A method and apparatus for reconstructing data, a medical imaging system and a storage medium

Abstract

Embodiments of the present application disclose a kind of reconstruction data processing method, device, medical imaging system and storage medium.The method comprises: obtaining at least two images collected to target object, wherein, at least partially overlap exists in the at least two images;According to the at least two images, splicing image is obtained by splicing;Every other preset scanning angle is scanned to the target object to obtain original projection data, and determine the truncated projection data in the original projection data;Based on the splicing image and the original projection data, extrapolation is carried out to the truncated projection data to obtain extrapolation projection data;Based on the original projection data and the extrapolation projection data, three-dimensional reconstruction is carried out.The technical scheme of the embodiment of the present application realizes the reconstruction of CBCT image more accurately.

Description

Technical Field

[0001] The present invention relates to biomedical imaging technology, and more particularly to a reconstruction data processing method, apparatus, medical imaging system and storage medium. Background Technology

[0002] In many cases requiring X-ray scanning for imaging, the object being scanned may be partially outside the field of view (FOV), such as due to improper positioning or obesity in the patient. In these situations, some of the projection data will fall outside the detector. Since only projection data within the FOV can be detected, discontinuities occur at the edges of the projection data. These discontinuities result in bright truncation artifacts at the image edges, blurring the reconstruction results of the FOV edge regions.

[0003] The main reason for the bright edges in the reconstructed image is the ramp filtering in the back projection of the classic reconstruction algorithm. When the projection data is truncated, the boundary projection data is not 0. After filtering by the filter kernel, the filtering result near the boundary is enhanced, and the value at the boundary is negative. A positive peak function with negative lobes is generated near the boundary, which causes the value at the edge of the FOV to produce a bright artifact in the back projection result.

[0004] like Figure 1 When scanning data using the cone-beam computed tomography (CBCT) system shown above, due to the ramp filtering method in the back projection, the truncation in the Z-axis direction does not need to be considered, and no truncation artifacts will be generated. Only the truncation in the X-axis direction needs to be considered during projection.

[0005] In existing technologies, methods for correcting truncation artifacts are mainly divided into two categories: The first category is based on projection consistency. This method is only applicable to CT systems that can scan at all angles. This type of method requires rearranging the projection data of the fan beam and cone beam to make them equivalent to parallel beam projection data. Then, the maximum value of the sum of the projection values ​​at each angle is found. If the projection data is less than a pre-set proportion of the maximum value, such as 90%, it is determined that the projection data is truncated. If the projection data is truncated, bilinear interpolation is performed using the untruncated projection data of the adjacent angles of the truncated projection data to calculate the sum of the projection data and the value of the missing data. Then, by assuming that the missing part is composed of cylindrical water, and the position and radius of the cylinder are determined by the value of the truncated data and the slope, the projection value of the cylindrical water is used to fit the truncated projection data. The fitted value is then compared with the actual missing value to further correct the derived projection value of the truncated data. The second type involves smooth truncation of the edge, without considering any consistency conditions. Common methods include the symmetric mirror method, the water column extrapolation method, and the linear extrapolation method. The symmetric mirror method is representative. It also requires rearranging the projection data of the fan and cone beams to make them equivalent to parallel beam data. A pre-defined position is sought where the projection value reaches twice the boundary value, and the distance between this position and the boundary is set as the extrapolation width. The projection value twice the boundary value is then subtracted from all values ​​within this interval, and the result is used as supplementary data for extrapolation. The first type of algorithm requires calculating the slope at the truncation point, which is computationally complex, has poor real-time performance, and requires prior knowledge. While the second type of method may be more practical and has better real-time performance than the first type's consistency condition-based solution, the results are often not accurate enough. Summary of the Invention

[0006] This invention provides a reconstruction data processing method, apparatus, medical imaging system, and storage medium to achieve more accurate reconstruction of CBCT images.

[0007] In a first aspect, embodiments of the present invention provide a reconstruction data processing method, the method comprising:

[0008] Acquire at least two images of the target object, wherein the at least two images have at least partial overlap;

[0009] A stitched image is obtained by stitching together at least two images;

[0010] The target object is scanned at preset scanning angles to obtain original projection data, and truncated projection data in the original projection data is determined.

[0011] Based on the stitched image and the original projection data, extrapolation is performed on the truncated projection data to obtain extrapolated projection data.

[0012] Based on the original projection data and the extrapolated projection data, a three-dimensional reconstruction is performed.

[0013] Secondly, embodiments of the present invention also provide a reconstruction data processing apparatus, the apparatus comprising:

[0014] The acquisition module is used to acquire at least two images of a target object, wherein the at least two images have at least partial overlap;

[0015] The image stitching acquisition module is used to stitch together the at least two images to obtain a stitched image;

[0016] The truncated projection data acquisition module is used to scan the target object at preset scanning angles to obtain the original projection data, and to determine the truncated projection data in the original projection data.

[0017] The extrapolation projection data acquisition module is used to extrapolate the truncated projection data based on the stitched image and the original projection data to obtain extrapolated projection data;

[0018] The reconstruction module is used to perform three-dimensional reconstruction based on the original projection data and the extrapolated projection data.

[0019] Thirdly, embodiments of the present invention also provide a medical imaging system, the system comprising:

[0020] The device includes an X-ray tube, a detector, and a reconstruction data processing apparatus, wherein the reconstruction data processing apparatus is used to execute any of the reconstruction data processing methods described in the embodiments of the present invention.

[0021] Fourthly, embodiments of the present invention also provide a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements any of the reconstruction data processing methods described in the embodiments of the present invention.

[0022] The technical solution of this invention acquires at least two images of a target object, wherein the at least two images have at least partial overlap. By stitching the at least two images together to obtain a stitched image, the total projection image of the target object can be obtained, thus obtaining prior conditions for extrapolation. The target object is scanned at preset scanning angles to obtain original projection data, and truncated projection data is determined from the original projection data, identifying the truncated projection data. Based on the stitched image and the original projection data, the truncated projection data is extrapolated to obtain extrapolated projection data. Extrapolation based on the obtained prior conditions can accurately correct truncation artifacts. Based on the original projection data and the extrapolated projection data, three-dimensional reconstruction is performed, thereby realizing the reconstruction of medical images. The above technical solution solves the problems of traditional techniques requiring calculation of the slope at the truncated point, which is computationally complex, has poor real-time performance, and results in inaccurate extrapolation results. It achieves smoother extrapolated data and more accurate correction of truncated data without calculating the slope at the truncated point. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of CBCT projection in the prior art;

[0024] Figure 2a This is a flowchart of a reconstruction data processing method provided in Embodiment 1 of the present invention;

[0025] Figure 2b This is a schematic diagram of a flat scan acquisition process provided in Embodiment 1 of the present invention;

[0026] Figure 2c This is a schematic diagram of a scanning and acquisition process provided in Embodiment 1 of the present invention;

[0027] Figure 3 This is a flowchart of a data reconstruction processing method provided in Embodiment 2 of the present invention;

[0028] Figure 4 This is a schematic diagram of the structure of a reconstruction data processing device provided in Embodiment 3 of the present invention;

[0029] Figure 5 This is a schematic diagram of the structure of a medical imaging system provided in Embodiment 4 of the present invention. Detailed Implementation

[0030] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, the accompanying drawings show only the parts relevant to the present invention, and not all of the structures.

[0031] It should also be noted that, for ease of description, the accompanying drawings show only the parts relevant to the invention and not all of them. Before discussing exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations (or steps) as sequential processes, many of the operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but it may also have additional steps not included in the drawings. The process may correspond to a method, function, procedure, subroutine, subprogram, etc.

[0032] Example 1

[0033] Figure 2a The flowchart below provides a reconstruction data processing method according to Embodiment 1 of the present invention. This embodiment is applicable to medical imaging, particularly to medical imaging using CBCT equipment. The method can be executed by a reconstruction data processing device, which can be implemented in hardware and / or software and integrated into the device (e.g., a CBCT imaging system). Specifically, it includes the following steps:

[0034] Step 101: Acquire at least two images of the target object, wherein the at least two images have at least partial overlap.

[0035] The target object can be a patient, etc. Optionally, at least two images of the target object can be acquired by performing a planar scan in the X-axis direction (i.e., parallel to the detector). The at least two images must have at least partial overlap, meaning they include the same area of ​​the target object; using at least two images allows for the acquisition of a complete projected image of the target object. It should be noted that the C-arm and the target object need to be positioned before image acquisition.

[0036] Optionally, the stitched image can be obtained by having the detector move in an arc around the X-ray tube while the tube remains stationary to acquire a complete projection image of the target. However, the method for obtaining a complete projection image of the target object in this step is not limited to this.

[0037] Step 102: Stitch together the at least two images to obtain a stitched image.

[0038] In one embodiment, the initialization parameters of the medical device are first set, such as inputting parameters like SAD (distance from the X-ray source to the rotation center) and SDD (vertical distance from the X-ray source to the detector). Then, at least two images of the target object are acquired at the initial position by a horizontal scan along a first direction. The image regions of these at least two images partially overlap, allowing all information of the target object to be acquired in the X-direction. The process then returns to the initial position. The horizontal scan acquisition process is as follows: Figure 2bAs shown; based on the mechanical feedback displacement Δx of the X-axis translation and the input parameters, feature point extraction and matching are calculated, and then image registration and fusion are performed to generate a stitched image, where the stitched image is a complete projection image of the target object.

[0039] Step 103: Scan the target object at preset scanning angles to obtain original projection data, and determine the truncated projection data in the original projection data.

[0040] Raw projection data can be obtained using CT or CBCT equipment. The CT or CBCT equipment includes an imaging component, which comprises a radiation source and a detector. The radiation source is typically an X-ray tube. The object to be imaged is placed between the X-ray tube and the detector. X-rays emitted from the X-ray tube pass through the object, and the detector receives the X-rays and forms projection data. Generally, at a certain imaging angle, if the outline of the object is not completely within the field of view (FOV) of the radiation source, truncation of the projection data may be observed. Truncation of the projection data may occur on one side or both sides. Here, CT equipment refers to equipment that uses a fan-beam computed tomography (CT) and CBCT equipment refers to equipment that uses a cone-beam computed tomography (CBCT), which may include, but is not limited to, mobile C-arm equipment and digital subtraction angiography (DSA) equipment.

[0041] The truncated projection data is the original projection data that has been truncated.

[0042] Optionally, the scanning equipment may include a mobile C-arm or a digital subtraction angiography (DSA) device.

[0043] Specifically, the target object is scanned at equal intervals of 180° + θ, where θ represents the sector angle. 180° + θ represents the scanning angle. The preset scanning angle interval can be 0.5°, 1°, etc., and can be set according to actual needs. A scan is performed every preset scanning angle interval to obtain the projection data for each scanning angle. The original projection data at each angle is then checked to determine if truncation has occurred. A schematic diagram of the scanning acquisition process is shown below. Figure 2c As shown.

[0044] Optionally, determining the truncated projection data in the original projection data includes:

[0045] The truncated projection data in the original projection data is determined according to each scanning angle.

[0046] At each scanning angle, it is determined whether the original projection data at each scanning angle is truncated, and the corresponding truncated projection data is determined according to the truncation determination method.

[0047] Optionally, determining the truncated projection data in the original projection data includes:

[0048] Determine whether the original projection data has been truncated line by line;

[0049] The original projection data is judged line by line. When it is determined that the original projection data of the current line is truncated, the projection data within a preset range of the original projection data of the current line is determined as truncated projection data.

[0050] Optionally, determining the truncated projection data in the original projection data according to each scanning angle includes:

[0051] Determine line by line whether the projection data within a preset range corresponding to each scanning angle of the original projection data is truncated;

[0052] If the projection value of the projection data within the preset range of the current row is not equal to 0, and the average of the projection values ​​of a preset number of projection data in the row direction adjacent to the preset range is greater than the set projection threshold, then the projection data within the preset range of the current row is determined as truncated projection data.

[0053] Optionally, the projection data within the preset range can be the projection data near the boundary of the original projection data, that is, the range close to the first and last columns of the original projection data.

[0054] Optionally, if the projection value of the projection data at the boundary (boundary point) of the current row is not equal to 0, and the average of the projection values ​​of a preset number of projection data in the row direction adjacent to the boundary is greater than a set projection threshold, then the projection data within a preset range (projection data adjacent to the boundary) of the current row is determined to be truncated projection data. Alternatively, if the projection value of the projection data at the boundary (boundary point) of the current row is not equal to 0, and the average of the projection values ​​of a preset number of projection data in the row direction adjacent to the boundary is greater than a set projection threshold, then the current row data is considered to have been truncated, and subsequent extrapolation of the current row data can be performed based on the extrapolation width.

[0055] For example, the 2D projection data is processed row by row, and it is determined whether truncation has occurred based on the left and right boundaries. The determination method is as follows:

[0056] If the projected value of the projected data at the boundary (including both left and right boundaries, where the left boundary could be the first column of the projected data and the right boundary could be the last column of the projected data) is not equal to 0, and the average of the projected values ​​of a preset number (which can be m) of the nearby projected data along the row direction at the boundary is greater than a set projection threshold, then the boundary is considered to be truncated, and the row of data is considered to be truncated. Subsequent extrapolation of the row of data is then performed based on the extrapolation width. It can be understood that, to determine whether truncation has occurred at the left boundary, for each row of data, m (e.g., m = 5) of the nearest points at the left boundary along the row direction are taken. For example, the weighted average of the projected values ​​of the first to fifth projected data points in the current row (i.e., m = 5) is taken. Similarly, for the right boundary, the weighted average of the projected values ​​of the last nearest column of the projected data along the row direction in the current row can be taken.

[0057] Optionally, when calculating the mean near the boundary, a weighted average can be calculated using m projection points at the boundary.

[0058]

[0059] Where m can be 5 to 15, for example, 10; P(i,j) represents the projected data in the i-th row and j-th column; the formula is the left boundary projection truncation judgment formula. The right boundary projection truncation judgment formula is similar, only requiring m and j to be adjusted to the corresponding positions of the right boundary.

[0060] Furthermore, a projection threshold K is set. Taking X-rays passing through a water phantom as an example (since the human body can be approximated as water), the water attenuation coefficient is set to 0.02. It is assumed that the attenuation value of X-rays passing through 5 to 10 mm of water is 0.1 to 0.2. For example, 0.15 is taken as the threshold, i.e., when...

[0061]

[0062] Then it is determined that a truncation occurs at the left boundary. The derivation of the right boundary is similar; the first column in the formula needs to be adjusted to the last column. For example, if the last column is 1024, then the formula for the right boundary derivation is:

[0063]

[0064] Step 104: Based on the stitched image and the original projection data, extrapolate the truncated projection data to obtain extrapolated projection data.

[0065] Extrapolation is a method of calculating approximate values ​​of the same object outside the observation range based on a set of observations.

[0066] In one embodiment, the generated stitched image can be calculated row by row for pixels and T. i(i = 1, ..., j, where j is the detector row number) serves as the extrapolation standard for the truncated projection data. Calculations can be performed based on the original projection data of the stitched image, and then the extrapolated projection data can be obtained by extrapolating the truncated projection data using the extrapolation function.

[0067] Step 105: Perform three-dimensional reconstruction based on the original projection data and the extrapolated projection data.

[0068] Optionally, the reconstruction method may include a filtered back-projection reconstruction method.

[0069] Filtered backprojection reconstruction is a commonly used algorithm for medical image reconstruction. The untrunculated projection data remains unchanged and is combined with the extrapolated projection data to form the complete projection data for filtered backprojection reconstruction, resulting in the reconstructed data.

[0070] It is understood that, based on CBCT scanning, the embodiments of the present invention can obtain a stitched image of the target object (for example, by acquiring and stitching accurate and uninterrupted total projection data (i.e., stitched image) through pre-scanning) and use this as a priori condition to evaluate the degree of truncation of the data acquired at each angle and to calculate the extrapolation width, thereby achieving the purpose of accurate truncation artifact correction.

[0071] Optionally, the step of performing three-dimensional reconstruction based on the original projection data and the extrapolated projection data includes:

[0072] Three-dimensional reconstruction is performed based on the extrapolated projection data and the projection data in the original projection data excluding the truncated projection data.

[0073] The technical solution of this invention acquires at least two images of a target object, wherein the at least two images have at least partial overlap. By stitching the at least two images together to obtain a stitched image, the total projected image of the target object can be obtained, thus obtaining prior conditions for extrapolation. The target object is scanned at preset scanning angles to obtain original projection data, and truncated projection data is determined from the original projection data, identifying the truncated projection data. Based on the stitched image and the original projection data, the truncated projection data is extrapolated to obtain extrapolated projection data. Extrapolation based on the obtained prior conditions can accurately correct truncation artifacts. Based on the original projection data and the extrapolated projection data, three-dimensional reconstruction is performed, thereby realizing the reconstruction of medical images. The above technical solution solves the problems of traditional techniques requiring calculation of the slope at the truncated point, which is computationally complex, has poor real-time performance, and results in inaccurate extrapolation. It achieves smoother extrapolated data and more accurate correction of truncated data without calculating the slope at the truncated point.

[0074] Example 2

[0075] Figure 3 This is a flowchart of a reconstruction data processing method provided in Embodiment 2 of the present invention. Based on the above embodiments, this embodiment optionally includes the step of extrapolating the truncated projection data based on the stitched image and the original projection data to obtain extrapolated projection data, including:

[0076] The sum of the first pixels of all pixels in each row of the stitched image is calculated row by row, and each row in the original projection data corresponds to each row in the stitched image.

[0077] Calculate the second pixel sum of all pixels in each row of the original projection data, row by row;

[0078] The pre-fill amount is obtained by subtracting the sum of the first pixel and the sum of the second pixel in the corresponding row.

[0079] Extrapolated projection data is obtained by extrapolating the truncated projection data based on the pre-fill amount and the target extrapolation function.

[0080] Based on this, further, the step of extrapolating the truncated projection data according to the pre-fill amount and the target extrapolation function to obtain the extrapolated projection data includes:

[0081] The extrapolation width is determined based on the pre-fill amount and the initial extrapolation function;

[0082] The target extrapolation function is determined based on the extrapolation width;

[0083] Extrapolated projection data is obtained by extrapolating the truncated projection data according to the target extrapolation function.

[0084] Based on this, the extrapolation width is further determined according to the pre-fill amount and the initial extrapolation function, including:

[0085] The extrapolation width is determined based on the following formula:

[0086]

[0087] Where f(x) represents the initial extrapolation function, x represents the variable of f(x), and ΔP i N represents the pre-fill amount for the current row. ext Indicates the extrapolated width.

[0088] like Figure 3 As shown, the method of this embodiment of the invention may specifically include the following steps:

[0089] Step 201: Acquire at least two images of the target object, wherein the at least two images have at least partial overlap.

[0090] Step 202: Stitch together the at least two images to obtain a stitched image.

[0091] Step 203: Scan the target object at preset scanning angles to obtain original projection data, and determine the truncated projection data in the original projection data.

[0092] Step 204: Calculate the sum of the first pixels of all pixels in each row of the stitched image.

[0093] Specifically, it is necessary to calculate the pixel count and the sum of pixels and T for each row in the stitched image. i (i = 1, ..., j, where j is the number of detector rows), where i represents the specific row number. Since each row can obtain the corresponding pixel sum, a vector is obtained.

[0094] Step 205: Calculate the second pixel sum of all pixels in each row of the original projection data, where each row of the original projection data corresponds to each row in the stitched image.

[0095] Specifically, it is necessary to calculate the pixel sum for each pixel in the original projection data collected by scanning, row by row. Since the corresponding pixel sum can be obtained for each row, a vector is obtained.

[0096] Step 206: Subtract the sum of the first pixel and the sum of the second pixel in the corresponding row to obtain the pre-fill amount.

[0097] Step 207: Extrapolate the truncated projection data according to the pre-fill amount and the target extrapolation function to obtain extrapolated projection data.

[0098] In this process, the pixel and P values ​​are calculated line by line for the truncated projection image. i Its value is the same as T in step 204. i Subtract the values ​​to obtain the prefill amount ΔP i That is, ΔP i =T i -P i , where i = 1, ..., m, m is the detector row number, and i indicates which row it is. The pre-fill amount is used as prior knowledge and as the extrapolation standard.

[0099] Optionally, the step of extrapolating the truncated projection data according to the pre-fill amount and the target extrapolation function to obtain extrapolated projection data includes: determining the extrapolation width according to the pre-fill amount and the initial extrapolation function; determining the target extrapolation function according to the extrapolation width; and extrapolating the truncated projection data according to the target extrapolation function to obtain extrapolated projection data.

[0100] Optionally, determining the extrapolation width based on the pre-fill amount and the initial extrapolation function includes:

[0101] The extrapolation width is determined based on the following formula:

[0102]

[0103] wherein, f(x) represents the initial extrapolation function, x represents the variable of f(x), ΔP i represents the pre-filling amount of the current row, N ext represents the extrapolation width, wherein i=1,…,j, j is the number of detector rows, and i represents the specific row number.

[0104] Wherein, the initial extrapolation function is not limited herein, for example, it may include first-order straight lines, second-order curves, sine and cosine curves, log curves, etc., and the initial extrapolation function may contain unknown parameters. The target extrapolation function refers to a function used for extrapolating projection data, which is a function obtained by solving the unknown parameters of the initial extrapolation function, for example, it may include first-order functions, second-order functions, etc.

[0105] Taking a first-order straight line as the initial extrapolation function as an example, the process of extrapolating projection data at the right boundary of a certain row is described. Let the detector size be 1416*1416, and let the projection data of a certain row be Y(n), wherein 0<n≤1416, calculate the extrapolation width ΔP i =Y(1416)*l / 2, obtain the extrapolation width l, calculate the slope a=Y(1416) / l of the first-order straight line according to the extrapolation width, so as to determine the target extrapolation function, and extrapolate the truncated projection data according to the determined target extrapolation function f(x).

[0106] Optionally, when it is determined that the original projection data at a certain scanning angle (for example, 20 degrees) is truncated, it can be determined which specific rows at this angle are truncated. For example, if the first row is truncated, the pixel sum of this row can be calculated, and then the pre-filling amount ΔP can be calculated by comparing with the pixel sum of the first row of the corresponding stitched image i , and then extrapolate the truncated projection data of this row according to the pre-filling amount.

[0107] It should be noted that the method of this embodiment sequentially determines whether the projection data is truncated at each scanning angle. If truncation occurs, the missing amount is calculated by comparing the projection integral value at this angle with the total projection value of the corresponding stitched image, then the extrapolation width is calculated according to the preset extrapolation function, and extrapolation filling is performed.

[0108] Then extrapolated projection data p′(i,j) is obtained,

[0109] p′(i,j)=Y(1416)-Y(1416)*a(1416-j),j∈(1417,1416+l), which is used for the subsequent filtered back projection reconstruction step.

[0110] Step 208: Perform three-dimensional reconstruction based on the original projection data and the extrapolated projection data.

[0111] The technical solution of this invention calculates the first pixel sum of all pixels in each row of the stitched image; calculates the second pixel sum of all pixels in each row of the original projection data; subtracts the first pixel sum from the second pixel sum of the corresponding row to obtain the pre-fill amount; and extrapolates the truncated projection data according to the pre-fill amount and the target extrapolation function to obtain extrapolated projection data. The pre-fill amount can be used as a priori condition to evaluate the degree of truncation and the extrapolation width of the data acquired at each scanning angle, thereby achieving the purpose of accurate truncation artifact correction.

[0112] Example 3

[0113] Figure 4 This is a schematic diagram of a reconstruction data processing device provided in Embodiment 3 of the present invention. The reconstruction data processing device provided in this embodiment of the present invention can execute the reconstruction data processing method provided in any embodiment of the present invention. The specific structure of the device is as follows: acquisition module 31, stitched image acquisition module 32, truncated projection data acquisition module 33, extrapolated projection data acquisition module 34, and reconstruction module 35.

[0114] Acquisition module 31 is used to acquire at least two images of a target object, wherein the at least two images have at least partial overlap;

[0115] The image stitching acquisition module 32 is used to stitch together the at least two images to obtain a stitched image;

[0116] The truncated projection data acquisition module 33 is used to scan the target object at preset scanning angles to obtain the original projection data, and to determine the truncated projection data in the original projection data.

[0117] The extrapolation projection data acquisition module 34 is used to extrapolate the truncated projection data based on the stitched image and the original projection data to obtain extrapolated projection data.

[0118] The reconstruction module 35 is used to perform three-dimensional reconstruction based on the original projection data and the extrapolated projection data.

[0119] The technical solution of this invention acquires at least two images of a target object, wherein the at least two images have at least partial overlap. By stitching the at least two images together to obtain a stitched image, the total projection image of the target object can be obtained, thus obtaining prior conditions for extrapolation. The target object is scanned at preset scanning angles to obtain original projection data, and truncated projection data is determined from the original projection data, identifying the truncated projection data. Based on the stitched image and the original projection data, the truncated projection data is extrapolated to obtain extrapolated projection data. Extrapolation based on the obtained prior conditions can accurately correct truncation artifacts. Based on the original projection data and the extrapolated projection data, three-dimensional reconstruction is performed, thereby realizing the reconstruction of medical images. The above technical solution solves the problems of traditional techniques requiring calculation of the slope at the truncated point, which is computationally complex, has poor real-time performance, and results in inaccurate extrapolation results. It achieves smoother extrapolated data and more accurate correction of truncated data without calculating the slope at the truncated point.

[0120] For example, the scanning equipment includes a mobile C-arm or a digital subtraction angiography (DSA) device.

[0121] For example, the reconstruction method includes a filtered back projection reconstruction method.

[0122] Optionally, the rebuild module can be used specifically for:

[0123] Three-dimensional reconstruction is performed based on the extrapolated projection data and the projection data in the original projection data excluding the truncated projection data.

[0124] Optionally, the extrapolation projection data acquisition module can be used specifically for:

[0125] Calculate the sum of the first pixels of all pixels in each row of the stitched image, row by row.

[0126] The second pixel sum of all pixels in each row of the original projection data is calculated row by row, and each row of the original projection data corresponds to each row in the stitched image;

[0127] The pre-fill amount is obtained by subtracting the sum of the first pixel and the sum of the second pixel in the corresponding row.

[0128] Extrapolated projection data is obtained by extrapolating the truncated projection data based on the pre-fill amount and the target extrapolation function.

[0129] Optionally, the extrapolation projection data acquisition module can be used specifically for:

[0130] The extrapolation width is determined based on the pre-fill amount and the initial extrapolation function;

[0131] The target extrapolation function is determined based on the extrapolation width;

[0132] Extrapolated projection data is obtained by extrapolating the truncated projection data according to the target extrapolation function.

[0133] Optionally, the extrapolation projection data acquisition module can be used specifically for:

[0134] The extrapolation width is determined based on the following formula:

[0135]

[0136] Where f(x) represents the initial extrapolation function, x represents the variable of f(x), and ΔP i N represents the pre-fill amount for the current row. ext Indicates the extrapolated width.

[0137] Optionally, the truncated projection data acquisition module can be used specifically for:

[0138] The truncated projection data in the original projection data is determined according to each scanning angle.

[0139] Optionally, the truncated projection data acquisition module can be used specifically for:

[0140] Determine line by line whether the projection data within a preset range corresponding to each scanning angle of the original projection data is truncated;

[0141] If the projection value of the projection data within the preset range of the current row is not equal to 0, and the average of the projection values ​​of a preset number of projection data in the row direction adjacent to the preset range is greater than the set projection threshold, then the projection data within the preset range of the current row is determined as truncated projection data.

[0142] The reconstruction data processing apparatus provided in the embodiments of the present invention can execute the reconstruction data processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of executing the method.

[0143] Example 4

[0144] Figure 5 This is a schematic diagram of a medical imaging system provided in Embodiment 4 of the present invention. The medical imaging system provided in this embodiment of the present invention can execute the reconstruction data processing method provided in any embodiment of the present invention. The specific structure of the medical imaging system is as follows: X-ray tube 41, detector 42, and reconstruction data processing device 43.

[0145] The reconstruction data processing device 43 is used to execute any of the reconstruction data processing methods described in the embodiments of the present invention.

[0146] Optionally, the system in this embodiment is particularly suitable for CBCT systems.

[0147] The technical solution of this invention acquires at least two images of a target object, wherein the at least two images have at least partial overlap. By stitching the at least two images together to obtain a stitched image, the total projection image of the target object can be obtained, thus obtaining prior conditions for extrapolation. The target object is scanned at preset scanning angles to obtain original projection data, and truncated projection data is determined from the original projection data, identifying the truncated projection data. Based on the stitched image and the original projection data, the truncated projection data is extrapolated to obtain extrapolated projection data. Extrapolation based on the obtained prior conditions can accurately correct truncation artifacts. Based on the original projection data and the extrapolated projection data, three-dimensional reconstruction is performed, thereby realizing the reconstruction of medical images. The above technical solution solves the problems of traditional techniques requiring calculation of the slope at the truncated point, which is computationally complex, has poor real-time performance, and results in inaccurate extrapolation results. It achieves smoother extrapolated data and more accurate correction of truncated data without calculating the slope at the truncated point.

[0148] Example 5

[0149] Embodiment 5 of the present invention also provides a storage medium containing computer-executable instructions, which, when executed by a computer processor, are used to perform a data reconstruction processing method, the method comprising:

[0150] Acquire at least two images of the target object, wherein the at least two images have at least partial overlap;

[0151] A stitched image is obtained by stitching together at least two images;

[0152] The target object is scanned at preset scanning angles to obtain original projection data, and truncated projection data in the original projection data is determined.

[0153] Based on the stitched image and the original projection data, extrapolation is performed on the truncated projection data to obtain extrapolated projection data.

[0154] Based on the original projection data and the extrapolated projection data, a three-dimensional reconstruction is performed.

[0155] Of course, the computer-executable instructions provided in the embodiments of the present invention are not limited to the method operations described above, but can also perform related operations in the reconstruction data processing method provided in any embodiment of the present invention.

[0156] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0157] It is worth noting that in the above embodiments of the data reconstruction processing device, the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy differentiation and are not used to limit the scope of protection of the present invention.

[0158] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for reconstructing data processing, characterized in that, include: Acquire at least two images of the target object; wherein the at least two images have at least partial overlap; A stitched image is obtained by stitching together at least two images; The target object is scanned at preset scanning angles to obtain original projection data, and truncated projection data in the original projection data is determined. Calculate the sum of the first pixels of all pixels in each row of the stitched image, row by row. The second pixel sum of all pixels in each row of the original projection data is calculated row by row; wherein each row of the original projection data corresponds to each row in the stitched image; The pre-fill amount is obtained by subtracting the sum of the first pixel and the sum of the second pixel in the corresponding row; wherein, the pre-fill amount is used to evaluate the degree of truncation and extrapolation width of the original projection data acquired at each scanning angle; The extrapolation width is determined based on the pre-fill amount and the initial extrapolation function; the unknown parameters contained in the initial extrapolation function are calculated based on the extrapolation width to determine the target extrapolation function; the truncated projection data is extrapolated based on the target extrapolation function to obtain the extrapolated projection data; the target extrapolation function includes a first-order function and a second-order function, which are functions that calculate the unknown parameters of the initial extrapolation function; Based on the original projection data and the extrapolated projection data, a three-dimensional reconstruction is performed; The step of determining the truncated projection data in the original projection data includes: Determine whether the original projection data has been truncated line by line; When it is determined that the original projection data of the current row is truncated, the projection data within a preset range of the original projection data of the current row is determined as truncated projection data; wherein, the projection data within the preset range is the projection data within the range closest to the first and last columns of the original projection data.

2. The method according to claim 1, characterized in that, The scanning equipment used includes a mobile C-arm or a digital subtraction angiography (DSA) device.

3. The method according to claim 1, characterized in that, The reconstruction method includes the filtered back projection reconstruction method.

4. The method according to claim 1, characterized in that, The three-dimensional reconstruction based on the original projection data and the extrapolated projection data includes: Three-dimensional reconstruction is performed based on the extrapolated projection data and the projection data in the original projection data excluding the truncated projection data.

5. The method according to claim 1, characterized in that, The step of determining the extrapolation width based on the pre-fill amount and the initial extrapolation function includes: The extrapolation width is determined based on the following formula: ， Where f(x) represents the initial extrapolation function, and x represents the variable of f(x). N represents the pre-fill amount for the current row. ext This indicates the extrapolated width.

6. The method according to claim 1, characterized in that, Determining the truncated projection data in the original projection data includes: The truncated projection data in the original projection data is determined according to each scanning angle.

7. The method according to claim 6, characterized in that, The step of determining the truncated projection data in the original projection data according to each scanning angle includes: Determine line by line whether the projection data within a preset range corresponding to each scanning angle of the original projection data is truncated; If the projection value of the projection data within the preset range of the current row is not equal to 0, and the average of the projection values ​​of a preset number of projection data in the row direction adjacent to the preset range is greater than the set projection threshold, then the projection data within the preset range of the current row is determined as truncated projection data.

8. A data reconstruction processing apparatus, characterized in that, include: The acquisition module is used to acquire at least two images of a target object; wherein the at least two images have at least partial overlap. The image stitching acquisition module is used to stitch together the at least two images to obtain a stitched image; The truncated projection data acquisition module is used to scan the target object at preset scanning angles to obtain the original projection data, and to determine the truncated projection data in the original projection data. The extrapolation projection data acquisition module is used for: Calculate the sum of the first pixels of all pixels in each row of the stitched image, row by row. The second pixel sum of all pixels in each row of the original projection data is calculated row by row; wherein each row of the original projection data corresponds to each row in the stitched image; The pre-fill amount is obtained by subtracting the sum of the first pixel and the sum of the second pixel in the corresponding row; wherein, the pre-fill amount is used to evaluate the degree of truncation and extrapolation width of the original projection data acquired at each scanning angle; The extrapolation width is determined based on the pre-fill amount and the initial extrapolation function; the unknown parameters contained in the initial extrapolation function are calculated based on the extrapolation width to determine the target extrapolation function; the truncated projection data is extrapolated based on the target extrapolation function to obtain the extrapolated projection data; the target extrapolation function includes a first-order function and a second-order function, which are functions that calculate the unknown parameters of the initial extrapolation function; The reconstruction module is used to perform three-dimensional reconstruction based on the original projection data and the extrapolated projection data; The step of determining the truncated projection data in the original projection data includes: Determine whether the original projection data has been truncated line by line; When it is determined that the original projection data of the current row is truncated, the projection data within a preset range of the original projection data of the current row is determined as truncated projection data; wherein, the projection data within the preset range is the projection data within the range closest to the first and last columns of the original projection data.

9. A medical imaging system, characterized in that, The system includes: The X-ray tube, the detector, and the reconstruction data processing device, wherein the reconstruction data processing device is used to perform the reconstruction data processing method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the reconstruction data processing method as described in any one of claims 1-7.

Images