Automatic detection of the optimal X-ray emitter position for mobile X-rays.

Abstract

A computer-implemented method for determining a position of an X-ray emitter of a mobile X-ray device to obtain high-quality mobile X-ray images is provided, the method comprising the steps of receiving a camera-acquired image resulting from a camera monitoring a patient during an X-ray imaging session, a reference X-ray image of an internal body structure to be examined in the X-ray examination, and X-ray detector position information, detecting and locating a plurality of anatomical landmarks in the camera-acquired image, determining a distance from the X-ray emitter to the patient based on the camera-acquired image, generating a virtual X-ray image of the internal body structure based on the detected plurality of anatomical landmarks, the distance from the X-ray emitter to the patient, and the X-ray detector position information, registering the virtual X-ray image of the internal body structure to the reference X-ray image of the internal body structure, and determining at least one parameter for adjusting the position of the X-ray emitter based on the registration result.

Description

Technical Field

[0001] The present invention relates to a computer-implemented method for determining the position of an X-ray emitter of a mobile X-ray device, an X-ray emitter positioning device, an X-ray imaging system, a computer program product, and a computer-readable medium.

Background Art

[0002] Mobile X-ray, also called portable X-ray, is an important imaging modality. It enables X-ray imaging in places where stationary X-ray imaging is not practical, such as in the emergency department (ED), intensive care unit (ICU), etc. Mobile X-ray mainly targets patients lying on a bed. In contrast to stationary X-ray, acquiring an X-ray image with portable X-ray is more difficult because it has a higher degree of freedom than a stationary X-ray image.

[0003] FIG. 1 shows the basic operating principle of a mobile X-ray device 10. First, a technician places an X-ray detector 12 behind the patient shown as the bed patient in FIG. 1. Next, the technician manually sets the position of the X-ray emitter 14 so that the desired field of view can be captured by the X-ray detector 12. Finally, the technician acquires an image by activating the X-ray emitter 14. There are many parameters that can significantly affect image quality, such as the position of the X-ray detector 12 (e.g., coordinates and rotation angle) and the position of the X-ray emitter (e.g., distance to the patient, rotation angle, and voltage).

Summary of the Invention

Problems to be Solved by the Invention

[0004] Therefore, it may be necessary to acquire higher-quality mobile X-ray images.

Means for Solving the Problems

[0005] The object of the present invention is solved by the subject matter of the independent claim, and further embodiments are incorporated into the dependent claims.

[0006] According to a first aspect of the present invention, a computer-aided method for determining the position of an X-ray emitter in a mobile X-ray apparatus is provided, and the method is a) (i) Camera-acquired images generated from a camera that monitors the patient during an X-ray imaging session, wherein the camera is registered to the origin of the X-ray emitter; (ii) Reference X-ray images of internal structures to be examined in the X-ray examination; and (iii) Position information of the X-ray detector. b) A step of detecting and localizing multiple anatomical landmarks in the camera-acquired image, c) A step of determining the distance from the X-ray emitter to the patient based on the camera image, d) A step of generating a virtual X-ray image of internal structures based on multiple detected anatomical landmarks, the distance from the X-ray emitter to the patient, and the position information of the X-ray detector, e) The step of aligning a virtual X-ray image of an internal structure with a reference X-ray image of the internal structure, f) A step of determining at least one parameter for adjusting the position of the X-ray emitter based on the alignment results, It holds.

[0007] In other words, this disclosure proposes a method for automatically determining the optimal X-ray emitter position for an image using either a reference X-ray image, such as a previous high-quality image from the same patient, or a reference image from an atlas that is deemed suitable for this imaging case. In the method disclosed herein, a reduction in the radiation dose received by the patient can be achieved by reducing the number of retakes. Furthermore, a reduced turnaround time can be achieved by avoiding the transmission of low-quality images to the PACS and their rejection during review. In addition, costs can be reduced as the number of job iterations is reduced.

[0008] This will be explained in more detail below, particularly with regard to the example shown in Figure 3.

[0009] According to one embodiment of the present invention, the step of generating a virtual X-ray image of an internal structure is further: The steps include generating a pseudo-density image of the patient based on multiple detected anatomical landmarks, The process involves projecting a pseudo-density image of the patient onto an X-ray detector based on the camera's position, the distance from the X-ray emitter to the patient, and the position information of the X-ray detector, which uses cone-beam projection to acquire a virtual X-ray image of the internal structure. It holds.

[0010] This will be explained in more detail below, particularly with regard to the example shown in Figure 5.

[0011] According to one embodiment of the present invention, the camera comprises a three-dimensional (3D) camera and / or a two-dimensional (2D) camera. The distance from the X-ray emitter to the patient is determined based on depth information in the camera-acquired image or a neural network-based depth estimation.

[0012] This will be explained in detail below, particularly with respect to step 230 shown in Figure 3.

[0013] According to one embodiment of the present invention, the step of aligning a virtual X-ray image of an internal structure with a reference X-ray image of the internal structure is further: A step of applying a pre-trained neural network to align a virtual X-ray image of an internal structure with a reference X-ray image of the internal structure, wherein the pre-trained neural network is trained to generate residual parameters for the camera position from the virtual X-ray image, the reference X-ray image, and multiple detected anatomical landmarks, and the residual parameters for the camera position are available to convert the virtual X-ray image to the reference X-ray image. It holds.

[0014] This will be described in detail below, and in particular with respect to the example shown in FIG. 3.

[0015] According to one embodiment of the present invention, at least one parameter for adjusting the position of the X-ray emitter is a parameter for adjusting the distance from the X-ray emitter to the patient, and a parameter for adjusting the rotation angle of the X-ray emitter, including one or more of. According to one embodiment of the present invention, the computer-implemented method further comprises providing an instruction signal to guide the user to manually adjust the position of the X-ray emitter based on at least one determined parameter, having.

[0016] According to one embodiment of the present invention, the computer-implemented method further comprises providing a control signal that can be used to control the mobile X-ray robotic arm of the mobile X-ray device to adjust the position of the X-ray emitter based on at least one determined parameter, having.

[0017] According to one embodiment of the present invention, the reference X-ray image is a previously acquired X-ray image of the internal structure of the patient, and a reference X-ray image from an atlas, including one or more of.

[0018] According to a second aspect of the present invention, there is provided an X-ray emitter positioning device having a processing unit configured to perform the steps of the method according to the first aspect and any related examples.

[0019] This will be described in detail below, and in particular with respect to the example shown in FIG. 2.

[0020] According to a third aspect of the present invention, A mobile X-ray device having an X-ray emitter, A camera that is attachable to the mobile X-ray device and is configured to capture a camera-acquired image from a patient in an X-ray imaging session, the camera being aligned with the origin of the X-ray emitter, the camera, An X-ray detector, A tracker device attached to or embedded in the X-ray detector and configured to provide position information of the X-ray detector, An X-ray emitter positioning device according to a second aspect and any related examples configured to determine at least one parameter for adjusting the position of the X-ray emitter, An X-ray imaging system having is provided.

[0021] This is described in detail below and is described particularly with respect to the example shown in FIG. 2.

[0022] According to an embodiment of the present invention, the tracker device has one or more of a marker device, a gyroscope, and an antenna.

[0023] According to an embodiment of the present invention, the mobile X-ray device further has a mobile X-ray robot arm that supports the X-ray emitter. The X-ray emitter positioning device is configured to provide a control signal for controlling the mobile X-ray robot arm of the mobile X-ray device so as to adjust the position of the X-ray emitter.

[0024] According to an embodiment of the present invention, the mobile X-ray device further has a display configured to display instructions provided by an X-ray emitter positioning device that guides the user to manually adjust the position of the X-ray emitter.

[0025] The predicted direction can be used in the X-ray device and displayed for manual correction by a technician, or automatic correction by the mobile X-ray robot arm (if present) can be realized.

[0026] According to another aspect of the present invention, a computer program product is provided which, when the program is executed by a processing unit, has instructions causing the processing unit to perform steps of the method disclosed herein.

[0027] A further aspect of the present invention provides a computer-readable medium on which a computer program product is stored.

[0028] It should be understood that all combinations of the aforementioned concepts and additional concepts discussed in more detail below (as long as such concepts do not contradict each other) are considered to be part of the subject matter of the invention disclosed herein. In particular, all combinations of claimed subject matter appearing at the end of this disclosure are considered to be part of the subject matter of the invention disclosed herein.

[0029] These and other aspects of the present invention will become apparent from and be explained with reference to the embodiments described below.

[0030] In drawings, the same reference numerals generally refer to the same part across different drawings. It should also be emphasized that the drawings are not necessarily to scale and instead generally illustrate the principles of the invention. [Brief explanation of the drawing]

[0031] [Figure 1] This shows the basic operating principle of a mobile X-ray device. [Figure 2] An exemplary X-ray imaging system is shown. [Figure 3] This flowchart illustrates a computer-based method for determining the position of the X-ray emitter in a mobile X-ray system. [Figure 4] This illustrates an exemplary alignment process. [Figure 5] A flowchart illustrating the generation of virtual X-ray images of internal body structures using the provided examples is shown. [Figure 6] Figure 2 illustrates the exemplary operating principle of the X-ray imaging system shown in the example. [Modes for carrying out the invention]

[0032] In clinical practice, mobile X-ray systems have limitations in image quality compared to stationary X-ray systems, meaning they can produce poor images, missed or cropped anatomical structures. Poor images can lead to re-imaging, which increases the patient's radiation dose.

[0033] For this purpose, methods, apparatus, and systems are provided to improve the image quality of mobile X-rays during the image acquisition phase and to overcome one or more of the problems described above. In particular, a system capable of automatically acquiring higher quality mobile X-ray images is proposed. The system disclosed herein may automatically locate the optimal X-ray emitter position for the image using either a previous high-quality image from the same patient or a reference image from an atlas deemed suitable for this imaging case.

[0034] Figure 2 shows an exemplary X-ray imaging system 100 according to one embodiment of the present invention. The X-ray imaging system 100 has a mobile X-ray apparatus 10 having an X-ray emitter 14. As shown in Figure 2, the mobile X-ray apparatus 10 further has a chassis 16 supporting an arm, for example, a robotic arm 18, and a system having wheels 20 for manual or electric movement, enabling the apparatus to be transported. The robotic arm 18 moves horizontally and / or vertically, and at its end, it may support a head assembly 22 in which the X-ray emitter 14 is located.

[0035] The X-ray imaging system 100 is attachable to the mobile X-ray apparatus 10 and further includes a camera 24 configured to capture camera-acquired images from a patient during an X-ray imaging session. The camera 24 is aligned to the origin of the X-ray emitter 14. For example, as shown in Figure 2, the camera 24 may be located in the head assembly 22. In some examples, the camera 24 may be an embedded camera. In some other examples, the camera 24 may be detachably mounted to the head assembly 22. In some examples, the camera 24 may be a two-dimensional (2D) camera configured to capture one scene by using one imaging lens and one image sensor. The resulting image is referred to as a 2D image. In some examples, the camera 24 may be a three-dimensional (3D) camera for capturing a 3D image. The 3D camera may be a range camera that produces a 2D image showing the distance from a specific point to a point in the scene. The 3D camera may be, for example, a stereo camera, which is a type of camera having two or more lenses, each having a separate image sensor or film frame.

[0036] The X-ray imaging system 100 shown in Figure 2 further includes a portable X-ray detector 12, which may be positioned behind the patient to measure the X-ray flux, spatial distribution, spectrum, and / or other properties. The portable X-ray detector 12 may be equipped with a tracker device 26, such as a gyroscope, antenna, marker, or other device that helps locate the position of the portable X-ray detector relative to the X-ray emitter 14.

[0037] The X-ray imaging system 100 shown in Figure 2 further comprises an X-ray emitter positioning device 30 configured to determine at least one parameter for adjusting the position of the X-ray emitter 14. Generally, the X-ray emitter positioning device 30 may have various physical and / or logical components that communicate and manipulate information, which can be implemented as desired for a given set of design parameters or performance constraints, as hardware components (e.g., computing devices, processors, logic devices), executable computer program instructions executed by various hardware components (e.g., firmware, software), or any combination thereof.

[0038] In some implementations, the X-ray emitter positioning device 30 may be embodied as a device or within a device, such as the mobile X-ray apparatus 10 or mobile device shown in Figure 2. The X-ray emitter positioning device 30 may include one or more microprocessors or computer processors that run appropriate software. The processing units of the apparatus 10 may be embodied by one or more of these processors. The software may be downloaded and / or stored in corresponding memory, for example, volatile memory such as RAM or non-volatile memory such as flash. The software may include instructions that configure one or more processors to perform the functions described herein.

[0039] It should be noted that the X-ray emitter positioning apparatus 30 may be implemented with or without a processor, and may be implemented as a combination of dedicated hardware that performs some functions and a processor (e.g., one or more programmed microprocessors and associated circuits) that performs other functions. For example, the functional units of the X-ray emitter positioning apparatus 30 may be implemented in the device or apparatus in the form of programmable logic, such as a field-programmable gate array (FPGA). In general, each functional unit of the apparatus may be implemented in the form of a circuit.

[0040] Figure 2 may show that the X-ray emitter positioning device 30 is implemented within the mobile X-ray apparatus 10, but it will be understood that in some implementations, the X-ray emitter positioning device 30 may be implemented as a mobile device, such as a tablet computer, or within one.

[0041] The X-ray emitter positioning apparatus 30 is configured to perform the method disclosed herein. The method is described below in detail, along with the flowchart shown in Figure 3.

[0042] Figure 3 shows a flowchart illustrating a computer implementation method 200 for determining the position of the X-ray emitter of a mobile X-ray apparatus. Method 200 can be implemented as a device, module, or related component within a set of logic instructions stored in a non-temporary mechanical or computer-readable storage medium such as random access memory (RAM), read-only memory (ROM), programmable ROM (PROM), firmware, flash memory, etc., in configurable logic such as programmable logic arrays (PLAs), field-programmable gate arrays (FPGAs), and composite programmable logic devices (CPLDs), in fixed-function hardware logic using circuit technologies such as application-specific integrated circuits (ASICs), complementary metal-oxide-semiconductor (CMOS), or transistor-transistor logic (TTL) technology, or as any combination thereof. For example, computer program code that performs the operations shown in Method 200 can be written in any combination of one or more programming languages, including symmetric programming languages ​​such as Java, Smalltalk, C++, and Python, and conventional procedural programming languages ​​such as the C programming language or similar programming languages. For example, the exemplary method can be implemented as the apparatus 30 shown in Figure 2.

[0043] In step 210, the method 200 includes receiving (i) a camera-acquired image generated from a camera that monitors the patient during an X-ray imaging session, wherein the camera is aligned to the origin of the X-ray emitter; (ii) a reference X-ray image of the internal structure being examined in the X-ray examination; and (iii) position information of the X-ray detector.

[0044] Camera-acquired images may originate from a camera 24, as shown in Figure 2, which monitors the patient during an X-ray imaging session. In some examples, camera 24 may be a video camera configured to transmit a real-time video stream to a device 30, which may have a video processing unit that processes the real-time video stream. In some examples, camera 24 may be a camera configured to capture images and transmit those images to the device 30, which may have an image processing unit that processes the images. In some examples, camera-acquired images may include one or more 3D images. In some examples, camera-acquired images may include one or more 2D images.

[0045] Camera 24 is aligned to the origin of the X-ray emitter 14. Figure 4 shows an exemplary alignment process. When an object is imaged, its representation is stored in a matrix of pixels that can be addressed by coordinates x, y. Typically, the origin (i.e., the 0,0 point) is located in the upper left corner of the matrix, with the x-axis running from left to right and the y-axis running from top to bottom. Unless the object and camera are rigidly mounted to each other, imaging an object twice results in two different matrices with two different coordinate systems. In the example in Figure 4, the patient's nose in image "a" is located lower and to the right than in image "b". It is possible to align the two images together to evaluate the coordinate transformation that allows one image to be transformed into the other. Since the imaging plane is perpendicular to the camera axis, the image transformation translates directly into the camera movement required to reacquire one image if the second image is available.

[0046] In this disclosure, the use of image alignment is proposed in two separate examples.

[0047] The transformation from the video camera to the emitter coordinate system is constant and can be explicitly expressed when designing the relative positions of two objects.

[0048] The transformation between emitter positions during two different imaging sessions can be more complex than the translation and rotation shown in Figure 4. Therefore, it is proposed to learn this transformation using a neural network, such as a convolutional network. This will be explained in detail below.

[0049] Reference X-ray images of internal structures examined in an X-ray examination can be downloaded from a picture archiving and communication system (PACS). The reference X-ray images may be good quality previous scans of the patient, or some good quality reference image or atlas. In some cases, the reference X-ray images may be obtained from different patients.

[0050] The position information of the X-ray detector 12 can be obtained from a tracker device 26 attached to the X-ray detector 12, such as a marker device, a gyroscope, an antenna, or any combination thereof.

[0051] In step 220, method 200 further comprises the step of detecting and localizing multiple anatomical landmarks in the camera-acquired images. In some examples, a patient model may be determined before acquisition to fit the imaging parameters to the patient's anatomical structure. The model may include the locations of anatomical landmarks such as the shoulders, pelvis, torso, and knees. Acquired surface images of the patient may be compared to a library of pre-modeled surface images to determine a model corresponding to the patient. This determination may be performed by a neural network trained on a library of pre-modeled surface images. In some examples, the locations of the head, shoulders, torso, knees, and ankles may be determined based on 2D or 3D images. The neural network may be trained to automatically detect landmarks. For example, supervised learning can be applied to anatomical landmark localization. See, for example, Gite, S., Mishra, A., and Kotecha, K. (2022). Enhanced lung image segmentation using deep learning. Neural Computing and Applications, 1-15.

[0052] In step 230, method 200 further includes the step of determining the distance from the X-ray emitter to the patient based on the camera-acquired image. If the camera is a 3D camera, the distance from the X-ray emitter to the patient may be determined based on the depth information of the camera-acquired image. If the camera is a 2D camera, the distance from the X-ray emitter to the patient may be determined based on neural network-based depth estimation. For example, self-supervised learning may be applied to distance or depth estimation. See, for example, Bian, J., Li, Z., Wang, N., Zhan, H., Shen, C., Cheng, MM, & Reid, I. (2019). Unsupervised scale-consistent depth and ego-motion learning from monocular video. Advances in neural information processing systems, 32.

[0053] In step 240, method 200 further comprises the step of generating a virtual X-ray image of an internal structure based on multiple detected anatomical landmarks, the distance from the X-ray emitter to the patient, and the position information of the X-ray detector.

[0054] Figure 5 shows a flowchart illustrating one embodiment of step 240.

[0055] In step 310 of step 240, a pseudo-density image of the patient is generated based on multiple detected anatomical landmarks.

[0056] In step 320 of step 240, a pseudodensity image of the patient is projected onto an X-ray detector based on the camera position of the camera, the distance from the X-ray emitter to the patient, and the position information of the X-ray detector, which uses cone beam projection to acquire a virtual X-ray image of the internal structure.

[0057] In other words, a virtual X-ray image can be generated in two steps. First, anatomical landmarks are used, for example, by a neural network to generate a pseudo-density image of the patient. Such a network can be trained using pairs of annotated images, e.g., CT images and corresponding masks. The neural network may be a convolutional network. Adversarial training can be used to train this model. See, for example, Park, T., Liu, MY, Wang, TC, & Zhu, JY (2019). Semantic image synthesis with spatially-adaptive normalization. In Proceedings of the IEEE / CVF conference on computer vision and pattern recognition (pp. 2337-2346). The resulting pseudo-density image is projected onto a detector, assuming the camera position, distance to the patient, and relative detector position using cone-beam projection. The resulting image is a virtual X-ray image.

[0058] Returning to Figure 3, in step 250, method 200 further includes the step of aligning a virtual X-ray image of the internal structure with a reference X-ray image of the internal structure.

[0059] In some examples, a pre-trained neural network can be applied to align virtual X-ray images of internal structures with reference X-ray images of those structures. The pre-trained neural network is trained to generate residual parameters for the camera's position from the virtual X-ray image, the reference X-ray image, and several detected anatomical landmarks. These residual parameters for the camera's position can be used to transform the virtual X-ray image to the reference X-ray image. For example, the neural network may be trained as follows: The neural network takes three inputs: a reference X-ray image, a virtual X-ray image, and an anatomical landmark map. The output of the model is residual parameters for the camera's position. Such parameters can be applied to transform the virtual image toward the reference image. The images required for training are artificially sampled from pseudo-CT projections using static detector parameters and a flexible emitter. The alignment results are used to predict the direction in which to adjust the emitter's position. The neural network may be a convolutional network. Self-supervised learning can be applied to train this model. See, e.g., Dalca, AV, Balakrishnan, G., Guttag, J., & Sabuncu, MR (2018, September). Unsupervised learning for fast probabilistic diffeomorphic registration. In International Conference on Medical Image Computing and Computer-Assisted Intervention (pp. 729-738). 88Springer, Cham.

[0060] In step 260, method 200 further includes the step of determining at least one parameter for adjusting the position of the X-ray emitter based on the alignment result. That is, the alignment result is used to predict the direction in which to adjust the position of the emitter. The at least one parameter may include one or more of the following: a parameter for adjusting the distance from the X-ray emitter to the patient and a parameter for adjusting the rotation angle of the X-ray emitter.

[0061] In some examples, the device 30 may provide command signals to guide the user to manually adjust the position of the X-ray emitter based on at least one determined parameter. In some examples, the command signals may be audio signals to guide the technician to manually correct the position of the X-ray emitter. In some examples, the command signals may be displayed commands used to guide the technician to manually correct the position of the X-ray emitter. In some other examples, the device 30 may provide control signals that can be used to control the mobile X-ray robot arm 18 of the mobile X-ray apparatus 10 to adjust the position of the X-ray emitter 14 based on at least one determined parameter.

[0062] Figure 6 shows an exemplary operating principle of the exemplary X-ray imaging system 100 shown in Figure 2.

[0063] The mobile X-ray device 10 is positioned in front of the patient (indicated by "a"). The X-ray detector 12 is positioned behind the patient.

[0064] The technician may download a reference X-ray image (indicated as "b") from the PACS. This may be a previous scan of good quality, or a reference image or atlas of good quality. The reference X-ray image is provided to the device 30. The camera 24 transmits camera-acquired images, such as a real-time video stream or images, to the device 30.

[0065] The device 30 may have a first neural network, also referred to as neural network A, to predict anatomical landmarks and the distance to the patient (indicated by "c") based on camera-acquired images in the patient video.

[0066] The apparatus 30 may have a second neural network, also called neural network B, to predict a virtual X-ray image using anatomical landmarks (indicated by "d"), an X-ray emitter (indicated by "f"), and a detector position (indicated by "e"). The virtual X-ray image is generated in two steps. 3D anatomical landmarks from neural network A are used by neural network B to generate a pseudo-density image of the patient. Such a network can be trained using annotated image pairs, e.g., CT images and corresponding masks. The resulting pseudo-density image is projected onto a detector, assuming the camera position, distance to the patient, and relative detector position using cone-beam projection. The resulting image is a virtual X-ray image.

[0067] The device 30 may have a third neural network, also called neural network C, to align a virtual X-ray image (indicated as "h") to a reference X-ray image (indicated as "i") using anatomical landmarks (indicated as "g"). Neural network C is trained as follows: It takes three inputs: a reference X-ray image, a virtual X-ray image, and an anatomical landmark map. The output of the model is a residual parameter of the camera position. Such parameters can be applied to transform the virtual image toward the reference image. The images required for training are artificially sampled from a pseudo-CT projection using static detector parameters and flexible emitter parameters. The alignment results are used to predict the direction of how to adjust the emitter position.

[0068] The predicted direction (indicated by "j") is used in the X-ray apparatus and can be displayed for manual correction by a technician, or automatic correction can be achieved by a mobile X-ray robotic arm (if one exists).

[0069] Methods, apparatus, and systems disclosed herein can automatically position the optimal X-ray emitter relative to the image using either a previous high-quality image from the same patient or a reference image from an atlas deemed suitable for the imaging case, thereby automatically acquiring higher-quality mobile X-ray images. Methods, apparatus, and systems disclosed herein can reduce the number of retakes and thus achieve a reduction in the radiation dose received by the patient. Methods, apparatus, and systems disclosed herein can avoid sending low-quality images to the PAC and rejecting them during review, thus achieving reduced turnaround time. Methods, apparatus, and systems disclosed herein can also reduce costs because multiple job iterations can be reduced.

[0070] In another exemplary embodiment of the present invention, a computer program or computer program element is provided, which is configured to perform a method step of a method according to one of the above embodiments on a suitable system.

[0071] Accordingly, the computer program elements may be stored in a computer unit which may be part of an embodiment of the present invention. This computing unit may be configured to perform or induce the execution of the steps of the method described above. Furthermore, it may be configured to operate the components of the apparatus described above. The computing unit may be configured to operate automatically and / or to perform user orders. The computer program may be loaded into the working memory of a data processor. Accordingly, the data processor may be equipped to perform the method of the present invention.

[0072] This exemplary embodiment of the present invention encompasses both computer programs that use the present invention from the outset and computer programs that modify an existing program to use the present invention through means of updating.

[0073] Furthermore, the computer program elements may provide all the steps necessary to satisfy the procedures of the exemplary embodiment of the method described above.

[0074] According to a further exemplary embodiment of the present invention, a computer-readable medium such as a CD-ROM is presented, and the computer-readable medium has computer program elements stored therein, which are described in the preceding section.

[0075] Computer programs may be stored and / or distributed on suitable media such as optical storage media or solid-state media supplied together with or as part of other hardware, but may also be distributed in other forms, such as via the Internet or other wired or wireless telecommunications systems.

[0076] However, computer programs may be presented on a network such as the World Wide Web and can be downloaded from such a network into the working memory of a data processor. According to a further exemplary embodiment of the present invention, a medium for making a computer program element available for download is provided, and this computer program element is configured to perform a method according to one of the aforementioned embodiments of the present invention.

[0077] It should be noted that embodiments of the present invention are described with reference to different subject matter. In particular, some embodiments are described with reference to method-type claims, and other embodiments are described with reference to apparatus-type claims. However, those skilled in the art will understand from the above and below descriptions that, unless otherwise notified, any combination of features belonging to one type of subject matter, as well as any combination of features relating to different subject matter, are also disclosed in this application. However, all features can be combined to provide a greater synergistic effect than the simple sum of the features.

[0078] Although the present invention has been illustrated and described in detail in the drawings and the foregoing description, such illustrations and descriptions should be considered illustrative or descriptive and not limiting. The present invention is not limited to the disclosed embodiments. Other variations of the disclosed embodiments can be understood and achieved by those skilled in the art in carrying out the claimed invention, based on an examination of the drawings, disclosure and dependent claims.

[0079] In the claims, the word “have” does not exclude other elements or steps, and the indefinite article “a” or “an” does not exclude plurality. A single processor or other unit may fulfill the functions of several items described in the claims. The mere fact that certain means are described in different dependent claims does not imply that combinations of these means cannot be used advantageously. No reference numeral in the claims should be construed as limiting in scope.

Claims

1. A computer-based method for determining the position of the X-ray emitter of a mobile X-ray apparatus, (i) Camera-acquired images generated from a camera that monitors the patient during an X-ray imaging session, wherein the camera is aligned to the origin of the X-ray emitter; (ii) Reference X-ray images of internal structures to be examined in the X-ray examination; and (iii) Position information of the X-ray detector. The steps include detecting and locating multiple anatomical landmarks in the camera-acquired image, The steps include determining the distance from the X-ray emitter to the patient based on the camera image, A step of generating a virtual X-ray image of the internal structure, A neural network generates a pseudo-density image of the patient based on the multiple anatomical landmarks detected, Based on the camera position of the camera, the distance from the X-ray emitter to the patient, and the position information of the X-ray detector that uses cone beam projection to acquire the virtual X-ray image of the internal structure, the pseudo-density image of the patient is projected onto the X-ray detector. The steps that are generated by, The steps include aligning the virtual X-ray image of the internal structure with the reference X-ray image of the internal structure, A step of determining at least one parameter for adjusting the position of the X-ray emitter based on the alignment result, A computer implementation method having

2. The aforementioned camera has a 3D camera and a 2D camera, The distance from the X-ray emitter to the patient is determined based on depth information in the camera-acquired image or a neural network-based depth estimation. The computer implementation method according to claim 1.

3. The step of aligning the virtual X-ray image of the internal structure with the reference X-ray image of the internal structure further includes: A step of applying a pre-trained neural network to align the virtual X-ray image of the internal structure with the reference X-ray image of the internal structure, wherein the pre-trained neural network is trained to generate residual parameters of the camera position of the camera from the virtual X-ray image, the reference X-ray image, and the detected anatomical landmarks, and the residual parameters of the camera position are available for converting the virtual X-ray image to the reference X-ray image. A computer implementation method according to claim 1, comprising:

4. The at least one parameter for adjusting the position of the X-ray emitter is, Parameters for adjusting the distance from the X-ray emitter to the patient, and A parameter for adjusting the rotation angle of the X-ray emitter, Includes one or more of the following: The computer implementation method according to claim 1.

5. A step of providing a command signal that guides the user to manually adjust the position of the X-ray emitter based on the at least one determined parameter, A computer implementation method according to claim 1, comprising:

6. A step of providing a control signal that can be used to control the mobile X-ray robot arm of the mobile X-ray apparatus to adjust the position of the X-ray emitter based on the at least one determined parameter, A computer implementation method according to claim 1, comprising:

7. The aforementioned reference X-ray image is, Previously acquired X-ray images of the internal structures of the patient, and Reference X-ray images from the atlas, Includes one or more of the following: The computer implementation method according to claim 1.

8. An X-ray emitter positioning apparatus having a processing unit configured to perform the steps of the method according to any one of claims 1 to 7.

9. A mobile X-ray device having an X-ray emitter, A camera that can be attached to the aforementioned mobile X-ray apparatus and is configured to capture camera-acquired images from a patient during an X-ray imaging session, wherein the camera is aligned to the origin of the X-ray emitter, X-ray detector and A tracker device that can be attached to or embedded in the X-ray detector and is configured to provide positional information of the X-ray detector, An X-ray emitter positioning device according to claim 8, configured to determine at least one parameter for adjusting the position of the X-ray emitter, An X-ray imaging system having the following features.

10. The aforementioned tracker device is Marker device and Gyroscope and, Antenna and, Includes one or more of the following: The X-ray imaging system according to claim 9.

11. The mobile X-ray apparatus further comprises a mobile X-ray robot arm that supports the X-ray emitter. The X-ray emitter positioning device is configured to provide control signals for controlling the mobile X-ray robot arm of the mobile X-ray apparatus to adjust the position of the X-ray emitter. The X-ray imaging system according to claim 9.

12. A display configured to show instructions provided by the X-ray emitter positioning device to guide the user to manually adjust the position of the X-ray emitter. The X-ray imaging system according to claim 9, further comprising the following:

13. A computer program having an instruction that, when executed by a processing unit, causes the processing unit to perform a step according to any one of claims 1 to 7.

14. A computer-readable medium storing the computer program described in claim 13.

Images