Flat panel detector signal crosstalk correction device and method

Abstract

The present application provides a device and method for correcting flat-panel detector signal crosstalk, the device comprising an X-ray source, a flat-panel detector, and a correction unit disposed between the flat-panel detector and the X-ray source and placed close to the flat-panel detector, the correction unit comprising an annular scale plate, a rotating assembly, and a rectangular metal sheet. The method comprises the following steps: fixing the device; determining the order of the matrix, and calculating the number of imaging times and the rotation angle; rotating the correction device sequentially according to a predetermined angle, collecting projection images, and calculating the line spread vector of the projection image; iteratively calculating the deconvolution correction matrix using the line spread vector; and removing the projection image crosstalk using the deconvolution correction matrix. The present application applies a correction device and correction method to obtain an accurate flat-panel detector PSF through simple steps to correct the flat-panel detector signal crosstalk and improve the projection image resolution.

Description

Technical Field

[0001] The present invention relates to the field of CT scanning imaging, and in particular to a device and method for correcting signal crosstalk of a cone-beam CT scanning flat panel detector. Background Art

[0002] The indirect conversion flat-panel detector (FPD) is currently the most widely used in cone-beam CT scanning systems. It consists of a scintillating screen, an image sensor, and back-end processing circuitry. Due to signal crosstalk between FPD pixels, the X-ray projection image of a point-like object is not an ideal point projection. Instead, it is a projection of a circular area covering a certain pixel radius, with the signal gradually attenuating from the center to the edge. This circular projection is the FPD's point spread function (PSF). The larger the area covered by the PSF, the more severe the signal crosstalk between FPD pixels, and the greater the gap between the FPD's actual spatial resolution and its theoretical spatial resolution. The FPD's PSF is obtained by measurement or calculation, and deconvolution of the projection image using the PSF can improve its resolution. Currently, PSF can be obtained through simulation and pinhole imaging, but both methods have significant drawbacks. Due to the difficulty in measuring the radiation absorption coefficient of the inter-pixel isolation layer of flat-panel detectors and X-ray hardening, many parameters in simulation methods are estimated, resulting in poor accuracy. Furthermore, higher X-ray energy increases the penetrating power, but as X-ray energy increases, the difficulty of fabricating the standard components for pinhole imaging increases, until it becomes unfeasible. Therefore, a highly accurate and easily implemented PSF acquisition method is urgently needed to address the challenge of flat-panel detector signal crosstalk correction. Summary of the Invention

[0003] An object of the present invention is to provide a device for correcting signal crosstalk in a flat panel detector.

[0004] The object of the present invention is achieved by such a technical solution, comprising an X-ray source and a flat panel detector, characterized in that it further comprises a correction unit arranged between the flat panel detector and the X-ray source and placed close to the flat panel detector;

[0005] The calibration unit includes an annular scale plate, a rotating assembly, and a rectangular metal sheet. The rotation center axis of the rotating assembly coincides with the central axis of the annular scale plate, and the rectangular metal sheet is fixedly mounted on the rotating assembly.

[0006] The circular ring scale is engraved with angle markings, and the rotating component can rotate 360° along the central axis of the circular ring scale. A position pointer is provided on the rotating component. The measuring edge of the rectangular metal sheet coincides with the center of the position pointer, and the rotation center of the measuring edge is located on the central axis of the circular ring scale.

[0007] An object of the present invention is to provide a method for correcting signal crosstalk in a flat panel detector.

[0008] The object of the present invention is achieved through such a technical solution, using the above-mentioned flat panel detector signal crosstalk correction device, the specific steps are as follows:

[0009] 1) Device initialization: Fix the calibration unit at the center of the flat panel detector, align the 0° position on the circular scale with the vertical or horizontal direction, and make the position pointer point to the 0° position;

[0010] 2) Parameter initialization: Preset the order N of the deconvolution correction matrix used for signal crosstalk removal, calculate the number of required measurements M and the corresponding single rotation angle a of the rotation component;

[0011] 3) Projection Image Acquisition: Rotate the rotating assembly by an angle a / 2, perform an X-ray imaging test, and obtain a first projection image of the calibration device. Then, rotate the rotating assembly by an angle a, perform an X-ray imaging test after each rotation, and obtain the second to Mth projection images.

[0012] 4) Calculate the line spread vector: Calculate the line spread vector Y of the first to Mth projection images m (i), where: (i=1,2...K, m=1,2...M);

[0013] 5) Calculate the correction matrix: use the line diffusion vector Y m (i) Obtain the deconvolution correction matrix through iterative calculation;

[0014] 6) Crosstalk correction: Deconvolution correction matrix is ​​used to deconvolve the projected image to complete the signal crosstalk correction of the flat panel detector projected image.

[0015] Furthermore, in step 2), the order N of the signal crosstalk deconvolution correction matrix is ​​an odd number, which can be preset by observing the edge grayscale change of the flat panel detector X-ray imaging image. The number of imaging times M is measured, where M is a divisor of 360 and is not less than the matrix order N. The rotation angle a is:

[0016] a=360 / M (1)

[0017] Furthermore, the specific steps of projected image acquisition in step 3) are:

[0018] 3-1) Rotating the rotating assembly by an angle a / 2, emitting an X-ray beam, and obtaining a first projection image of the correction device;

[0019] 3-2) Rotating the rotating assembly by angle a, emitting an X-ray beam, and obtaining the mth projection image of the correction device;

[0020] 3-3) Repeat step 3-2) until m=M, and obtain the second to Mth projection images.

[0021] Furthermore, the specific steps of calculating the first to M-th projection image line diffusion vectors in step 4) are:

[0022] 4-1) reversely interpolating and rotating the first to Mth projection images by an angle a / 2+ja, (j=0, 1, 2, ..., M-1), so that the first to Mth projection images are identical to the projection image when the metal sheet is at position 0;

[0023] 4-2) Select a certain area of ​​interest in the center of the image, draw multiple edge response curves along the edge of the rectangular metal sheet, and then calculate the average value to obtain the average edge response curve. The average edge response curve is derived to obtain the line diffusion response curve. Select K points at the center of the curve according to the pixel size to obtain Y m (i), (i=1,2...K,m=1,2,...M), where: K is an odd number.

[0024] Furthermore, the specific steps for calculating the deconvolution correction matrix in step 5) are:

[0025] 5-1) Set a K-order square matrix S;

[0026] 5-2) S is rotated by interpolation a / 2 to obtain the square matrix S1, then the sum of the square matrix S1 along the vertical direction is Y1(i), (i=1,2...K), S1 continues to interpolate and rotate a to obtain the square matrix S2, then the sum along the vertical direction is Y2(i), (i=1,2...K), and the sum of the matrices obtained by continuing M-2 interpolation and rotation a in the vertical direction is Y m (i), (i=1,2...K,m=3,4,...M), the point diffusion matrix is ​​obtained by iterative calculation:

[0027] S(j,k) n+1 =S(j,k) n +λ(Y n (j)-∑S(j,:) n ) / K (2)

[0028] Where S(j,k) n+1 is the point diffusion matrix obtained in the n+1th iteration, λ is the correction coefficient λ∈(0,1), Y n (j) is the nth line diffusion vector measured in step 4, S(j,:) n is the j-th row vector after the n-th iteration, where j is the row and k is the column.

[0029] 5-3) Repeat step 5-2) until the 2-norm of the difference between the deconvolution matrix before and after the iterative calculation is less than the preset value δ;

[0030] |S n+1 -S n |2<δ (3)

[0031] 5-4) Select the central N-order square matrix of the point diffusion matrix as the deconvolution correction matrix.

[0032] Furthermore, before deconvolution of the projected image using the deconvolution correction matrix in step 6), normalization processing needs to be performed so that the sum of all elements in the deconvolution correction matrix is ​​1.

[0033] Due to the adoption of the above technical solution, the present invention has the following advantages:

[0034] 1. This application designs a device specifically for measuring the point spread matrix of a flat-panel detector. It can be set at any angle to measure the signal diffusion of the flat-panel detector in any angular direction.

[0035] 2. This application compares the line diffusion curve to the projection of the back-projection correction matrix at a certain angle. Through iteration, the signal diffusion of the flat-panel detector in all directions can be accurately obtained, and the deconvolution correction matrix for signal crosstalk removal can be obtained, thereby realizing signal crosstalk correction of the projected image and improving the resolution of the projected image.

[0036] Other advantages, objectives, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objectives and other advantages of the present invention may be realized and obtained through the following description and claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings of the present invention are described below.

[0038] Figure 1 The figure is a flow chart of the flat panel detector signal crosstalk correction method of the present invention.

[0039] Figure 2 This is a top view of the correction unit in the flat panel detector signal crosstalk correction device of the present invention.

[0040] Figure 3 (a) and (b) are the reconstructed CT images before and after the projection image crosstalk correction according to the present invention, respectively.

[0041] Figure 4 (a) and (b) are the grayscale curves of the region of interest before and after the projection image crosstalk correction of the present invention, respectively.

[0042] In the figure: 1 - dial fixed component; 2 - rotating component; 3 - rectangular metal sheet; 201 - position pointer; 301 - measuring edge. DETAILED DESCRIPTION

[0043] The present invention will be further described below with reference to the accompanying drawings and examples.

[0044] Example 1:

[0045] like Figure 2 A flat panel detector signal crosstalk correction device shown in the figure includes an X-ray source (not shown in the figure) and a flat panel detector (not shown in the figure), and is characterized by further including a correction unit arranged between the flat panel detector and the X-ray source;

[0046] The calibration unit includes an annular scale plate 1, a rotating assembly 2, and a rectangular metal sheet 3. The rotation center axis of the rotating assembly 2 coincides with the central axis of the annular scale plate 1, and the rectangular metal sheet 3 is fixedly mounted on the rotating assembly 2.

[0047] The circular ring scale 1 is engraved with an angle mark 4, and the rotating component 2 can rotate 360° along the central axis of the circular ring scale 1. The rotating component 2 is provided with a position pointer 201. The measuring edge 301 of the rectangular metal sheet 3 coincides with the center of the position pointer 201, and the rotation center of the measuring edge 301 is located on the central axis of the circular ring scale 1.

[0048] In the embodiment of the present invention, the rotating assembly 2 is driven by an external force to rotate about the central axis of the annular scale plate 1. When the external force is removed, the rotating assembly 2 can be maintained at any position, thereby achieving angle adjustment. The external force can be manual or powered by a power component, which is not limited in this application.

[0049] Example 2:

[0050] like Figure 1 A flat panel detector signal crosstalk correction method is shown, characterized in that the flat panel detector signal crosstalk correction device according to claim 1 is used, and the specific steps are as follows:

[0051] 1) Device initialization: Fix the calibration unit at the center of the flat panel detector, align the 0° position on the annular scale plate 1 with the vertical or horizontal direction, and make the position pointer 201 point to the 0° position.

[0052] In the embodiment of the present invention, the 0° position on the annular scale plate 1 is initially aligned with the vertical direction.

[0053] 2) Parameter initialization: Preset the order N of the deconvolution correction matrix for signal crosstalk removal, calculate the number of required measurements M and the corresponding single rotation angle a of the rotating component 2; the specific method is:

[0054] The order N of the signal crosstalk deconvolution correction matrix is ​​an odd number and can be preset by observing the edge grayscale changes of the flat panel detector X-ray imaging image. The number of imaging times M (M is a divisor of 360) measured is not less than the matrix order N, and the rotation angle a is:

[0055] a=360 / M (4)

[0056] In the example of the present invention, the order N of the deconvolution correction matrix is ​​7, and the number of measurements M is 360.

[0057] 3) Projection Image Acquisition: Rotate the rotating assembly 2 by an angle a / 2, perform an X-ray imaging test, and obtain a first projection image of the calibration device. Then, rotate the rotating assembly 2 by an angle a, perform an X-ray imaging test after each rotation, and obtain the second to Mth projection images. The specific steps are as follows:

[0058] 3-1) Rotating the rotating assembly 2 by an angle a / 2, emitting an X-ray beam, and obtaining a first projection image of the correction device;

[0059] 3-2) Rotating the rotating assembly 2 by angle a, emitting an X-ray beam, and obtaining the mth projection image of the correction device;

[0060] 3-3) Repeat step 3-2) until m=M, and obtain the second to Mth projection images.

[0061] 4) Calculate the line spread vector: Calculate the line spread vector Y of the first to Mth projection images m (i), where: (i=1,2...K, m=1,2...M), the specific steps are:

[0062] 4-1) reversely interpolating and rotating the first to Mth projection images by an angle a / 2+ja, (j=0, 1, 2, ..., M-1), so that the first to Mth projection images are identical to the projection image when the metal sheet is at position 0;

[0063] 4-2) Select a certain area of ​​interest in the center of the image, draw multiple edge response curves along the edge of the rectangular metal sheet 301, and then calculate the average value to obtain the average edge response curve. The average edge response curve is derived to obtain the line diffusion response curve. Select K points at the center of the curve according to the pixel size to obtain Y m (i), (i=1,2...K,m=1,2,...M), where: K is an odd number.

[0064] In the example of the present invention, the value of K is 11.

[0065] 5) Calculate the correction matrix: use the line diffusion vector Y m (i) Obtain the deconvolution correction matrix through iterative calculation; the specific steps are:

[0066] 5-1) Set a K-order square matrix S;

[0067] 5-2) S is rotated by interpolation a / 2 to obtain the square matrix S1, then the sum of the square matrix S1 along the vertical direction is Y1(i), (i=1,2...K), S1 continues to interpolate and rotate a to obtain the square matrix S2, then the sum along the vertical direction is Y2(i), (i=1,2...K), and the sum of the matrices obtained by continuing M-2 interpolation and rotation a in the vertical direction is Y m (i), (i=1,2...K,m=3,4,...M), the point diffusion matrix is ​​obtained by iterative calculation:

[0068] S(j,k) n+1 =S(j,k) n +λ(Y n (j)-∑S(j,:) n ) / K (5)

[0069] Where S(j,k) n+1 is the point diffusion matrix obtained in the n+1th iteration, λ is the correction coefficient λ∈(0,1), Y n (j) is the nth line diffusion vector measured in step 4, S(j,:) n is the j-th row vector after the n-th iteration, where j is the row and k is the column.

[0070] 5-3) Repeat step 5-2) until the 2-norm of the difference between the deconvolution matrix before and after the iterative calculation is less than the preset value δ;

[0071] |S(j,:) n+1 -S(j,:) n |2<δ (6)

[0072] In the example of the present invention, the value of δ is 0.001.

[0073] 5-4) Select the central N-order square matrix of the point diffusion matrix as the deconvolution correction matrix.

[0074] In the example of the present invention, the value of N is 7.

[0075] 6) Crosstalk correction: Deconvolution correction matrix is ​​used to deconvolve the projected image to complete the signal crosstalk correction of the flat panel detector projected image.

[0076] In the embodiment of the present invention, before deconvolution is performed on the projection image using the deconvolution correction matrix, normalization processing is required to make the sum of all elements in the deconvolution correction matrix equal to 1.

[0077] like Figure 3 As shown in the figure, after the crosstalk correction of the projection image, the recognition between the metal sheet and the gap between the metal sheets on the reconstructed CT image is significantly enhanced.

[0078] like Figure 4 As shown in FIG, after the crosstalk of the projection image is corrected, the grayscale contrast between the metal pieces and the gaps between the metal pieces on the reconstructed CT image is improved, and the image spatial resolution is enhanced.

[0079] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0080] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0081] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0082] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The steps for the function specified in one or more boxes.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for correcting signal crosstalk in a flat panel detector, characterized in that: Crosstalk correction is performed using a flat panel detector signal crosstalk correction device, the flat panel detector signal crosstalk correction device comprising: an X-ray source, a flat panel detector, and a correction unit disposed between the flat panel detector and the X-ray source and placed close to the flat panel detector; The calibration unit comprises an annular scale plate (1), a rotating assembly (2), and a rectangular metal sheet (3); the rotation center axis of the rotating assembly (2) coincides with the central axis of the annular scale plate (1); and the rectangular metal sheet (3) is fixedly mounted on the rotating assembly (2); The annular scale plate (1) is engraved with an angle mark (4), the rotating assembly (2) can rotate 360 ​​degrees along the central axis of the annular scale plate (1), the rotating assembly (2) is provided with a position pointer (201), the measuring edge (301) of the rectangular metal sheet (3) coincides with the center of the position pointer (201), and the rotation center of the measuring edge (301) is located on the central axis of the annular scale plate (1). The specific steps of the crosstalk correction method are: 1) Device initialization: fix the correction unit at the center of the flat panel detector, align the 0° position on the annular scale plate (1) with the vertical direction or the horizontal direction, and make the position pointer (201) point to the 0° position; 2) Parameter initialization: presetting the order N of the deconvolution correction matrix for signal crosstalk removal, calculating the required number of measurements M and the corresponding single rotation angle a of the rotation component (2); 3) Projection image acquisition: rotating the rotating component (2) by an angle a / 2, performing an X-ray imaging test, obtaining a first projection image of the correction device, and then rotating the rotating component (2) by an angle a in sequence, performing an X-ray imaging test after each rotation, obtaining second to Mth projection images; 4) Calculate the line spread vector: Calculate the line spread vector Y of the first to Mth projection images m (i), where: (i=1,2...K, m=1,2...M); 5) Calculate the correction matrix: use the line diffusion vector Y m (i) Obtain the deconvolution correction matrix through iterative calculation; 6) Crosstalk correction: Deconvolution correction matrix is ​​used to deconvolve the projected image to complete the signal crosstalk correction of the flat panel detector projected image.

2. A flat panel detector signal crosstalk correction method according to claim 1, characterized in that: In step 2), the order N of the signal crosstalk deconvolution correction matrix is ​​an odd number and can be preset by observing the edge grayscale changes of the flat panel detector X-ray imaging image. The number of imaging times M is measured, where M is a divisor of 360 and is not less than the matrix order N. The rotation angle a is: a=360 / M (1).

3. A flat panel detector signal crosstalk correction method according to claim 1, characterized in that: The specific steps of projected image acquisition in step 3) are: 3-1) rotating the rotating assembly (2) by an angle a / 2, emitting an X-ray beam, and obtaining a first projection image of the correction device; 3-2) rotating the rotating assembly (2) by an angle a, emitting an X-ray beam, and obtaining an mth projection image of the correction device; 3-3) Repeat step 3-2) until m=M, and obtain the second to Mth projection images.

4. A flat panel detector signal crosstalk correction method according to claim 1, characterized in that: The specific steps of calculating the first to M-th projection image line diffusion vectors in step 4) are: 4-1) reversely interpolating and rotating the first to Mth projection images by an angle a / 2+ja, (j=0, 1, 2, ..., M-1), so that the first to Mth projection images are identical to the projection image when the metal sheet is at position 0; 4-2) Select a certain area of ​​interest in the center of the image, draw multiple edge response curves along the edge of the rectangular metal sheet (301) and calculate the average value to obtain the average edge response curve. The average edge response curve is derived to obtain the line diffusion response curve. Select K points at the center of the curve according to the pixel size to obtain Y m (i), (i=1,2...K,m=1,2,...M), where: K is an odd number.

5. The method for correcting flat panel detector signal crosstalk according to claim 1, wherein: The specific steps for calculating the deconvolution correction matrix in step 5) are: 5-1) Set a K-order square matrix S; 5-2) S is rotated by interpolation a / 2 to obtain the square matrix S1, then the sum of the square matrix S1 along the vertical direction is Y1(i), (i=1,2...K), S1 continues to interpolate and rotate a to obtain the square matrix S2, then the sum along the vertical direction is Y2(i), (i=1,2...K), and the sum of the matrices obtained by continuing M-2 interpolation and rotation a in the vertical direction is Y m (i), (i=1,2...K,m=3,4,...M), the point diffusion matrix is ​​obtained by iterative calculation: S(j,k) n+1 =S(j,k) n +λ(Y n (j)-∑S(j,:) n ) / K (2) Where S(j,k) n+1 is the point diffusion matrix obtained in the n+1th iteration, λ is the correction coefficient λ∈(0,1), Y n (j) is the nth line diffusion vector measured in step 4, S(j,:) n is the j-th row vector after the n-th iteration, where j is the row and k is the column. 5-3) Repeat step 5-2) until the 2-norm of the difference between the deconvolution matrix before and after the iterative calculation is less than the preset value δ; |S n+1 -S n |2<δ (3) 5-4) Select the central N-order square matrix of the point diffusion matrix as the deconvolution correction matrix.

6. The method for correcting crosstalk of a flat panel detector signal according to claim 1, wherein: Before deconvolving the projected image using the deconvolution correction matrix in step 6), normalization processing is required so that the sum of all elements in the deconvolution correction matrix is ​​1.

Images