Image representation in limited angle tomography methods

Abstract

For image representation in a limited angle tomography method, a plurality of projection images representing an object (8) are obtained and a three-dimensional image reconstruction is generated based on the plurality of projection images. A first slice image (7) for passing through a first section plane of the object (8) is generated based on the image reconstruction, and a region of interest (9) of the object (8) is determined in the first slice image (7). A line of intersection (10, 10a, 10b, 10c, 10d) of a second cutting plane is determined in the first cutting plane, the second cutting plane extending through the region of interest (9). A calculation region (11, 11a, 11b, 11c, 11d) around the second sectioning plane is determined as a function of the position of the line of intersection and / or as a function of the geometric extension of the region of interest (9) in the first slice image (7). And generating a second slice image for the second sectioning plane according to the calculation area based on the plurality of projection images.

Description

Technical Field

[0001] This invention relates to a computer-implemented method for image representation in a finite-angle tomography method, in which multiple projected images of a represented object corresponding to multiple different projection directions are obtained, and a three-dimensional image reconstruction is generated based on the multiple projected images. The invention also relates to a data processing system for performing such a computer-implemented method, an imaging system having such a data processing system, and a corresponding computer program product. Background Technology

[0002] Limited angle tomography (LAT) is an X-ray imaging method in which a series of slice images can be generated from a finite number of two-dimensional projected images. LAT is also referred to herein and hereinafter as a tomographic synthesis method.

[0003] Tomographic synthesis methods enable three-dimensional visualization of anatomical structures with lower radiation doses than, for example, computed tomography (CT) or cone-beam CT. By recording multiple projected images around the patient from different angles, i.e., with different projection directions, a series of thin-slice images can be generated. These slice images can be reconstructed into a three-dimensional image reconstruction, also known as a volumetric dataset. This allows for better visualization of complex structures, such as the lungs, and allows radiologists to identify and characterize lesions with lower radiation doses.

[0004] In traditional tomographic synthesis methods, the image quality achieved is often insufficient for image-guided interventions, such as bronchoscopy. This limitation stems from the limited angle acquisition and typically small number of projection directions, which can lead to geometric distortion and streak artifacts.

[0005] Methods for using deep learning algorithms to compensate for missing projections have been published, for example in Y. Huang et al., “Data consistent artifact reduction for limited angle tomography with deep learning prior.”, International workshop on machine learning for medical image reconstruction. Cham: Springer International Publishing, 2019.

[0006] One drawback is the need for a large amount of annotated training data, which can be problematic in a clinical setting due to data protection regulations and other reasons.

[0007] As an alternative, F. Saad et al., “Deformable 3D / 3D CT-to-digital-tomosynthesis image registration in image-guided bronchoscopy interventions.” Comput. Biol. Med. 171: 108199, proposed a registration and reconstruction method that improves the image quality of tomosynthesis by using diagnostic CT scans recorded prior to the tomosynthesis scan. In this case, the previous CT image is matched with the surgical tomosynthesis image and used as an initial estimate for reconstruction. The additional CT scan increases the total applied radiation dose. Summary of the Invention

[0008] The technical problem to be solved by the present invention is to provide a feasibility for image representation in finite angle tomography, which can improve image quality and overcome or reduce the above-mentioned disadvantages.

[0009] This invention is based on the understanding that, due to the principle of finite-angle data acquisition, artifacts do not have the same degree of prominence across all viewing directions or cutting planes passing through the mapped object. Therefore, it is proposed to determine the region of interest (ROI) of the object within a first cutting plane passing through the object, and to determine a computational region based on this ROI, which is used to generate a slice image based on a second cutting plane different from the first cutting plane.

[0010] According to one aspect of the invention, a computer-implemented method for image representation in a finite-angle tomography method is provided. This method involves obtaining multiple projected images representing an object and corresponding to multiple different projection directions, and generating a three-dimensional image reconstruction based on the multiple projected images. Based on the image reconstruction, a first slice image is generated for passing through a first cutting plane, particularly a predetermined first cutting plane, and a region of interest (ROI) of the object is determined in the first slice image. A line of intersection of a second cutting plane, particularly a predetermined second cutting plane, is determined in the first cutting plane, the second cutting plane intersecting the first cutting plane and extending through the ROI, particularly within the first cutting plane. A computational region surrounding the second plane is determined based on the position of the line of intersection in the first cutting plane, particularly the position of the line of intersection relative to the ROI, and / or based on the geometric extension of the ROI in the first slice image, wherein the computational region has a position and a slice thickness. Based on the multiple projected images, a second slice image for the second cutting plane is generated according to the computational region, and the second slice image is displayed, for example, particularly on a display device.

[0011] Unless otherwise stated, all steps of the computer-implemented method can be performed by a data processing system comprising at least one data processing device. In particular, the at least one data processing device is configured or adapted to perform the steps of the computer-implemented method. For this purpose, the at least one data processing device may, for example, store a computer program containing instructions that, when executed by the at least one data processing device, cause the at least one data processing device to perform the computer-implemented method. The computer-implemented method can also be implemented wholly or partially in hardware. The terms "data processing system" and "at least one data processing device" are used interchangeably herein and hereinafter. This also applies to their derived terms.

[0012] If the at least one data processing device comprises two or more data processing devices, then a particular step performed by the at least one data processing device can also be understood as different steps being performed by different data processing devices or different parts of a step. It is particularly unnecessary that each data processing device performs these steps. In other words, the execution of these steps can be distributed among two or more data processing devices.

[0013] From each implementation of the computer-implemented method, a corresponding implementation of a method for image representation in a finite-angle tomography method is generated, which is not purely computer-implemented, but includes corresponding steps for generating the plurality of projected images, particularly by means of an imaging device.

[0014] Receiving or acquiring data or information can include, for example, receiving or acquiring the data from a sending entity, particularly by a data processing system, or reading data from a data storage device, or receiving or acquiring a data stream containing the data, or extracting data from the data stream, etc. For this purpose, wired or wireless data transmission can be used in particular. Specifically, data transmission can occur between the hardware and / or software interface of the sending entity and the hardware and / or software interface of the data processing system.

[0015] These multiple projection images are, for example, multiple X-ray projection images. However, the method according to the invention is also applicable in principle to other imaging modalities that are based on generating projection images from different projection directions and reconstructing three-dimensional images based thereon.

[0016] These multiple projected images are generated based on finite-angle data acquisition. In other words, these multiple different projection directions do not cover the full angular range required for accurate 3D reconstruction, but only a portion of it. For example, for complete reconstruction, an angular range of at least 180° is required, or 180° plus the so-called fan angle of the X-ray system. In finite-angle tomography, the swept angular range is less than 180°, for example, less than 120° or less than 60°. For example, the entire swept angular range is in the range of [30°, 150°], or in the range of [30°, 90°], or in the range of [30°, 60°]. The swept angular range here involves a predetermined swing axis, around which the corresponding X-ray source and, if necessary, the X-ray detector are rotated to achieve different projection directions. Circular trajectories, i.e., pure rotational motion, and helical trajectories are all feasible, in which, in addition to rotational motion, translational motion is simultaneously or alternately performed along the swing axis.

[0017] In finite angle computed tomography (FCT) methods, the number of multiple projection images is typically significantly less than in CT or CBCT methods, and is, for example, in the range of 30 to 150 images, or in the range of 30 to 80 images.

[0018] Three-dimensional image reconstruction can be generated using known reconstruction methods for finite-angle tomography. The 3D image reconstruction is provided by a three-dimensional voxel grid, where a corresponding attenuation value is calculated for each voxel of the voxel grid based on the multiple projected images using the reconstruction method. The voxels of the voxel grid are typically cubes or cuboids, but in some implementations they may have other geometries.

[0019] The projection direction here and below can be understood as the normal direction of the corresponding projection plane. However, the cutting plane should be understood as the cutting plane that passes through the reconstructed 3D image. Therefore, for the first and second cutting planes, there may not necessarily be a projected image with a projection direction perpendicular to the corresponding cutting plane.

[0020] The cutting plane is given by its position and orientation, for example, in the form of a corresponding normal direction to the cutting plane. The intersection line specifically corresponds to a straight line or line segment. Therefore, the position of the intersection line can be understood as its location within the first slice image relative to the region of interest, perpendicular to the direction of extension of the straight line or line segment. The position of the intersection line can also correspond to the perpendicular distance from a point defined on the contour of the region of interest to the intersection line.

[0021] The first and second cutting planes can be perpendicular to each other, but this is not necessary.

[0022] The oscillation axis is also referred to below as the z-axis of the Cartesian coordinate system. If the object is a human patient, the body's longitudinal axis, i.e., the intersection of the patient's sagittal and frontal planes, is usually oriented parallel to the z-axis. However, this is not mandatory.

[0023] The projection planes of these multiple projected images are therefore all parallel to the z-axis, or in other words, the z-axis lies within all projection planes. The projection directions of these multiple projection directions therefore lie within the xy-plane of the coordinate system.

[0024] Based on the fundamental principle of finite-angle tomography, due to the angle limitation, artifacts are least noticeable in the cutting plane that is parallel or approximately parallel to the central or intermediate projection direction of the scanned angle range. For example, the angle range can be represented by […]. ] indicates that the 0° angle corresponds to the y-axis of the coordinate system, and and .For example, Therefore, the artifacts are expected to be minimal on the cutting plane perpendicular to the y-axis, while relatively strong artifacts are expected on the cutting plane perpendicular to the z-axis or x-axis.

[0025] The region of interest (ROI) corresponds to the area within an object that is of particular interest to the application or user. This could be a specific region within an organ, or a lesion, tumor, etc.

[0026] To generate a slice image corresponding to a specific cutting plane, which can then be displayed to the user and / or used for further processing, a computational region is defined around that cutting plane. Data reconstructed from the 3D image within this computational region, i.e., the data of the corresponding voxels, is then used to generate the slice image. For example, the data reconstructed from the 3D image within the computational region can be averaged. In a corresponding implementation, the computational region can also be referred to as the averaging region. Alternatively, data from the multiple projected images can be used, and the slice image can be reconstructed based on the computational region. In this case, the computational region can be considered as the parameters themselves being reconstructed. The larger the computational region, the higher the signal-to-noise ratio of the slice image, for example.

[0027] According to the present invention, it is now understood that regions of interest (ROIs) can be determined particularly accurately and reliably due to the low artifact intensity in certain cutting planes. If the ROI is determined based on a first slice image, the computational region used to generate a second slice image, particularly the position of the computational region and / or the slice thickness, can be precisely adjusted according to the position of the intersection line relative to the ROI and / or the extension of the ROI, in order to improve the image quality of the second slice image.

[0028] The computational region specifically corresponds to an infinitely extending slice of constant thickness, which corresponds to the slice thickness of the computational region or the intersection of the slice with a spatial region containing 3D reconstructed data. However, this is not mandatory. In particular, instead of a slice, a wedge-shaped figure can also be specified as the computational region, i.e., a region defined by two non-parallel planes or other geometric shapes. The computational region may also encompass the entire region of interest. The location of the computational region corresponds, for example, to its position along the normal direction of the second cutting plane or relative to the intersection line.

[0029] It should be noted that, also in order to generate the first slice image, a corresponding additional computational region may be used. This additional computational region may be predetermined in a conventional manner, i.e., having a predetermined slice thickness and a predetermined position relative to the first cutting plane, such that, for example, the first cutting plane is centrally located within this additional computational region.

[0030] Determining the region of interest includes, for example, determining the location of the region of interest within the first slice image and its geometric extension within the first slice image. To do this, for example, the contour of the region of interest within the first slice image can be determined. The determination of the region of interest can be performed, for example, by applying a known segmentation algorithm or by computer-aided detection (CAD), or it can be performed manually by a user, who is then shown the first slice image on a display device.

[0031] The geometric extension of the region of interest corresponds, for example, to a geometric extension perpendicular to the direction of the intersection line, i.e., a linear geometric extension perpendicular to the direction of the intersection line.

[0032] The extension of the second cutting plane through the region of interest can be understood as the second cutting plane at least contacting the region of interest, that is, sharing at least one point with the region of interest.

[0033] According to at least one embodiment, the respective projection directions of the plurality of projection directions lie in the xy plane of a Cartesian coordinate system. The plurality of projection directions sweep across an angular range around a oscillation axis corresponding to the z-axis of the coordinate system. For example, this angular range is symmetric about the y-axis of the coordinate system.

[0034] According to at least one embodiment, the plurality of projection directions includes a projection direction perpendicular to the first cutting plane.

[0035] Vertical orientation here also includes approximately vertical orientation, which in particular has... The tolerance range, of which Where N is the total number of projection directions, and α is the range of angles swept, especially... .

[0036] A particular advantage of this type of implementation is that the image quality of the first slice image based on such a first cutting plane is particularly high, or the intensity of artifacts in such a first slice image is particularly low. For example, the first cutting plane is parallel or approximately parallel to the xz plane of the coordinate system, or, in the case of the patient's corresponding position, parallel to the frontal or coronal plane passing through the patient's body.

[0037] For example, the projection direction perpendicular to the first cutting plane corresponds to the central or intermediate projection direction within the swept angular range. In particular, the projection directions of these plurality of projection directions are symmetrically distributed around the central or intermediate projection direction and / or uniformly distributed within the swept angular range.

[0038] According to at least one embodiment, the second cutting plane is perpendicular to the first cutting plane.

[0039] This is particularly advantageous because, in this case, it is possible to determine with particular accuracy from the first slice image whether the second cutting plane extends through the interior of the region of interest, whether it contacts the region of interest at its contour, or precisely at which point the second cutting plane extends through the region of interest. Therefore, the location of the intersection line can also be determined with particular precision.

[0040] In some embodiments, the plurality of different projection directions include a projection direction perpendicular to the first cutting plane, and furthermore, the second cutting plane is perpendicular to the first cutting plane.

[0041] This is particularly advantageous because, in this case, the image quality of the first slice is exceptionally high, and the artifacts perpendicular to the first cutting plane are particularly noticeable.

[0042] According to at least one embodiment, the second cutting plane is parallel to all of the plurality of different projection directions.

[0043] In such embodiments, the concept according to the invention plays a particularly advantageous role because artifacts are particularly noticeable in such a cutting plane. For example, the second cutting plane is parallel to all of the plurality of different projection directions, and furthermore, the second cutting plane is perpendicular to the first cutting plane.

[0044] According to at least one embodiment, a first boundary position of the region of interest with the maximum extension in the direction perpendicular to the intersection line and / or a second boundary position of the region of interest with the maximum extension in the direction perpendicular to the intersection line are determined. The calculation region, especially the position of the calculation region and / or the slice thickness of the calculation region are determined based on the distance from the intersection line to the first boundary position, especially the vertical distance and / or the distance from the intersection line to the second boundary position, especially the vertical distance.

[0045] In other words, if the location of the first boundary is determined but the location of the second boundary is not, the computational region is determined based on the distance from the intersection line to the first boundary location. If the location of the second boundary is determined but the location of the first boundary is not, the computational region is determined based on the distance from the intersection line to the second boundary location. If both the first and second boundary locations are determined, the computational region is determined based on the distance from the intersection line to the first boundary location and / or based on the distance from the intersection line to the second boundary location.

[0046] The boundary locations specifically correspond to points on the contour of the region of interest in the first slice image, which extends between these points in a direction perpendicular to the intersection line. If the second cutting plane is perpendicular to the first cutting plane, the intersection line extends particularly horizontally in the first slice image. The first boundary location corresponds, for example, to the upper boundary location in the vertical direction, and the second boundary location corresponds, for example, to the lower boundary location in the vertical direction.

[0047] The computational region is determined by considering the locations of the first and / or second boundaries, which can advantageously always be chosen such that the second slice image can reproduce the relevant structures in the region of interest particularly well. In particular, this allows the computational region to lie entirely within the region of interest, at least if this can be determined in the first slice image.

[0048] According to at least one embodiment, the location of the calculation region is determined based on the location of the intersection line. For example, in this case, the second cutting plane is perpendicular to the first cutting plane.

[0049] Therefore, the position of the computation region relative to the intersection line can be dynamically determined based on the position of the intersection line within the region of interest. In particular, the position of the computation region can be chosen such that the intersection line is not located at the center of the computation region if it is advantageous for generating the second slice image, for example, if this allows a large portion or the entire computation region to be located within the region of interest.

[0050] For example, the location of the calculation region can be determined based on the distance from the intersection line to the first boundary location and / or the distance from the intersection line to the second boundary location.

[0051] According to at least one embodiment, the position of the intersection line corresponds to either a first boundary position or a second boundary position. The position and / or slice thickness of the computational region are determined such that the portion of the computational region outside the region of interest is smaller than the portion of the computational region inside the region of interest, or such that the computational region is entirely within the region of interest. For example, in this case, the second cutting plane is perpendicular to the first cutting plane.

[0052] This allows for the calculation of the second slice image by considering a larger portion of the region of interest and / or a smaller portion of the object located outside the region of interest, compared to the conventional method of determining the calculation region independently of the intersection location. Furthermore, this preserves the boundaries of the region of interest. The image quality of the second slice image can thus be further improved.

[0053] According to at least one embodiment, the location and slice thickness of the computational region are determined such that the portion of the computational region outside the region of interest, independent of the location of the intersection line, is smaller than the portion of the computational region inside the region of interest, or such that the computational region, independent of the location of the intersection line, is entirely within the region of interest. For example, in this case, the second cutting plane is perpendicular to the first cutting plane.

[0054] In other words, it is stipulated that for any position of the intersection line within the region of interest, i.e. between the first boundary position and the second boundary position, where the first boundary position and the second boundary position are included as possible positions of the intersection line, the computational region is always completely located within the region of interest, or at least the portion of the computational region outside the region of interest is always smaller than the rest of the computational region.

[0055] This allows for the calculation of the second slice image by considering a larger portion of the region of interest and / or a smaller portion of the object located outside the region of interest, compared to the conventional method of determining the calculation region independently of the intersection location. The image quality of the second slice image can thus be further improved.

[0056] According to at least one embodiment, the slice thickness of the computational region is determined based on the position of the intersection line. For example, in this case, the second cutting plane is perpendicular to the first cutting plane.

[0057] Therefore, the slice thickness of the computational region can be dynamically determined based on the position of the intersection line within the region of interest. A larger slice thickness may result in better image quality, for example, in the second slice image. By determining the slice thickness as described, the available region can be better utilized without having a large portion of the computational region located outside the region of interest.

[0058] For example, the slice thickness of the computational region can be determined based on the distance from the intersection line to the first boundary position and / or the distance from the intersection line to the second boundary position.

[0059] According to at least one embodiment, the slice thickness of the computational region is determined based on the geometric extension of the region of interest in the first cutting plane, for example, based on the distance of the first boundary position relative to the second boundary position in a direction perpendicular to the intersection line.

[0060] Therefore, the slice thickness of the computational region can be dynamically determined based on the geometric extension of the region of interest. A larger slice thickness may result in better image quality, for example, in the second slice image.

[0061] According to at least one embodiment, data corresponding to the calculation region of the plurality of projected images are extracted, and a second slice image is generated based on the extracted data.

[0062] In other words, in this type of implementation, the second slice image for the second cutting plane is generated directly based on the plurality of projected images according to the calculated region.

[0063] Generating a second slice image can correspond, for example, to a complete reconstruction, particularly a reconstruction of the 3D image, or to a partial reconstruction based solely on the computational region. This can also include a step of averaging the corresponding data within the computational region. In particular, the average value can be determined dynamically, thereby preserving the boundaries of the region of interest.

[0064] The advantage of generating the second slice image directly from these multiple projected images is that these multiple projected images exist with a higher spatial resolution than those reconstructed from 3D images. Therefore, the image quality of the second slice image can be further improved.

[0065] According to at least one implementation, data corresponding to the computational region of the image reconstruction is extracted, and a second slice image is generated based on the data thus extracted.

[0066] In other words, in this type of implementation, the second slice image is generated indirectly based on the plurality of projected images, since the 3D image reconstruction is also generated based on the plurality of projected images.

[0067] Generating the second slice image specifically involves averaging the image reconstruction data within the computational region. In particular, the average value can be determined dynamically, thus preserving the boundaries of the region of interest. The advantage of generating the second slice image based on the image reconstruction data is that this data already exists, and therefore the computational cost of generating the second slice image can be reduced.

[0068] According to another aspect of the present invention, a data processing system is provided, which is configured to perform a computer-implemented method according to the present invention.

[0069] In this disclosure, the terms "data processing system" and "at least one data processing device" are used interchangeably. A data processing device can be particularly understood as a data processing device containing processing circuitry. Therefore, a data processing device can, in particular, process data to perform computational operations. The computational operations may also include, where necessary, operations for performing index accesses to data structures, such as look-up tables (LUTs), and data processing procedures implemented in hardware.

[0070] Data processing devices may, in particular, include one or more computers, one or more microcontrollers, and / or one or more integrated circuits, such as one or more application-specific integrated circuits (ASICs), one or more field-programmable gate arrays (FPGAs), and / or one or more system-on-a-chip (SoCs). Data processing devices may also include one or more processors, such as one or more microprocessors, one or more central processing units (CPUs), one or more graphics processing units (GPUs), and / or one or more signal processors, particularly one or more digital signal processors (DSPs). Data processing devices may also include a physical or virtual collection of computers or other such units.

[0071] In different embodiments, the data processing device includes one or more hardware and / or software interfaces and / or one or more storage units.

[0072] The storage cell can be designed as volatile data memory, such as dynamic random access memory (DRAM) or static random access memory (SRAM), or as non-volatile data memory, such as read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), flash memory or flash EEPROM, ferroelectric random access memory (FRAM), magnetoresistive random access memory (MRAM), or phase-change random access memory (PCRAM).

[0073] According to another aspect of the invention, an imaging system for finite-angle tomography is provided. The imaging system includes a data processing system according to the invention, and an imaging device configured to generate the plurality of projected images.

[0074] According to at least one embodiment of the imaging system, the imaging device is an X-ray imaging device, such as a C-arm device.

[0075] According to at least one embodiment, the imaging system has a display device, wherein the data processing system is configured to display a second slice image on the display device.

[0076] In some implementations, the display device may also be considered part of the data processing system.

[0077] Further embodiments of the imaging system according to the invention derive directly from various configurations of the computer-implemented method according to the invention, and vice versa. In particular, individual features and corresponding explanations and advantages of various embodiments of the computer-implemented method according to the invention can be similarly transferred to corresponding embodiments of the imaging system according to the invention. In particular, the imaging system according to the invention is designed or programmed to perform the computer-implemented method according to the invention. In particular, the imaging system according to the invention performs the computer-implemented method according to the invention.

[0078] According to another aspect of the invention, a computer program having instructions is provided. When executed by a data processing system, the instructions cause the data processing system to perform a computer-implemented method according to the invention.

[0079] The instructions may exist, for example, as program code. The program code may be provided, for example, in the form of binary code or assembler and / or source code of a programming language, such as C, and / or program scripts, such as Python.

[0080] According to another aspect of the invention, a computer-readable storage medium, particularly a physical and / or non-volatile computer-readable storage medium, is provided that stores a computer program according to the invention.

[0081] Computer programs and computer-readable storage media are computer program products with instructions, respectively.

[0082] Further features and combinations thereof are derived from the accompanying drawings and description. Attached Figure Description

[0083] The present invention will be explained in more detail below with the aid of specific embodiments and corresponding schematic diagrams. In the drawings, the same or functionally identical elements may be provided with the same reference numerals. It is not necessary to repeat the description of the same or functionally identical elements in different figures if necessary.

[0084] In the attached diagram:

[0085] Figure 1 A schematic diagram illustrating an exemplary embodiment of an imaging system for limited-angle tomography according to the present invention is shown;

[0086] Figure 2 A schematic block diagram illustrating an exemplary embodiment of a computer-implemented method for image representation according to the present invention;

[0087] Figure 3 A schematic diagram illustrating a first slice image and a computational region for a second slice image, representing another exemplary embodiment of a computer-implemented method for image representation according to the present invention;

[0088] Figure 4 A schematic diagram illustrating a first slice image and a computational region for a second slice image, representing another exemplary embodiment of a computer-implemented method for image representation according to the present invention; and

[0089] Figure 5 A schematic diagram showing a first slice image and a computational region for a second slice image, illustrating another exemplary embodiment of a computer-implemented method for image representation according to the present invention. Detailed Implementation

[0090] exist Figure 1 An exemplary embodiment of an imaging system 1 for limited-angle tomography according to the present invention is illustrated schematically.

[0091] The imaging system 1 has an imaging device 3, which is configured to generate multiple projected images representing the object 8 and corresponding to multiple different projection directions.

[0092] Object 8 can be, for example, a patient or a part of a patient's body. The patient can be placed on the patient bed 2 of the imaging system 1.

[0093] For example, imaging device 3 is shown as a C-arm X-ray device with an X-ray source 4 and an X-ray detector 5. However, the following explanation can be similarly applied to other imaging devices 3 that can generate projected images representing object 8 from multiple different projection directions. Multiple different projection directions can be achieved, for example, by rotating the X-ray source 4 and X-ray detector 5 around a swing axis on a trajectory, such as a circular or helical trajectory, thereby sweeping across a predetermined angular range. The swing axis is particularly parallel to or equivalent to the longitudinal axis of the patient's body.

[0094] The imaging system 1 has a data processing system 6 according to the invention, which is configured to execute a computer-implemented method according to the invention for image representation in a finite-angle tomography method based on the plurality of projected images.

[0095] Figure 2 A block diagram schematically illustrates an exemplary embodiment of this computer-implemented method according to the present invention. Further aspects of various embodiments of the computer-implemented method are described below. Figures 3 to 5 As shown in the image.

[0096] In step 200, the plurality of projected images are obtained, and a three-dimensional image reconstruction is generated based on the plurality of projected images. In step 220, a first slice image 7 for passing through a first cutting plane of object 8 is generated based on the image reconstruction, and a region of interest 9 of object 8 is determined in the first slice image 7.

[0097] exist Figures 3 to 5 In the example, the first cutting plane is, for example, the coronal plane of the patient's lungs passing through object 8. The region of interest 9 may correspond, for example, to a tumor or other lesion.

[0098] In step 240, the intersection lines 10, 10a, 10b, 10c, and 10d of the second cutting plane are determined in the first cutting plane, the second cutting plane extending through the region of interest 9. The second cutting plane may, for example, be perpendicular to the first cutting plane xz. Figures 3 to 5 In the example, the second cutting plane xy corresponds, for example, to a cross section or axial plane passing through the patient's body.

[0099] In step 260, computational regions 11, 11a, 11b, 11c, and 11d are determined around the second cutting plane based on the positions of the intersection lines 10, 10a, 10b, 10c, and 10d and / or based on the geometric extension of the region of interest 9 in the first slice image 7. Specifically, the slice thickness of computational regions 11, 11a, 11b, 11c, and 11d and their positions relative to the intersection lines 10, 10a, 10b, 10c, and 10d are determined.

[0100] In step 280, based on the plurality of projected images, a second slice image for the second cutting plane is generated according to the calculation regions 11, 11a, 11b, 11c, and 11d. This can be done, for example, directly based on the plurality of projected images, or indirectly based on 3D image reconstruction, i.e., based on the plurality of projected images.

[0101] exist Figure 3In the example, a first slice image 7 is shown from left to right at four different locations with intersection lines 10a, 10b, 10c, and 10d. The slice thickness of computation regions 11a, 11b, 11c, and 11d is the same in all four cases, and the positions of computation regions 11a, 11b, 11c, and 11d are determined according to the positions of intersection lines 10a, 10b, 10c, and 10d, specifically such that in all four cases, computation regions 11a, 11b, 11c, and 11d are completely within the region of interest 9, wherein, in this sense, the outline of the region of interest 9 is also within the region of interest 9.

[0102] In the leftmost first slice image 7, the intersection line 10a corresponds to the upper boundary of the region of interest 9. The computational region 11a extends downwards from the intersection line 10a only into the region of interest 9. "Up" and "down" here refer to the positive or negative z-direction, respectively. In the second first slice image 7 from the left, the intersection line 10b is located below the upper boundary of the region of interest 9. The computational region 11b extends symmetrically upwards and downwards from the intersection line 10b. The situation is similar for the third first slice image 7 from the left, the intersection line 10c, and the computational region 11c. In the rightmost first slice image 7, the intersection line 10d corresponds to the lower boundary of the region of interest 9. The computational region 11d extends upwards from the intersection line 10a only into the region of interest 9.

[0103] like Figure 3 As shown, in Figure 4 In the example, the first slice image 7 is shown from left to right at four different locations with intersection lines 10a, 10b, 10c, and 10d. The positions of the intersection lines 10a, 10b, 10c, and 10d are related to... Figure 3 The results are consistent with those shown. However, the slice thickness of regions 11a, 11b, 11c, and 11d also depends on the positions of the intersection lines 10a, 10b, 10c, and 10d. Therefore, different parts of the region of interest 9 can be utilized according to the specific requirements of the second slice image.

[0104] exist Figure 5 The image shows two first slice images 7 with different geometric extensions of the region of interest 9, specifically showing different extensions H between the upper and lower boundary positions in the z-direction. In the case shown above, H is smaller than in the case shown below. Here, the slice thickness of the computational region 11 is determined based on the geometric extension of the region of interest 9. For the case with a larger extension H, the slice thickness of the computational region 11 is also greater.

[0105] As described above, the present invention achieves image representation with higher image quality in finite angle tomography.

[0106] In some implementations, the slice thickness of computational regions 11a, 11b, 11c, and 11d for axial slice images is automatically adjusted based on a region of interest 9 visible on the coronal slice image. Region of interest 9 corresponds, for example, to a lesion. This allows the utilization of differences in lesion identifiability across different anatomical planes. For example, due to the good identifiability of lesions in the sagittal plane, lesions can be identified using a CAD algorithm. The CAD algorithm provides the location of the lesion in the sagittal plane, thereby allowing the calculation of the slice thickness for the axial plane. Alternatively, lesions can also be manually marked. Assuming the head-to-tail axis is referred to as the z-axis, i.e., the axis perpendicular to the axial slice, the slice thickness can be adjusted based on the z-position of the intersection lines 10a, 10b, 10c, and 10d within region of interest 9.

[0107] Other advantages of the various exemplary embodiments include ease of integration into existing reconstruction and visualization methods without altering the image data, which would otherwise increase the risk of information loss or artifacts not present in the original data. Furthermore, no additional imaging data, such as imaging data obtained through prior CT scans, is required.

Claims

1. A computer-implemented method for image representation in a finite-angle tomography method, wherein... - Obtain the representation object (8) and multiple projection images corresponding to multiple different projection directions, and generate a three-dimensional image reconstruction based on the multiple projection images; - Based on the image, a first slice image (7) is generated for passing through the first cutting plane of the object (8), and the region of interest (9) of the object (8) is determined in the first slice image (7); - Determine the intersection lines (10, 10a, 10b, 10c, 10d) of the second cutting plane in the first cutting plane, which extends through the region of interest (9); -Based on the position of the intersection line (10, 10a, 10b, 10c, 10d) and / or based on the geometric extension of the region of interest (9) in the first slice image (7), determine the computational region (11, 11a, 11b, 11c, 11d) surrounding the second cutting plane, wherein, The computational region (11, 11a, 11b, 11c, 11d) has a location and slice thickness; and - Based on the multiple projected images, a second slice image for the second cutting plane is generated according to the calculation region (11, 11a, 11b, 11c, 11d).

2. The computer-implemented method according to claim 1, wherein, The second cutting plane is perpendicular to the first cutting plane.

3. The computer-implemented method according to claim 2, wherein... - Determine the first boundary position of the region of interest (9) in the direction of maximum extension perpendicular to the intersection line (10, 10a, 10b, 10c, 10d) and / or the second boundary position of the region of interest (9) in the direction of maximum extension perpendicular to the intersection line (10, 10a, 10b, 10c, 10d); and - The calculation region (11, 11a, 11b, 11c, 11d) is determined based on the distance from the intersection line (10, 10a, 10b, 10c, 10d) to the first boundary position and / or the distance from the intersection line (10, 10a, 10b, 10c, 10d) to the second boundary position.

4. The computer-implemented method according to any one of the preceding claims, wherein, The location of the calculation area (11, 11a, 11b, 11c, 11d) is determined based on the location of the intersection line (10, 10a, 10b, 10c, 10d).

5. The computer-implemented method according to claims 4 and 3, wherein, The position of the intersection line (10, 10a, 10b, 10c, 10d) corresponds to the first boundary position or the second boundary position, and the position of the calculation region (11, 11a, 11b, 11c, 11d) is determined such that the portion of the calculation region (11, 11a, 11b, 11c, 11d) outside the region of interest (9) is smaller than the portion of the calculation region (11, 11a, 11b, 11c, 11d) inside the region of interest (9), or such that the calculation region (11, 11a, 11b, 11c, 11d) is completely within the region of interest (9).

6. The computer-implemented method according to any one of claims 1 to 4, wherein, The position and slice thickness of the computational region (11, 11a, 11b, 11c, 11d) are determined such that the portion of the computational region (11, 11a, 11b, 11c, 11d) outside the region of interest (9) is smaller than the portion of the computational region (11, 11a, 11b, 11c, 11d) within the region of interest (9), independent of the position of the intersection line (10, 10a, 10b, 10c, 10d), or such that the computational region (11, 11a, 11b, 11c, 11d) is completely within the region of interest (9), independent of the position of the intersection line (10, 10a, 10b, 10c, 10d).

7. The computer-implemented method according to any one of the preceding claims, wherein, The slice thickness of the computational region (11, 11a, 11b, 11c, 11d) is determined based on the position of the intersection line (10, 10a, 10b, 10c, 10d).

8. The computer-implemented method according to any one of the preceding claims, wherein, The slice thickness of the computational region (11, 11a, 11b, 11c, 11d) is determined based on the geometric extension of the region of interest (9) in the first cutting plane.

9. The computer-implemented method according to any one of the preceding claims, wherein, Data corresponding to the calculation region (11, 11a, 11b, 11c, 11d) from the multiple projected images are extracted, and the second slice image is generated based on the extracted data.

10. The computer-implemented method according to any one of claims 1 to 8, wherein, Extract the data corresponding to the computational region (11, 11a, 11b, 11c, 11d) from the image reconstruction, and generate the second slice image based on the extracted data.

11. The computer-implemented method according to any one of the preceding claims, wherein, The plurality of different projection directions include a projection direction perpendicular to the first cutting plane, and / or the second cutting plane being parallel to all of the plurality of different projection directions.

12. A data processing system (6) configured to perform a computer-implemented method according to any one of the preceding claims.

13. An imaging system (1) for limited-angle tomography, comprising a data processing system (6) according to claim 12 and an imaging device (3) configured to generate the plurality of projected images.

14. The imaging system (1) according to claim 13, wherein, The imaging device (3) is configured as an X-ray C-arm device.

15. A computer program product having instructions that, when executed by a data processing system (6), cause the data processing system (6) to perform a computer-implemented method according to any one of claims 1 to 11.

Images