Radiology workflow

Abstract

Many potential interactions may occur during planning and execution of non-standard projections in radiology examinations, resulting in increased error risks and longer, more complex and expensive workflows. There is provided a method for use in a radiological workflow, the method comprising: in a planning step, enabling a first user to prepare an imaging examination request defining a non-standard projection of a radiographic image; a communication step in which the imaging inspection request is processed to generate a position instruction regarding how to implement the non-standard projection, the position instruction is output to the second user, and a radiographic image acquired using the non-standard projection is received; and in the verification step, determining whether the received radiographic image satisfies the imaging inspection request. Thus, the method facilitates the use of non-standard projection to plan and execute radiographic images while reducing the risk, complexity, duration and cost of workflow.

Description

Technical Field

[0001] This invention relates to methods and systems used in radiological workflows. Background Technology

[0002] Figure 1 The diagram illustrates a typical workflow 100 for radiographic imaging, which begins with a referring physician requesting X-ray images from the radiology department. Standard projections exist for all parts of the human anatomy, and these are well-documented. However, patient conditions sometimes require projections that do not conform to these standards. The referring physician may instruct the radiographer in the request how they should deviate from the nearest standard projection (e.g., for an X-ray of the ankle, “AP ankle mortise plus 10 degrees of internal rotation”). However, due to the number of degrees of freedom involved, such deviations can be difficult to describe clearly. Therefore, the referring physician may supervise the radiographer or even perform patient localization themselves. Figure 1 The diagram also illustrates the many potential interactions and negotiations that may occur during the planning and execution of non-standard projections (especially where explicitness is required), leading to increased risk of errors and longer, more complex, and more costly workflows. Summary of the Invention

[0003] It is desirable to provide a technical solution to solve one or more of the above-mentioned problems. Therefore, in a first aspect of the invention, a method for use in a radiological workflow is provided. The method may include: in a planning step, enabling a first user to prepare an imaging inspection request, the imaging inspection request defining a non-standard projection of a radiographic image. The method may include: in a communication step, processing the imaging inspection request to generate position instructions regarding how to implement the non-standard projection; outputting the position instructions to a second user; and receiving a radiographic image acquired using the non-standard projection. The method may include: in a verification step, determining whether the received radiographic image satisfies the imaging inspection request.

[0004] Therefore, the method facilitates the planning and execution of radiographic image acquisition using non-standard projections while reducing workflow risks, complexity, duration, and costs. When combined with patient positioning support technologies (e.g., using a 3D camera system), the method also benefits from the ability to acquire images using non-standard projections as directly as using standard projections. The method can be performed by planning tools as described herein, which can also be described as interactive 3D planning tools or virtual X-ray planning tools for non-standard X-ray projections.

[0005] As described above, an imaging examination request defines a non-standard projection for radiographic images. An imaging examination request may be simply referred to as an examination request. To define a non-standard projection, an examination request may specify values ​​for one or more patient position parameters relative to the projection geometry. The parameters of the projection may be specified in a manner that facilitates both the acquisition and verification of radiographic images. Parameters may be specified along with specific acceptance criteria to allow determination of whether the acquired images are valid, as described elsewhere herein. An examination request may specify a detector configuration, including the position and / or orientation of the detector. An examination request may specify a source configuration, including the position and / or orientation of the source. An examination request may specify an anatomical configuration, including the position and / or orientation of at least a portion of the anatomical structure to be imaged. In cases involving joints, the anatomical configuration may specify one or more hinge angles of the joint. An examination request may specify a beam configuration, including the position and / or orientation of the beam emanating from the detector. An examination request may specify a collimator configuration, which indicates one or more collimation parameters. An examination request may specify the intersection area between the beam and the anatomical structure. In any of these ways, a formal description of the acquisition may form an essential part of the imaging examination request.

[0006] A “non-standard” projection is one that deviates from at least one standard projection in at least one aspect, or more specifically, from every element of the standard projection set. A projection may also be referred to as a view. The standard projection set may be defined by a radiology manual or standard. An example manual defining an exemplary standard projection set is KATHY MQUILLEN-MARTENSEN, Radiographic Image Analysis, 5th Edition, April 18, 2018, ISBN: 9780323522816, Chapter I. Another exemplary non-exhaustive set of standard projections for the ankle joint includes AP (anteroposterior), PA (posterior-anterior), DP (dorsolateral), lateral (imaged with a central ray perpendicular to the midsagittal plane), oblique (imaged with a central ray at an angle to any body plane), flexion (the joint is imaged in flexion), extension (the joint is imaged in extension), prone (patient lying face down), supine (patient lying face up), cranial or cephalic (canal angle towards the head), and foot position (canal angle towards the foot). Another example of a standard projection that can be specified by a specific user or for a specific user is the internal standard projection in radiology, such as “Professor xxx - View”.

[0007] The first user to prepare an imaging examination request can be any user involved in planning image acquisition, such as a referring physician. The second user to receive location instructions can be any user involved in acquiring radiographic images, such as a radiology technician, radiologist, or radiology specialist.

[0008] It should be understood that patients can be humans or animals.

[0009] The method enables the preparation of imaging examination requests via a user interface, including defining non-standard projections for radiographic images. The user interface allows input of values ​​for parameters of the non-standard projection. The user interface can present a 3D model of the anatomical structure to be imaged. This 3D model can be patient-specific (e.g., the patient to be imaged) or a generic model. The 3D model allows for attitude manipulation of one or more of the anatomical structure of interest, detectors, sources, and collimators. The 3D model may include joint models. In this case, the joint models can be hinged to indicate the flexion of a specified joint. The user interface can indicate the non-standard projection, for example, by indicating its orientation or position and / or orientation relative to the 3D model of the anatomical structure. Similarly, the user interface can indicate one or more standard projections for reference. The user interface can indicate the position and / or orientation of the non-standard projection relative to one or more standard projections. Whether the projection is standard or non-standard, it can be indicated by means of lines, arrows, or beams (e.g., a central beam emanating from the source). The user interface can generate and display a synthetic preview radiographic image. The preview image can be generated using a model and optionally one or more physical assumptions. Preview images can show image features, such as outline highlighting (e.g., bone outline) and / or selected exemplary anatomical landmarks. For example, these landmarks might correspond to locations that can be easily palpated by a radiologist (e.g., the transverse ankle). Therefore, the proposed visualization supports radiologists in obtaining the necessary 3D understanding of bone geometry more quickly. Projected geometry can be used to generate projected outlines and / or landmarks. Different views of the model can be obtained via a user interface for individual or simultaneous display. Views can include 2D projections in a common anatomical plane, such as top, side, front, etc. The user interface can facilitate user interaction, enabling one or more interactive options, such as rotation, dragging, and zooming of virtual settings in 3D. Articulated 3D models can be constructed from MR or CT scans capturing a series of articulated states. Specialized models can be generated for gender or age.

[0010] The method may further include determining the feasibility of a requested non-standard projection by taking into account constraints imposed by the target imaging system. The imaging system may also be referred to as an image acquisition unit, etc. For this purpose, data defining such constraints for one or more imaging systems can be used. Exemplary constraints may relate to the mobility of one or more components of the imaging system (e.g., a source and / or a detector). Such constraints may relate to the degrees of freedom, range of motion, and / or size of such components. These constraints can be integrated into a 3D model to be expressed as limitations on the user's ability to reposition the imaging system using a user interface.

[0011] The method may include checking the desirability of a requested patient location. For example, guidance defining patient location or location ranges is available for a given examination type and categorized according to their desirability, such as acceptable, preferred, or optimal. These guidance may be provided by members of the radiology department or other experts. The guidance may be designed to avoid patient injury, pain, or discomfort. The method may include dynamically checking patient positioning and indicating the desirability of patient positioning during user manipulation of the 3D model. Similarly, the method may include seeking and obtaining manual confirmation of the requested patient positioning, such as from a radiology technician or radiologist, before completing the planning steps. The method may include storing such user input in guidance to be applied in future image planning. Such guidance may be used in conjunction with algorithms that automatically evaluate the positioning quality of radiographic images, such as SmartOrtho. The method may include using a user interface as described herein to define acceptable, preferred, or optimal patient positioning for a given examination type.

[0012] Once a non-standard projection is defined using the user interface, it can be stored in, for example, the patient's medical record for future reuse (e.g., in the case of a follow-up examination).

[0013] Positioning instructions enabling a second user to perform non-standard projections may include one or more patient positioning instructions. Patient positioning instructions may include instructions regarding patient orientation, such as whether the patient is prone, lying down, or in a left or right lateral position (supine, prone, etc.). Patient positioning instructions may include instructions or recommendations regarding the use of positional support materials (e.g., foam blocks, headrests, etc.). Positioning instructions may include instrument positioning instructions. Instrument positioning instructions may include instructions regarding how to position one or more components of the imaging system (such as the source (e.g., X-ray tube), detector, and collimator). Alternatively, positioning instructions may be referred to as guidance data.

[0014] Positioning instructions may include graphical instructions. Graphical instructions may show the requested patient positioning and / or the requested instrument positioning and / or the requested patient positioning relative to the imaging system. Graphical instructions may show one or more reference positions and / or orientations involved in the requested position and / or orientation. Graphical instructions may show one or more distances and / or angles between the requested position and / or orientation and the reference positions and / or orientations. Graphical instructions may include schematic representations of the patient (e.g., anatomical structures) and / or the imaging system or its components. Graphical instructions may be animated, for example, to indicate how (e.g., in which direction) the patient and / or imaging system should be moved to achieve a non-standard projection.

[0015] Location instructions can include text and / or audio instructions. Such instructions can be expressed in medical terminology.

[0016] The location commands described herein enable the requested non-standard location to be compared with the current actual patient location, thereby facilitating examination requests (by the referring physician).

[0017] Positioning instructions may include at least one visual representation of the current alignment and / or the target alignment. The visual representation may show the optical path generated by the alignment (e.g., an arrow). The visual representations of the current alignment and the target alignment may be displayed in a single window on the same screen or in separate windows. The visual representation of the current alignment and / or the target alignment can advantageously simplify the adjustment process from the current alignment to the target alignment because the user can see the difference on the screen. The current alignment can be determined using a patient positioning system known in the art, which may be camera-based. For example, the current alignment can be derived from at least one optical sensor and / or control signal of an imaging system. The optical sensor may measure one or more positions of the region of interest. Based on the measurement results, patient positioning data, such as angles, distances, etc., are derived. The control signal may indicate the position and / or orientation of one or more components of the imaging system (i.e., source and / or detector). This can be advantageous because each adjustment of the alignment is visible and used to improve the adjustment from the current alignment to the target alignment. The visual representation can be updated to reflect the current alignment. This is advantageous for improving the current alignment to fit the target alignment. The visual representation can be updated continuously or periodically.

[0018] Positioning instructions may include at least one visual representation of the region of interest (ROI). The visual representation may be an average representation of the ROI (e.g., a joint, such as the knee or ankle joint) (used for all patients), or a personalized representation of the ROI based on patient data. A visual representation of the ROI can advantageously simplify the adjustment process from the current alignment to the target alignment because the user can see on the screen how to adjust the position of the ROI, rather than relying on their imagination. Positioning instructions may also include a visual representation of the imaging system.

[0019] Positioning instructions can include absolute values. For example, an absolute value could be an angle displayed as a number and / or a table (e.g., a 10° difference between the current alignment and the target alignment). This reduces complexity and is beneficial for the user's adjustment process.

[0020] Positioning instructions may also include one or more sequence recommendations for adjusting patient and / or system positioning to achieve a non-standard projection. Sequence recommendations may involve the order in which patient body parts and / or imaging system components are moved or positioned. This is advantageous because parameters may interact adversely (e.g., when one positioning parameter has a target value, that parameter can constrain the adjustment of another positioning parameter).

[0021] Positioning instructions may include a preview image of the region of interest based on the current alignment. The preview image may include a simulated image representing what is acquired using the current alignment in actual image acquisition. The simulated image can be derived from a 3D model of the region of interest and the current alignment. In other words, based on the current alignment and the 3D model, a computational algorithm can calculate the forward projection for imaging in reality using the current alignment. This is advantageous for balancing whether the current alignment is sufficient. The method described herein is provided, wherein the preview image is preferably updated based on the current alignment. This is advantageous for improving the current alignment. The preview image can be adjusted continuously or periodically. This simplifies the preparation process and helps radiographers achieve target alignment. The simulated image representing the projection may be a pseudo-X-ray image generated from the 3D model using parallel or projected projections onto a virtual detector plane using the 3D model. This is advantageous for improving positioning data.

[0022] Positioning instructions may include palpation guidance from a radiologist. Palpation guidance may involve palpating one or more palpable bony landmarks, such as a region of interest, on the patient. To this end, the method may include obtaining positioning data for one or more anatomical landmarks, such as one or more palpable bony landmarks. This can be achieved by the radiologist palpating the region of interest with their hand while a positioning system measures the spatial position of the radiologist's fingertips to obtain positioning data for the palpable bony landmarks. Additionally or alternatively, the radiologist may utilize assistive devices, such as wearable tracking devices on the hand, wrist, or fingers (e.g., electromagnetic tracking devices or reflective markers for optical tracking), to obtain the positioning data. Positioning data may be tracked continuously or periodically. Positioning instructions may include one or more instructions on how to move the palpable bony landmarks to allow the patient and / or imaging system to move from their current system to a target position to achieve a specified non-standard projection.

[0023] As used herein, the term "tactile bone landmark" refers to bone landmarks that are typically obscured by the tissue and skin of an object, making them undetectable by direct detection—that is, unseen by, for example, a visual system (e.g., the human eye or a camera). Tactile bone landmarks can be suitable bone segments, joints, etc., such as the ankle, which may be obscured by excessive tissue in overweight individuals. Accordingly, the term palpation, as used herein, refers to tactile sampling. Palpation can be performed by a human (e.g., a radiographer) or by a soft tissue robot. Palpation can also be performed by a human using assistive devices (such as pointer devices).

[0024] Positioning instructions can include translation, rotation, or both, providing guidance to radiology technicians on how to achieve non-standard projections.

[0025] Generating position instructions may include using one or more generative models. These generative models may be configured to generate position instructions using schematic image generation (e.g., in the case of graphical instructions) and / or natural language processing (NLP, e.g., in the case of text and / or audio instructions). Additionally or alternatively, generating position instructions may include using aspects of the geometric specification of the projection to a pre-prepared database of position instructions.

[0026] Location instructions can be output to the second user using any suitable user interface technology (e.g., display devices and / or audio devices at the imaging site). It should be understood that the first and second users may not be in the same location. Therefore, the location instructions can be output as a data file and transmitted from the first user to the second user via their respective computing systems and interconnection networks (e.g., the Internet). The data can be stored in a computer-readable file format suitable for such transmission and ready for access by the second user. In one example, the second user could be a radiology technician. Location instructions can also be sent to other users, such as clinical stakeholders in the radiology value chain, such as radiologists and / or another physician. In other examples, the referring physician and the radiology technician are in the same location, and such transmission is not necessary.

[0027] In one example, where 3D reconstruction (e.g., from a CT or MR scan) exists for a specific patient, the output positional instructions may include (e.g., via 3D printing) generating a patient-specific projection around an anatomical structure of interest, wherein the projection is configured to provide one or more visual indications of the requested projection geometry. For example, the projection may be configured such that a proxy of the central beam (i.e., using a shutter or laser) is aligned with the projection in a predetermined manner to indicate the desired orientation.

[0028] In one example, radiographic images include X-ray images, but can also be images acquired using any imaging modality that employs planning and execution of a specific projection. Radiographic images can be stored in any suitable file format. In one example, the radiographic image is in the form of a DICOM image. Medical Digital Imaging and Communications (DICOM) is a standard for the communication and management of medical imaging information and related data. In cases where the first and second users are located in different locations, radiographic images can be received after transmission from another user (e.g., from a second user (e.g., a radiology technician) and / or from a third user (e.g., a radiologist)) to the first user. This transmission can be direct or indirect, for example, by passing through other stakeholders before returning to the first user. Radiographic images can be received as part of a radiology report, as is known in the art. Radiology reports can also include data indicating deviations from a standard projection. Such data can also be transmitted to radiologists to support the interpretation process and / or transmitted to other stakeholders in the clinical management process.

[0029] In the verification step, determining whether the received radiographic images meet the imaging inspection request may include comparing or cross-referencing the received images with reference data. The reference data may include reference images, such as synthetic preview images generated using a 3D model. Additionally or alternatively, the reference data may include non-image data defining non-standard projections, for example, for one or more attitude parameters. Comparing the received images with the reference data may include determining whether the received images meet one or more acceptance criteria used to determine the acceptability of deviations, which may include, for example, one or more ranges, tolerances, constraints, and / or rules. Acceptance criteria may be specified in the inspection request itself, or they may be stored separately, for example, in a model or database.

[0030] In one example, determining whether a received radiographic image meets an imaging inspection request may include performing a similarity check. Performing a similarity check may include calculating one or more similarity / distance metrics describing the similarity / distance between the received image and a reference image (e.g., a composite preview image). Performing a similarity check may include comparing one or more features (e.g., bone contours and / or landmarks) of the received image with corresponding features of the reference image; and calculating one or more similarity scores (e.g., contour similarity / distance score and / or landmark similarity / distance score). For this purpose, feature detection and / or image segmentation may be performed to identify features for the similarity check.

[0031] In another example, determining whether received radiographic images meet an imaging examination request may include, for example, using a model of the anatomical joint of interest to obtain an AI-based interpretation of the received images. For instance, a machine learning model may be trained to highlight deviations in the received images.

[0032] Determining whether a received radiographic image meets an imaging inspection request may include: deriving the attitude parameters of the projection used in the received image; and comparing the derived attitude parameters with the requested attitude parameters.

[0033] Determining whether a received radiographic image meets an imaging inspection request may include performing one or more verification checks. Verification checks may include one or more range checks to check that parameter values ​​fall within corresponding minimum-maximum ranges. Verification checks may include constraint checks to check that one or more constraints are met. Verification checks may include one or more rule-based checks to ensure that the received image conforms to one or more rules, requirements, or assumptions. Verification checks may include format checks to determine if the received image conforms to a required format. Verification checks may include one or more existence checks for the required image content.

[0034] In one example, deviations from a requested non-standard projection can be indicated by highlighting mismatched areas in the received image, such as when using a machine learning model to perform verification. Alternatively or additionally, verification can simply output a binary indication of success or failure for the overall image and / or individual acceptance criteria. Verification failure can trigger a request or suggestion to retake the image.

[0035] The first approach can be implemented using a computer.

[0036] According to the second aspect, a computing system is provided, which is configured to perform the method of the first aspect.

[0037] According to a third aspect, a computer program product including instructions is provided, which, when run by a computing system, enable the computing system to perform the method of the first aspect.

[0038] According to a fourth aspect, a computer-readable medium including instructions that, when run by a computing system, enable the computing system to perform the method of the first aspect is provided.

[0039] This invention may include one or more aspects, examples, or features in isolation or combination, whether or not specifically disclosed in that combination or isolation. Any optional feature or sub-aspect of one of the foregoing aspects is applicable to any other aspect.

[0040] These and other aspects of the invention will be apparent from the embodiments described below, and these and other aspects of the invention will be illustrated with reference to the embodiments described below. Attached Figure Description

[0041] A detailed description will now be given by way of example only, with reference to the accompanying drawings, in which:

[0042] Figure 1 The diagram illustrates a typical workflow for radiographic imaging examinations;

[0043] Figure 2 The illustration shows an improved workflow for performing radiographic imaging examinations using planning tools, according to this disclosure;

[0044] Figure 3 An example of a user interface for a planning tool used to define non-standard projections is shown;

[0045] Figure 4 The diagram illustrates the technical summary generated once a non-standard projection is specified;

[0046] Figure 5 The diagram illustrates the positional instructions for radiology technicians on how to achieve non-standard projections; and

[0047] Figure 6 The diagram illustrates a computing system that can be used based on the systems and methods disclosed herein. Detailed Implementation

[0048] Figure 2 The diagram illustrates an improved workflow 20 for performing radiographic imaging examinations using a planning tool 200, according to this disclosure. The planning tool 200 can be used with… Figure 1 The workflow 10 reduces communication to facilitate the planning, acquisition, and verification of non-standard radiographic images.

[0049] Using planning tool 200, the referring physician (e.g., an orthopedic surgeon) interactively configures a 3D model to indicate specific, non-standard projections of anatomical structures to be examined via radiography. Planning tool 200 allows the configuration to be visualized and simulated in 3D to produce pseudo or synthetic radiographic images. Once complete, the orientation parameters are transmitted to the radiology department, where they can be used to position the patient accordingly. After acquisition, planning tool 200 is again used to compare the actual radiographic images with the specified non-standard projections to verify that the referring physician's request has been met. This workflow 20 rationalizes communication between the referring physician and the radiographer through the use of precise and comprehensive specifications, 2D and 3D visualization support, and verification steps. Clear and easy communication becomes even more important, as is the case with radiology at the acquisition center, when the referring physician and radiographer are physically separated. Planning tool 200 uses outcomes to facilitate communication along the workflow to support information exchange. Outcomes in the non-limiting example shown include one or more of examination request 202, DICOM image 204, and radiology report 206. However, it should be understood that virtually all examination and imaging data, along with associated metadata, are potentially of interest (including, but not limited to, patient-specific data (e.g., age, sex, weight, body size, clinical status, pain, disability, prosthesis, etc.) or technical data (exposure time, kV, etc.). It may be useful that deviations from standard projections are relayed to the radiologist to support the interpretation process. These details can be obtained from the radiology report to support any further image evaluations conducted by the referring physician or other stakeholders during clinical management.

[0050] Figure 3 An example of a user interface 300 for a planning tool 200 used to define non-standard projections is shown. For defining non-standard projections, the user interface 300 allows the patient's position to be specified relative to X-ray geometry. Using the user interface 300, the referring physician can modify various parameters to specify the radiologist's projection. Specifically, 3D pose manipulation is possible using 3D models of the anatomy of interest 302 (ankle in this non-limiting example), detector 304, and source (not shown). The position and orientation of the source and detector 304 (including the central beam direction and the intersection of the central beam with the anatomy 302) can both be specified.

[0051] User interface 300 also allows specifying joint articulations 306 in a 3D model of anatomical structure 302. In this non-limiting example, the joint model can be articulated to indicate the flexion of the upper ankle joint. The articulation model used by planning tool 200 can be constructed from MR or CT scans capturing the range of articulation states. Specialized models can be generated for gender or age.

[0052] User interface 30 uses beam 308 to indicate a non-standard projection direction (center beam). This can be shown in conjunction with and relative to an established standard projection, which in this non-limiting example includes ankle lateral 310, ankle AP 312, and ankle mortise 314.

[0053] User interface 300 displays a synthetic X-ray preview image 316. The synthetic preview image 316 is generated using a 3D model and some physical assumptions (e.g., involving Beer-Lambert's law, statistical methods of ray convergence, and / or assumptions about material properties (e.g., bone density and attenuation)). In this example, the synthetic preview image shows a highlighted bone profile 318 and exemplary anatomical landmarks 320, such as key points of the bone. For example, these landmarks may correspond to locations that can be easily palpated by a radiologist (e.g., the lateral ankle). Projective geometry is used to generate the projected profiles or landmarks.

[0054] User interface 300 includes various elements (such as those shown at 322) that provide users with options for rotating, hinged, dragging, or scaling virtual settings in 3D.

[0055] Figure 4 The illustration shows the technical summary generated once a non-standard projection is specified. The requested specific projection can be summarized for a 2D projection in a common anatomical plane. In this non-limiting example, the technical summary includes a side view 402, a top view 404, and a front view 406 showing the non-standard projection, as well as a composite preview image 316.

[0056] Once the configuration is complete, positioning instructions for the radiographer are generated, for example, through speech processing or schematic image generation. These instructions enable the radiographer to perform non-standard projections. Positioning instructions can provide the radiographer with recommendations on how to position the patient (e.g., supine, prone, etc.), how to position support materials (e.g., foam blocks), and / or how to position the X-ray tube and detector.

[0057] Figure 5The illustration shows a non-limiting example of a position instruction 500 for a radiographer regarding how to implement a non-standard projection. The planning tool 200 can generate the position instruction using a database that links geometric projection specifications to a general guide for medical language or via a generative model such as Natural Language Processing (NLP). In this non-limiting example, the position instruction includes graphical instructions on how to position the ankle relative to the source and detector. In this example, the position instruction includes graphical instructions where the region of interest (in this case, the ankle) is schematically and image-based. The position instruction 500 includes a schematic side view (left image), an image side view (middle image), and a schematic rear view (right image), where all three views depict the ankle positioned on the detector. Dashed lines represent the central beam emanating from the source. In this example, the two side views instruct the radiographer to position the ankle in the middle of the detector in a target alignment with the target, which involves translation instructions for the radiographer. The rear view includes instructions to set the rotation angle to a recommended value of 30° relative to the radiation path. The ankle orientation is to be rotated 30° to achieve the specified non-standard projection.

[0058] Appropriate user interfaces and software, along with standardized views and descriptions, can be used to deliver location instructions to other clinical stakeholders in the radiology value chain, such as radiologists and other referring physicians.

[0059] After real X-ray images are acquired by a radiology technician, an automated review of the acquired images occurs. In one example, an AI-based interpretation of the image data, using a model of the anatomical joint of interest, is used to highlight mismatched areas on the image corresponding to deviations from the requested specific projection. An example of this AI-based interpretation technique is described in DOMINIK ET AL. An AI-based Framework for Diagnostic Quality Assessment of Ankle Radiographs. In Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, no. 143, pp. 484-496, PMLR, 7-9 July 2021. Based on clinical acceptance criteria, a request to re-encode the image can be suggested. In another example, a similarity score can be calculated using, for example, contour / landmark distance, by comparing image features (e.g., bone contours or landmarks) between a synthetic preview image and an actual acquired X-ray image.

[0060] Additionally or alternatively, one or more attitude estimation techniques can be used to derive the attitude parameters of the projection, which can be compared with the requested attitude parameters to detect deviations from non-standard projections and indicate, for example, positioning errors. In S... An example of a suitable pose estimation technique is described in CNN-based pose estimation for assessing quality of ankle-joint X-ray images. SPIE medical imaging conference. 2022. Paper 12032-37.

[0061] Many variations of the system and method described herein can be envisioned. For example, the planning tool 200 can be extended to examine the feasibility of a requested non-standard projection using constraints imposed by the desired target X-ray system. The specified non-standard projection can be stored in the patient's medical record (e.g., for follow-up). If the patient has a complex record of specific anatomical structures and a 3D reconstruction (from CT or MR) exists in advance, a patient-specific 3D-printed projection around the anatomical structure of interest can be generated and used for positioning the fixed joint. Specifically, a proxy for the central beam (i.e., using a shutter or laser) can be uniquely aligned with the projection to fix its attitude, similar to a guide light for maritime navigation or a precision approach path indicator (PAPI) for aviation navigation. In one such example, the projection may include two or more openings aligned such that the proxy beam passes through them. To achieve this, the attitude must be fixed in a unique manner. The openings are thus positioned and oriented to achieve the specified non-standard projection.

[0062] Some intelligent radiology assistants include algorithms (e.g., “SmartOrtho”) for automatically assessing the quality of positioning in musculoskeletal X-ray examinations. These algorithms measure 3D patient positioning parameters for a given image and compare their values ​​to clinically expected positioning parameters for a given examination type to assess positioning quality. The planning tool 200 described herein can be used as a graphical user interface for defining the expected optimal positioning for a given examination type (e.g., local head radiology). Given this optimal positioning, the planning tool 200 can display ray propagation and corresponding simulated X-ray images for deviations from the positioning (e.g., by initially rotating only around the leg axis). The user (such as a radiology specialist) can then indicate when the positioning quality becomes unacceptable. This user input can then be used to automatically configure the entire SmartOrtho algorithm to local preferences.

[0063] The methods and systems described in this paper can be applied in the fields of digital X-ray, bone radiography, image quality, patient positioning, pose estimation, intuitive user interfaces, and radiographer-guided learning.

[0064] Figure 6 An exemplary computing system 800 is illustrated for use with respect to the systems and methods disclosed herein. The computing system 800 may be part of or include any desktop computer, laptop computer, server, or cloud-based computing system. The computing system 800 includes at least one processor 802 that executes instructions stored in memory 804. For example, the instructions may be instructions for implementing functions described as being performed by one or more components described herein, or instructions for implementing one or more methods described herein. The processor 802 can access memory 804 via system bus 806. In addition to storing executable instructions, memory 804 may also store session inputs, scores assigned to session inputs, etc.

[0065] Additionally, the computing system 800 includes a data storage 808 accessible by the processor 802 via a system bus 806. The data storage 808 may include executable instructions, log data, etc. The computing system 800 also includes an input interface 810 that allows external devices to communicate with the computing system 800. For example, the input interface 810 may be used to receive instructions from external computer devices, users, etc. The computing system 800 also includes an output interface 812 that interfaces the computing system 800 with one or more external devices. For example, the computing system 800 may display text, images, etc., through the output interface 812.

[0066] It is conceivable that external devices communicating with the computing system 800 via input interface 810 and output interface 812 can be included in an environment that provides virtually any type of user interface to which the user can interact. Examples of user interface types include graphical user interfaces, natural user interfaces, and so on. For example, a graphical user interface can accept input from a user using an input device (such as a keyboard, mouse, remote control, etc.) and provide output on an output device (such as a display). Furthermore, a natural user interface allows the user to interact with the computing system 800 in a manner unconstrained by input devices (such as a keyboard, mouse, remote control, etc.). Instead, a natural user interface can rely on speech recognition, touch and stylus recognition, on-screen and near-screen gesture recognition, air gestures, head and eye tracking, voice and speech, vision, touch, gestures, machine intelligence, etc.

[0067] Furthermore, although shown as a single system, it should be understood that computing system 800 can be a distributed system. Therefore, for example, several devices can communicate via a network connection and collaboratively perform tasks described as being performed by computing system 800.

[0068] The various functions described herein can be implemented in hardware, software, or any combination thereof. If implemented in software, these functions can be stored or transmitted thereon as one or more instructions or code on a computer-readable medium. Computer-readable media include computer-readable storage media. A computer-readable storage medium can be any available storage medium that can be accessed by a computer. By way of example and not limitation, such computer-readable storage media can include flash storage media, RAM, ROM, EEPROM, CD-ROM or other optical disc storage devices, disk storage devices or other magnetic storage devices, or any other medium that can be used to carry or store desired program code in the form of instructions or data structures and that can be accessed by a computer. As used herein, disks and optical discs include compact discs (CDs), laser discs, optical discs, digital versatile optical discs (DVDs), floppy disks, and Blu-ray discs (BDs), wherein disks typically reproduce data magnetically, and optical discs typically reproduce data optically using lasers. Furthermore, the propagation of signals is not included in the scope of computer-readable storage media. Computer-readable media also include communication media, which includes any medium that facilitates the transfer of a computer program from one place to another. For example, a connection can be a communication medium. For example, if software is transmitted from a website, server, or other remote source using coaxial cable, fiber optic cable, twisted pair, digital subscriber line (DSL), or wireless technologies such as infrared, radio, and microwave, then coaxial cable, fiber optic cable, twisted pair, DSL, or wireless technologies such as infrared, radio, and microwave are included in the definition of communication media. Combinations of the above should also be included within the scope of computer-readable media.

[0069] Alternatively or additionally, the functions described herein may be performed at least in part by one or more hardware logic components. For example, and not limitingly, illustrative types of hardware logic components that may be used include field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), program-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and the like.

[0070] It should be understood that, in addition to the functions mentioned, the circuit described above may also have other functions, and these functions may be performed by the same circuit.

[0071] The applicant hereby discloses each individual feature described herein, as well as any combination of two or more such features, which, to a certain extent, can be performed based on this specification as a whole according to ordinary general knowledge of a person skilled in the art, regardless of whether such features or combinations of features solve any problem disclosed herein, and are not limited to the scope of the claims. The applicant indicates that aspects of the invention may include any such individual feature or combination of features.

[0072] It should be noted that embodiments of the invention are described with reference to different categories. Specifically, some examples are described with reference to methods, while other instances are described with reference to apparatus. However, those skilled in the art will infer from the description that, unless otherwise notified, any combination of features relating to different categories is also considered to be disclosed in this application, except for any combination of features belonging to one category. However, all features can be combined to provide more synergistic effects than a simple summary of the features.

[0073] Although the invention has been described and illustrated in detail with reference to the accompanying drawings and the foregoing description, such description and illustration should be considered exemplary rather than limiting. The invention is not limited to the disclosed embodiments. Other variations of the disclosed embodiments can be understood and implemented from a study of the drawings, this disclosure, and the appended claims.

[0074] The word "including" does not exclude other elements or steps.

[0075] The indefinite articles “one” or “a” do not exclude multiple. Furthermore, the articles “one” and “a” used in this article should generally be interpreted as meaning “one or more”, unless otherwise stated or clearly indicated from the context that a singular form is involved.

[0076] A single processor or other unit can perform the functions of several items recounted in the claims.

[0077] The fact that certain measures are described in mutually different dependent claims does not indicate that combinations of these measures cannot be used advantageously.

[0078] Computer programs can be stored / distributed on suitable media, such as optical storage media or solid-state media supplied together with or as part of other hardware, but can also be distributed in other forms, such as via the Internet or other wired or wireless communication systems.

[0079] Any reference numerals in the claims should not be construed as limiting the scope.

[0080] Unless otherwise stated or clearly understood from the context, as used herein, the phrases “one or more of A, B, and C,” “at least one of A, B, and C,” and “A, B, and / or C” are intended to refer to all possible permutations of one or more of the listed items. That is, any of the following instances satisfy the phrase “X includes A and / or B”: X includes A; X includes B; or X includes both A and B.

[0081] Depending on the specific implementation, the term "module" as used herein may be replaced by "system," "circuit," or "tool."

Claims

1. A computer-implemented method for use in a radiology workflow, the method comprising: In a planning step, enabling a first user to prepare an imaging examination request defining a non-standard projection of a radiographic image; In a communicating step, processing the imaging examination request to generate position instructions on how to implement the non-standard projection, outputting the position instructions to a second user, and receiving a radiographic image acquired using the non-standard projection; as well as In the verification step, it is determined whether the received radiographic image satisfies the imaging examination request.

2. The method according to claim 1, wherein: In the verification step, determining whether the received radiographic image satisfies the imaging examination request includes computing one or more similarity metrics describing similarity between the received image and a reference image.

3. The method according to claim 2, wherein: The reference image is a synthetic preview image generated during the planning step.

4. A method according to any preceding claim, wherein: In the verification step, determining whether the received radiographic image satisfies the imaging examination request includes: deriving posture parameters of the projection used in the received image; and comparing the derived posture parameters with the requested posture parameters.

5. A method according to any preceding claim, wherein: In the verification step, determining whether the received radiographic image satisfies the imaging examination request includes obtaining an AI-based interpretation of the received image, the AI-based interpretation of the received image highlighting mismatched areas in the received image.

6. A method according to any preceding claim, comprising: In the planning step, the imaging examination request can be prepared by means of a user interface presenting a 3D model of the anatomy to be imaged, wherein the 3D model enables pose manipulation of one or more of the anatomy of interest, a detector, and a source.

7. The method according to claim 6, wherein: The 3D model includes an articulated joint model, which can indicate the flexion degree of a specified joint.

8. The method according to any one of claims 6 or 7, wherein: The user interface indicates the non-standard projection relative to one or more standard projections.

9. The method according to any one of claims 6 to 8, wherein: The user interface can show a composite preview radiographic image.

10. A method according to any preceding claim, wherein: The position instructions include graphical instructions including a schematic representation of one or more of the anatomical structure of interest, the detector, and the source.

11. The method according to claim 10, wherein: The graphical instructions are animated to indicate how one or more of the anatomy of interest, the detector, and the source should be moved to achieve the non-standard projection.

12. A method according to any preceding claim, wherein: The location instructions include text instructions or audio instructions generated using natural language processing.

13. A method according to any preceding claim, wherein: Outputting the position instructions includes generating a patient-specific projection for surrounding the anatomical structure of interest, wherein the projection is configured to provide one or more visual indications of the non-standard projection.

14. A computing system configured to perform a method according to any preceding claim.

15. A computer readable medium comprising instructions which, when executed by a computing system, cause the computing system to perform the method according to any one of claims 1-13.

Images