panel detector static-radiation source circular motion, flat panel detector-radiation source reverse circular trajectory motion. These five scanning modes determine that the reconstruction area with high image quality depends on parameters such as the flat panel detector-radiation source spacing (usually set to 600-1000mm), the flat panel detector-scanning object spacing (usually set to 100-500mm), the radiation source opening angle (usually not more than 25°), and the rotation angle (usually not more than 60°). For the most widely used tomosynthesis imaging technology, the reconstruction space area is small, and the area used for orthopedic three-dimensional measurement and analysis is limited. At the same time, the scanning mode and back projection reconstruction method of the tomosynthesis technology determine that the center of the target area can obtain higher intra-layer resolution and tissue contrast, but the back projection information at the edge of the target area and the non-target area is less, and the image has serious aliasing.

[0010] (2) Low inter-layer resolution. Currently, tomosynthesis technology uses only a set of flat-panel detectors and radioactive sources for scanning, so it can only reconstruct tomographic images in one direction. Since tomosynthesis technology mainly relies on the back projection of processed X-ray fluoroscopic images to form target layer images, the frequency response of the flat-panel detector-radioactive source co-planar scanning system to the target area in the direction of the flat-panel detector normal vector is low, and the frequency response of the flat-panel detector-radioactive source parallel circular scanning system to the target area in the direction of the parallel circular trajectory plane normal vector is low, resulting in low inter-layer resolution of the reconstructed image and loss of depth information outside the target layer.

[0011] (3) Weight-bearing imaging is not possible. Currently available tomosynthesis reconstruction systems are designed based on the subject's supine or lateral position, and can be fully applied to surgical guidance, bone morphology analysis, and shoulder, elbow, and wrist joint status assessment. However, for bones and joints such as the spine and lower limbs that bear weight in daily activities, the clinical value of important orthopedic parameters obtained by supine position measurement (such as the degree of scoliosis, lower limb force line alignment, hip joint coverage, etc.) is reduced.

[0012] In summary, it is necessary to propose a tomographic reconstruction algorithm based on dual-plane long-stroke scanning to solve the defects of the existing technology, such as small spatial area of ​​traditional tomographic synthesis reconstruction, severe image aliasing in non-central areas, low inter-layer resolution, and inability to perform weighted imaging. Summary of the Invention

[0013] The purpose of the present invention is to provide a tomographic reconstruction method and device based on dual-plane long-stroke scanning, combining dual-plane X-ray technology and tomographic synthesis reconstruction technology based on single-plane long-stroke scanning to realize a tomographic reconstruction algorithm based on dual-plane long-stroke scanning, while realizing coronal and sagittal weighted-position tomographic image reconstruction, expanding the tomographic image reconstruction range, and improving the image inter-layer resolution.

[0014] The present invention provides a tomographic reconstruction method based on dual-plane long-stroke scanning, which is applied to a dual-plane X-ray system. The dual-plane X-ray system includes a first X-ray machine and a second X-ray machine.

[0015] The first X-ray machine includes a first radiation source for emitting X-rays toward a human body; a first detector disposed opposite to the first radiation source for receiving the X-rays after passing through the human body;

[0016] The second X-ray machine includes a second radiation source for emitting X-rays toward the human body; a second detector disposed opposite to the second radiation source for receiving the X-rays after passing through the human body;

[0017] The reconstruction method comprises the following steps:

[0018] Setting a dual-plane X-ray scanning mode and geometric environment, determining the relative positional relationship between the radiation source and the detector on the first X-ray machine and the second X-ray machine, and then performing a long-stroke scan to obtain images of the object to be reconstructed at different relative positional relationships to obtain an original image;

[0019] Preprocessing the original image to obtain grayscale values ​​of the original image, performing logarithmic transformation to convert the grayscale values ​​into corresponding X-ray absorptivity, and determining a tomographic reconstruction range based on a long-stroke scanning geometric environment;

[0020] Selecting a reconstruction width, traversing all reconstruction range points with the reconstruction width, and completing the reconstruction of all coronal and sagittal tomographic images by filtered back projection; and

[0021] The overlapping parts of the reconstructed tomographic images in the vertical direction are obtained, and the weighted superposition method is used to stitch the overlapping parts to form a long-stroke image.

[0022] Preferably, the setting of the dual-plane X-ray scanning mode and the geometric environment, and performing long-stroke scanning after determining the relative positional relationship between the radiation source and the detector on the first X-ray machine and the second X-ray machine includes:

[0023] Setting a first distance between the first flat panel detector and the first radiation source, and setting a second distance between the second flat panel detector and the second radiation source, wherein the first distance and the second distance are set to be the same and remain unchanged during a scanning process;

[0024] Setting a scanning range so that when the object to be imaged is located in a common area of ​​the beams formed by the first radiation source and the second radiation source, the first flat panel detector and the first radiation source, and the second flat panel detector and the second radiation source, move synchronously in a vertical direction;

[0025] Starting from the initial height h1, the first flat panel detector and the first radiation source and the second flat panel detector and the second radiation source move synchronously in the vertical direction at the same linear speed, and complete an X-ray exposure every △h until the entire scanning range is covered and the first X-ray machine and the second X-ray machine reach the end height h N , the whole process completes exposure N times; and

[0026] Get a specific height h k The first X-ray machine and the second X-ray machine are used as references for the image space of the second X-ray machine, and a rigid body orthogonal space coordinate system is established according to the relative position relationship between the first X-ray machine and the second X-ray machine at the initial position.

[0027] Preferably, obtaining the original image after acquiring the images of the object to be reconstructed at different relative position relationships comprises:

[0028] For a specific height h k Under the above situation, a rigid body orthogonal space coordinate system (R1, V 1,k ),satisfy:

[0029]

[0030] Where R1 is a 3×3 matrix, each column of which represents the unit vector of the coordinate axis of the first X-ray machine image coordinate system; V 1，k is a 3×1 column vector, representing the origin of the first X-ray image coordinate system; k is a scalar, indicating the vertical height between the center of the first detector and the first radiation source;

[0031] For a specific height h k Next, a rigid body orthogonal space coordinate system (R2, V 2，k ), where R 2，k V is a 3×3 matrix, each column of which represents the coordinate axis unit vector of the second X-ray machine image coordinate system; 2，k is a 3×1 column vector, representing the origin of the second X-ray machine image coordinate system;

[0032] Rigid body orthogonal space coordinate system (R2, V 2，k) is determined by the relative positional relationship between the first X-ray machine and the second X-ray machine in the initial position:

[0033] R 2,k =R2, V 2,k =V 21 +V 1,k .

[0034] Among them, V 21 is the relative position relationship between the second X-ray machine image space and the first X-ray machine image space at the initial height. Preferably, the grayscale value of the original image is obtained, and a logarithmic transformation is performed to convert the grayscale value into the corresponding X-ray absorption rate. The logarithmic transformation is:

[0035]

[0036] Wherein, I is the grayscale value of the original X-ray image captured by the first X-ray machine or the second X-ray machine at any height, which represents the residual intensity of the X-ray after penetrating the tissue; I0 is the intensity of the X-ray emitted by the radiation source; μ n is the tissue absorption rate of the photographed object, d n is the tissue thickness of the photographed object, and C is a constant after logarithmic transformation.

[0037] Preferably, the tomographic reconstruction range is determined according to the long-stroke scanning geometric environment.

[0038] The chromatography range includes a biplane chromatography region and a non-biplane chromatography region.

[0039] The dual-plane tomosynthesis region is the common region where cone-beam X-rays emitted by the first X-ray machine and the second X-ray machine intersect, and is capable of dual-plane tomosynthesis reconstruction;

[0040] The non-dual-plane tomosynthesis region is only penetrated by cone-beam X-rays emitted by the first X-ray machine or the second X-ray machine, and can only be reconstructed by single-plane tomosynthesis.

[0041] Preferably, the tomographic reconstruction range is determined according to the long-stroke scanning geometric environment.

[0042] For any point P within the reconstruction range, determine the heights H1, H2, ..., H n The X-ray penetration points P emitted by the first X-ray machine and the second X-ray machine are recorded at different positions of the image, where H1, H2, ..., H n satisfy:

[0043] H1,H2,…,H n ∈{h k |k=1,2,...,N},

[0044] According to the geometric environment (R1, V 1，k ), (R2, V 2，k ) and the distance between the flat panel detector and the radiation source is set to SID.

[0045] Determine a first projection point P1 of point P on the first detector plane and a second projection point P2 of point P on the second detector plane, where the local coordinates of the first projection point P1 and the second projection point P2 relative to the first detector and the second detector plane satisfy:

[0046]

[0047]

[0048] Among them, T is the operation of matrix in linear algebra, which means transpose, and V1 is V 1，k , V2 is V 2，k .

[0049] Preferably, the step of selecting a reconstruction width, traversing all reconstruction range points with the reconstruction width, and completing the reconstruction of all coronal and sagittal tomographic images by filtered back projection comprises:

[0050] Using the first X-ray machine to project any point set within a reconstruction range with a height range equal to the selected reconstruction width onto the first detector plane based on the geometric environment to form a first rearranged image;

[0051] The second X-ray machine projects any point set within the reconstruction range having a height range of the selected reconstruction width onto the second detector plane based on the geometric environment to form a second rearranged image;

[0052] performing one-dimensional Fourier transform on the first rearranged image and the second rearranged image by column to obtain a primary transformed image, the primary transformed image including an image of the first rearranged image after a single Fourier transform and an image of the second rearranged image after a single Fourier transform;

[0053] Setting a ramp filter according to the selected reconstruction width, performing a one-dimensional Fourier transform on the ramp filter and superimposing a window function, filtering each column of the primary transformed image, and performing a one-dimensional inverse Fourier transform on each column to obtain a secondary transformed image, wherein the secondary transformed image includes an image of the first rearranged image after a secondary Fourier transform and an image of the second rearranged image after a secondary Fourier transform;

[0054] Back-projecting the secondary transformed image to the space where the point set is located according to the geometric projection environment to obtain a first filtered back-projection image and a second filtered back-projection image respectively;

[0055] The first filtered back-projection image and the second filtered back-projection image are weightedly superimposed to obtain a final reconstructed tomographic strip image.

[0056] Preferably, the obtaining of overlapping portions of the reconstructed tomographic images in the vertical direction and stitching the overlapping portions using a weighted superposition method to form a long-run image comprises:

[0057] For height h i Tomographic image ff with width w i , its adjacent height h i-1 With h i+1 There is a tomographic image ff with width w i-1 With ff i+1 , ff i With ff i-1 , ff i With ff i+1 The overlapping height is d, and the weighted superposition function W(n) used is:

[0058]

[0059] Where f(n) is an arbitrary strictly monotonically increasing function defined on [0,1] with a range of [0,1] and a derivative of 0 at n=0,1;

[0060] For ff i-1 , ff i with ff i+1 Then we have:

[0061] ff(n)=W(nd)ff i-1 (n)+W(n)ff i (n)+W(n+d)ff i+1 (n),

[0062] Among them, ff i-1 is at height h i-1 The reconstructed image under ff i+1 is at height h i+1 The reconstructed image below.

[0063] The present invention also provides a tomographic reconstruction device based on dual-plane long-stroke scanning, which is applied to a dual-plane X-ray system. The dual-plane X-ray system includes a first X-ray machine and a second X-ray machine.

[0064] The first X-ray machine includes a first radiation source for emitting X-rays toward a human body; a first detector disposed opposite to the first radiation source for receiving the X-rays after passing through the human body;

[0065] The second X-ray machine includes a second radiation source for emitting X-rays toward the human body; a second detector disposed opposite to the second radiation source for receiving the X-rays after passing through the human body;

[0066] The reconstruction device comprises:

[0067] a scanning mode and geometric environment setting module, configured to set a dual-plane X-ray scanning mode and geometric environment, determine the relative positional relationship between the radiation source and the detector on the first X-ray machine and the second X-ray machine, perform a long-stroke scan, and obtain images of the object to be reconstructed at different relative positional relationships to obtain the original image;

[0068] An image preprocessing module is used to preprocess the original image, obtain the grayscale value of the original image, perform logarithmic transformation to convert the grayscale value into the corresponding X-ray absorptivity, and determine the tomographic reconstruction range according to the long-stroke scanning geometric environment;

[0069] The tomographic image reconstruction module is used to select the reconstruction width, traverse all reconstruction range points with the reconstruction width, and complete the reconstruction of all coronal and sagittal tomographic images through filtered back projection;

[0070] The long-stroke image stitching module is used to obtain the overlapping parts of the reconstructed tomographic images in the vertical direction, and stitch the overlapping parts using the weighted superposition method to form a long-stroke image.

[0071] The present invention further provides a computer-readable storage medium, wherein the computer program for implementing the tomographic reconstruction method based on dual-plane long-stroke scanning as described in an embodiment of the present invention is loaded in the computer-readable storage medium.

[0072] With respect to the prior art, the present invention has the following beneficial effects:

[0073] The tomographic reconstruction method based on dual-plane long-stroke scanning provided by the present invention combines dual-plane X-ray technology with tomosynthesis reconstruction technology based on single-plane long-stroke scanning to implement a tomographic reconstruction algorithm based on dual-plane long-stroke scanning. This method can simultaneously reconstruct tomographic images in the coronal and sagittal weighted positions, expand the range of tomographic image reconstruction, and improve the inter-layer resolution of images.

[0074] The present invention adopts a vertical scanning method, which can scan a long range from top to bottom, and output images of the patient's whole body including the coronal plane and sagittal plane in the weight-bearing position, forming a long-range image. When scanning a long range from top to bottom, the reconstruction space area is large;

[0075] The tomosynthesis reconstruction method adopted by the present invention is not only applicable to the subject's supine or lateral position, but also can be used for weight-bearing imaging. For bones and joints such as the spine and lower limbs that bear weight in daily activities, important orthopedic parameters obtained by measuring in the supine and weight-bearing positions, such as the degree of scoliosis, lower limb force line alignment, and hip joint coverage, have high clinical value. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 Schematic diagram of the flat-panel detector and radiation source coplanar reverse scanning method in the background technology of the present invention, A is the flat-panel detector and radiation source reverse linear motion; B is the flat-panel detector and radiation source reverse circular motion; C is the flat-panel detector linear and radiation source reverse circular motion; D is the flat-panel detector stationary and radiation source circular motion;

[0077] Figure 2 This is a schematic diagram of parallel circular reverse scanning of a flat panel detector and a radiation source in the background technology of the present invention;

[0078] Figure 3 This is a schematic diagram of the flat panel detector and radiation source scanning in the same plane and in the same direction in the background technology of the present invention;

[0079] Figure 4 Schematic diagram of the steps of a tomographic reconstruction method based on dual-plane long-stroke scanning in an embodiment of the present invention;

[0080] Figure 5 Schematic diagram of a dual-plane long-stroke scanning structure and coordinate system in an embodiment of the present invention;

[0081] Figure 6 Schematic diagram of the tomographic reconstruction range and projection relationship based on dual-plane long-stroke scanning in an embodiment of the present invention;

[0082] Figure 7 Schematic diagram of dual-plane data rearrangement and spatial geometric relationship in an embodiment of the present invention. DETAILED DESCRIPTION

[0083] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0084] Example 1

[0085] The present invention provides a tomographic reconstruction method based on dual-plane long-stroke scanning, which is applied to a dual-plane X-ray system. The dual-plane X-ray system includes a first X-ray machine and a second X-ray machine. The first X-ray machine and the second X-ray machine are arranged at a certain angle. The first X-ray machine includes a first radiation source and a first detector. The first radiation source is used to emit X-rays toward the human body. The first detector is arranged opposite to the first radiation source and is used to receive the X-rays after they pass through the human body.

[0086] The second X-ray machine includes a second radiation source and a second detector. The second radiation source is used to emit X-rays toward the human body. The second detector is arranged opposite the second radiation source and is used to receive the X-rays after passing through the human body. The dual-plane X-ray machine system used in this embodiment includes a first X-ray machine (hereinafter referred to as F1) and a second X-ray machine (hereinafter referred to as F2), as well as a rack connecting the hardware devices, a control motor, etc. When reconstructing in a computer system, the first and second X-ray machines are primarily considered. Other hardware devices such as the rack are only used to ensure the geometric environment required for scanning and reconstruction. The first and second detectors used in this embodiment can be flat-panel detectors.

[0087] like Figure 4 As shown, the reconstruction method includes the following steps:

[0088] S1: Set the dual-plane X-ray scanning mode and geometric environment, determine the relative position relationship between the radiation source and the detector on the first X-ray machine and the second X-ray machine, and then perform a long-stroke scan to obtain the image of the object to be reconstructed at different relative position relationships (i.e., different height positions) to obtain the original image; those skilled in the art will understand that in this embodiment, before long-stroke stitching, the relative position relationship between the first and second X-ray machines must be determined first, and then a long-stroke scan is performed. The method for determining the relative position relationship between the first and second X-ray machines here is already disclosed and will not be repeated here.

[0089] S2: Preprocessing the original image to obtain grayscale values ​​of the original image, performing logarithmic transformation to convert the grayscale values ​​into corresponding X-ray absorptivity, and determining a tomographic reconstruction range based on a long-stroke scanning geometric environment;

[0090] S3: Select the reconstruction width, traverse all reconstruction range points with the reconstruction width, and complete the reconstruction of all coronal and sagittal tomographic images through filtered back projection; the range of tomographic reconstruction includes depth, i.e., z-axis position and width, i.e., the upper and lower boundary positions of the y-axis, which is a narrow rectangle.

[0091] S4: Obtain overlapping portions of the vertically reconstructed tomographic images and stitch them together using a weighted superposition method to form a long-run image. The long-run image formed in this embodiment refers to a full-body scan output of the patient, including coronal and sagittal weight-bearing positions, from top to bottom.

[0092] The tomographic reconstruction algorithm based on dual-plane long-stroke scanning of the present invention can simultaneously realize the reconstruction of coronal and sagittal weight-bearing tomographic images, expand the tomographic image reconstruction range, and improve the image inter-layer resolution.

[0093] Specifically, setting the dual-plane X-ray scanning mode and geometric environment in step S1 includes:

[0094] A first distance between the first flat panel detector and the first radiation source is set, and a second distance between the second flat panel detector and the second radiation source is set, the first distance and the second distance are set to be the same, and the first distance and the second distance remain unchanged during the scanning process; for F1 and F2, the distance between the flat panel detector and the radiation source is set to SID and remains unchanged during the scanning process.

[0095] The scanning range is set. When the object to be imaged is located in the common area of ​​the beam formed by the first radiation source and the second radiation source, the first flat panel detector and the first radiation source and the second flat panel detector and the second radiation source operate synchronously and move along a vertical trajectory. When the object to be imaged stands in the common area of ​​the cone beam formed by the F1 and F2 radiation sources, Figure 5 As shown, the directions of the y1 axis and the y2 axis are the same, which are the directions of the motion trajectories of F1 and F2.

[0096] After setting the vertical scanning range, the first flat panel detector and the first radiation source of F1, the second flat panel detector and the second radiation source of F2 all start from the initial height h1, move synchronously and in the same direction along the y1 axis at the same linear speed, and complete an X-ray exposure at an interval of △h to obtain a specific height h k The first X-ray machine F1 and the second X-ray machine F2 are used as references for the lower double-plane X-ray image, and a rigid body orthogonal space coordinate system is established according to the relative position relationship between the first X-ray machine F1 and the second X-ray machine F2 at the initial position.

[0097] The first X-ray machine F1 and the second X-ray machine F2 maintain uniform motion and exposure for long-stroke scanning until the entire scanning range is covered and the first X-ray machine F1 and the second X-ray machine F2 reach the end height h N , the whole process is exposed N times.

[0098] Specifically, obtaining the original image after acquiring the images of the object to be reconstructed at different relative position relationships includes:

[0099] For a specific height h k Under the condition that the image space of the first X-ray machine F1 is used as a reference to establish a rigid body orthogonal space coordinate system (R1, V 1,k ),satisfy:

[0100]

[0101] Where R1 is a 3×3 matrix, each column of which represents the unit vector of the coordinate axis of the image coordinate system of the first X-ray machine F1; V 1，k is a 3×1 column vector, representing the origin of the image coordinate system of the first X-ray machine F1; k is a scalar indicating the vertical height between the center of the first detector and the first radiation source in the current F1.

[0102] For a specific height h k Next, a rigid body orthogonal space coordinate system (R2, V 2，k ), where R 2，k V is a 3×3 matrix, each column of which represents the coordinate axis unit vector of the image coordinate system of the second X-ray machine F2; 2，k It is a 3×1 column vector, representing the origin position of the image coordinate system of the second X-ray machine F2.

[0103] Since the relative positional relationship between the first detector and the first radiation source of F1 and the second detector and the second radiation source of F2 does not change during the long-stroke scanning process, the rigid body orthogonal space coordinate system (R2, V 2，k ) is determined by the relative positional relationship between the first X-ray machine F1 and the second X-ray machine F2 in the initial position:

[0104] R 2,k =R2, V 2,k =V 21 +V 1,k .

[0105] In the formula, V 21 is the relative position relationship between the second X-ray machine image space and the first X-ray machine image space at the initial height.

[0106] Specifically, in step S2, the grayscale value of the original image is obtained, and a logarithmic transformation is performed to convert the grayscale value into the corresponding X-ray absorption rate. Since the data obtained by the flat panel detector is the residual ray intensity after the X-ray penetrates the object being photographed, in order to optimize the subsequent linear operation, the obtained image is preprocessed and a logarithmic transformation is performed:

[0107]

[0108] Wherein, I is the grayscale value of the original X-ray image captured by the first X-ray machine or the second X-ray machine at any height, which represents the residual intensity of the X-ray after penetrating the tissue; I0 is the intensity of the X-ray emitted by the radiation source; μ nis the tissue absorption rate of the photographed object, d n is the tissue thickness of the photographed object, and C is a constant after logarithmic transformation.

[0109] Specifically, in step S2, the tomographic reconstruction range is determined based on the long-stroke scanning geometric environment. The tomographic reconstruction range includes a dual-plane tomographic area and a non-dual-plane tomographic area. The dual-plane tomographic area is the common area where the cone-beam X-rays emitted by the first X-ray machine F1 and the second X-ray machine F2 intersect, and dual-plane tomographic synthesis reconstruction can be performed. In the non-dual-plane tomographic area, only the cone-beam X-rays emitted by the first X-ray machine or the second X-ray machine pass through, and only single-plane tomographic synthesis reconstruction can be performed. It can be understood by those skilled in the art that for the long-stroke scanning geometric environment of F1 and F2, the tomographic reconstruction range can be determined, such as Figure 6 The gray area represents the common area where cone-beam X-rays emitted by F1 and F2 intersect, and dual-plane tomosynthesis reconstruction can be performed; the non-gray area represents the area where only cone-beam X-rays emitted by F1 or F2 pass through, and only single-plane tomosynthesis reconstruction can be performed.

[0110] Specifically, for any point P within the reconstruction range, determine the heights H1, H2, ..., H n The X-ray penetration points P emitted by the first X-ray machine and the second X-ray machine are recorded at different positions of the image, where H1, H2, ..., H n satisfy:

[0111] H1,H2,…,H n ∈{h k |k=1,2,...,N},

[0112] According to the geometric environment (R1, V 1，k ), (R2, V 2，k ) and the distance between the flat panel detector and the radiation source is set to SID, and the first projection point P1 of point P on the first detector plane and the second projection point P2 of point P on the second detector plane are determined, such as Figure 6 As shown. The local coordinates of the first projection point P1 and the second projection point P2 relative to the planes of the first detector and the second detector satisfy:

[0113]

[0114]

[0115] Among them, T is the operation of matrix in linear algebra, which means transpose, and V1 is V 1，k , V2 is V 2，k .

[0116] The first flat panel detector of F1 and the first radiation source are at different heights H1, H2, ..., H n The projection P1 of a specific spatial point P is stored in different rows and the same column, and the second flat panel detector of F2 and the second radiation source are at different heights H1, H2, ..., H n The projection P2 of a specific spatial point P is stored in different rows and the same column.

[0117] like Figure 7 As shown, Figure 7 The grid points in A represent the pixel centers of the preprocessed image, and the black dots represent the local coordinate positions of projection points P1 and P2 relative to the image. Because projection points P1 and P2 do not necessarily coincide with the pixel centers at the grid points, two-dimensional linear interpolation is used to obtain the data at the projection points.

[0118] In order to expand the reconstruction range of tomographic images, realize the reconstruction of coronal and sagittal weighted-position tomographic images, and improve the inter-layer resolution of images, the reconstruction width is selected in step S3, and all reconstruction range points are traversed with the reconstruction width. The reconstruction of all coronal and sagittal tomographic images is completed by filtered back projection, including:

[0119] The first X-ray machine F1 projects any point set within the reconstruction range of the selected reconstruction width w onto the first detector plane based on the geometric environment to form a first rearranged image f1;

[0120] The second X-ray machine F2 projects any point set within the reconstruction range of the selected reconstruction width w onto the second detector plane based on the geometric environment to form a second rearranged image f2;

[0121] performing a one-dimensional Fourier transform on each of the first rearranged image f1 and the second rearranged image f2 by column to obtain a primary transformed image, wherein the primary transformed image includes an image FF1 after the first rearranged image f1 undergoes a single Fourier transform, and an image FF2 after the second rearranged image undergoes a single Fourier transform;

[0122] The reconstruction width w is selected in the vertical direction, and the F1 geometric environment will reconstruct any point set P with a height range of w within the range. Pset Projected onto the first flat detector plane to form a vertical strip image f1, F2 geometric environment will reconstruct the point set P within any height range w Pset Projected onto the second flat-panel detector plane to form a vertical strip image f2, the images f1, f2 and the filter h[n] are transformed from physical space to frequency space to satisfy:

[0123]

[0124]

[0125] where f 1,Hn is the height H n Under F1 point set P Pset The projected image is a w×1 vector; f 2,Hn is the height H n Under F2 point set P Pset The projected image is a w×1 vector; f1 and f2 are w×n matrices. Perform a one-dimensional Fourier transform on f1 and f2 column by column to obtain FF1 and FF2.

[0126] To filter out noise and enhance image edges, a ramp filter is set based on the selected reconstruction width w. A one-dimensional Fourier transform is performed on the ramp filter, and a window function is superimposed. Each column of the first transformed image is filtered, and a one-dimensional inverse Fourier transform is performed to obtain a second transformed image. The second transformed image includes a second Fourier transformed image of the first rearranged image and a second Fourier transformed image of the second rearranged image. The second transformed images are back-projected into the space containing the point set according to the geometric projection environment to obtain a first filtered back-projected image and a second filtered back-projected image, respectively.

[0127] Set the shelving filter h[n] according to the selected reconstruction width w:

[0128]

[0129] Where mod(n, 2) is the remainder function, and the calculation result is the remainder when n is divided by 2. When n is an odd number, the calculation result is 1, and when n is an even number, the calculation result is 0.

[0130] Perform a one-dimensional Fourier transform on the ramp filter h[n] to obtain H[v] and superimpose the Shepp-Logan window to obtain:

[0131]

[0132] Where A(v) is the window function in one-dimensional Fourier space. filter Filter each column of FF1 and FF2, perform one-dimensional inverse Fourier transform, and back-project to point set P according to the geometric projection environment Pset In the space, a first filtered back-projection image ff1 and a second filtered back-projection image ff2 are obtained respectively.

[0133] The first filtered back-projection image ff1 and the second filtered back-projection image ff2 obtained in two planes are weightedly superimposed to obtain a final reconstructed tomographic strip image:

[0134] ff=ωff1+(1-ω)ff2,ω∈(0,1)

[0135] All reconstruction range points are traversed with the reconstruction width w to complete the reconstruction of all coronal and sagittal tomographic images.

[0136] Specifically, in step S4, obtaining the overlapping portion of the reconstructed tomographic images in the vertical direction and stitching the overlapping portion using the weighted superposition method to form a long-run image includes:

[0137] For height h i Tomographic image ff with width w i , its adjacent height h i-1 With h i+1 There is a tomographic image ff with width w i-1 with ff i+1 , ff i with ff i-1 , ff i with ff i+1 The overlapping height is d, and the weighted superposition function W(n) used is:

[0138]

[0139] Where f(n) is an arbitrary strictly monotonically increasing function defined on [0,1] with a range of [0,1] and a derivative of 0 at n=0,1;

[0140] For ff i-1 , ff i with ff i+1 Then we have:

[0141] ff(n)=W(nd)ff i-1 (n)+W(n)ff i (n)+W(n+d)ff i+1 (n),

[0142] Among them, ff i-1 is at height h i-1 The reconstructed image under ff i+1 is at height h i+1 The reconstructed image below.

[0143] The tomographic reconstruction method based on dual-plane long-stroke scanning provided by the present invention has a vertical scanning range of 2m and a total long-stroke scanning time of less than 15s; the intra-layer resolution of the reconstructed tomographic image reaches 0.2mm, and the inter-layer resolution reaches 1mm, which expands the tomographic image reconstruction range and improves the image inter-layer resolution.

[0144] Example 2

[0145] The present invention also provides a tomographic reconstruction device based on dual-plane long-stroke scanning, which is applied to a dual-plane X-ray system. The dual-plane X-ray system includes a first X-ray machine and a second X-ray machine.

[0146] The first X-ray machine includes a first radiation source for emitting X-rays toward the human body; and a first detector, disposed opposite to the first radiation source, for receiving the X-rays after passing through the human body.

[0147] The second X-ray machine includes a second radiation source for emitting X-rays toward the human body; and a second detector, disposed opposite to the second radiation source, for receiving the X-rays after passing through the human body.

[0148] The reconstruction device includes a scanning mode and geometric environment setting module, an image preprocessing module, a tomographic image reconstruction module, and a long-stroke image stitching module. The scanning mode and geometric environment setting module is used to set the dual-plane X-ray scanning mode and geometric environment, determine the relative positional relationship between the radiation source and detector on the first and second X-ray machines, perform long-stroke scanning, and obtain images of the object to be reconstructed in different relative positional relationships to obtain the original image.

[0149] The image preprocessing module is used to preprocess the original image, obtain the grayscale value of the original image, perform logarithmic transformation to convert the grayscale value into the corresponding X-ray absorption rate, and at the same time, determine the tomographic reconstruction range according to the long-stroke scanning geometric environment.

[0150] The tomographic image reconstruction module is used to select a reconstruction width, traverse all reconstruction range points with the reconstruction width, and complete the reconstruction of all coronal and sagittal tomographic images through filtered back projection.

[0151] The long-stroke image stitching module is used to obtain the overlapping parts of the reconstructed tomographic images in the vertical direction, and stitch the overlapping parts using the weighted superposition method to form a long-stroke image.

[0152] The present invention provides a tomographic reconstruction algorithm based on dual-plane long-stroke scanning, which can simultaneously realize the reconstruction of coronal and sagittal weight-bearing tomographic images, expand the tomographic image reconstruction range, and improve the image inter-layer resolution.

[0153] The specific contents and implementation methods of the above-mentioned scanning mode and geometric environment setting module, image preprocessing module, tomographic image reconstruction module and long-stroke image stitching module are as described in the first embodiment and will not be repeated here.

[0154] Finally, in order to apply the above-mentioned tomographic reconstruction method based on dual-plane long-stroke scanning to an image acquisition and generation system, device or equipment with relevant hardware conditions, the present application also provides a computer-readable storage medium, in which a computer program is loaded. When the computer program is executed by a computer, it realizes the functions of the corresponding method embodiment as described above.

[0155] Likewise, the present application also protects a computer-readable storage medium loaded with a computer program for implementing the tomographic reconstruction method based on dual-plane long-stroke scanning.

[0156] In each of the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When software is used for implementation, it can be implemented in whole or in part in the form of a computer program. The computer program includes one or more computer programs. When the computer program is loaded and executed on a computer, the process or function described in the embodiment of the present disclosure is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer program can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer program can be transmitted from a website, computer, server or data center to another website, computer, server or data center by wired (such as coaxial cable, optical fiber, DDL (Digital Subscriber Line, Digital Subscriber Line)) or wireless (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a high-density DVD (Digital Video DiDD, digital video disc)), or a semiconductor medium (eg, a solid-state drive (SSD)).

[0157] It should be noted that, in the above-mentioned embodiment, the terms used herein are only for the purpose of describing specific exemplary embodiments, and are not intended to be restrictive. As used herein, the singular forms "one", "an" and "the or described" may be intended to also include plural forms, unless the context clearly indicates otherwise. The terms "comprise", "include" and "have" are inclusive and therefore specify the presence of stated features, wholes, steps, operations, elements and / or parts, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, parts and / or their groups. The method steps, processes and operations described herein should not be interpreted as necessarily requiring the method steps, processes and operations to be performed in the specific order discussed or shown, unless otherwise specified, they must be performed in the set step order. It should also be understood that additional or alternative steps may be adopted.

[0158] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A tomographic reconstruction method based on dual-plane long-stroke scanning, characterized in that: Applicable to a dual-plane X-ray system, the dual-plane X-ray system comprising a first X-ray machine and a second X-ray machine, The first X-ray machine includes a first radiation source for emitting X-rays toward a human body; a first detector disposed opposite to the first radiation source for receiving the X-rays after passing through the human body; The second X-ray machine includes a second radiation source for emitting X-rays toward the human body; a second detector disposed opposite to the second radiation source for receiving the X-rays after passing through the human body; The reconstruction method comprises the following steps: Setting a dual-plane X-ray scanning mode and geometric environment, determining the relative positional relationship between the radiation source and the detector on the first X-ray machine and the second X-ray machine, and then performing a long-stroke scan to obtain images of the object to be reconstructed at different relative positional relationships to obtain an original image; Preprocessing the original image to obtain grayscale values ​​of the original image, performing logarithmic transformation to convert the grayscale values ​​into corresponding X-ray absorptivity, and determining a tomographic reconstruction range based on a long-stroke scanning geometric environment; Selecting a reconstruction width, traversing all reconstruction range points with the reconstruction width, and completing the reconstruction of all coronal and sagittal tomographic images by filtered back projection; and The overlapping parts of the reconstructed tomographic images in the vertical direction are obtained, and the weighted superposition method is used to stitch the overlapping parts to form a long-stroke image.

2. The tomographic reconstruction method based on dual-plane long-stroke scanning according to claim 1, characterized in that: The step of setting a dual-plane X-ray scanning mode and a geometric environment, and determining the relative positional relationship between the radiation source and the detector on the first X-ray machine and the second X-ray machine before performing a long-stroke scan comprises: Setting a first distance between the first detector and the first radiation source, and setting a second distance between the second detector and the second radiation source, wherein the first distance and the second distance are set to be the same and remain unchanged during a scanning process; Setting a scanning range so that when the object to be imaged is located in a common area of ​​the beams formed by the first radiation source and the second radiation source, the first detector and the first radiation source and the second detector and the second radiation source move synchronously in a vertical direction; Starting from the initial height h1, the first detector and the first radiation source and the second detector and the second radiation source move synchronously in the vertical direction at the same linear speed, and complete an X-ray exposure every △h until the entire scanning range is covered and the first X-ray machine and the second X-ray machine reach the end height h N , the whole process completes exposure N times; and Get a specific height h k The first X-ray machine and the second X-ray machine are used as references for the image space of the second X-ray machine, and a rigid body orthogonal space coordinate system is established according to the relative position relationship between the first X-ray machine and the second X-ray machine at the initial position.

3. The tomographic reconstruction method based on dual-plane long-stroke scanning according to claim 2, characterized in that: The obtaining of the original image after acquiring the images of the object to be reconstructed in different relative position relationships comprises: For a specific height h k Under the above situation, a rigid body orthogonal space coordinate system (R1, V 1,k ),satisfy: , where R1 is a 3×3 matrix, each column of which represents the coordinate axis unit vector of the first X-ray machine image coordinate system; V 1,k is a 3×1 column vector, representing the origin of the first X-ray image coordinate system; k is a scalar, indicating the vertical height between the center of the first detector and the first radiation source; For a specific height h k Under the condition that the image space of the second X-ray machine is used as the reference, a rigid body orthogonal space coordinate system (R2, V 2,k ), where R 2,k V is a 3×3 matrix, each column of which represents the coordinate axis unit vector of the second X-ray machine image coordinate system; 2,k is a 3×1 column vector, representing the origin of the second X-ray machine image coordinate system; Rigid body orthogonal space coordinate system (R2, V 2,k ) is determined by the relative positional relationship between the first X-ray machine and the second X-ray machine in the initial position: R2=R 2，k ;V 2，k =V 21 +V 1，k ; Among them, V 21 is the relative positional relationship between the second X-ray machine image space and the first X-ray machine image space at the initial height.

4. The tomographic reconstruction method based on dual-plane long-stroke scanning according to claim 1, characterized in that: The grayscale value of the original image is obtained, and a logarithmic transformation is performed to convert the grayscale value into the corresponding X-ray absorption rate. The logarithmic transformation formula is: ， Wherein, I is the grayscale value of the original X-ray image captured by the first X-ray machine or the second X-ray machine at any height, which represents the residual intensity of the X-ray after penetrating the tissue; I0 is the intensity of the X-ray emitted by the radiation source; μ n is the tissue absorption rate of the photographed object, d n is the tissue thickness of the photographed object, and C is a constant after logarithmic transformation.

5. The tomographic reconstruction method based on dual-plane long-stroke scanning according to claim 1, characterized in that: The tomographic reconstruction range is determined based on the long-stroke scanning geometric environment. The tomographic reconstruction range includes a dual-plane tomographic region and a non-dual-plane tomographic region; The dual-plane tomosynthesis region is the common region where cone-beam X-rays emitted by the first X-ray machine and the second X-ray machine intersect, and is capable of dual-plane tomosynthesis reconstruction; The non-dual-plane tomosynthesis region is only penetrated by cone-beam X-rays emitted by the first X-ray machine or the second X-ray machine, and can only be reconstructed by single-plane tomosynthesis.

6. The tomographic reconstruction method based on dual-plane long-stroke scanning according to claim 3, characterized in that: The tomographic reconstruction range is determined based on the long-stroke scanning geometric environment. For any point P within the reconstruction range, determine the heights H1, H2, …, H n The X-ray penetration points P emitted by the first X-ray machine and the second X-ray machine are recorded at different positions of the image, where H1, H2, ..., H n satisfy: ， According to the rigid body orthogonal space coordinate system (R1, V 1,k ), (R2, V 2,k ) and the distance between the flat panel detector and the radiation source is set to SID. Determine a first projection point P1 of point P on the first detector plane and a second projection point P2 of point P on the second detector plane, where the local coordinates of the first projection point P1 and the second projection point P2 relative to the first detector and the second detector plane satisfy: ， ， Among them, T is the operation of matrix in linear algebra, which means transpose, and V1 is V 1,k , V2 is V 2,k .

7. The tomographic reconstruction method based on dual-plane long-stroke scanning according to claim 1, characterized in that: The step of selecting a reconstruction width, traversing all reconstruction range points with the reconstruction width, and completing the reconstruction of all coronal and sagittal tomographic images by filtered back projection includes: Using the first X-ray machine to project any point set within a reconstruction range with a height range equal to the selected reconstruction width onto the first detector plane based on the geometric environment to form a first rearranged image; The second X-ray machine projects any point set within the reconstruction range having a height range of the selected reconstruction width onto the second detector plane based on the geometric environment to form a second rearranged image; performing one-dimensional Fourier transform on the first rearranged image and the second rearranged image by column to obtain a primary transformed image, the primary transformed image including an image of the first rearranged image after a single Fourier transform and an image of the second rearranged image after a single Fourier transform; Setting a ramp filter according to the selected reconstruction width, performing a one-dimensional Fourier transform on the ramp filter and superimposing a window function, filtering each column of the primary transformed image, and performing a one-dimensional inverse Fourier transform on each column to obtain a secondary transformed image, wherein the secondary transformed image includes an image of the first rearranged image after a secondary Fourier transform and an image of the second rearranged image after a secondary Fourier transform; Back-projecting the secondary transformed image to the space where the point set is located according to the geometric projection environment to obtain a first filtered back-projection image and a second filtered back-projection image respectively; The first filtered back-projection image and the second filtered back-projection image are weightedly superimposed to obtain a final reconstructed tomographic strip image.

8. The tomographic reconstruction method based on dual-plane long-stroke scanning according to claim 1, characterized in that: The obtaining of the overlapping portion of the reconstructed tomographic images in the vertical direction and stitching the overlapping portion using a weighted superposition method to form a long-stroke image comprises: For height h i The tomographic image ff reconstructed at width w i (n), its adjacent height h i-1 With h i+1 Tomographic image ff with width w i-1 (n) with ff i+1 (n), ff i (n) with ff i-1 (n), ff i (n) with ff i+1 The overlapping height of (n) is d, and the weighted superposition function W(n) is used: Where f(n) is an arbitrary strictly monotonically increasing function defined on [0,1] with a range of [0,1] and a derivative of 0 at n=0,1; For ff i-1 (n), ff i (n) with ff i+1 (n) then: ， Among them, ff i-1 (n) is at height h i-1 The reconstructed image under ff i+1 (n) is at height h i+1 The reconstructed image below.

9. A tomographic reconstruction device based on dual-plane long-stroke scanning, characterized in that: Applicable to a dual-plane X-ray system, the dual-plane X-ray system comprising a first X-ray machine and a second X-ray machine, The first X-ray machine includes a first radiation source for emitting X-rays toward a human body; a first detector disposed opposite to the first radiation source for receiving the X-rays after passing through the human body; The second X-ray machine includes a second radiation source for emitting X-rays toward the human body; a second detector, disposed opposite to the second radiation source, for receiving the X-rays after passing through the human body; The reconstruction device comprises: a scanning mode and geometric environment setting module, configured to set a dual-plane X-ray scanning mode and geometric environment, determine the relative positional relationship between the radiation source and the detector on the first X-ray machine and the second X-ray machine, perform a long-stroke scan, and obtain images of the object to be reconstructed at different relative positional relationships to obtain the original image; An image preprocessing module is used to preprocess the original image, obtain the grayscale value of the original image, perform logarithmic transformation to convert the grayscale value into the corresponding X-ray absorptivity, and determine the tomographic reconstruction range according to the long-stroke scanning geometric environment; The tomographic image reconstruction module is used to select the reconstruction width, traverse all reconstruction range points with the reconstruction width, and complete the reconstruction of all coronal and sagittal tomographic images through filtered back projection; The long-stroke image stitching module is used to obtain the overlapping parts of the reconstructed tomographic images in the vertical direction, and stitch the overlapping parts using the weighted superposition method to form a long-stroke image.

10. A computer-readable medium, characterized in that The readable storage medium is loaded with a computer program for implementing the tomographic reconstruction method based on dual-plane long-stroke scanning as claimed in any one of claims 1 to 8.

Images