Medical image processing method, medical image processing apparatus, and storage medium

Abstract

The embodiments relate to medical image processing methods, medical image processing apparatuses, and storage media. The aim is to provide medical image processing methods, medical image processing apparatuses, and storage media capable of improving the accuracy of corrections for defective views. The medical image processing method of the embodiments includes: acquiring scan data including multiple views collected in a computed tomography (CT) scan of a subject; determining at least one complementary X-ray for multiple X-rays corresponding to defective views not obtained in the CT scan of the subject; and reconstructing image data of the subject based on the scan data and the determined at least one complementary X-ray.

Description

[0001] This application claims priority based on U.S. Provisional Application No. 63 / 295,035, filed December 30, 2021, and U.S. Patent Application No. 17 / 723,894, filed April 19, 2022, the entire contents of which are incorporated herein by reference. Technical Field

[0002] This invention relates to medical image processing methods, medical image processing devices, and storage media.

[0003] This invention relates to a method and system (apparatus) for compensating for continuously missing views in computed tomography (CT) reconstruction, and in one embodiment, it relates to the use of complementary light for interpolating the missing views. Background Technology

[0004] In X-ray computed tomography (CT), a CT image consists of multiple views of the subject as X-rays pass through it and illuminate the X-ray detector. Typically, the X-ray irradiator and detector move in a circular motion around the subject, capturing views at various angles. Furthermore, the subject is usually horizontal, and by moving the table that holds the subject horizontally, a wider area of ​​the subject can be scanned. Helical scans of the subject are achieved through the movement of the subject as a result of the table movement and the movement of the X-ray irradiator / detector.

[0005] Here, CT scanners sometimes determine during a scan that the data associated with the X-ray detector for a corresponding view is unreliable. In such cases, the CT scanner treats the data for the unreliable view as "defective".

[0006] Here, in order to fill in the missing views mentioned above, there are many cases where interpolation is used in image reconstruction. For example, as... Figure 19A As shown, multiple channels of a CT scanner should be able to produce multiple consecutive defective views 1000 (depicted by a black bar in the center of the view). In this case, to compensate for the defective views, for example... Figure 19B As shown, interpolation is performed using correctly obtained data from adjacent views to "fill in" the missing views. However, in many cases, when there are a large number of missing views, interpolation, especially linear interpolation, is insufficient to generate a reconstructed image without artifacts.

[0007] Figure 20(A) represents the original reconstructed image generated using all views obtained from a scan of soft tissue, without any views considered as "defective". Figure 20 (B) indicates that although using Figure 20 (A) data, but a test reconstructed image generated by deleting multiple consecutive views and simulating the missing views before generating the reconstructed image. According to Figure 20 As can be seen from the circled portion in (B), the resulting image is similar to the original. Figure 20 (A) Compared to this, it has deteriorated (e.g., streaking). Figure 20 (C) indicates that although using Figure 20 (A) data, but a test reconstructed image generated by deleting multiple consecutive views and simulating the missing views. Here, in Figure 20 In (C), the view was linearly interpolated to fill in the missing view data before generating the reconstructed image. For example... Figure 20 The part enclosed by the circle in (C) is visible, and the resulting image is the same as the original. Figure 20 (A) Compared to the previous version, it has deteriorated (e.g., streaks).

[0008] Figure 21 (A) represents the original reconstructed image generated using all views obtained from a scan of one part of a human lung, without considering it as a "defective" view. Figure 21 (B) indicates that although using Figure 21 (A) data, but a test reconstructed image generated by deleting multiple consecutive views and simulating the missing views before generating the reconstructed image. According to Figure 21 As can be seen from the circled portion in (B), the resulting image is similar to the original. Figure 21 (A) Compared to the previous version, it has deteriorated (including shading). Figure 21 (C) indicates that although using Figure 21 (A) data, but a test reconstructed image generated by deleting multiple consecutive views and simulating the missing views. Here, in Figure 21 In (C), the view was linearly interpolated to fill in the missing view data before generating the reconstructed image. According to Figure 21 (C) shows that the resulting image is identical to the original. Figure 21 (A) Compared to the previous version, it has deteriorated (e.g., streaks). Summary of the Invention

[0009] The purpose of this invention is to provide a medical image processing method, a medical image processing device, and a storage medium that can improve the accuracy of correcting defective views.

[0010] The medical image processing method of the relevant technical solution includes: acquiring scan data including multiple views collected in a computed tomography (CT) scan of a subject; determining at least one complementary X-ray for multiple X-rays corresponding to defective views not obtained in the CT scan of the subject; and reconstructing image data of the subject based on the scan data and the determined at least one complementary X-ray.

[0011] Invention Effects

[0012] The medical image processing method according to the technical solution can improve the accuracy of correcting defective views. Attached Figure Description

[0013] Figure 1 This is a schematic diagram illustrating one embodiment of an X-ray CT apparatus having an X-ray source and a detector for collecting CT projection data.

[0014] Figure 2A It is a diagram that uses terms to illustrate the X-ray beam incident on the detector and the focal position, the beam's projection angle, and the angles of the X-rays within the beam.

[0015] Figure 2B This is a diagram illustrating an example of the spiral path of an X-ray source surrounding an object being imaged.

[0016] Figure 3 It is a diagram showing the directions of the first and second X-rays detected by a set of curved detectors from the X-ray source S, and the directions of the complementary X-rays (α', γ') relative to the second X-ray.

[0017] Figure 4A This is a diagram showing an X-ray tube and an X-ray detector arranged vertically so that the viewing angle α = 0 radians.

[0018] Figure 4B This indicates a vertical arrangement so that the view angle α = π radians (or distance). Figure 4A A diagram of the X-ray tube and X-ray detector (originally oriented 180 degrees).

[0019] Figure 5A It is a diagram representing a series of views, including at least one defective view 500, which shall be taken as part of the process of imaging a part of the subject.

[0020] Figure 5BIt is a diagram representing a set of images obtained as a result of replacing the missing view 500 with a view 520 that contains at least information from the preceding supplementary view 500'.

[0021] Figure 5C It is a diagram that shows the ability to correct a defective view using a preliminary view, or a combination of a preliminary view and a subsequent view, which have at least one complementary X-ray to each of the multiple X-rays missing from the defective view.

[0022] Figure 5D It is a diagram representing a replacement view calculated by a weighted combination of at least two supplementary views.

[0023] Figure 5E It represents a graph that can interpolate a missing view using an asymmetric group of supplementary views.

[0024] Figure 5F It is a diagram representing a replacement view calculated by a weighted combination of at least two supplementary views.

[0025] Figure 6A It is a diagram showing how a portion of a missing view is captured by a detector element when it has been properly photographed.

[0026] Figure 6B This indicates that interpolation is performed to reconfigure complementary X-rays to the original X-ray location. Figure 6A An illustrative group of images showing complementary X-rays corresponding to the defective areas.

[0027] Figure 7 This is a diagram representing three possible weighted groups used to combine complementary X-rays to generate alternative X-rays.

[0028] Figure 8 It is a graph representing a group of known weights that function as a sliding window used for image reconstruction.

[0029] Figure 9 This indicates that 100 views in the center are missing. Figure 8 The chart.

[0030] Figure 10 This indicates that the weights that should have been applied to the missing view are now applied to the supplementary view. Figure 8 The revised version of the chart.

[0031] Figure 11A It is an exemplary numerically weighted graph representing a subset of views to which corrected weights have been applied for the missing view.

[0032] Figure 11B It is an exemplary numerically weighted graph representing a subset of views to which corrected weights have been applied for the missing views.

[0033] Figure 11C It is an exemplary numerically weighted graph representing a subset of views to which corrected weights have been applied for the missing views.

[0034] Figure 11D It is an exemplary numerically weighted graph representing a subset of views to which corrected weights have been applied for the missing views.

[0035] Figure 12 It is a diagram representing the original reconstructed slice, the reconstructed slice with 50 consecutive defective views in the center, and the reconstructed slice after the 50 consecutive defective views in the center have been replaced by weighted supplementary views.

[0036] Figure 13 It is a diagram representing the original reconstructed slice, the reconstructed slice with 100 consecutive missing views in the center, and the reconstructed slice after the 100 consecutive missing views in the center are replaced by weighted supplementary views.

[0037] Figure 14 This refers to the case where the weights that should have been applied to the missing view are instead applied to the supplementary view, centered on the view with an angle of approximately -7θ / 8 around the missing view. Figure 8 The revised version of the chart.

[0038] Figure 15 This indicates the original reconstructed slice, and will Figure 14 A graph of an image of a reconstructed slice after replacing a continuous defective view with a weighted supplementary view.

[0039] Figure 16 It is a graph representing a group of redundant weights that are smoothed out in order to make the transition between data from a defective view and data without errors smooth.

[0040] Figure 17 This is a graph showing the effect of smoothing on the reconstructed slices obtained under the first set of scanning conditions.

[0041] Figure 18 It means about and Figure 17 The same slice, and the effect of smoothing on the reconstructed slice obtained under the second set of scanning conditions.

[0042] Figure 19A It is a diagram representing an image containing multiple consecutive missing views (depicted with black bars in the center of the views) captured on multiple channels.

[0043] Figure 19B It means Figure 19A A graph of a group of images of the view and a speculative view generated by interpolation.

[0044] Figure 20 The graph represents the original reconstructed image generated from all views obtained in a soft tissue scan, indicating the absence of views considered as "defective"; the test reconstructed image generated by deleting multiple consecutive views and simulating defective views before generating the reconstructed image; and the test reconstructed image generated by using linear interpolation after initially deleting multiple consecutive views and simulating defective views.

[0045] Figure 21 The graph represents the original reconstructed image generated from all views obtained in a scan of a human lung, indicating the absence of views considered as "defective"; the test reconstructed image generated by deleting multiple consecutive views and simulating the defective view before generating the reconstructed image; and the test reconstructed image generated by using linear interpolation after initially deleting multiple consecutive views and simulating the defective view. Detailed Implementation

[0046] As described in this specification, methods, apparatus (systems), and storage media interpolate continuous defect views in computed tomography (CT) reconstruction. Specifically, single or multiple defect views can be filled from previous or subsequent views using at least one complementary X-ray. In the presence of multiple complementary X-rays, defect views can be filled using a linear or nonlinear combination of X-rays, and the weights used in the combination can be smoothed to prevent overemphasis on the replacement views.

[0047] Here, referring to the attached diagrams, the same reference numerals indicate the same or corresponding structures across multiple diagrams. Figure 1 This describes an implementation of a radiographic gantry included in an X-ray CT apparatus or scanner. For example... Figure 1 As shown, the radiography gantry 100, viewed from the side, includes an X-ray tube 101, a ring frame 102, and a multi-row or two-dimensional array type X-ray detector 103. The X-ray tube 101 and X-ray detector 103 sandwich a subject OBJ mounted radially on the ring frame 102, which is supported and capable of rotating about a rotation axis RA. The rotation unit 107 rotates the ring frame 102 at a high speed of 0.4 seconds per revolution while the subject OBJ moves along the axis RA in the long side direction towards the inside or front of the paper.

[0048] Furthermore, there are various types of X-ray CT devices, such as rotating / rotating type where both the X-ray tube and X-ray detector rotate around the subject being examined, and fixed / rotating type where multiple detection elements are arranged in a ring or planar shape and only the X-ray tube rotates around the subject being examined. This embodiment can be applied to any type. Hereinafter, the currently mainstream rotating / rotating type is illustrated.

[0049] The multi-slice X-ray CT apparatus also includes a high-voltage generator 109 that generates a tube voltage applied to the X-ray tube 101 via a slip ring 108, causing the X-ray tube 101 to generate X-rays. The X-rays irradiate the subject OBJ, whose cross-sectional area is represented by a circle. An X-ray detector 103, clamping the subject OBJ, is positioned on the opposite side of the X-ray tube 101 to detect emitted X-rays that have passed through the subject OBJ. The X-ray detector 103 also includes individual detector elements or modules.

[0050] The X-ray CT apparatus also includes other devices for processing the detection signals from the X-ray detector 103. The data acquisition circuit or data acquisition system (DAS) 104 converts the signals output from the X-ray detector 103 of each channel into voltage signals, amplifies these signals, and then converts them into digital signals. The X-ray detector 103 and DAS 104 are configured to process a predetermined total number of projections per rotation (TPPR). Examples of TPPR include 800 TPPR, 900 TPPR, 900–1800 TPPR, and 900–3600 TPPR, but are not limited to these.

[0051] The aforementioned data is transmitted via a non-contact data transmitter 105 to a pre-processing unit 106 housed in a console located outside the radiography gantry 100. The pre-processing unit 106 performs certain corrections on the raw data, such as sensitivity correction. The memory 112 stores the resulting data, also known as projection data, in the stage preceding reconstruction processing. The memory 112, reconstruction unit 114, input unit 115, and display 116 are all connected to the system controller 110 via a data / control bus 111. The system controller 110 controls a current regulator 113 that limits the current to a level sufficient for the CT system's drive.

[0052] The detector rotates and / or is fixed relative to the patient in each generation of CT scanner systems. In one embodiment, the CT system described above can be an example of a system combining third-generation and fourth-generation geometries. In a third-generation system, the X-ray tube 101 and X-ray detector 103 are radially mounted on an annular frame 102, rotating around the subject OBJ via the annular frame 102 about the rotation axis RA. In a fourth-generation geometries, the detector is fixedly positioned around the patient, and the X-ray tube rotates around the patient. In another embodiment, the radiography gantry 100 has multiple detectors disposed on an annular frame 102 supported by a C-arm and a stand.

[0053] The memory 112 can store measurement values ​​representing the X-ray exposure dose of the X-ray detector 103. Furthermore, the memory 112 can store, for example, a dedicated program for executing the CT image reconstruction method 300 described in this specification.

[0054] The reconstruction apparatus 114 is capable of performing the CT image reconstruction method 300 described in this specification. Furthermore, the reconstruction apparatus 114 can perform image processing such as volume rendering and image difference processing as needed. The pre-reconstruction processing of projection data performed by the pre-processing apparatus 106 may include detector calibration, correction of detector nonlinearity, polarity effects, noise balancing, and material discrimination. The post-reconstruction processing performed by the reconstruction apparatus 114 may include image filtering and smoothing, volume rendering, and image difference processing as needed. Image reconstruction processing can be performed using filtered back-projection (FBP), successive approximation image reconstruction, or probabilistic image reconstruction. The reconstruction apparatus 114 can use the memory 112 to store, for example, projection data, reconstructed images, correction data and parameters, and computer programs.

[0055] The reconstruction device 114 may have processing circuitry (e.g., as discrete logic gates, capable of being implemented as a CPU in an Application Specific Integrated Circuit (ASIC), Field Programmable Gate Array (FPGA), or other Complex Programmable Logic Device (CPLD)). The FPGA or CPLD implementation can be coded using VHDL, Verilog, or other hardware description languages, or the code can be directly stored in the electronic memory within the FPGA or CPLD, or stored as other electronic memory. Furthermore, the memory 112 can be a non-volatile memory such as ROM, EPROM, EEPROM, or flash memory. The memory 112 can also be a volatile memory such as static RAM or dynamic RAM, and can be equipped with a processor such as a microcontroller or microprocessor to manage the electronic memory and the interaction between the FPGA or CPLD and the memory.

[0056] Furthermore, the processing circuitry (e.g., CPU) within the reconstruction device 114 is capable of executing a computer program containing a set of computer-readable commands that perform the functions described in this specification (e.g., acquisition function, decision function, and reconstruction function). The program is stored in one of the aforementioned non-volatile electronic memories and / or hard disk drives, CDs, DVDs, flash drives, or other known storage media. Furthermore, the computer-readable commands can also be provided as utility applications, background daemons, operating system components, or combinations thereof, executed with processors such as Intel's Xenon processors or i3, i5, i7, i9 processors, or AMD's Opteron or Ryzen processors, and operating systems such as Microsoft Vista, UNIX (registered trademark), Solaris, LINUX (registered trademark), Apple, MAC-OS, and other operating systems known to those skilled in the art. Furthermore, the CPU can execute commands in a coordinated and parallel manner for multiple processors. Additionally, the acquisition function is an example of an acquisition unit. The decision function is an example of a decision unit. The reconstruction function is an example of the reconstruction department.

[0057] In one embodiment, the reconstructed image can be displayed on a display 116. The display 116 can be an LCD display, a CRT display, a plasma display, an OLED, an LED, or other displays known in the art.

[0058] The memory 112 may be a hard disk drive, CD-ROM drive, DVD drive, flash drive, RAM, ROM, or other electronic storage device known in the art.

[0059] Figure 2A This represents the cone-shaped beam shape of the X-rays traveling from the X-ray tube 101 toward the X-ray detector 103. Typically, the projected measurement of the X-rays can be represented as a line integral expressed by the following equation (1).

[0060] [Formula 1]

[0061]

[0062] Here, f(r) represents the reconstructed specimen, R represents the radius of the helical track, H represents the helical pitch (the feed per revolution of the bed), and (β, γ, α) represent the projection angle, X-ray angle, and cone angle, respectively (refer to...). Figure 1 ), φβ, γ, α represent the unit vectors of points (γ, α) from the X-ray focus s(13) of β toward the surface of the cylindrical detector, as expressed by the following equation (2).

[0063] [Formula 2]

[0064]

[0065] Here, when β = 0, the focus is the plane of interest Z = z0 of the projection β0.

[0066] Figure 2B This illustrates an example of the helical path of the X-ray tube for imaging the subject OBJ. In this embodiment, the subject OBJ is positioned on a table, and the table translates linearly as the X-ray tube 101 and the X-ray detector 103 rotate along their respective circular paths, so that the path of the X-ray tube 101 relative to the subject OBJ traverses a helical path. Furthermore, in this embodiment, the projection angle β is correlated with the variables time t and position z using the following equation (3), such that t = β = 0 when the online source is in the slice of interest z0.

[0067] [Formula 3]

[0068] t / T rot =β / 2π=(z+Z0) / H…(3)

[0069] In one embodiment, projection data is obtained without translating the imaged subject OBJ relative to the X-ray tube 101 and the X-ray detector 103. In this scenario, since the track is circular (not spiral), the analysis is simplified by setting H=0. Figure 2A The terminology used to represent X-rays is generally applicable to all X-ray beams, including cone beams and fan beams. For fan beam shapes (e.g., parallel fan beams where X-rays diverge in one dimension and are parallel in other dimensions), setting α = 0 simplifies the analysis.

[0070] Figure 3 This indicates that the first and second X-rays, detected by a set of curved detectors, are irradiated from X-ray source S. It shows that the first X-ray 300A reaches the center of the detector at position 0, and the second X-ray 300B reaches position σ. In cases where these X-rays are unusable or part of a missing view, data regarding how the X-rays are absorbed by the object at the angles shown in the diagram is lost.

[0071] In cases where a view is unavailable or incomplete, interpolation can be performed instead, using at least one complementary X-ray from a previous or subsequent view to replace the data. For example... Figure 3 As shown, the second X-ray 300B corresponds to the channel of a detector that attempts to detect X-rays, defined by the angle pair (α, γ) where α (relative to the y-axis) is the tube angle (or view angle) and γ is the sector angle. In views that the second X-ray (α, γ) 300B cannot detect because it corresponds to a defective view, the defective view data (also referred to as "data about a single defective view" or "data about multiple defective views") can be replaced by a complementary X-ray (with direction (α', γ')) that travels along the same path as the second X-ray 300B but in a complementary direction.

[0072] like Figure 4A and Figure 4B As shown, the X-ray tube 101 and the X-ray detector 103 move with a series of views. Figure 4A In the image, the X-ray tube 101 and X-ray detector 103, arranged vertically and represented by dense shadows, are configured with a view angle α = 0 radians (as opposed to...). Figure 3 (Also using the y-axis as the reference point). Figure 4A The X-ray tube 101 and X-ray detector 103 are shown in a grey version. These versions indicate the orientation of the X-ray tube 101 and X-ray detector 103 when the preview and follow-up views are obtained. For comparison, Figure 4B Indicates and Figure 4A Compared to the X-ray tube 101 and X-ray detector 103 located on the opposite side of the subject being imaged. Figure 4B The orientation of the X-ray tube / X-ray detector structure is α = π radians (or 180 degrees from the original orientation).

[0073] In addition, Figure 4A The image shows X-rays 200C that irradiate the subject at the center of the X-ray detector 103 with a fan-shaped angle γ = 0 radians. Figure 4B This represents complementary X-ray 200C' that also irradiates the subject at a sector angle γ' = 0 radians and irradiates the center of the X-ray detector 103. Typically, X-rays with angle pairs (α, γ) and complementary X-rays with angle pairs (α', γ') satisfy the two equations shown in Equation (4) below.

[0074] [Formula 4]

[0075]

[0076] If using the above Figure 4A and Figure 4B For example, equation (4) can be expressed as equation (5) below.

[0077] [Formula 5]

[0078]

[0079] By using these equations, it is possible to process X-rays from at least one other subsequent or preceding view to obtain the missing data from the data generated for image reconstruction.

[0080] like Figure 5A As shown, the series of views includes at least one defective view 500, which, as part of the process of imaging a part of the subject, should have been captured by at least one X-ray with an angle pair (α, γ). For illustration, the part that should have been captured is indicated by the character "B" to emphasize the inverted characteristics of the captured image. By using at least one X-ray with an angle pair (α', γ') from at least one supplementary view 500' (represented as a subsequent view of the inverted image), the system is able to fill in the defect data of at least one defective view 500. Figure 5B This refers to a group of images obtained by replacing the missing view 500 with a replacement view 520 that contains at least information from the supplementary view 500'.

[0081] like Figure 5CAs shown, the defective view 500 can also be corrected using a preliminary view 500A', which has at least one complementary X-ray relative to the multiple X-rays missing from view 500, or a combination of a preliminary view 500A' and a subsequent view 500B'. For example... Figure 5D As shown, the replacement view 520 is a mask that differs from both the missing view 500 and the supplementary views 500A' and 500B'. This difference in masking is intended to indicate that the replacement view 520 can also be calculated by a weighted combination of at least two supplementary views 500A' and 500B'.

[0082] like Figure 5E As shown, the supplementary views 500A', 500B', and 500C' used to fill in the missing view 500 can be obtained from any part of the existing supplementary view, or symmetrically from the existing supplementary view. In fact, if a missing view is generated at the beginning of imaging, previous supplementary views cannot be used; only subsequent supplementary views can be used. Figure 5F This indicates that a replacement view 520 is used instead of the missing view 500. The replacement view is preferably created as a weighted combination of at least two supplementary views 500A' and 500B', and in one embodiment, a weighted combination of at least three supplementary views 500A', 500B', and 500C' is created. In such an embodiment, the weighted combination is formed by all supplementary views containing X-rays that can replace the data from the missing view 500 with the values ​​of the supplementary views, but are normalized to give all X-rays the same redundancy (e.g., by setting it to 0.5 in the case of two complementary X-rays, and by weighting it by 1 / 3 in the case of three complementary X-rays).

[0083] Those skilled in the art will understand that the complementary X-rays used to fill in the missing data from the missing view 500 may not all be in the same view, but may span multiple views depending on the degree of rotation occurring between the views. Furthermore, in the case of multiple consecutive missing views, a greater number of supplementary views are used to provide replacement X-rays for the missing view 500. Moreover, due to the amount of rotation between views, it is unlikely that the channel / detector elements will be correctly positioned to find complementary X-rays with angle pairs (α', γ') to compensate for the missing X-rays with angle pairs (α, γ). Therefore, in the system and method described in this specification, the shift of the detected X-rays can be further adjusted by interpolation between adjacent channel / detector elements. For example, Figure 6A The defective view shown, which should be obtained by X-ray, can also be used to generate Figure 6B The image shown is a complementary X-ray. However, due to the shift of the X-rays, Figure 6AThe image portion is shared across adjacent channels / detector elements. Therefore, the method and system use a weighted average of adjacent pixels to shift the X-ray detected in the supplementary view back to its corresponding position in the defective view. In the remainder of this description, it is assumed that the use of complementary X-rays is based on interpolated complementary X-rays; however, in some embodiments, complementary X-rays without interpolation may also be used (e.g., in cases where the X-rays are smoothed or otherwise filtered at a certain timing).

[0084] When generating replacement views from supplementary views, several different methods can be applied as indicated in this specification, and the generation of these replacement views can occur at least at two different times. According to one embodiment, it is possible to perform a group of replacements that do not depend on the slice. In such an embodiment, the replacement X-ray is calculated regardless of which slice the missing data is used by. For example, if data is acquired first, the system detects multiple consecutive view defects, temporarily replaces the defective view with the replacement view, thereby creating an extended group of views. Then, the extended group of views can be processed in the same way as the group of views without errors, and image reconstruction is performed normally. In such a case, any of the above techniques can be performed. For example, (1) if there is only one complementary X-ray in the supplementary view for each missing X-ray, the missing X-ray in each missing view is replaced by the corresponding X-ray from the supplementary view; (2) if there are multiple complementary X-rays for each missing ray, the missing X-ray in each missing view is replaced by a weighted combination of the corresponding rays from the supplementary view. The weighted combination may include any one of the following: (1) an average of a fixed number or a fixed percentage of the corresponding complementary X-rays; (2) an average of all the corresponding complementary X-rays; (3) a linear or nonlinear combination of a fixed number or a fixed percentage of the corresponding complementary X-rays; and (4) a linear or nonlinear combination of the corresponding complementary X-rays within a fixed angular range from the defective view. The combination may include certain weights, or it may include weights based on the distance between the complementary X-rays and the defective ray to give greater weight to complementary X-rays closer to the defective X-ray. Figure 7 This is a chart representing the three possible weighted averages inherent in the X-rays within the view. In the absence of erroneous scans, the X-ray redundancy (repeatability) is the same for all views used for reconstruction. The complementary X-rays within the maximum fixed range (centered on the defective X-ray) are combined in the first combination using half the weight 700 (i.e., by setting each X-ray to 0.5 times). In the second weighting 710, π+fan... max Complementary X-rays within the arc are weighted in a manner that generates half the weight, where fanmax This refers to the change between the maximum and minimum sector angles used in the view. In the 3rd weighted 720, this applies to situations involving more than half a circle + fan. max In the case of data less than one lap, a special function with weighting based on "more than half a lap" is used.

[0085] According to other implementations, in order to be used in helical reconstruction, it is possible to perform a group of replacements depending on the slice. Figure 8 This is a graph representing a group of known weights used as a sliding window for X-rays within a series of views used to reconstruct the slices. The illustrated graph shows how the weights slide between views if the reconstructed slices change. Image reconstruction can be performed by selecting the data range (-θ to θ) that can be used by each slice, based on pitch (circular scan: pitch = 0), detector Z-coverage, desired noise level, and temporal resolution. Figure 8 This illustration illustrates an example of redundant weights used across a data range, but the invention is not limited to the mapping of redundant weights shown. Values ​​that are brighter (i.e., close to white) within the graph are assigned higher weights for the corresponding X-rays of the view. Values ​​that are darker (i.e., close to black) within the graph are assigned lower weights for the corresponding X-rays of the view. In the graph, the weighting, which includes 100% at the center (corresponding to angle 0), decreases linearly towards 0% at the ends (i.e., views corresponding to angles θ / 2 and -θ / 2 have a 50% weight, and views corresponding to angles 3θ / 4 and -3θ / 4 have a 25% weight). Views corresponding to angles greater than θ or less than -θ are assigned a weight of 0%. Furthermore, while the weights in the illustration are not channel-dependent, in other embodiments, the weights may vary depending on the channel.

[0086] Figure 9 express Figure 8 The chart corresponds to the case where a portion of the slice has been reconstructed, i.e., a situation where many consecutive views (e.g., 100 views) are missing in the center of the sliding window. In the illustrated example, although there is more data than within the range (-θ, θ), the data outside this range is weighted at 0%. In the illustrated implementation, the missing X-rays are each contained in at least two supplementary views. If the weights that would normally be applied to the X-rays of the missing views were instead applied to the rays of the supplementary views, then... Figure 10As shown, new weighted groups are generated in units of light rays. This changes the previously high weights associated with the defective view, splitting it into two slanted bands (the slope of which depends on the pitch of the helical scan). By weighting the other supplementary views, the defective view is effectively "filled in," overcoming the zero effect of the defective view. In the illustrated embodiment, because the weight of the central defective view is split between two complementary X-rays, the corresponding parts of the slanted bands are brighter than their original weighted values ​​(i.e., given higher weights). That is, the central view's weight is now divided in half and allocated to the supplementary views. Figures 11A to 11D In this context, the change in that weight with respect to a smaller number of values ​​is expressed numerically. Figure 11A In this case, the damaged view is represented with the damaged view centered. For example... Figure 11B As seen in the first row, the 100% weighted average from the central view is divided into 50% / 50%, and the 50% boxes are assigned separately, resulting in two boxes in the first row with 100% weight. In the second row, the central 100% weighted average is divided into 50% / 50%, and the 40% and 60% boxes are assigned separately, resulting in two boxes in the second row with increased weights of 90% and 110%. For comparison, in... Figure 11C In the middle, the missing view is shown to be positioned further forward than the center. In such cases, as... Figure 11D As seen in the first row, the 60% weighted average from the missing view is divided into 50% / 50% and assigned to the first 10% box and the second 90% box, resulting in two boxes in the first row with weights of 40% and 120% respectively. In the second row, the 60% weighted average from the missing view is divided into 50% / 50% and assigned to the 0% box and the 100% box respectively, resulting in two boxes in the second row with increased weights of 30% and 130% respectively. Figures 11A to 11D The comparison shows that the weighted combination of complementary X-rays is a weighted combination of the reconstructed image data based on the angular position of the missing X-rays.

[0087] Figure 12 (A)~ Figure 12 (C) are images representing the original reconstructed slice, the reconstructed slice with 50 consecutive missing views in the center, and the reconstructed slice after replacing the 50 consecutive missing views in the center with weighted supplementary views. Figure 13 (A)~ Figure 13 (C) are images representing the original reconstructed slice, the reconstructed slice with 100 consecutive missing views in the center, and the reconstructed slice with the 100 consecutive missing views in the center replaced by weighted supplementary views.

[0088] and Figure 11D Similarly, Figure 14 This indicates the case where the missing view is centered around a view at an angle of approximately -7θ / 8, and the weights that should have been applied to the missing view are instead applied to the supplementary view. Figure 8 A revised version of the chart. Figure 15 (A) and Figure 15 (B) represents the original reconstructed slice and the... Figure 14 The image of the reconstructed slice is replaced by a weighted supplementary view of the continuous missing view.

[0089] If using Figure 10 , Figure 11A 11D and Figure 14 That kind of weighting, then with Figure 8 Compared to the original weighting, there are sharp changes between adjacent weights. This can sometimes produce stripes in the resulting reconstructed image, but it can also be smoothed using other weighting methods. Figure 16 This refers to a group of smoothed redundant weights used to make the transition between data from a defective view and data without errors smoother.

[0090] Figure 17 (A)~ Figure 17 (C) indicates the effect of smoothing on the reconstructed slices obtained under the first set of scanning conditions. Figure 17 (A) is the original reconstructed image without gaps in the view. Figure 17 (B) is used Figure 10 The reconstructed image without smoothing weights. Compared to the original, there are bright shadows in the circle on the left and stripes in the circle on the right. If using Figure 16 After smoothing and weighting, Figure 17 (C) The slice is reconstructed without masking variations or stripes, thus more faithfully representing the original slice.

[0091] Figure 18 (A)~ Figure 18 (C) indicates regarding and Figure 17 (A)~17(C) The same slices obtained under the second set of scanning conditions, and the effect of smoothing the reconstructed slices. Figure 18 (A) is the original reconstructed image without gaps in the view. Figure 18 (B) is used Figure 10 The reconstructed image without smoothing weights. Compared to the original, there are bright shadows in the circle on the left and stripes in the circle on the right. If using Figure 16 After smoothing and weighting, Figure 18 (C) The slice is reconstructed without masking variations or stripes, thus more faithfully representing the original slice.

[0092] The implementation methods also include those specified in the appendices shown below.

[0093] (1) A medical image processing method, comprising: obtaining scan data acquired in a CT scan of a subject and containing data corresponding to multiple views; obtaining an indication of a defective view not acquired in the CT scan of the subject; determining at least one complementary X-ray for a plurality of X-rays corresponding to the defective view not acquired in the CT scan of the subject; filling the at least one complementary X-ray for the plurality of X-rays corresponding to the defective view not acquired in the CT scan of the subject; and reconstructing image data of the subject based on the obtained scan data and the at least one complementary X-ray filled for the plurality of X-rays corresponding to the defective view not acquired in the CT scan of the subject; but not limited thereto.

[0094] (2) The medical image processing method as described in (1) for filling the at least one complementary X-ray with the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject includes: determining for each X-ray that the scan data for the X-ray of the defect view contains a single complementary X-ray; filling the X-ray with the single complementary X-ray; but is not limited thereto.

[0095] (3) The medical image processing method as described in any one of (1) and (2), for filling the at least one complementary X-ray with the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject, comprises: for each X-ray, determining that the scan data contains a plurality of complementary X-rays for one X-ray of the defect view; filling the one X-ray by a weighted combination of the plurality of complementary X-rays for the one X-ray; but is not limited thereto.

[0096] (4) In the medical image processing method described in (3), the weighted combination is a weighted combination based on the angular position of the X-ray relative to the reconstructed image data.

[0097] (5) The medical image processing method as described in any one of (1) to (4), for filling the at least one complementary X-ray with the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject, comprises: for each X-ray, determining that the scan data for one X-ray of the defect view contains a plurality of complementary X-rays; filling the one X-ray with the average of the plurality of complementary X-rays for the one X-ray; but is not limited thereto.

[0098] (6) The medical image processing method as described in any one of (1) to (5), wherein the number of X-rays in the plurality of complementary X-rays exceeds 2.

[0099] (7) The medical image processing method as described in any one of (1) to (6), for the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject, determining the at least one complementary X-ray includes: determining the at least one complementary X-ray based on the sector angle and tube angle of the plurality of X-rays corresponding to the defect view; but is not limited thereto.

[0100] (8) A medical image processing apparatus comprising a processing circuit configured to: obtain scan data acquired in a CT scan of a subject and containing data corresponding to multiple views; obtain an indication of a defective view not acquired in the CT scan of the subject; determine at least one complementary X-ray for a plurality of X-rays corresponding to the defective view not acquired in the CT scan of the subject; fill the at least one complementary X-ray for the plurality of X-rays corresponding to the defective view not acquired in the CT scan of the subject; and reconstruct image data of the subject based on the obtained scan data and the at least one complementary X-ray filled for the plurality of X-rays corresponding to the defective view not acquired in the CT scan of the subject; but not limited thereto.

[0101] (9) The medical image processing apparatus as described in (8) is configured to fill at least one complementary X-ray with the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject, including the processing circuit configured to perform the following processing: for each X-ray, determine that the scan data contains a single complementary X-ray for one X-ray of the defect view; fill the single X-ray with the single complementary X-ray; but is not limited thereto.

[0102] (10) The medical image processing apparatus as described in any one of (8) and (9) is configured to fill the at least one complementary X-ray with the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject, including the processing circuit configured to perform the following processing: for each X-ray, determine that the scan data contains a plurality of complementary X-rays for one X-ray of the defect view; fill the one X-ray by a weighted combination of the plurality of complementary X-rays for the one X-ray; but is not limited thereto.

[0103] (11) The medical image processing apparatus as described in (10) is a weighted combination based on the angular position of the X-ray relative to the reconstructed image data.

[0104] (12) The medical image processing apparatus as described in any one of (8) to (11), configured to fill the at least one complementary X-ray with the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject, includes a processing circuit configured to perform the following processing: for each X-ray, determine that the scan data contains a plurality of complementary X-rays for one X-ray of the defect view; fill the one X-ray with the average value of the plurality of complementary X-rays for the one X-ray; but is not limited thereto.

[0105] (13) The medical image processing apparatus as described in any one of (8) to (12), wherein the number of X-rays in the plurality of complementary X-rays exceeds 2.

[0106] (14) The medical image processing apparatus as described in any one of (8) to (13) is configured to determine at least one complementary X-ray for the plurality of X-rays corresponding to the defect view not obtained in the CT scan of the subject, including the processing circuit configured to determine the at least one complementary X-ray based on the sector angle and tube angle of the plurality of X-rays corresponding to the defect view; but is not limited thereto.

[0107] (15) A computer storage device having a non-transitory computer-readable medium containing a stored command, which, if read and executed by a computer processor, causes the computer processor to perform: a step of obtaining scan data acquired in a CT scan of an object being imaged, containing data corresponding to multiple views; a step of obtaining an indication of a defective view not acquired in the CT scan of the object; a step of determining at least one complementary X-ray for a plurality of X-rays corresponding to the defective view not acquired in the CT scan of the object; a step of filling the at least one complementary X-ray for the plurality of X-rays corresponding to the defective view not acquired in the CT scan of the object; and a step of reconstructing image data of the object based on the obtained scan data and the at least one complementary X-ray filled for the plurality of X-rays corresponding to the defective view not acquired in the CT scan of the object; but not limited thereto.

[0108] (16) The computer storage device as described in (15) causes the computer processor to perform any one of the methods described in (2) to (7) if the computer command stored in the non-transitory computer-readable medium is read and executed.

[0109] According to at least one embodiment described above, the accuracy of correcting a defective view can be improved.

[0110] Several embodiments have been described, but these embodiments are merely illustrative and not intended to limit the scope of the invention. These new embodiments can be implemented in a wide variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope or spirit of the invention, and are included in the scope of the invention as described in the claims and its equivalents.

Claims

1. A medical image processing method comprising: acquiring scan data including a plurality of views collected in a computed tomography (CT) scan of an object; deciding at least one complementary X-ray for a plurality of X-rays corresponding to a missing view not acquired in the CT scan of the object; setting a weight for the at least one complementary X-ray based on an angular position of the missing view with respect to reconstructed image data; and reconstructing image data of the object based on the scan data and the at least one complementary X-ray set with the weight.

2. The medical image processing method according to claim 1, further comprising: performing a smoothing process on the scan data including the at least one complementary X-ray set with the weight.

3. The medical image processing method according to claim 1 or 2, further comprising: filling the at least one complementary X-ray for the plurality of X-rays corresponding to the missing view.

4. The medical image processing method according to claim 3, wherein filling the at least one complementary X-ray for the plurality of X-rays corresponding to the missing view includes deciding a single complementary X-ray corresponding to one X-ray of the missing view included in the scan data for each X-ray included in the missing view, and filling the one X-ray by the single complementary X-ray.

5. The medical image processing method according to claim 3, wherein filling the at least one complementary X-ray for the plurality of X-rays corresponding to the missing view includes deciding a plurality of complementary X-rays corresponding to one X-ray of the missing view included in the scan data for each X-ray included in the missing view, and filling the one X-ray by a weighted combination of the plurality of complementary X-rays corresponding to the one X-ray.

6. The medical image processing method according to claim 5, wherein the weighted combination is a combination weighted based on an angular position of the one X-ray with respect to the reconstructed image data.

7. The medical image processing method according to claim 3, wherein filling the at least one complementary X-ray for the plurality of X-rays corresponding to the missing view includes deciding a plurality of complementary X-rays corresponding to one X-ray of the missing view included in the scan data for each X-ray included in the missing view, and filling the one X-ray by an average of the plurality of complementary X-rays corresponding to the one X-ray.

8. The medical image processing method according to claim 5, wherein a number of the plurality of complementary X-rays is two or more.

9. The medical image processing method according to claim 1, wherein deciding the at least one complementary X-ray for the plurality of X-rays corresponding to the missing view includes deciding the at least one complementary X-ray based on a fan angle and a tube angle of the plurality of X-rays corresponding to the missing view.

10. A medical image processing apparatus comprising: an acquisition unit that acquires scan data including a plurality of views collected in a computed tomography (CT) scan of an object; a determination unit that determines at least one complementary X-ray for a plurality of X-rays corresponding to a missing view not acquired in the CT scan of the object; a setting unit that sets a weighting for the at least one complementary X-ray based on an angular position of the missing view with respect to reconstructed image data; and a reconstruction unit that reconstructs image data of the object based on the scan data and the at least one complementary X-ray to which the weighting is set.

11. A storage medium in which a program that causes a computer to execute the following processes is non-volatile stored: acquire scan data including a plurality of views collected in a computed tomography (CT) scan of an object; determine at least one complementary X-ray for a plurality of X-rays corresponding to a missing view not acquired in the CT scan of the object; set a weighting for the at least one complementary X-ray based on an angular position of the missing view with respect to reconstructed image data; and reconstruct image data of the object based on the scan data and the at least one complementary X-ray to which the weighting is set.

Images