The provision of the result data set
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- SIEMENS HEALTHINEERS AG
- Filing Date
- 2023-01-18
- Publication Date
- 2026-07-21
Smart Images
Figure CN116503265B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to a computer-implemented method for providing a result dataset, a medical dual-plane imaging device, and a computer program product. Background Technology
[0002] In cone-beam computed tomography (CBCT), subject movement, such as head movement, can limit the image quality of the recorded slices and / or cause artifacts, particularly motion artifacts. These artifacts can adversely obscure image regions important for the diagnostic assessment of image data, especially slices, such as images of hemorrhage. The latest generation of biplane angiography systems is typically technically capable of simultaneously mapping the subject onto two probe planes. This advantageously reduces acquisition time for recording projected images and minimizes the risk of motion-related complications.
[0003] The publication "Rigid motion compensation in interventional C-arm CT using consistency measure on projection data" by R. Frysch and G. Rose (International Society for Medical Imaging Computation and Computer-Aided Intervention, Springer, Cham, 2015) discloses that the motion of an examined object can be compensated, for example, by adapting the geometric parameters of the imaging system to the motion using a consistency metric based on projection images. The publication also discloses that while this method can achieve robust and computationally efficient compensation for motions outside the plane of rotation, it is significantly limited when compensating for motions within the plane of rotation.
[0004] Furthermore, projection images recorded by medical X-ray equipment often contain scattered radiation artifacts and / or radiation hardening artifacts. In such cases, correction is often time-consuming and / or error-prone, as scattered radiation artifacts and / or radiation hardening artifacts may occur simultaneously with motion artifacts. Summary of the Invention
[0005] Therefore, the technical problem to be solved by the present invention is to achieve improved artifact and / or motion correction in medical projection images.
[0006] This technical problem is solved according to the present invention by a method for providing a result dataset, a medical dual-plane imaging device, and a computer program product. Advantageous embodiments with suitable extended designs are given in the specification.
[0007] The invention relates, in a first aspect, to a computer-implemented method for providing a result dataset (or data set). Here, at least one pair of projected images of an examination subject is recorded using a medical dual-plane imaging device. The at least one pair of projected images has a first projected image and a second projected image of the examination subject, which simultaneously abbilize (or image) the examination subject onto first and second probe planes. The first and second probe planes are arranged non-parallel to each other. Furthermore, a correction model is determined for correcting artifacts and / or motion, wherein the artifacts and / or motion are simultaneously mapped in at least one first projected image and at least one second projected image. The at least one pair of projected images is pre-defined (or pre-stipulated) with consistency conditions for determining the correction model. Furthermore, the result dataset is reconstructed from at least the at least one first projected image and based on the correction model. The result dataset is then provided.
[0008] The subjects of examination may be, for example, female patients and / or male patients (human and / or animal). Advantageously, at least one pair of projected images, particularly multiple pairs of projected images, of the subjects can be recorded using a medical dual-plane imaging device. The medical dual-plane imaging device may have a first imaging unit and a second imaging unit. Here, the first imaging unit may be designed to record at least one first projected image of at least one pair of projected images, particularly multiple first projected images of multiple pairs of projected images. For this purpose, the first imaging unit may have a first source and a first detector, particularly a first flat panel detector. Furthermore, the second imaging unit may be designed to record at least one second projected image of at least one pair of projected images, particularly multiple second projected images of multiple pairs of projected images. For this purpose, the second imaging unit may have a second source and a second detector, particularly a second flat panel detector. The medical dual-plane imaging device may be advantageously designed to move and / or position the first and second imaging units in a coordinated manner, particularly in a defined arrangement relative to each other and / or independently. The first detection plane may describe a sensor surface of the first detector, particularly a substantially planar sensor surface, designed to detect radiation and / or waves emitted by the first source after interacting with the subject. The second detection plane can describe the sensor surface of the second detector, which is essentially planar and designed to detect radiation and / or waves emitted by the second source after interacting with the object being inspected.
[0009] Advantageously, the first and second projection images of the at least one projection image pair map at least a portion, particularly a completely common area of the object to be examined. The first and second projection images of the at least one projection image pair can map the object to be examined, particularly in two-dimensional (2D) spatial resolution. Furthermore, the first and second projection images of the at least one projection image pair can simultaneously, particularly at the same recording time point, map the object to be examined from two different, particularly non-parallel and not antiparallel, projection directions, particularly angles, particularly artifacts and / or motion. At least one first projection image and at least one second projection image can, for example, include X-ray projection images and / or ultrasound images and / or positron emission tomography images. The projection direction can describe the imaging source, such as an X-ray source and / or the space ray between the ultrasound emitter and the detector, particularly the central ray.
[0010] If multiple pairs of projected images of the object being inspected are recorded, the defined arrangement of the first and second detection planes, particularly the first source and the first detector, and the defined arrangement of the second source and the second detector, can remain substantially stationary relative to the object being inspected during the recording of the projected image pairs. Alternatively, the defined arrangement of the first and second detection planes, particularly the first source and the first detector, and the defined arrangement of the second source and the second detector, can move, in particular, along a trajectory, particularly a circular track, around the object being inspected during the recording of multiple pairs of projected images of the object, particularly in relation to each other.
[0011] The correction model may include presets for particularly rigid transformations, such as translation and / or rotation, and / or correction functions, particularly nonlinear image value corrections, and / or filtering for at least the at least one first projected image. In particular, the correction model may have presets for transformations and / or correction functions and / or filtering for at least one first projected image and at least one second projected image. Here, the correction model may be parameterized. Determining the correction model may include adjusting at least one parameter of a general or original correction model, particularly the correction model, to adapt to artifacts and / or motion mapped in at least one pair of projected images.
[0012] Advantageously, the resulting dataset can be reconstructed from at least one first projection image, particularly from multiple first projection images, based on a correction model. Here, the reconstruction of the resulting dataset may include applying a transformation and / or correction function and / or filter, pre-defined by the correction model, particularly a rigid one, to at least a portion or each of the at least one first projection image, particularly multiple first projection images. Advantageously, the resulting dataset can map the object of inspection in 2D and / or 3D spatial resolution. Furthermore, the resulting dataset can be temporally resolved. Mapping the object for inspection.
[0013] If, during the recording of multiple pairs of projected images of the object under inspection, the first probe plane is positioned substantially stationary relative to the object, and if the resulting dataset is reconstructed solely from the multiple first projected images, then the resulting dataset can advantageously map the object in 2D spatial resolution. If, during the recording of multiple pairs of projected images of the object under inspection, the first probe plane moves along a trajectory relative to the object, and if the resulting dataset is reconstructed solely from the multiple first projected images, then the resulting dataset can map the object in 3D spatial resolution.
[0014] The provision of the result dataset may include, for example, storage on a computer-readable storage medium and / or display on a display unit and / or transmission to the providing unit. In particular, a graphical view of the result dataset may be displayed on the display unit.
[0015] The proposed implementation can achieve improved artifact correction and / or motion correction based on preset consistency conditions using at least one projected image. Furthermore, by simultaneously mapping the object under inspection onto two probe planes, isolation of different artifacts to be corrected, such as motion artifacts and / or scattered ray artifacts and / or ray hardening artifacts, can be achieved.
[0016] In another advantageous embodiment of the proposed method for providing the result dataset, multiple pairs of projected images of the object under inspection can be recorded, these pairs of projected images mapping the motion of the object under inspection in a time-resolved manner. Furthermore, artifacts may include motion artifacts. Advantageously, the multiple pairs of projected images may be pre-defined, at least along the time dimension, with consistency conditions for determining the correction model.
[0017] Multiple projection image pairs can time-resolvedly map the object of examination onto two probe planes, particularly at multiple sequential recording time points, and especially from two different projection directions. Multiple projection image pairs, particularly multiple first and second projection images, can time-resolvedly map the motion of the object of examination, particularly the motion of at least one part of the object, such as body parts and / or anatomical structures. The motion of the object of examination, particularly physiological motion, can include, for example, respiratory motion, cardiac motion, and / or head motion. Here, the first and second projection images of each projection image pair can simultaneously map the state of the object of examination, particularly its motion state, onto the two probe planes at the corresponding recording time points of the projection image pair, and especially from different projection directions.
[0018] Furthermore, multiple pairs of projected images can map the motion of the object under examination in a time-resolved manner. When reconstructing the resulting dataset from at least the first projected image, this can lead to motion artifacts in the resulting dataset without appropriate correction, particularly motion correction. Determining the correction model can include modeling, in particular, the motion of the object under examination mapped in the pair of projected images in a three-dimensional (3D) spatial and temporal resolution. Determining the correction model can, in particular, include identifying and / or tracing the geometric and / or anatomical features of the object under examination, which are mapped in the first and second projected images of the pair.
[0019] When determining a correction model for the motion of the object being examined mapped in a pair of projected images, the projected image pairs can be pre-defined with consistency conditions for determining the correction model, at least along the time dimension. It can be assumed here that the state of the object being examined mapped in the first and second projected images of each projected image pair, particularly the state of the motion mapped by the object, is particularly consistent in pairs. The consistency condition can be pre-defined, particularly along the time dimension, as the consistency of one of the projected image pairs relative to all the other projected image pairs. The second projected image of the projected image pair can be pre-defined with respect to depth information regarding the motion of the object being examined mapped in the first projected image for determining the correction model. Advantageously, the correction model can be determined such that the motion of the object being examined, modeled in the correction model, satisfies the consistency conditions pre-defined by the projected image pairs, at least along the time dimension, at multiple recording time points of the projected image pairs.
[0020] Therefore, especially when the object being inspected moves parallel to the first detection plane, motion correction reconstruction of the result dataset can be achieved.
[0021] In another advantageous embodiment of the proposed method for providing the results dataset, physiological motion signals that map the motion of the object being examined can be received. Here, the determination of the correction model can be additionally based on the physiological motion signals.
[0022] The reception of motion signals may in particular include the detection and / or reading from a computer-readable data storage device and / or reception by a data storage unit, such as a database. Furthermore, motion signals may be provided by a physiological sensor providing unit, such as an electrocardiogram (ECG) and / or a respiratory sensor and / or a pulse sensor and / or a motion sensor and / or a sensor providing unit for detecting the positioning of the object being examined, such as electromagnetic and / or optical and / or acoustic and / or mechanical sensors.
[0023] Advantageously, the correction model can be additionally determined based on physiological motion signals. This allows for a more robust and accurate reconstruction of the result dataset. In another advantageous embodiment of the proposed method for providing the result dataset, the medical biplane imaging device can be designed as a medical biplane X-ray device.
[0024] Here, the first source can be a first X-ray source, the second source can be a second X-ray source, the first detector can be a first X-ray detector, especially a first X-ray flat panel detector, and the second detector can be a second X-ray detector, especially a second X-ray flat panel detector.
[0025] To record at least one pair of projected images, particularly at least one first and at least one second projected image, first and second X-ray sources can respectively emit X-ray beams, particularly in the form of conical beams and / or fan-shaped beams and / or parallel beams, to irradiate the object to be examined, particularly the first and second X-ray detectors. Here, the first and second X-ray sources can emit corresponding X-ray beams with the same or different X-ray doses. The first X-ray detector can detect the X-ray beam emitted by the first X-ray source after interacting with the object to be examined and provide a first projected image. Furthermore, the second X-ray detector can detect the X-ray beam emitted by the second X-ray source after interacting with the object to be examined and provide a second projected image.
[0026] This allows for time-efficient and high-resolution imaging of the object in at least one pair of projected images.
[0027] In another advantageous embodiment of the proposed method for providing the result dataset, at least one first projection image can be recorded at a higher X-ray dose than at least one second projection image.
[0028] The signal-to-noise ratio and / or contrast-to-noise ratio of at least one first projection image and at least one second projection image are related to the X-ray dose during the recording of at least one first and at least one second projection image. Higher X-ray doses generally result in higher signal-to-noise ratios and / or contrast-to-noise ratios. However, higher X-ray doses during the recording of at least one first and at least one second projection image may adversely increase the radiation burden on the subject being examined.
[0029] Advantageously, at least one first projection image, and in particular multiple first projection images, can be recorded at a first X-ray dose, and at least one second projection image, and in particular multiple second projection images, can be recorded at a second X-ray dose, wherein the first X-ray dose is higher than the second X-ray dose. This advantageously allows at least one first projection image to have an optimal signal-to-noise ratio and / or contrast-to-noise ratio, particularly sufficient for diagnostic evaluation. Furthermore, the second X-ray dose can be minimized to such an extent that at least one second projection image adequately maps the geometric and / or anatomical features of the examined object to determine the correction model, but has an insufficient signal-to-noise ratio and / or contrast-to-noise ratio for diagnostic evaluation. This can be particularly advantageous when the resulting dataset consists only of at least one first projection image and is reconstructed based on the correction model.
[0030] This allows for the advantageous minimization of the radiation burden on the object being examined during the recording of the projected image pairs.
[0031] In another advantageous embodiment of the proposed method for providing the result dataset, the artifacts may include scattered ray artifacts and / or ray hardening artifacts. Here, at least one first projected image and at least one second projected image of at least one projection image pair can be pre-defined in pairs to pre-set consistency conditions for determining the correction model.
[0032] The first and second projection images of each projection image pair can simultaneously map scattered ray artifacts and / or ray hardening artifacts onto the first and second detection planes.
[0033] Scattered ray artifacts may be designed as rings and / or stripes in at least one first projected image and at least one second projected image. The correction model may include presets, particularly filter kernels, for filtering of at least one first projected image and at least one second projected image. Determining the correction model may include adapting (or adjusting) at least one parameter of the preset for filtering, particularly at least one parameter of the filter kernel.
[0034] Ray-hardening artifacts can be designed as deviations from pure line integrals in the image values of at least one first and at least one second projected image. The correction model can advantageously include correction functions, particularly polynomial correction functions, for achieving nonlinear image correction of the image values of at least one first and at least one second projected image.
[0035] The adjustment, particularly optimization, of at least one parameter and / or correction function for filtering can be performed iteratively, for example, according to the Nelder-Mead-Simplex method. Applying the final determined preset and / or correction function for filtering to at least one first and at least one second projected image can provide at least one corrected first projected image and at least one corrected second projected image. To correct ray hardening artifacts, a correction function, particularly a polynomial correction function, can be applied to the image values of at least one first and at least one second projected image with at least one parameter, particularly at least one coefficient, that is final adjusted. To correct scattered ray artifacts, a preset for filtering, particularly a filter kernel, such as a Gaussian kernel, can be applied to at least one first and at least one second projected image with at least one parameter, particularly the variance and / or height of the filter kernel, that is final adjusted.
[0036] At least one first and at least one second projected image, and in particular, the first and second projected images of each of a plurality of projected image pairs, can be paired and preset with consistency conditions for determining the correction model, and in particular for adjusting at least one parameter of the preset and / or correction function used for filtering, especially preset as epipolar consistency conditions. By simultaneously mapping the object under inspection onto the first and second detection planes in at least one first projected image and at least one second projected image, scattered ray artifacts and / or ray hardening artifacts can be advantageously isolated from possible motion artifacts.
[0037] Advantageously, at least one parameter of the correction model can be adjusted based on a pairwise consistency condition, particularly an epipolar consistency condition, between at least one corrected first projection image and at least one corrected second projection image. The pairwise consistency can be quantified, particularly by an epipolar consistency index. Advantageously, by adjusting, and particularly optimizing, at least one parameter of the correction model, the epipolar inconsistency between at least one corrected first projection image and at least one corrected second projection image can be minimized. This allows for improved artifact correction.
[0038] In another advantageous implementation of the proposed method for providing the result dataset, the result dataset may additionally be reconstructed from at least one second projected image.
[0039] Advantageously, the resulting dataset can be reconstructed from at least one first projection image and at least one second projection image, particularly from multiple first projection images and multiple second projection images, based on a correction model. Thus, the resulting dataset can advantageously map the object under inspection in 3D spatial resolution. The reconstruction of the resulting dataset can, for example, include filtered back projection. Furthermore, the reconstruction of the resulting dataset can include applying particularly rigid transformations, such as translation and / or rotation, to the at least one first projection image, particularly multiple first projection images, and / or the at least one second projection image, particularly multiple second projection images. Here, the transformation can be determined according to the correction model.
[0040] The proposed implementation method can advantageously achieve 3D spatial resolution reconstruction of the resulting dataset while simultaneously reducing the acquisition time.
[0041] In another advantageous implementation of the proposed method for providing the result dataset, multiple pairs of projected images of the object under examination can be recorded around a common axis of rotation.
[0042] Advantageously, multiple first projected images can be recorded around a first isocenter, and multiple second projected images can be recorded around a second isocenter. The first isocenter can be a first rotation center, around which the first imaging unit, particularly the first source and the first detector, moves, particularly rotates, during the recording of the first projected images. Furthermore, the second isocenter can be a second rotation center, around which the second imaging unit, particularly the second source and the second detector, moves, particularly rotates, during the recording of the second projected images. Here, the first and second isocenters can be arranged on a common axis of rotation. Advantageously, during the recording of the first projected image of the object under inspection, the spatial distance between the first source and the first detector can be constant relative to the first isocenter. Furthermore, during the recording of the second projected image of the object under inspection, the spatial distance between the second source and the second detector can be constant relative to the second isocenter. Advantageously, the first and / or second isocenters can be arranged within the object under inspection.
[0043] The proposed implementation can achieve improved, especially more consistent, determination of the calibration model and improved reproducibility of the resulting dataset.
[0044] In another advantageous embodiment of the proposed method for providing the result dataset, multiple pairs of projected images of the object under examination can be recorded. Here, the first and second probe planes of each pair of projected images can have a constant angle relative to each other.
[0045] Advantageously, the first imaging unit, particularly the first source and the first detector, and the second imaging unit, particularly the second source and the second detector, can have a constant spatial arrangement relative to each other during the recording of multiple pairs of projected images. Furthermore, the first imaging unit and the second imaging unit can always have a defined spatial arrangement relative to each other during the recording of multiple pairs of projected images.
[0046] Therefore, the consistency conditions used to determine the correction model can be preset at different recording time points of multiple projected image pairs with the same relative recording geometry of the first and second probe planes, especially along the time dimension.
[0047] In another advantageous embodiment of the proposed method for providing the result dataset, the first and second probe planes of each pair of projected images have an angle between 50° and 130° relative to each other, particularly an angle of 90°.
[0048] This advantageously ensures that artifacts and / or the motion of the object under inspection, especially the motion of the object parallel to one of the probe planes, can be reliably and simultaneously mapped in the corresponding other probe plane. This allows for the reliable pre-setting of consistency conditions used to determine the correction model.
[0049] In another advantageous embodiment of the proposed method, multiple pairs of projected images of the object being examined can be recorded around a common isocenter.
[0050] Advantageously, during the recording of multiple pairs of projected images, the spatial distances between the first source, the first detector, the second source, and the second detector can each be constant relative to a common isocenter. This common isocenter can be a common rotation center around which the first imaging unit, particularly the first source and the first detector, and the second imaging unit, particularly the second source and the second detector, move, particularly rotate, during the recording of the first projected image. Here, the first and second imaging units can have a common axis of rotation or different axes of rotation. The common isocenter can be arranged on the common axis of rotation or at the intersection of different axes of rotation.
[0051] The common isocenter can be stationary or repositioned during the recording of multiple pairs of projected images. For example, the common isocenter can be moved along a predetermined trajectory during the recording of a pair of projected images, particularly between the recording time points of the pair. Advantageously, the predetermined trajectory can extend at least partially, and particularly completely, within the object being examined.
[0052] The proposed implementation can achieve improved, especially more consistent, determination of the calibration model and improved reproducibility of the resulting dataset.
[0053] In another advantageous embodiment of the proposed method for providing the result dataset, multiple pairs of projected images of the object under examination can be recorded. Here, a first projected image can be recorded around a first isocenter, and a second projected image can be recorded around a second isocenter. The first and second isocenters can be different.
[0054] The first and second isocenters can be spatially spaced apart from each other, either constantly or variably, during the recording of multiple pairs of projected images. Therefore, during the recording of the projected image pairs, the first and second imaging units can move, and in particular rotate, around their respective isocenters, especially the first and second isocenters. Using different isocenters allows for a more flexible mapping of the object being examined, especially motion and / or artifacts, within the projected image pairs.
[0055] In another advantageous implementation of the proposed method for providing the result dataset, determining the calibration model may include optimizing the consistency metric.
[0056] A consistency metric can assess the consistency, particularly the coherence, between artifacts and / or motion modeled by the correction model and those mapped in at least one pair of projected images, particularly simultaneously in at least one first and at least one second projected image. Specifically, the consistency metric can quantify the deviation between the artifacts and / or motion modeled by the correction model and those mapped in at least one pair of projected images. In particular, the consistency metric can measure the pairwise inconsistency of the first and second projected images of each pair relative to all other projected image pairs, especially along the time dimension. In this case, the correction model can be determined by optimizing the consistency metric, particularly by minimizing the inconsistency of each projected image pair relative to all other projected image pairs, especially along the time dimension. This can be particularly advantageous for motion correction.
[0057] Optimization of the consistency index may include determining, and in particular generating, first and second virtual, especially corrected, projected images relative to at least one pair of projected images, and especially first and second virtual projected images respectively relative to multiple pairs of projected images, based on a correction model. The correction model may, for example, include a particularly deformable 3D spatially resolved volumetric model, such as a volumetric mesh model, of at least a portion of the object being examined. Here, at least one first and at least one second virtual projected image can be determined by virtually projecting the correction model, especially the volumetric model, onto virtual first and virtual second probe planes. The virtual projections on the virtual first and virtual second probe planes can be determined based on the recording geometry used to record at least one first and at least one second projected image mapping the object being examined onto the first and second probe planes. This can be particularly advantageous for correcting motion mapped in multiple pairs of projected images. To correct for ray hardening artifacts and / or scattered ray artifacts, the correction model can model the anatomical structure of the object being examined, such as the skeletal structure, which can be mapped by virtual projections onto the virtual first and virtual second probe planes. Thus, improved, particularly specialized, correction of artifacts generated by the anatomical structure can be achieved, for example, by additional correction polynomials in the correction function used for image value correction.
[0058] Alternatively or additionally, at least one virtual, in particular corrected first projection image and at least one virtual, in particular corrected second projection image can be provided by applying a correction model, in particular a preset and / or correction function for filtering, to at least one first and at least one second projection image.
[0059] The consistency index can advantageously evaluate both the conformity between at least one first projected image and at least one first virtual projected image, and the conformity between at least one second projected image and at least one second virtual projected image. In particular, the consistency index can be determined as the sum of two separate indices, which respectively evaluate the conformity between at least one first projected image and at least one first virtual projected image, and the conformity between at least one second projected image and at least one second virtual projected image. Advantageously, the consistency index can be optimized such that these two separate indices are optimized. This satisfies the pre-defined consistency conditions for at least one first and at least one second projected image of each projection image pair.
[0060] Alternatively or additionally, the consistency index can assess pairwise consistency, particularly polewise consistency, between images in at least one virtual, in particular corrected first projected image and at least one virtual, in particular corrected second projected image in each pair of projected images.
[0061] Optimization of the consistency metric can, for example, involve iteratively minimizing the cost function. The proposed implementation can achieve particularly accurate determination of the calibration model.
[0062] In another advantageous embodiment of the proposed method, at least one additional projection image can be recorded for the at least one projection image pair using an additional imaging unit, which simultaneously maps the object under inspection onto an additional probe plane. Here, the additional probe plane can be arranged non-parallel to the first and second probe planes. Furthermore, consistency conditions for determining the correction model can be additionally preset based on the at least one additional projection image. Alternatively or additionally, the resulting dataset can be additionally reconstructed from the at least one additional projection image.
[0063] The additional imaging unit may have all the features and characteristics described with respect to the first and / or second imaging unit, and vice versa. In particular, the additional imaging unit may include an additional source and an additional detector. The additional imaging unit may be arranged, in particular, movably fixed to the dual-plane imaging device, or may be arranged spaced apart from the dual-plane imaging device, in particular, independently movable. At least one additional projected image may have all the features and characteristics described with respect to the at least one first projected image and / or the at least one second projected image, and vice versa. Advantageously, at least one additional projected image may simultaneously map the object under inspection onto an additional detection plane with at least one first and at least one second projected image of a corresponding projected image pair. In particular, additional projected images may be recorded for multiple first projected images and multiple second projected images, which simultaneously map the object under inspection, in particular artifacts and / or motion, onto the additional detection plane. Advantageously, the additional detection plane is arranged neither parallel to the first detection plane nor parallel to the second detection plane.
[0064] By additionally mapping the object under inspection onto another probe plane, at least one pair of projected images can be expanded into a triplet of projected images. This advantageously improves the robustness of the pre-defined consistency conditions used to determine the correction model from at least one pair of projected images.
[0065] The resulting dataset can advantageously be reconstructed from at least one first projection image, at least one additional projection image, and based on a correction model. In particular, the resulting dataset can be reconstructed from at least one first projection image, at least one second projection image, and at least one additional projection image, and based on a correction model.
[0066] In a second aspect, the present invention relates to a medical dual-plane imaging device designed for implementing the proposed method for providing a result dataset.
[0067] The advantages of the proposed dual-plane imaging device essentially correspond to the advantages of the proposed method for providing the result dataset. The features, advantages, or alternative embodiments mentioned herein can also be applied to other claimed technical solutions, and vice versa.
[0068] A medical dual-plane imaging device may include a first imaging unit and a second imaging unit. The first imaging unit may include a first source and a first detector. Furthermore, the second imaging unit may include a second source and a second detector. The first imaging unit may be designed to record at least one first projected image of the object being examined. Furthermore, the second imaging unit may be designed to record at least one second projected image of the object being examined. Advantageously, the first and second imaging units may be, in particular, movable, especially rotatable, relative to the object being examined, in a robotic manner and / or in a defined arrangement relative to each other. Advantageously, the medical dual-plane imaging device may also have a display unit, such as a screen and / or monitor, designed to display a graphical view of the result dataset.
[0069] Advantageously, the medical dual-plane imaging device can have a first C-arm and a second C-arm, which can move in a coordinated manner and / or independently of each other. Here, a first source and a first detector can be arranged, in particular, in a defined arrangement relative to each other on the first C-arm. Furthermore, a second source and a second detector can be arranged, in particular, in a defined arrangement relative to each other on the second C-arm. The medical dual-plane imaging device can also be designed to allow the first C-arm to move, in particular, rotate, about a first isocenter and to allow the second C-arm to move, in particular, rotate, about a second isocenter. Here, the first and second isocenters can be the same or different.
[0070] In another advantageous embodiment of the proposed medical dual-plane imaging apparatus, the dual-plane imaging apparatus may include a first X-ray unit and a second X-ray unit. The first X-ray unit may include a first X-ray source and a first X-ray detector. Here, the first X-ray unit may be designed to record at least one first projected image of the subject being examined. The second X-ray unit may include a second X-ray source and a second X-ray detector. Here, the second X-ray unit may be designed to record at least one second projected image of the subject being examined.
[0071] In a third aspect, the present invention relates to a computer program product having a computer program that can be directly loaded into the memory of a providing unit, the computer program having program segments that, when executed by the providing unit, implement all steps of a method for providing a result dataset.
[0072] The present invention may also relate to a computer-readable storage medium storing program segments that can be read and executed by a providing unit, so that when the program segments are executed by the providing unit, all steps of the method for providing a result dataset are implemented.
[0073] The largely software-based implementation has the following advantages: currently used supply units can also be easily updated and modified via software to operate in accordance with the invention. In addition to the computer program, such a computer program product may, if necessary, include additional components, such as documentation and / or additional parts, and hardware components, such as hardware keys (dongles, etc.) for using the software. Attached Figure Description
[0074] Embodiments of the present invention are shown in the accompanying drawings and described in more detail below. In the different drawings, the same reference numerals are used for the same features. In the drawings:
[0075] Figures 1 to 3 Various advantageous implementations of the proposed method for providing the resulting dataset are illustrated schematically;
[0076] Figure 4 A schematic diagram of a medical dual-plane X-ray device is shown. Detailed Implementation
[0077] exist Figure 1 The diagram schematically illustrates an advantageous implementation of a proposed method for providing a PROV-ED results dataset ED. Here, at least one pair of projected images (PPs) of an ACQ-PP examination subject can be recorded using a medical dual-plane imaging device. The at least one pair of projected images (PPs) may include a first projected image (PP1) and a second projected image (PP2) of the subject, which simultaneously map the subject onto first and second probe planes. Here, the first and second probe planes are arranged non-parallel to each other. Furthermore, a DET-CM correction model (CM) can be determined to correct artifacts and / or motion simultaneously mapped in at least one first projected image (PP1) and at least one second projected image (PP2). Here, at least one pair of projected images (PPs) may pre-set consistency conditions for determining the DET-CM correction model (CM). Furthermore, the PROV-ED results dataset ED can be reconstructed from at least the at least one first projected image (PP1) and based on the correction model (CM). The PROV-ED results dataset ED can then be provided.
[0078] Advantageously, the medical dual-plane imaging device can be designed as a medical dual-plane X-ray device. Furthermore, artifacts may include scattered ray artifacts and / or radiation hardening artifacts. Here, at least one first projection image PP1 and at least one second projection image PP2 of at least one projection image pair PP can be paired and preset with consistency conditions for determining the DET-CM correction model CM.
[0079] Determining the DET-CM correction model CM may include optimizing the consistency index. Advantageously, the resulting dataset ED can be additionally reconstructed from at least one second projected image PP2 to reconstruct RECO-ED.
[0080] Advantageously, multiple projection image pairs PP of the object being examined can be recorded ACQ-PP around a common axis of rotation, particularly a common isocenter. Furthermore, the first and second detection planes of each projection image pair can have a constant angle relative to each other, for example, between 50° and 130°, particularly 90°. For example, multiple projection image pairs PP, particularly the first projection image PP1 and the second projection image PP2, can be recorded along equidistant projection directions within a preset projection angle range. For a total of N first projection images PP1 and second projection images PP2, N / 2 projection image pairs can be formed. The i-th of the N / 2 projection image pairs can include a projection image marked i as the first projection image and a projection image marked i+N / 2 as the second projection image. In this case, the first and second detection planes of the N / 2 projection image pairs can each have a constant angle relative to each other equal to half the projection angle range.
[0081] Figure 2 A schematic diagram illustrates another advantageous embodiment of the proposed method for providing a PROV-ED result dataset ED. Here, for the at least one projection image pair PP, at least one additional projection image PF is recorded using an additional imaging unit, which simultaneously maps the object under inspection onto an additional probe plane. Here, the additional probe plane is arranged non-parallel to the first and second probe planes. Advantageously, the consistency conditions for determining the DET-CM correction model CM can be additionally preset based on the at least one additional projection image PF. Alternatively or additionally, the result dataset ED can be additionally reconstructed from the at least one additional projection image PF to reconstruct RECO-ED.
[0082] exist Figure 3The diagram schematically illustrates another advantageous implementation of the proposed method for providing a PROV-ED results dataset (ED). Multiple projection image pairs (PPs) of the ACQ-PP examination subject can be recorded, these PPs mapping the motion of the examination subject in a time-resolved manner. Here, the multiple projection image pairs (PPs) can be pre-defined, at least along the time dimension, with consistency conditions for determining the DET-CM correction model (CM). Furthermore, RECO-MS physiological motion signals (MS) mapping the motion of the examination subject can be received. Advantageously, the determination of the correction model (CM), DET-CM, can be additionally based on the physiological motion signals (MS).
[0083] Advantageously, the first projected image PP1 can be recorded around a first isocenter (ACQ-PP), and the second projected image PP2 can be recorded around a second isocenter. Here, the first and second isocenters can be different.
[0084] Figure 4 A schematic diagram of a medical dual-plane X-ray device is shown, exemplarily representing a proposed medical dual-plane imaging device. The dual-plane X-ray device 7 includes a first X-ray unit 37.1 and a second X-ray unit 37.2. The first X-ray unit 37.1 includes, for example, a 6-axis articulated robot 16 on which a C-arm 38.1 is fixed, carrying a first X-ray source 33.1 and a first X-ray detector 34.1. In the illustrated embodiment, the second X-ray unit 37.2 includes a track-movable support 17 carrying the movable C-arm 38.2, on which the second X-ray source 33.2 and the second X-ray detector 34.2 are fixed.
[0085] The medical biplane X-ray device 7 may also include a providing unit PRVS. Advantageously, the medical biplane X-ray device 7 is designed to implement an embodiment of the proposed method for providing a PROV-ED result dataset ED.
[0086] To record at least one pair of projected images PP of the ACQ-PP, particularly at least one first projected image PP1 and at least one second projected image PP2, the C-arm 38.1 of the first X-ray unit 37.1 and the C-arm 38.2 of the second X-ray unit 37.2 can be movably supported about one or more axes, respectively. To record at least one pair of projected images PP of the examination subject 31 arranged on the patient support device 32, the providing unit PRVS can send signals 24.1 and 24.2 to the first X-ray source 33.1 and the second X-ray source 33.2. The first X-ray source 33.1 can then emit a first X-ray beam, particularly a conical beam and / or a fan-shaped beam and / or a parallel beam. When the first X-ray beam strikes the surface of the first X-ray detector 34.1 after interacting with the examination subject 31, the first X-ray detector 34.1 can provide signal 21.1 to the providing unit PRVS. Similarly, the second X-ray source 33.2 can emit a second X-ray beam. When the second X-ray beam strikes the surface of the second X-ray detector 34.2 after interacting with the object under inspection 31, the second X-ray detector 34.2 can send a second signal 21.2 to the providing unit PRVS. The providing unit PRVS can receive, for example, at least one pair of projected images PP, particularly at least one first projected image PP1 and at least one second projected image PP2, according to the second signal 21.1 and the second signal 21.2.
[0087] Here, the dual-plane X-ray device 7, especially the first X-ray unit 37.1 and the second X-ray unit 37.2 arranged at an angle to each other, enables the simultaneous recording of at least one first projection image PP1 and at least one second projection image PP2 from different projection directions, and in particular, the object to be inspected 31 is simultaneously mapped onto the first and second detection planes.
[0088] Furthermore, the medical biplane X-ray device 7 may include an input unit 42, such as a keyboard, and / or a display unit 41, such as a monitor and / or display. For example, in the case of a capacitive and / or resistive input display, the input unit 42 may preferably be integrated into the display unit 41. Here, control of the medical biplane X-ray device 7, and especially the proposed method for providing the PROV-ED result dataset ED, can be achieved through input by the user, especially a medical operator, on the input unit 42.
[0089] Furthermore, the display unit 41 can be designed to display information and / or a graphical view of information about the medical biplane X-ray device 7 and / or the providing unit PRVS and / or other components. For this purpose, the providing unit PRVS can, for example, send a signal 25 to the display unit 41. In particular, the display unit 41 can be designed to display a graphical view of the result dataset ED.
[0090] To reduce the radiation dose during recording at least one projection image of the ACQ-PP examination subject 31 for PP, at least one first projection image PP1 can be recorded with a higher X-ray dose than at least one second projection image PP2. The X-ray dose used to record at least one second projection image PP2 can here be lower than the X-ray dose required for diagnostic evaluation.
[0091] The schematic diagrams included in the accompanying drawings do not depict scale or size proportions in any way.
[0092] Finally, it should be reiterated that the methods and apparatus described in detail above are merely embodiments, and those skilled in the art can modify these embodiments in different ways without departing from the scope of the invention. Furthermore, the use of the indefinite article "a" does not preclude the existence of multiple related features. Similarly, the terms "unit" and "element" do not preclude the possibility that a related component consists of multiple sub-components that function together, and these sub-components may, if necessary, be spatially distributed.
Claims
1. A computer-implemented method for providing a result dataset, comprising: - Record multiple pairs of projected images of the subject (31) being examined using a medical dual-plane imaging device. The plurality of projection image pairs respectively have a first projection image and a second projection image of the object to be inspected (31), and the first and second projection images simultaneously map the object to be inspected (31) onto the first and second detection planes. The first and second detection planes are arranged non-parallel to each other. The plurality of projected images are time-resolvedly mapped to the motion of the object (31) under inspection. - Determine the correction model to correct artifacts and / or motion. The artifacts and / or motion are simultaneously mapped in the first and second projected images of each projection image pair. Among them, the plurality of projected image pairs are pre-defined with consistency conditions for determining the correction model at least along the time dimension. - The resulting dataset is reconstructed from at least one first projected image and based on a correction model. - Provides the results dataset.
2. The method according to claim 1, in, Receive physiological motion signals, which are mapped to the motion of the object being examined (31). The determination of the correction model is additionally based on physiological motion signals.
3. The method according to any one of the preceding claims, in, The medical dual-plane imaging device is designed as a medical dual-plane X-ray device (7).
4. The method according to claim 3, in, At least one first projection image is recorded at a higher X-ray dose than at least one second projection image.
5. The method according to any one of the preceding claims, in, The artifacts include scattered ray artifacts and / or ray hardening artifacts. In this process, at least one first projection image and at least one second projection image of at least one projection image pair are pre-set to determine the consistency conditions for the correction model.
6. The method according to any one of the preceding claims, in, The resulting dataset is additionally reconstructed from at least one second projected image.
7. The method according to any one of the preceding claims, in, Multiple projected images of the object (31) were recorded around a common axis of rotation.
8. The method according to any one of the preceding claims, in, Record multiple pairs of projected images of the object under inspection (31). In each pair of projected images, the first and second detection planes have a constant angle relative to each other.
9. The method according to claim 8, wherein, The first and second detection planes of each projection image pair are at an angle between 50° and 130° relative to each other.
10. The method according to claim 9, wherein, The first and second detection planes of each projection image pair are at a 90° angle relative to each other.
11. The method according to any one of the preceding claims, in, Multiple projected images of the object (31) were recorded around a common isocenter.
12. The method according to any one of claims 1 to 10, in, Record multiple pairs of projected images of the object under inspection (31). In this configuration, a first projected image is recorded around a first isocenter, and a second projected image is recorded around a second isocenter. The first-order centers and the second-order centers are different.
13. The method according to any one of the preceding claims, wherein, Determining the calibration model includes optimizing the consistency index.
14. The method according to any one of the preceding claims, in, For the at least one pair of projected images, at least one additional projected image is recorded using another imaging unit, which simultaneously maps the object under inspection (31) onto another detection plane. The additional detection plane is arranged non-parallel to the first and second detection planes. Additionally, consistency conditions for determining the correction model are preset based on the at least one additional projection image, and / or the result dataset is reconstructed from the at least one additional projection image.
15. A medical dual-plane imaging device designed for performing the method according to any one of the preceding claims.
16. The medical dual-plane imaging device according to claim 15, comprising a first X-ray unit (37.1) and a second X-ray unit (37.2), in, The first X-ray unit (37.1) includes a first X-ray source (33.1) and a first X-ray detector (34.1). The first X-ray unit (37.1) is designed to record at least one first projection image of the object under inspection (31). The second X-ray unit (37.2) includes a second X-ray source (33.2) and a second X-ray detector (34.2). The second X-ray unit (37.2) is designed to record at least one second projection image of the object under inspection (31).
17. A computer program product having a computer program that can be directly loaded into the memory of a providing unit, the computer program having program segments for implementing all steps of the method according to any one of claims 1 to 14 when the program segments are executed by the providing unit.