Generation of maxillofacial imaging volumes with motion

Abstract

This invention provides a system and method for generating a dynamic maxillofacial imaging volume from a three-dimensional (3D) extraoral X-ray system. [Solution] The method includes acquiring imaging data of the patient captured during the movement of the patient's maxilla and mandible, establishing a motion model that defines the expected movement of the maxilla relative to the mandible, segmenting the imaging data into at least a maxillary segment corresponding to the maxilla and a mandibular segment corresponding to the mandible based on the expected movement defined by the motion model, and outputting an imaging volume that includes artifact-corrected representations of the maxillary segment and artifact-corrected representations of the mandibular segment.

Description

TECHNICAL FIELD

[0001] This application claims the benefit of priority to U.S. Provisional Patent Application No. 63 / 770,626, filed on March 12, 2025, entitled "GENERATION OF MAXILLOFACIAL IMAGING VOLUME WITH MOTION", the entire contents of which are incorporated herein by reference. BACKGROUND ART

[0002] In some contexts, it may be necessary to capture and depict jaw movements in connection with imaging procedures for dental and maxillofacial diagnosis. For example, the patient's jaw movement and the range of motion experienced during mastication and occlusion may be relevant to the diagnosis and treatment of dental conditions.

[0003] Currently, several limited techniques are used to measure or reproduce jaw movement. One of the existing techniques includes static cone beam computed tomography (CBCT) depicting different positions of the patient's jaw, such as images captured by a first scan when the patient's jaw is open and a second scan when the patient's jaw is closed, and digital intraoral images (e.g., acquired with an intraoral optical scanner). This is combined with data captured by a mechanical articulator equipped with sensors for measuring movement and jaw position. Matching the movement observed by the mechanical articulator with the positions captured in digital imaging can be time-consuming and may give rise to numerous potential sources of error.

[0004] Another existing technique for replicating jaw movement involves the use of a virtual articulator, which includes software to process CBCT (and optionally optical scan images) to model the relationship between the maxilla and mandible and create a virtual representation that simulates each jaw movement. However, the use of a virtual articulator often involves manually modifying and defining the parameters of jaw movement in order to properly position and operate the jaw model in virtual space. As a result, the use of mechanical and virtual articulators often relies on human modification and may not fully model the accurate range of motion of the patient's temporomandibular joint.

[0005] In drawings that are not necessarily to scale, similar numbers may represent similar components in different drawings. Similar numbers with different suffixes may represent different examples of similar components. Drawings are generally illustrative, not restrictive, examples of the various embodiments described herein. [Brief explanation of the drawing]

[0006] [Figure 1] An example of a computerized extraoral X-ray system is shown. [Figure 2] A flowchart of an algorithm for capturing an image and outputting a representation of the patient's jaw, as in one example, is shown. [Figure 3] This shows the type of segmentation motion used in images of a patient's jaw movements, as in one example. [Figure 4] This shows an example of motion artifact correction operation used in images of a patient's jaw movement. [Figure 5] This shows an example of how to display a corrected representation of the jaw segment. [Figure 6] An example of jaw segmentation is shown. [Figure 7] A flowchart illustrating an example of a method for generating imaging volumes of the maxillofacial region with movement is shown. [Figure 8]A block diagram of an example computing system in which one or more of the techniques described herein may be performed is shown. [Modes for carrying out the invention]

[0007] The following concerns improvements in image processing techniques for capturing CBCT scans involving intentional movements, such as masticatory movements or jaw opening and closing movements. Specifically, the following approaches introduce techniques for generating motion-corrected 3D imaging volumes, while also performing functional evaluations of movements that occur during image acquisition. This includes correcting non-rigid movements through several rigid-body assumed movements of partial volumes, such as the respective volumes defined for the maxilla and mandible. As used herein, “maxilla” imaging refers to imaging of the maxilla and structures that are rigid in themselves and all rigidly connected, such as the upper dentition, crowns, and implants, and “mandible” imaging refers to imaging of the mandible and structures that are rigid in themselves and all rigidly connected, such as the lower dentition, crowns, and implants.

[0008] Motion occurring during X-ray or CBCT scans can result in numerous artifacts that degrade the quality of the underlying image, making accurate capture and reproduction of motion very difficult in many types of imaging procedures. According to the following approach, motion artifact correction (MAC) is performed in conjunction with the segmentation process, which separates the upper and lower parts of the jaw and converts each part into a subvolume assumed to be rigid (e.g., the respective volumes of the maxilla and mandible). These subvolumes, defining smaller portions of the overall scanned volume, are essentially treated as not deforming or changing shape during the image scan.

[0009] The segmentation process may be performed on a sinogram (e.g., a representation of raw projection images acquired during an X-ray or CBCT scan) or on a reconstructed volume (e.g., a 3D representation reconstructed from X-ray or CBCT projection images). Functional motion analysis may be performed during motion artifact correction, for example, with the use of a parameterized motion model, or functional motion analysis may be performed after the completion of motion artifact correction. As used herein, “motion model” refers to a computerized representation of jaw movement characteristics used to determine expected motion (e.g., how the patient’s jaw is expected to move during imaging) and estimated motion (e.g., how the patient’s jaw actually moved during imaging). This motion model can be instantiated with specific parameters (e.g., model values ​​or settings) and is useful for modeling the physical joint behavior for a particular patient based on incomplete information obtained from patient-specific imaging.

[0010] In some cases, parameterized motion models may correspond to articulator models. Relative (rigid) motion between the maxilla and mandible can be mathematically described by three translational parameters and three rotational parameters that change over time. However, due to anatomical constraints, this flexible jaw movement is impossible. Articulator models may limit the degrees of freedom of motion to rotation around an axis that may change position slightly while the jaw is open. Therefore, the parameterization of motion models described herein may include the angle of rotation around the axis and the position of the axis of rotation.

[0011] The motion models described herein can be implemented with different parameters to represent jaw movement. For example, the three implementations of the motion models shown below can be used individually or in combination, or each motion model can be converted to the others.

[0012] In the first implementation, the motion model includes the minimum and maximum jaw opening angles. This model describes the relationship between the opening angle and the relative position of the jaw. This model may include relative assumptions about the velocity of jaw movement (e.g., acceleration and deceleration during chewing) related to a series of opening states / jaw movements.

[0013] In the second implementation, the motion model is based on anatomical conditions such as jaw size or shape or other characteristics, and describes the possibilities of jaw movement.

[0014] In the third implementation, the motion model is given basic features, but it has numerous parameters. The position of each jaw can be described over time with rotation and translation. This may include three rotation parameters and three translation parameters (or fewer) per time point or per opening angle.

[0015] Various aspects of these motion models can be improved. When motion models are improved, parameters such as minimum or maximum opening angles, the relationship between opening angle and the relative position of the jaw, typical velocities, and anatomical conditions of the joints are updated. This improvement allows the motion model to better describe the estimated motion (e.g., with less error), and / or fit better with image data (e.g., with less error), and / or maintain a small deviation from the standard model, as the patient's anatomical structure and movement are expected to follow certain norms.

[0016] The approach described herein can be used to accurately capture and digitally reproduce imaging volumes that depict the unique movements of a patient's jaw. This may be useful in a variety of diagnostic and therapeutic situations, including, but not limited to, reviewing the temporomandibular joint in motion, creating occlusal splints, performing temporomandibular joint pain examinations, or verifying the opposing occlusion when creating dentures.

[0017] Figure 1 shows a computerized extraoral X-ray system 100 available for use in this technology. As shown in Figure 1, the X-ray system 100 includes an X-ray device 101 for patient imaging. Before exposure, the patient's head (not shown) is preferably positioned at the center of the X-ray device 101 by, for example, a head restraint 106. As will be described later, the X-ray device 101 includes an X-ray emitter 103 and an X-ray detector 104 that rotate around the patient's head to acquire X-ray images from different angles. The X-ray device 101 may include an operator bus 105, which includes a basic user interface or operator control (e.g., touchscreen, buttons) for activating and controlling various functions of the X-ray system 100. The X-ray system 100 is capable of generating 2D and 3D X-ray images, including the use of a CBCT X-ray beam, and the resulting images are reconstructed to generate a 3D volume of the patient's craniofacial region, particularly the maxillofacial region. As used herein, the term “imaging data” refers to projection images (e.g., generated from an X-ray detector), sinograms (e.g., generated from projection images), reconstructed imaging volumes (e.g., generated from a reconstruction process using projection images or sinograms), or any combination thereof.

[0018] As shown in Figure 1, the X-ray system 100 may operably connect the X-ray device 101 to another computing system 110 (or set of computers) and another output device 142 (e.g., a display device) for visualizing the dataset. The computing system 110 may be connected to the X-ray device 101 via a local network (not shown) or via the Internet. The computing system 110 may be part of a cloud computing environment. In other examples, the functions of the computing system 110 may be integrated into the X-ray device 101. Thus, the data processing operations described herein may be performed on the X-ray device 101, on the computing system 110, or within connected computing resources in the cloud.

[0019] For example, the computing system 110 may run various computer programs and manipulate the dataset to create different visualizations and representations of the imaging data. In some examples, the computing system 110 may provide commands for controlling the X-ray device 101 via the input device 144 or via program control. Alternatively, another computer may be used for image processing and device control operations. The imaging dataset generated herein may be presented to a clinician (e.g., physician, dentist, trained personnel, etc.) for visualization, particularly for diagnostic purposes, by using the output device 142 or by communication to another type of output device, such as a printer or screen.

[0020] The X-ray system 100 is adapted to perform dental imaging procedures for evaluation or diagnostic purposes, including generating imaging volumes of the maxillofacial region of a moving patient using techniques described later. As shown in Figure 1, the X-ray system 100 comprises an X-ray emitter 103. The X-ray device 101 further comprises an X-ray detector 104 for at least partially detecting the X-ray radiation emitted by the X-ray emitter during rotation.

[0021] The X-ray device 101 may further include an aperture mechanism (not shown) for collimating the X-ray radiation emitted by the X-ray emitter into each irradiation area. The X-ray device 101 may further include a control device or mechanism (not shown) for moving the X-ray emitter 103 and the X-ray detector 104, and for controlling the X-ray emitter and reading out a series of projected images of each irradiation area of ​​the X-ray detector 104 during rotation. These and other features may be controlled by onboard logic and processing circuits on the X-ray device 101, which in some examples may be controlled by an operator device 105.

[0022] The computing system 110 is shown as including processing circuitry 112, memory 114, and a storage device 116. The processing circuitry 112 may execute machine-readable software instructions (e.g., provided from the memory 114 or the storage device 116) to perform operations that implement CBCT image data processing 122, motion artifact correction processing 124, image segmentation processing 126, motion display processing 128, and a graphical user interface 140, as described later. The image processing methods described herein may be computer-implementable methods executable via the cooperation of the X-ray system 100 including the X-ray device 101, the computing system 110, or subsystems of both. Further implementation examples of the processing circuitry 112, the memory 114, and the storage device 116 will be described later with reference to Figure 8.

[0023] The image processing method performed by the computing system 110 may include CBCT image data processing 122 used to capture and process CBCT projection images, and reconstruct an image volume from the projection images. The motion artifact correction processing 124 is used to correct distortion caused by motion during the acquisition of CBCT imaging data, and is used in cooperation with the image segmentation processing 126 that provides segmentation and separation of rigid anatomical structures (e.g., the upper jaw and the lower jaw) within the imaging data. The motion display processing 128 is used to generate a display and representation of the patient's motion, such as animated motion between two segmented image volumes defined for the upper jaw and the lower jaw respectively. Further details regarding the creation of CBCT image data involving motion are described below in Figures 2 to 7. The results of the motion display processing 128 can be used to create a 3D or 4D imaging volume that is presented via the graphical user interface 140, such as an image display user interface that is controlled for example via the input device 144 and presented via the output device 142. Other image data processing functions of the computing system 110 may be invoked via the user interface 140 or the operator device 105.

[0024] The present image processing method for capturing and reconstructing an imaging volume of the jaw with intentional controlled movements provides significant advantages over existing approaches using mechanical or virtual articulators. In particular, the present image processing method enables motion artifact correction simultaneous with segmentation, which allows correction of non-rigid motion in CBCT imaging data before reconstructing the imaging volume. This enables simultaneous acquisition of static and dynamic CBCT imaging data, in addition to simultaneous functional acquisition and evaluation of jaw motion. Furthermore, the present approach can provide a simplified workflow and result in time savings, as there is no need to attach a mechanical actuator to the patient or to perform calibration and tracking within the system. Additionally, since there is no need to use an occlusal block to fix the patient's jaw, it is possible to fully simulate an accurate representation of the patient's occlusion within the 3D image volume.

[0025] Figure 2 shows a flowchart of an example algorithm for capturing images and outputting a representation of a patient's jaw. Although this flowchart shows a series of operations, some of the operations may be repeated, omitted, or performed in an alternative order.

[0026] Operation 201 involves acquiring CBCT projection imaging data while the patient's jaw is moving. For example, acquisition may involve the use of an extraoral X-ray unit, such as X-ray device 101, which acquires individual CBCT projection images while the X-ray emitter rotates around the patient's head while the patient's jaw is moving. Acquisition of projection imaging data may involve capturing multiple 2D projection images as the X-ray device 101 rotates multiple times around the patient's head while the patient continues jaw movement. In some examples, this may involve capturing CBCT images while rotating two or more times, depending on the scanning and rotation speed, the sampling rate of the projection images, and the maximum dose delivered to the patient from the X-rays. As is understood, the image quality of the resulting reconstructed image volume may be improved by increasing the number of projection images from the same geometric arrangement. In some examples, CBCT images are captured during half a rotation or less than two rotations, which may result in a decrease in the image quality of the resulting image volume, but this can be improved by further techniques such as sparse sampling or artificial intelligence.

[0027] Various techniques may be used to control or improve the acquisition of CBCT imaging data related to the patient's jaw movements. For example, the patient may be instructed to open and close their jaw a specific number of times, or to move it at a constant speed or rate. Such instructions may encourage the capture of redundant data used to improve the resulting CBCT imaging data and functional analysis of the movements. For example, instructions guiding the patient's jaw movements may be communicated to a human user using audible acoustic signals or displayed visual indicators (e.g., information displayed on operator device 105 or another screen) that include signals or indicators indicating when the patient should open and close their mouth, how far to open their jaw, and at what speed to open and close their jaw. Additional software-based measurements may be used to determine whether the patient's jaw movements are accurate, too slow, too fast, or have an incomplete range of motion. Measurements of the patient's jaw movements may be based on evaluation of data from X-ray projection images, X-ray sinograms, or external sensing devices (e.g., cameras that capture images or videos of the patient's jaw movements).

[0028] Operation 202 involves performing an initial reconstruction of the 3D image volume obtained from the acquired CBCT images (e.g., multiple projection images from a geometric arrangement compiled into a sinogram). In some examples, various image correction techniques may be used before or after reconstruction, such as dark-field correction, gain or flat-field correction, or defect correction (e.g., scatter removal, noise reduction, contrast enhancement). In other examples, such image correction techniques may be used after the application of segmentation and motion artifact correction (described later in operations 203, 204, and 205).

[0029] Operation 203 involves segmenting the mandible and maxilla within the reconstructed volume using an initial implementation of a motion model. Figure 3, described later, shows an example of segmentation techniques applied to projected images or reconstructed image voxels to perform and improve segmentation. When segmentation is applied to projected images, the motion model can provide information about where structures (e.g., mandibles) are expected to be located within the projected images, as the location of structures changes over time and between one projected image and the next. Figure 6, described later, also shows an example of a segmentation diagram from a diagram of a reconstructed CBCT volume, including segmentation of imaging data into a maxillary segment corresponding to the patient's maxilla and a mandibular segment corresponding to the patient's mandible. When segmentation is applied to the imaging volume, the motion model can provide information about where structures (e.g., mandibles) are expected to be located according to geometric conditions. However, initial implementations of motion models (e.g., motion models with time-varying parameters) can be used during the reconstruction of the imaging volume, which can help reduce motion artifacts in the volume and improve the segmentation results.

[0030] Various techniques can be used to determine or refine the parameters of a motion model. As a first example, the parameters of the motion model may be determined or modified from generally valid assumptions about the patient's anatomical structure, parameterized from one or more landmarks detected in the sinogram. As a second example, the parameters of the motion model may be determined or refined from the guided jaw movements performed by the patient. As a third example, the parameters of the motion model may be determined or refined based on a selection from several standard jaw movements. As a fourth example, the parameters of the motion model may be determined or refined based on measurements of the start and end points of the patient's movement, corresponding, for example, to a tracking system or associated sensors (e.g., sensors located in the occlusal block). As a result of this refinement of the motion model parameters, the quality of the reconstructed image can be improved by more accurate motion artifact correction and segmentation operations performed in the motion model, thus improving the display of the jaw.

[0031] The motion model used in action 203 preferably describes a series of movements of both jaws relative to each other. The motion model can be used to simulate typical jaw movements over time or to reproduce movements observed in image data. The motion model can also be used to describe the results of motion artifact correction, for example, by describing a series of movements related to the image data. Furthermore, the motion model can be improved to match movements or anatomical conditions observed in the image data.

[0032] Operation 204 includes performing motion artifact correction on the mandibular and maxillary segments generated from the segmentation of operation 203 individually. Figure 4, described later, shows an example of motion artifact correction techniques applied to the mandibular and maxillary segments. These motion artifact correction techniques may be modified or adapted based on the expected jaw movements indicated by the motion model. Also, as described later, motion artifact correction techniques may be performed individually, sequentially, or simultaneously on the partial volumes corresponding to each jaw segment. Motion artifact correction may focus on correcting the motion of rigid bodies.

[0033] In some cases, motion models can be used to describe motion within image data during motion artifact correction. This simplifies motion artifact correction by limiting the possible motions to a set of highly probable motions or ranges of motion. Therefore, motion models may be used to reduce the number of possible motions, reduce the number or complexity of parameters to optimize, or provide weights for determining the possible motions.

[0034] Operation 205 includes an optional step of adjusting the segmentation of the mandibular and maxillary segments based on the results of motion artifact correction. Operation 450 in Figure 4 provides an optional example of how the maxillary and / or mandibular segments are reconstructed into improved segments and motion artifact correction may be repeated on the improved segments as needed.

[0035] Operation 206 includes generating motion artifact-corrected representations of the mandibular and maxillary segments. The motion artifact-corrected representations of the mandibular and maxillary segments may be output in a new or updated imaging volume. Thus, the new or updated imaging volume can provide, in addition to motion information (e.g., changes over time recorded in four dimensions), artifact-corrected representations of the maxillary segment and artifact-corrected representations of the mandibular segment.

[0036] Operation 207 includes displaying corrected representations of the mandibular and maxillary segments, accompanied by motion (e.g., time-based playback or motion simulation). The display of such motion between the mandibular and maxillary segments may be improved as a result of refining the motion model parameters, as described above. Figure 5, described later, shows an example of a display technique for displaying the respective volumes corresponding to the mandibular and maxillary segments. In further examples, other information related to motion or structure may be displayed or output to the user, for example, in the display of the relative motion trajectory of the mandible to the maxilla.

[0037] The operations described above (202-207) refer to examples of motion artifact correction for reconstructed 3D image volumes, but other variations of segmentation and motion artifact correction may be based on projected image data. For example, a variation may include performing initial segmentation of the sinogram before generating initial subvolumes, and then performing motion artifact correction (and adjustments to the segmentation as needed) to provide motion estimates for each subvolume.

[0038] Figure 3 shows additional techniques implemented as part of action 203 for mandibular and maxillary segmentation in the reconstructed volume. These segmentation approaches are performed using initial implementations of motion models that define the expected movements and positions of the maxilla and mandible, respectively. Action 203 may provide alternative methods for performing the initial segmentation in actions 310, 320, and 330, respectively. It is understood that other variations may be used, including segmentation techniques, or improvements to segmentation techniques using any combination or modification of these actions.

[0039] Operation 310 provides a first approach to segmentation by determining continuous image regions of rigid structure motion applied to voxels of a reconstructed image volume. This approach may include detecting the displacement of a particular voxel's motion within a subvolume compared to other voxels (e.g., compared to a majority of other voxels). The segmentation performed on the reconstructed image volume is designed to compensate for distortion caused by motion. In one example, the segmentation of the image volume compensates for motion by first reconstructing the volume considering motion in the projection geometry, and then performing segmentation. If motion is not considered in the volume reconstruction, the volume will be blurred, severe motion artifacts will occur, and the quality of the resulting segmentation will be very poor (or impossible).

[0040] Operation 320 provides a second approach to segmentation by determining continuous regions of motion of a rigid structure based on pixels of one or more projected images. Such pixels can be assigned to several subvolumes. In some examples, the contour of a jaw can be segmented and then transformed for pixel segmentation. Also in some examples, landmarks detected in the projected images can be used as a guide for segmentation. It is understood that the initial reconstruction of the image volume (e.g., in operation 202) does not need to be performed before segmentation is performed on the projected images.

[0041] Operation 330 provides a third approach to performing semantic segmentation using one or more trained models (e.g., trained artificial intelligence models), such as neural networks. Semantic segmentation may be performed on voxels of an image volume or pixels of one or more projected images. As described above, the initial reconstruction of the image volume (e.g., in operation 202) does not need to be performed before segmentation is performed on the projected images.

[0042] The results of operations 310, 320, and 330 are used to generate initial segmentation of the imaging data (e.g., pixel or voxel segmentation) that defines segments corresponding to the maxilla and mandible. Operation 350 includes refining this initial segmentation of the imaging data based on the estimated anatomical state of the patient. Examples of estimated anatomical states may be based on known shapes or regions of the image (e.g., the upper region of the image represents the maxilla and the lower region of the image represents the mandible). The segmentation refined based on the estimated anatomical state can be further refined by operations 310, 320, 330, and 350. Operation 360 includes optionally refining the segmentation based on user input. The segmentation optionally refined based on user input can be further refined by operations 310, 320, 330, and 350. The resulting segmentation of the imaging data provides a starting point for motion artifact correction in operation 204.

[0043] Figure 4 shows additional techniques implemented as part of operation 204 for motion artifact correction on two partial volumes of the imaging data (for example, a first partial volume generated from the maxillary segment corresponding to the maxilla and a second partial volume generated from the mandibular segment corresponding to the mandible). Operation 204 may provide alternative methods for motion artifact correction in operations 420, 430, and 440, respectively. It is understood that combinations of correction techniques, or other modifications including improvements to correction techniques using combinations or modifications of these operations, may be used.

[0044] Firstly, in operation 410, motion estimates are generated, each having separate motion estimates corresponding to the mandibular segment and / or maxillary segment.

[0045] Operation 420 provides a first approach to motion artifact correction using motion estimates for independent correction of partial volumes of the imaging data (e.g., a first partial volume corresponding to the maxilla and a second partial volume corresponding to the mandible). Thus, this correction technique is based on independent estimations of two absolute motions of the maxillary and mandibular segments.

[0046] Operation 430 provides a second approach to motion artifact correction using motion estimates for continuous correction of partial volumes of imaging data. This may include correcting a first partial volume (e.g., corresponding to the maxilla) and then correcting a second partial volume (e.g., corresponding to the mandible). A similar approach may include correcting the partial volume corresponding to the mandible and then correcting the partial volume corresponding to the maxilla. Thus, this correction technique is based on estimation of relative motion between partial volumes.

[0047] Operation 440 provides a third approach to motion artifact correction using motion estimates in addition to motion models for correcting partial volumes of imaging data. Therefore, this correction technique is based on the functional dependencies of the partial volumes.

[0048] In further examples, specific metrics of motion artifact correction may be optimized in projected images or reconstructed imaging volumes. As is understood, motion artifact correction may alter the underlying shape or size of anatomical objects (e.g., maxilla or mandible) depicted in the reconstructed volume; therefore, various approaches to motion artifact correction may involve optimizing one or more metrics. This may necessitate iterative reconstruction of partial volumes as needed.

[0049] Therefore, the results of motion artifact correction for a partial volume (or multiple partial volume) can be used to optimize or improve the segmentation. Operation 450 provides an optional approach to reconstructing the lower and / or upper segments, such as by adjusting (e.g., improving) the segmentation of one or both segments. Additional motion artifact correction may then be performed by repeating operation 410 and motion correction operations (e.g., operations 420, 430, or 440).

[0050] Figure 5 shows additional techniques implemented as part of operation 207 to display artifact-corrected representations of the mandibular and maxillary segments in motion. Other types of outputs or measurements from the mandibular segments may be provided. Operation 207 may provide alternative methods for displaying the corrected representations in operations 510, 520, and 530, respectively. It is understood that other modifications, including combinations of display techniques, may be used.

[0051] In operation 510, all partial volumes are displayed as a single volume representation generated from artifact-corrected partial volumes. In other words, this combined volume is calculated from a combination of partial volumes of the maxillary segment and partial volumes of the mandibular segment.

[0052] In operation 520, partial volumes are displayed individually, for example, in representations generated from partial volumes of the maxillary segment or partial volumes of the mandibular segment.

[0053] In operation 530, an animated movement of a partial volume is displayed. This animated movement may be provided in conjunction with the display of several functional assessments of the jaw, and optionally in conjunction with the display of measurements.

[0054] In operation 540, various animated movements (e.g., movements over time) may be presented, either corresponding to movements initially captured from the patient or to some motion model. Variations of the display of jaw segments and reproduced jaw movements may be provided for specific dental or orthodontic applications, such as the creation of dental prostheses (e.g., to simulate how prostheses interact with jaw movements and occlusion) or the visualization of occlusal contact points (e.g., to simulate how teeth contact each other during occlusion).

[0055] Some variations of the display may include showing animated movement by indicating controlled jaw movement during 3D or 4D data visualization and manipulation, and by showing aspects of movement relative to the surface of anatomical structures. In addition to displaying animations of jaw movement, these displays may also show parameters or numerical measurements, such as general parameters related to occlusal contact amount or range of motion (e.g., the width of the patient's jaw opening). Such parameters may also be used to improve the polishing or mechanical optimization of crown shape or to finalize the design shape of crowns or other prostheses.

[0056] Figure 6 shows an example of segmentation of the jaw segment into mandibular segment 620 and maxillary segment 610. This figure shows how each segment can be defined based on the shape of the rigid anatomical structure initially captured from the X-ray data. Specifically, the segmentation generates the shape of maxillary segment 610 based on the shape of maxilla 601, and the mandibular segment 620 based on the shape of mandible 602.

[0057] Figure 6 also shows how the movement between rigid structures may be based on specific points or positions within the image volume. For example, the movement of the maxillary and mandibular segments may rotate around an axis located within region 630. This region 630 may correspond to the approximate location of the patient's temporomandibular joint. In some examples, the amount of movement of the maxillary and mandibular segments may correspond to the application of motion artifact correction, which generates the estimated movement of the corrected volume segments.

[0058] Other information related to jaw movement may support additional digital imaging or diagnostic applications. For example, after calculating the jaw movement of a particular patient, the jaw movement information can be transferred to, or combined with, a digital impression of the patient's dentition obtained by an optical scan (e.g., a surface scan performed on the teeth and oral cavity). Optical scans can provide high accuracy in a coordinate system correlated with aspects of X-ray imaging data. Therefore, extensions of the aforementioned techniques may include integrating motion and motion representation into optical scan representations and various other imaging and diagnostic applications.

[0059] Figure 7 shows a flowchart 700 of an example method for generating an imaging volume with captured motion, according to one example.

[0060] Operation 701 includes acquiring imaging data depicting the maxillofacial region of the patient while the patient's maxilla and mandible are moving. For example, the imaging data may be CBCT imaging data. For instance, the CBCT imaging data may consist of multiple CBCT projection images captured while the CBCT scanner rotates around the patient multiple times. Alternatively, the CBCT imaging data may consist of a three-dimensional imaging volume.

[0061] Operation 702 includes establishing a motion model that defines the expected movement of the maxilla relative to the mandible. This motion model may define the expected temporal positions of the maxilla and mandible, respectively, in response to the expected movement. The parameters of the motion model may be determined or modified based on one or more landmarks detected in the imaging data, the estimated maxillary and mandibular movements in motion artifact correction, the movement corresponding to guided jaw movement, the movement corresponding to standard jaw movement, one or more measurements from the tracking system, and / or user input related to the motion model. In a specific example, the expected movement of the maxilla relative to the mandible is determined or modified based on the estimated maxillary and mandibular movements in motion artifact correction.

[0062] Operation 703 includes segmenting the imaging data into maxillary and mandibular segments based on the expected motion defined by a motion model. In some examples, the motion model is used to establish the expected motion of the maxillary segment relative to the mandibular segment based on joints (e.g., temporomandibular joints) that are partially located in the maxillary segment and partially in the mandibular segment. In specific examples, the segmentation of the imaging data is based on segmentation of continuous image regions of rigid structures between voxels in the image volume of the imaging data if the image volume has been reconstructed with motion in mind, segmentation of continuous image regions of motion of rigid structures between pixels in the projected image of the imaging data, and / or semantic segmentation of the image volume or projected image if semantic segmentation is performed with at least one trained segmentation model. In specific examples, the segmentation of the imaging data is also refined based on the estimated anatomical state of the maxilla, the estimated anatomical state of the mandible, and / or user input related to segmentation.

[0063] Operation 704 includes performing motion artifact correction on the imaging data separately for the maxillary segment and the mandibular segment. This motion artifact correction may be based on estimated motion corresponding to the mandibular segment and / or maxillary segment, and includes independent motion artifact correction of the maxillary and mandibular segments based on estimated motion, sequential motion artifact correction based on estimated motion, correcting one of the maxillary or mandibular segments followed by the other of the maxillary or mandibular segment, and / or motion artifact correction of the maxillary and mandibular segments based on estimated motion and motion model.

[0064] Operation 705 includes outputting an imaging volume containing artifact-corrected representations of the maxillary segment and the mandibular segment. In some examples, this output includes displaying corrected representations of the maxillary and mandibular segments with motion, based on a motion model (for example, by applying the results of a functional analysis).

[0065] The techniques described herein can be used to generate or enhance many types of images, including 2D images, 3D images, 4D images (for example, dynamic CBCT images showing movement and changes over time as four dimensions), and related representations. Such imaging can be used to support dental treatments such as restorative treatment, root canal treatment, periodontal treatment, implant treatment, aligner treatment, and orthodontic treatment. Thus, such images can be used to generate therapeutic and diagnostic information related to teeth, tooth roots, nerve canals, temporomandibular joints, jawbones, etc. Furthermore, the embodiments described herein are not limited to CBCT imaging techniques but can also be applied to other forms of X-ray and radiographic imaging.

[0066] While the segmentation and artifact correction techniques described above were explained with reference to patient jaw movements, it is understood that other types of movements of rigid anatomical structures in CBCT data can be simulated and captured by variations of these techniques. For example, patient airway movements resulting from respiration can be identified and corrected by similar methods used to reconstruct X-ray images of rigid anatomical structures. Furthermore, the techniques described herein are not limited to dental and maxillofacial clinical settings, and other anatomical regions may be captured, segmented, corrected, and represented.

[0067] To ensure understanding, the embodiments described herein can be implemented using a single computer or using a computer system comprising multiple computers, each programmed with control logic to perform the various functions described above. Embodiments can be implemented in hardware (e.g., circuitry), firmware, software, or any combination thereof. Embodiments may also be implemented as instructions stored in a machine-readable medium that can be read and executed by one or more processors (e.g., an implementation of processor circuitry).

[0068] Figure 8 shows a block diagram of an example computing system 800 (or appropriate apparatus, device, system, or machine) in which one or more of the techniques (e.g., methods) described herein may be performed. In alternative embodiments, system 800 may operate as a standalone device or may be connected to other machines (e.g., network-connected). In a network-connected environment, system 800 may function as a server machine, a client machine, or both in a server-client network environment. System 800 may be a personal computer (PC), a tablet PC, a server computer, a mobile phone, a web appliance, a network-connected device, or any machine capable of executing instructions (sequential instructions or otherwise) that specify the actions to be performed by that machine. Furthermore, although only a single machine is illustrated, the term “machine” shall be considered to include any set of machines that individually or collectively execute a set (or more sets) of instructions to perform one or more of the methodologies described herein, such as cloud computing, software as a service (SaaS), or other computer cluster configurations.

[0069] System 800 may receive user commands by any of the following: input device 812, UI navigation device 814, display device 810, or signals from a remote system (e.g., received via network 826). System 800 may output an indication that an action has been performed by either the display device 810 or the signal generation device 818. System 800 may also provide an output of an imaging volume generated in a machine-readable medium 822 based on techniques for generating imaging volumes from artifact-corrected representations of the maxillary segment and artifact-corrected representations of the mandibular segment, as described herein. System 800 may be used to implement actions performed by methods for generating imaging volumes of the maxillofacial region (e.g., as described with reference to Figure 7), and related data processing actions relating to processing, reconstructing, segmenting, modeling, improving, representing, and visualizing the associated data.

[0070] System 800 (for example, a computer system) may include processing circuits such as a hardware processor 802 (for example, a central processing unit (CPU), a graphics processing unit (GPU), a hardware processor core, a field-programmable gate array (FPGA), or any combination thereof), main memory 804, and static memory 806, some or all of which may communicate with each other via an interconnection (for example, a bus) 830. System 800 may further include a display device 810, an input device 812 (for example, a keyboard or other alphanumeric input device), and a user interface (UI) navigation device 814 (for example, a mouse). As an example, the display device 810, the input device 812, and the UI navigation device 814 may be integrated into a touchscreen display. System 800 may further include a mass storage device 816 (for example, a drive unit or other similar storage device or unit), a signal generating device 818 (for example, a speaker), a network interface device 820 connected to a network 826, and one or more sensors 808. The system 800 may include an output controller 828 for communicating with or controlling one or more peripheral devices (e.g., a printer, a card reader, etc.) via, for example, a serial connection (e.g., Universal Serial Bus (USB)), a parallel connection, or other wired or wireless connections (e.g., infrared (IR), near-field communication (NFC), etc.).

[0071] The mass storage device 816 may include a machine-readable medium 822 containing one or more sets of data structures or instructions 824 (e.g., software) that embody or are used by any one or more of the technologies or functions described herein. The instructions 824 may reside in whole or at least partially in the main memory 804, static memory 806, or hardware processor 802 while the system 800 is executing the instructions. For example, one or a combination of the hardware processor 802, main memory 804, and static memory 806, or the mass storage device 816, may constitute the machine-readable medium.

[0072] Although the machine-readable medium 822 is illustrated as a single medium, the term “machine-readable medium” may include a single or multiple mediums (e.g., a centralized or distributed database, and / or associated caches and servers) configured to store one or more instructions 824. The term “machine-readable medium” may include any medium capable of storing, encoding, or carrying instructions for execution by System 800, causing System 800 to execute one or more of the technologies of this disclosure, or storing, encoding, and carrying data structures used by or associated with such instructions. Examples of non-limiting machine-readable mediums may include solid-state memory, and optical and magnetic media. Specific examples of machine-readable media (e.g., one or more non-temporary storage media) may include read-only memory (ROM), random access memory (RAM), non-volatile memory such as semiconductor memory devices (e.g., electrically programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM)) and flash memory devices, magnetic disks such as internal hard disks and removable disks, magneto-optical disks, and CD-ROM and DVD-ROM disks.

[0073] The examples described herein include, or operate by, a logic or a set of components or mechanisms, including circuits. A circuit includes a collection of circuits (e.g., simple circuits, gates, logic, etc.) implemented in a tangible entity, including hardware. The components of a circuit may be flexible to changes over time and to changes in the underlying hardware. A circuit includes components that can perform specified operations individually or in combination during operation. For example, the hardware of a circuit may be designed immutably (e.g., hardwired) to perform or implement a particular operation. For example, the hardware of a circuit may include variable-connected physical components (e.g., execution units, transistors, simple circuits, etc.) including a computer-readable medium that is physically modified (e.g., magnetically, electrically, or through the movable arrangement of invariant mass particles) to encode instructions for a particular operation. By instruction, the incorporated hardware (e.g., an execution unit or loading mechanism) can create components of the circuit within the hardware via variable connections to perform or implement a portion of a particular operation during operation. Thus, the computer-readable medium is communicatively coupled to other components of the circuit during the operation of the device. For example, any one of the physical components may be used in multiple components of multiple circuit units. For instance, under operation, an execution unit may be used in a first circuit of a first circuit unit at one point in time, and reused at a different point in time by a second circuit within the first circuit unit or a third circuit within the second circuit unit.

[0074] Therefore, embodiments can be implemented in hardware (e.g., circuitry), firmware, software, or any combination thereof. Embodiments may also be implemented as instructions stored in a machine-readable medium that can be read and executed by one or more processors. Furthermore, firmware, software, routines, and instructions may be described herein as performing specific operations. However, such descriptions are merely for convenience, and it should be understood that in practice, such operations are the result of computing devices, processors, controllers, or other devices that execute the firmware, software, routines, instructions, etc. Moreover, any of the implementation variations can be performed by a general-purpose computer.

[0075] The various components described herein may be referred to as “modules,” “units,” “devices,” “functions,” or similar terms. Such components may be implemented through any suitable combination of hardware and / or software components that are applicable and / or known to achieve their respective intended functions. This may include, in addition to or instead of, mechanical and / or electrical components, processors, processing circuits, or other suitable hardware components. Such components may be configured to operate independently or to execute instructions or computer programs stored in a suitable computer-readable medium. Regardless of the specific implementation, such components may be referred to herein as “circuits,” “controllers,” “processors,” or “processing circuits,” or by any alternative terms provided herein, where applicable and relevant.

[0076] Additional aspects and features of this disclosure are described in the following numbered examples.

[0077] Example 1 is a method for generating an imaging volume of the maxillofacial region, comprising: acquiring imaging data depicting the maxillofacial region of a patient during the movement of the patient's maxilla and mandible; establishing a motion model that defines the expected movement of the maxilla relative to the mandible; segmenting the imaging data into at least a maxillary segment corresponding to the maxilla and a mandibular segment corresponding to the mandible based on the expected movement defined by the motion model; and performing motion artifact correction on the imaging data, wherein the motion artifact correction is performed separately for the maxillary segment and the mandibular segment; and outputting an imaging volume that includes the artifact-corrected representation of the maxillary segment and the artifact-corrected representation of the mandibular segment.

[0078] In Example 2, the subject matter of Example 1 optionally includes the fact that the imaging data is cone-beam computed tomography (CBCT) imaging data.

[0079] In Example 3, the subject matter of Example 2 optionally includes the CBCT imaging data including multiple CBCT projection images captured while the CBCT scanner rotates multiple times around the patient.

[0080] In Example 4, one or more of the themes from Examples 2 to 3 optionally include the CBCT imaging data having a three-dimensional imaging volume.

[0081] In Example 5, one or more of the themes from Examples 1 to 4 optionally include the motion model defining the expected position over time for the maxilla and mandible, respectively, in response to the expected movement.

[0082] In Example 6, the subject matter of Example 5 optionally includes determining or improving the parameters of the motion model based on at least one of the following: one or more landmarks detected in the imaging data, maxillary and mandibular movements estimated in motion artifact correction, movements corresponding to guided jaw movements, movements corresponding to standard jaw movements, one or more measurements from the tracking system, or user input related to the motion model.

[0083] In Example 7, one or more of the themes from Examples 1 to 6 optionally include the fact that the expected motion is determined or improved based on the estimated maxillary and mandibular movements in motion artifact correction.

[0084] In Example 8, one or more thematic items from Examples 1 to 7 optionally include being used to establish the expected movement of the maxillary segment relative to the mandibular segment based on joints that are partly located in the maxillary segment and partly in the mandibular segment.

[0085] In Example 9, one or more of the themes from Examples 1 to 8 optionally include that the segmentation of the imaging data is based on segmentation of continuous image regions of rigid structures between voxels in an image volume of the imaging data that is reconstructed with motion in mind, segmentation of continuous image regions of motion of rigid structures between pixels in a projected image of the imaging data, or semantic segmentation of the imaging volume or projected image performed by at least one trained segmentation model.

[0086] In Example 10, one or more of the themes from Examples 1 to 9 optionally include that the segmentation of imaging data is improved based on at least one of the estimated anatomical state of the maxilla, the estimated anatomical state of the mandible, or user input related to segmentation.

[0087] In Example 11, one or more thematic features from Examples 1 to 10 are optionally included in that motion artifact correction is performed based on estimated motion corresponding to at least one of the mandibular or maxillary segments, and that independent motion artifact correction of the maxillary and mandibular segments is based on estimated motion, and sequential motion artifact correction correcting one of the maxillary or mandibular segments followed by the other is based on estimated motion, or that motion artifact correction of the maxillary and mandibular segments is based on estimated motion and motion model.

[0088] In Example 12, one or more thematic items from Examples 1 to 11 optionally include outputting an imaging volume that displays a corrected representation of the maxillary and mandibular segments with motion based on a motion model.

[0089] In Example 13, one or more of the themes from Examples 1 to 12 optionally include outputting auditory or visual instructions to guide the patient to move their maxilla or mandible during the acquisition of imaging data.

[0090] In Example 14, one or more of the themes from Examples 1 to 13 optionally include the imaging volume being a four-dimensional volume that depicts the movement of the maxilla and mandible over time.

[0091] Example 15 is at least one machine-readable medium that, when executed by the processing circuit, contains instructions causing the processing circuit to perform one of the methods of Examples 1 to 12 to generate an imaging volume of the maxillofacial region with motion.

[0092] Example 16 is an X-ray system comprising an X-ray source and an X-ray detector for acquiring imaging data from the maxillofacial region of a patient, acquired during the movement of the maxilla relative to the mandible of the patient, and a computing unit for performing any of the methods of Examples 1 to 12 to generate an imaging volume of the maxillofacial region in motion.

[0093] Example 17 is a computer system comprising a memory for storing imaging data from the maxillofacial region of a patient acquired during the movement of the maxilla relative to the mandible of the patient, and at least one processor configured to perform any of the methods of Examples 1 to 12 to generate an imaging volume of the maxillofacial region with movement.

[0094] The various embodiments described above are presented illustratively, not limitingly. Those skilled in the art will see that various modifications in form and detail (e.g., different hardware, communication protocols, etc.) are possible without departing from the scope of these embodiments. Furthermore, it should be understood that the accompanying drawings, which highlight the functions described herein, are presented as typical examples. The architecture of this embodiment is highly flexible and can be configured to allow for use and operation in ways other than those shown in the drawings.

Claims

1. A method for generating imaging volumes of the maxillofacial region, Acquiring imaging data depicting the maxillofacial region of the patient during the movement of the patient's maxilla and mandible, To establish a motion model that defines the expected movement of the maxilla relative to the mandible, Based on the predicted motion defined by the motion model, the imaging data is segmented into at least the maxillary segment corresponding to the maxilla and the mandibular segment corresponding to the mandible. The method involves performing motion artifact correction on the aforementioned imaging data, wherein the motion artifact correction is performed separately on the maxillary segment and the mandibular segment. Outputting an imaging volume that includes the artifact-corrected representation of the maxillary segment and the artifact-corrected representation of the mandibular segment. A method that includes this.

2. The method according to claim 1, wherein the imaging data is cone-beam computed tomography (CBCT) imaging data.

3. The method according to claim 2, wherein the CBCT imaging data includes a plurality of CBCT projection images captured while the CBCT scanner rotates multiple times around the patient.

4. The method according to claim 2, wherein the CBCT imaging data includes a three-dimensional imaging volume.

5. The method according to claim 1, wherein the motion model defines the expected position over time for each of the maxilla and the mandible in accordance with the expected movement.

6. The parameters of the aforementioned motion model are: One or more landmarks detected in the aforementioned imaging data, The estimated movements of the maxilla and mandible in motion artifact correction, Movements corresponding to guided jaw movements, Movements corresponding to standard jaw movements, One or more measurements from the tracking system, or User input related to the aforementioned motion model The method according to claim 5, which is determined or improved based on at least one of the following.

7. The method according to claim 1, wherein the expected movement is determined or improved based on the estimated movement of the maxilla and mandible in motion artifact correction.

8. The method according to claim 1, wherein the motion model is used to establish the expected movement of the maxillary segment relative to the mandibular segment based on joints that are partly located in the maxillary segment and partly located in the mandibular segment.

9. The segmentation of the aforementioned imaging data is as follows: Segmentation of continuous image regions of rigid structures between voxels within the image volume of the imaging data, which is reconstructed taking motion into consideration. Segmentation of continuous image regions of rigid body movement between pixels in the projected image of the aforementioned imaging data, or Semantic segmentation of the image volume or the projected image performed by at least one trained segmentation model. The method according to claim 1, based on the present invention.

10. The segmentation of the aforementioned imaging data is as follows: The estimated anatomical state of the maxilla, The estimated anatomical state of the mandible, or User input related to the aforementioned segmentation The method according to claim 1, improved upon based on at least one of the following.

11. The motion artifact correction is performed based on the estimated motion corresponding to at least one of the mandibular segment or the maxillary segment. The independent motion artifact correction of the maxillary segment and the mandibular segment is based on the estimated motion. The continuous motion artifact correction that corrects one of the maxillary segment or the mandibular segment followed by the other of the maxillary segment or the mandibular segment is based on the estimated motion, or The method according to claim 1, wherein the motion artifact correction of the maxillary segment and the mandibular segment is based on the estimated motion and the motion model.

12. The method according to claim 1, wherein outputting the imaging volume includes displaying a corrected representation of the moving maxillary segment and the mandibular segment based on the motion model.

13. The method according to claim 1, further comprising outputting auditory or visual instructions to guide the patient to move the maxilla or the mandible while acquiring the imaging data.

14. The method according to claim 1, wherein the imaging volume is a four-dimensional volume that depicts the movement of the maxilla and the mandible over time.

15. When executed by the processing circuit of a computer system, the processing circuit will Acquiring imaging data depicting the maxillofacial region of the patient during the movement of the patient's maxilla and mandible, To establish a motion model that defines the expected movement of the maxilla relative to the mandible, Based on the predicted motion defined by the motion model, the imaging data is segmented into at least the maxillary segment corresponding to the maxilla and the mandibular segment corresponding to the mandible. The method involves performing motion artifact correction on the aforementioned imaging data, wherein the motion artifact correction is performed separately on the maxillary segment and the mandibular segment. Outputting an imaging volume that includes the artifact-corrected representation of the maxillary segment and the artifact-corrected representation of the mandibular segment. A non-temporary machine-readable medium containing instructions for performing an action including the following.

16. The non-temporary machine-readable medium according to claim 15, wherein the imaging data is cone-beam computed tomography (CBCT) imaging data.

17. The CBCT imaging data mentioned above is Multiple CBCT projection images captured while the CBCT scanner rotates multiple times around the patient, or 3D imaging volume A non-temporary machine-readable medium according to claim 16, including the following:

18. A memory for storing imaging data from the maxillofacial region of the patient, acquired during the movement of the maxilla relative to the mandible of the patient, At least one processor, A motion model is established that defines the expected movement of the maxilla relative to the mandible. Based on the predicted motion defined by the motion model, the imaging data is segmented into at least the maxillary segment corresponding to the maxilla and the mandibular segment corresponding to the mandible. Motion artifact correction is performed on the aforementioned imaging data, and the motion artifact correction is performed separately for the maxillary segment and the mandibular segment. The system outputs an imaging volume that includes the artifact-corrected representation of the maxillary segment and the artifact-corrected representation of the mandibular segment. At least one processor configured as follows and A computer system equipped with the following features.

19. The computer system according to claim 18, wherein the imaging data is cone-beam computed tomography (CBCT) imaging data.

20. The CBCT imaging data mentioned above is Multiple CBCT projection images captured while the CBCT scanner rotates multiple times around the patient, or 3D imaging volume The computer system according to claim 19, including the computer system according to claim 19.

Images