Apparatus and method for selecting imaging view to optimize tracking volume detectability and model quality

Abstract

An image of a region of interest (ROI) is identified from a first perspective, the image having an uncertainty value associated with the ROI. A model is generated based on the image, the model including one or more parameters associated with the ROI. A plurality of tracking quality indicators is determined for a plurality of perspectives from which subsequent images of the ROI can be captured by the imaging system, the plurality of tracking quality indicators indicating a reduction in the uncertainty value associated with the ROI for the subsequent images. A second perspective of the plurality of perspectives is selected for the subsequent image based on a respective tracking quality indicator of the plurality of tracking quality indicators indicating a reduction in the uncertainty value for the second perspective.

Description

[0001] Related applications

[0002] This application claims the benefit of U.S. Patent Application No. 17 / 014,021, filed September 8, 2020, pursuant to 35 U.S. SC §119(e), the entire contents of which are incorporated herein by reference. Technical Field

[0003] This disclosure relates to selecting an imaging perspective to optimize tracking volume detectability and model quality. Background Technology

[0004] In radiotherapy, radiation delivery systems can utilize motion tracking to determine the correlation between direct measurements of the motion of the intended target or region of interest and the position of the tracked structure. This correlation is determined by fitting a motion model that records the target or region of interest and predicts its motion relative to the tracked structure. Summary of the Invention

[0005] This disclosure will be more fully understood from the following detailed description and from the accompanying drawings of various embodiments thereof. According to various aspects of this disclosure, an imaging perspective is selected to optimize tracking volume detectability and model quality, wherein processing logic may select a second perspective from multiple perspectives for capturing subsequent images based on a tracking quality metric. The processing logic may select the second perspective based on a corresponding tracking quality metric indicating that subsequent images from the second perspective will reduce the uncertainty value associated with the ROI. Selecting the second perspective ensures that subsequent images will reduce the location uncertainty of the ROI, prevents the capture of subsequent images that may not contain useful information (e.g., do not improve the location uncertainty of the ROI), and reduces or eliminates oversampling. By reducing or eliminating oversampling, the non-therapeutic imaging dose that the patient may be exposed to is also reduced, improving the performance of the radiation delivery system. Attached Figure Description

[0006] Figure 1A A helical radiation delivery system according to an embodiment described herein is shown.

[0007] Figure 1B A robotic radiotherapy system that can be used according to the embodiments described herein is shown.

[0008] Figure 1C A C-arm gantry-based radiotherapy system according to an embodiment described herein is shown.

[0009] Figure 2 A flowchart is depicted illustrating a method for optimizing tracking volume detectability and model quality by selecting an imaging viewpoint according to embodiments of the present disclosure.

[0010] Figure 3AThis is an illustration of an example of an imaging system with multiple viewpoints having a region of interest, according to embodiments of the present disclosure.

[0011] Figure 3B This is an illustration of an example of selecting a viewpoint for capturing subsequent images based on a tracking quality metric according to an embodiment of the present disclosure.

[0012] Figure 4 A flowchart is depicted illustrating a method for updating a model based on receiving subsequent images of the region of interest, according to embodiments of the present disclosure.

[0013] Figure 5 This is a block diagram of an example computing device that can perform one or more of the operations described herein, according to some embodiments. Detailed Implementation

[0014] This document describes embodiments of methods and apparatus for selecting imaging perspectives to optimize the detectability of the tracking volume and the quality of the model. Radiation delivery systems can utilize high-frequency online motion tracking. High-frequency online motion tracking depends on the correlation between the expected target motion and the position of the tracking structure (assuming it moves one in front of the target) using low-frequency but accurate and direct measurements (e.g., two-dimensional (2D) X-ray images) and high-frequency motion surrogates (e.g., tracking breathing or other quasi-periodic motions or motionless LED markers in the case of a assumed static target).

[0015] To obtain low-frequency direct measurements, radiation delivery systems may include imaging systems, such as kilovolt (kV) or magnetic resonance (MR) imaging systems, to capture 2D X-ray images of regions of interest (ROIs) that may include the tracked structure.

[0016] When acquiring these images, there is a trade-off between the amount of non-therapeutic imaging dose (and / or imaging time / frequency) and location uncertainty. For example, taking fewer images reduces the non-therapeutic imaging dose exposed to the patient but increases location uncertainty (hereinafter also referred to as "undersampling"). Conversely, taking a large number of images reduces location uncertainty but increases the non-therapeutic imaging dose exposed to the patient (hereinafter also referred to as "oversampling"). Therefore, it is important to balance these two considerations when determining when and / or from what perspective these images should be captured.

[0017] The aspects of this disclosure can remedy the above and other deficiencies by selecting an imaging viewpoint and / or time that optimizes the tracking volume and model quality. When an image of the ROI is acquired from a first viewpoint, the image may have an uncertainty value corresponding to the positional uncertainty associated with the ROI. For example, the uncertainty value may correspond to the positional uncertainty of the tracking structure or target within the ROI.

[0018] The processing logic of the processing device can generate a model associated with the Region of Interest (ROI) based on the image. This model may include one or more parameters associated with the ROI. In embodiments, the one or more parameters may be any feature or value that can affect the location uncertainty of the ROI. Examples of parameters associated with the ROI may include, but are not limited to, respiratory motion trajectories associated with the ROI, radiosensitive structures within the ROI, the visibility of the ROI, or the viewpoint of the ROI.

[0019] The processing logic can use this model to determine a tracking quality metric for multiple viewpoints from which subsequent images of the ROI can be captured. The tracking quality metric can indicate the reduction in uncertainty associated with the ROI by subsequent images captured from each viewpoint. The tracking quality metric can be used to identify one or more viewpoints with a high reduction in uncertainty relative to other viewpoints.

[0020] The processing logic can select a second perspective from multiple perspectives for capturing subsequent images based on a tracking quality metric. The selection of the second perspective is based on a corresponding tracking quality metric that indicates subsequent images from the second perspective will reduce the uncertainty value associated with the ROI. Selecting the second perspective ensures that subsequent images will reduce the location uncertainty of the ROI, prevents the capture of subsequent images that may not contain useful information (e.g., fail to improve the location uncertainty of the ROI), and reduces or eliminates oversampling. By reducing or eliminating oversampling, the non-therapeutic imaging dose that the patient may be exposed to is also reduced, improving the performance of the radiation delivery system.

[0021] Although embodiments of this disclosure are described in the context of radiation delivery systems, such description is provided for illustrative purposes only. Aspects of this disclosure can be utilized by any type of imaging system that benefits from optimized ROI imaging perspective selection. For example, aspects of this disclosure can be utilized by various types of surgical guidance systems that include imaging systems. Furthermore, while embodiments of this disclosure can be described using kV imaging systems, aspects of this disclosure can be utilized by other types of imaging systems, such as MR imaging systems or megavolt (MV) imaging systems.

[0022] In embodiments, aspects of this disclosure can provide an improved MR imaging system. Because MR imaging does not involve radiation dose, there is no need to reduce patient exposure to radiation when using an MR imaging system. However, aspects of this disclosure can be used to optimize the location, orientation, and / or timing of images of a region of interest (ROI) acquired using an MR imaging system. For example, an MR imaging system may not be fast enough to capture three-dimensional (3D) images to capture motion across the entire ROI. Instead of attempting to capture a 3D image of the ROI, multiple one-dimensional (1D) or two-dimensional MR images (which can be acquired faster than 3D images) can be acquired from various locations and orientations within the ROI. Aspects of the present invention can be used to optimize the location, orientation, and / or timing of these images to minimize the uncertainties associated with using an MR imaging system to track critical targets and sensitive structures within the ROI.

[0023] Figure 1A A helical radiation delivery system 800 according to an embodiment of the present disclosure is shown. The helical radiation delivery system 800 may include a linear accelerator (LINAC) 850 mounted on a ring gantry 820. The linear accelerator 850 can be used to generate a radiation beam (i.e., a treatment beam) by guiding an electron beam to an X-ray emission target. The treatment beam can deliver radiation to a target area (i.e., a tumor). The treatment system also includes a multi-leaf collimator (MLC) 860 coupled to the distal end of the linear accelerator 850. The multi-leaf collimator includes a housing housing a plurality of leaflets that are movable to adjust the aperture of the multi-leaf collimator to shape the treatment beam. In an embodiment, the multi-leaf collimator 860 may be a binary multi-leaf collimator comprising a plurality of leaflets disposed in two opposing groups, wherein the leaflets of the two opposing groups are staggered and openable or closed to form an aperture. In some embodiments, the multi-leaf collimator 860 may be an electromagnetically actuated multi-leaf collimator. In an embodiment, the multi-leaf collimator 860 may be any other type of multi-leaf collimator. The annular gantry 820 has an annular shape, through which the patient 830 passes. The linear accelerator 850 is mounted around the periphery of the annulus and rotates about an axis passing through its center to irradiate the target area with beams delivered from one or more angles around the patient. During treatment, the patient 830 can be simultaneously moved through the annulus of the gantry of the treatment bed 840.

[0024] The helical radiation delivery system 800 includes an imaging system comprising a linear accelerator 850 as an imaging source and an X-ray detector 870. The linear accelerator 850 can be used to generate a megavolt X-ray image (MVCT) of a region of interest (ROI) of a patient 830 by guiding a series of X-ray beams incident on the X-ray detector 870, opposite the linear accelerator 850, to the ROI to image the patient 830 for setting and generating a pre-treatment image. In one embodiment, the helical radiation delivery system 800 may further include a secondary imaging system comprising a kV imaging source 810 orthogonally (e.g., 90 degrees apart) mounted on a ring gantry 820 relative to the linear accelerator 850, and can be aligned to project an imaging X-ray beam onto the target region and, after passing through the patient 830, to illuminate the imaging plane of the detector.

[0025] Figure 1B A radiotherapy system 1200 that can be used according to the alternative embodiments described herein is shown. As shown in the figure... Figure 1B The configuration of a radiotherapy system 1200 is illustrated. In the illustrated embodiment, the radiotherapy system 1200 includes a linear accelerator (LINAC) 1201 serving as a radiotherapy source and a multi-leaf collimator 1205 coupled to the distal end of the linear accelerator 1201 to shape the treatment beam. In one embodiment, the linear accelerator 1201 is mounted on the end of a robotic arm 1202 having multiple (e.g., five or more) degrees of freedom to position the linear accelerator 1201 to irradiate pathological anatomical structures (e.g., targets) around the patient in multiple planes with operative amounts of beam delivered from multiple angles. Treatment may involve beam paths with a single isocenter, multiple isocenters, or a non-isocenter configuration.

[0026] During treatment, the linear accelerator 1201 can be positioned at multiple different nodes (predetermined locations where the linear accelerator 1201 stops and radiation can be delivered) via a movable robotic arm 1202. At each node, the linear accelerator 1201 can deliver one or more radiation therapy beams to the target, wherein the shape of the radiation beams is determined by the position of the blades of the multi-leaf collimator 1205. The nodes can be arranged in an approximately spherical distribution around the patient. The specific number of nodes and the number of treatment beams applied to each node may vary depending on the location and type of the pathological anatomy being treated.

[0027] In another embodiment, the robotic arm 1202 and the linear accelerator 1201 can move continuously between nodes at their ends while delivering radiation. During the continuous movement of the linear accelerator 1201, the shape of the radiation beam and a two-dimensional intensity map are determined by the rapid movement of the blades in the multi-leaf collimator 1205.

[0028] The radiotherapy system 1200 includes an imaging system 1210 having a processing device 1230 connected to X-ray sources 1203A and 1203B (i.e., imaging sources) and fixed X-ray detectors 1204A and 1204B. Alternatively, the X-ray sources 1203A, 1203B and / or the X-ray detectors 1204A, 1204B may be movable, in which case they can be repositioned to maintain alignment with the target, or alternatively, to image the target from different orientations, or to acquire multiple X-ray images and reconstruct three-dimensional (3D) cone-beam CT. In one embodiment, the X-ray source is not a point source, but an array of X-ray sources, as understood by those skilled in the art. In one embodiment, a linear accelerator 1201 is used as an imaging source, wherein the linear accelerator power level is reduced to a level acceptable for imaging.

[0029] Imaging system 1210 can perform computed tomography (CT) scans such as cone-beam CT or spiral megavolt computed tomography (MVCT), and the images generated by imaging system 1210 can be two-dimensional (2D) or three-dimensional (3D). Two X-ray sources 1203A and 1203B can be mounted in fixed positions on the ceiling of the operating room and can be aligned to project X-ray imaging beams from two different angular positions (e.g., 90 degrees apart) to intersect at a central point such as the machine (here referred to as the treatment center, which provides a reference point for positioning the patient on treatment bed 1206 during treatment) and irradiate the imaging planes of corresponding detectors 1204A and 1204B after passing through the patient. In one embodiment, imaging system 1210 provides stereoscopic imaging of the target and the surrounding volume of interest (VOI). In other embodiments, imaging system 1210 may include more or fewer than two X-ray sources and more or fewer than two detectors, and any of the detectors may be movable rather than fixed. In yet another embodiment, the positions of the X-ray sources and detectors may be interchanged. Detectors 1204A and 1204B can be made of a scintillation material (e.g., amorphous silicon) that converts X-rays into visible light, and a CMOS (complementary metal-oxide-semiconductor) or CCD (charge-coupled device) imaging unit that converts the light into a digital image. This digital image can be compared with a reference image during image registration processing, which transforms the digital image coordinate system into a reference image coordinate system, as is known to those skilled in the art. The reference image can, for example, be a digitally reconstructed radiographic image (DRR), which is a virtual X-ray image generated from a three-dimensional CT image based on simulating the X-ray image formation process by projecting rays through the CT image.

[0030] In one embodiment, the IGRT delivery system 1200 further includes a secondary imaging system 1239. The imaging system 1239 is a cone-beam computed tomography (CBCT) imaging system, such as the medPhoton imaging ring system. Alternatively, other types of volumetric imaging systems may be used. The secondary imaging system 1239 includes a rotatable gantry 1240 (e.g., a ring) connected to an arm and track system (not shown), which moves the rotatable gantry 1240 along one or more axes, such as an axis extending from the head to the tail of the treatment bed 1206. An imaging source 1245 and a detector 1250 are mounted to the rotatable gantry 1240. The rotatable gantry 1240 can rotate 360 ​​degrees about the axis extending from the head to the tail of the treatment bed. Therefore, the imaging source 1245 and the detector 1250 can be positioned at many different angles. In one embodiment, the imaging source 1245 is an X-ray source, and the detector 1250 is an X-ray detector. In one embodiment, the secondary imaging system 1239 includes two rings that can rotate independently. Imaging source 1245 can be mounted to the first ring, while detector 1250 can be mounted to the second ring. In one embodiment, rotatable gantry 1240 is rested at the tail of the treatment bed during radiotherapy delivery to avoid collision with robotic arm 1202.

[0031] like Figure 1B As shown, the image-guided radiotherapy system 1200 can also be associated with a treatment delivery workstation 150. The treatment delivery workstation can be located remotely from the radiotherapy system 1200 in a different room from the treatment room where the radiotherapy system 1200 and the patient are located. The treatment delivery workstation 150 may include processing devices (which may be processing device 1230 or other processing devices) and memory, which modify the treatment delivery to the patient 1225 based on the detection of target motion based on one or more image registrations described herein.

[0032] Figure 1CA C-arm radiation delivery system 1400 is illustrated. In one embodiment, in the C-arm system 1400, the beam energy of the linear accelerator can be adjusted during treatment, allowing the linear accelerator to be used for both X-ray imaging and radiotherapy. In another embodiment, system 1400 may include an onboard kV imaging system for generating X-ray images and a separate linear accelerator for generating a higher-energy therapeutic radiation beam. System 1400 includes a gantry 1410, a linear accelerator 1420, a multi-leaf collimator 1470 coupled to the distal end of the linear accelerator 1420 to shape the beam, and a field imaging detector 1450. The C-arm gantry 1410 can be rotated to a selected projection angle and is used to acquire X-ray images of the VOI of a patient 1430 on a treatment table 1440. In embodiments including a field imaging system, the linear accelerator 1420 can generate an X-ray beam that passes through a target of the patient 1430 and is incident on the field imaging detector 1450, thereby creating an X-ray image of the target. After generating an X-ray image of the target, the beam energy of the linear accelerator 1420 can be increased, allowing the linear accelerator 1420 to generate a radiation beam to treat the target area of ​​the patient 1430. In another embodiment, the kV imaging system can generate an X-ray beam through the target of the patient 1430 while simultaneously creating an X-ray image of the target. In some embodiments, the field imaging system can acquire field images during treatment delivery. The field imaging detector 1450 can measure the exit radiation flux after the beam has passed through the patient 1430. This allows internal or external reference points or anatomical blocks (e.g., tumors or bones) to be located within the field image.

[0033] Alternatively, the kV imaging source or field imager and operating method described herein can be used with other types of gantry-based systems. In some gantry-based systems, the gantry allows the kV imaging source and linear accelerator to rotate about an axis passing through an isocenter. Gantry-based systems include a ring-shaped gantry having a generally annular shape, with the patient's body extending through holes in the ring / annular structure, and the kV-level imaging source and linear accelerator mounted around the periphery of the ring and rotating about an axis passing through an isocenter. Gantry-based systems may also include a C-arm gantry, in which the kV-level imaging source and linear accelerator are cantilevered above and about an axis passing through an isocenter. In another embodiment, the kV-level imaging source and linear accelerator can be used in a robotic arm-based system comprising a robotic arm with the kV-level imaging source and linear accelerator mounted as described above. Aspects of this disclosure can also be used in other such systems, such as gantry-based linear accelerator systems, static imaging systems associated with radiotherapy and radiosurgery, proton therapy systems using integrated image guidance, interventional radiology and intraoperative X-ray imaging systems, etc.

[0034] Figure 2A flowchart depicts a method 200 for optimizing tracking volume detectability and model quality by selecting an imaging viewpoint according to embodiments of the present disclosure. Method 200 can be executed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, processor, processing device, central processing unit (CPU), system-on-a-chip (SoC), etc.), software (e.g., instructions that run / execute on the processing device), firmware (e.g., microcode), or a combination thereof. In embodiments, various portions of method 200 may be generated as previously described in... Figure 1A The processing logic of the processing device of the radiation delivery system described in -1C is executed.

[0035] refer to Figure 2 Method 200 illustrates example functionality used in various embodiments. Although specific functional blocks (“blocks”) are disclosed in method 200, such blocks are exemplary. That is, embodiments are well-suited for performing various other blocks or variations thereof described in method 200. It should be understood that the blocks in method 200 may be performed in a different order than presented, and not all blocks in method 200 may be performed.

[0036] Method 200 begins at block 210, where processing logic identifies an image of the ROI from a first-view perspective, having an uncertainty value associated with the ROI. In embodiments, the uncertainty value may correspond to location uncertainty associated with the ROI and / or one or more objects within the ROI. For example, the uncertainty value may correspond to location uncertainty associated with a tracking structure within the ROI.

[0037] In some embodiments, the image of the ROI can be acquired by an imaging system during a treatment phase administered by a radiation delivery system. In embodiments, the image of the ROI can be a previously acquired image. For example, the image could be an image of the ROI previously captured as part of a treatment planning phase.

[0038] At box 220, the processing logic generates a model based on the image that includes one or more parameters associated with the ROI. The one or more parameters can be any feature or value that can affect the location uncertainty of the ROI. In an embodiment, the parameters may include the respiratory motion trajectory of the patient associated with the ROI. The respiratory motion trajectory may describe the movement of the ROI during different phases of the patient's respiratory cycle and include the time of each phase of the respiratory cycle. In an embodiment, the parameters may include radiosensitive structures within or near the ROI. For example, the parameters may indicate whether the patient's rectum is within or near the ROI. In one embodiment, the parameters may include the visibility of the ROI by the imaging system. In an embodiment, the parameters may include the viewing angle of the ROI. In some embodiments, other parameters may be used in the model.

[0039] At box 230, the processing logic determines a tracking quality metric for the viewpoint from which subsequent images can be captured. The tracking quality metric indicates whether subsequent images captured from each viewpoint will reduce the uncertainty value of the image identified at box 210. In an embodiment, the tracking quality metric can be determined by examining higher-order terms of the Taylor series of the model-fit objective function L, as follows:

[0040]

[0041] in:

[0042] t is the time taken during the measurement. For rotating gantry systems, this time can also specify the position of the imaging system and patient support relative to the treatment beam.

[0043] It is a set of measurements of the tracking structure or the detection or candidate location of the tracking structure in a two-dimensional X-ray image.

[0044] High-frequency substitute motion data

[0045] It is a motion model that converts high-frequency data into the position of the tracking structure.

[0046] These are the parameters of the motion model.

[0047] This is a projection operator in X-ray images that associates 3D data with its corresponding 2D data. For points, this is the view projection matrix. For image data, this is a DRR generator.

[0048] It measures 2D measurement data according to the model. A distance function relating the similarity to the tracked structure at its predicted 2D location. In embodiments, the distance function may reflect the amplification of errors due to the 3D position of the tracked structure relative to the imaging geometry. In some embodiments, the distance function may reflect prior uncertainty in the measurement. In one embodiment, the distance function may reflect the age-related low weighting of older measurements. In embodiments, the distance function may reflect the spatial probability distribution of candidate tracked structures in image matching, weighting the error vector based on the quality of the 2D image data. In some embodiments, the distance function may reflect other factors reflecting the statistical quality of the model as an estimate of its ability to describe the motion of the tracked structure.

[0049] The value of the objective function for model fitting can be used to evaluate the model quality for the ROI, where the objective function... A smaller value indicates a better model fit to the ROI. However, this does not necessarily describe the quality of the model when predicting motion associated with the ROI. For example, this may not describe how the model predicts respiratory motion of a patient associated with the ROI.

[0050] Additional measurements of the quality of the model fit can be obtained by examining relative to... At the minimum value The shape is obtained from the optimal model parameters. Given. Consider. about Taylor series expansion:

[0051]

[0052] The first term is the value of the objective function at its minimum, the second term is the gradient of the objective function, and the third term is the Hessian matrix—the matrix of the second derivatives of the objective function.

[0053]

[0054] in Let be the Hessian operator with respect to the parameters associated with the model. This matrix can describe the curvature of the function. Large curvature implies large orientation in the model parameter space. A small step change in the function will result in a large change in its value. If the curvature is very large in a particular direction, there is greater certainty about the location of the minimum in that direction because the function / measurement is more sensitive to changes in the parameter in that direction.

[0055] When Hessian is at its minimum When the Hessian is minimized, it can be a symmetric positive definite (SPD) matrix. Such a matrix can be considered as describing an elliptic (centered at zero) or a quadratic domain, where the islikelihood objective function values ​​are elliptic. The SPD matrix is ​​in the form of a covariance matrix, and the Hessian matrix at its minimum can be considered as describing the covariance of a (normal) distribution from which the model parameters are derived as follows:

[0056]

[0057] The equation above describes the distribution of the parameters of the generated model. The covariance of the distribution is the inverse of the Hessian, where the Hessian curvature matrix can be considered a deterministic matrix, and its inverse can be considered an uncertainty matrix, where small uncertainty is expected. If the variance of the distribution of model parameters is small (e.g., corresponding to a large Hessian matrix), the model quality is likely to be high. Therefore, the optimal next imaging time is the time when the image maximizes the size of the Hessian of the objective function. Alternatively (or additionally), the next optimal imaging time can also be forced to maintain a minimum amount of expected error (given by uncertainty).

[0058] To determine the optimal viewing angle for capturing images, another angle can be added to the image set:

[0059]

[0060] Where t' corresponds to the time when new tracking measurements will be acquired in the future, and This corresponds to the predicted value of the high-frequency motion substitute at time t'. The Hessian can be calculated with respect to the model parameters to determine the determinism of subsequent images:

[0061]

[0062] The determinant can be considered a measure of the size of a deterministic matrix. The determinant of an SPD is proportional to the volume of the elliptic it represents. Therefore, the optimal imaging time can be:

[0063]

[0064] The unknown value at time t' can be predicted using reasonable extrapolation. This problem is tractable because it is a one-dimensional optimization of a discrete set of imaging viewpoints / times. In embodiments, this problem can be made more general by solving for multiple future time points. For example, if the points used for a single complete rack rotation are estimated all at once, higher quality time points can be estimated, taking into account any time constraints on the point set.

[0065] Each term in the determinant can be an SPD matrix that defines an ellipse. The sum of the SPD matrices is the SPD matrix, and thus the sum also represents an ellipse. The optimal time / viewpoint for subsequent images can correspond to the time / viewpoint for generating a deterministic ellipse that generates the largest ellipse when added to the first ellipse.

[0066] At box 240, the processing logic selects a second view from multiple views based on a corresponding tracking quality metric indicating a reduction in the uncertainty value of the second view. In one embodiment, the processing logic may select a second view based on a corresponding tracking quality metric indicating a maximum reduction in the uncertainty value of multiple views. In one embodiment, the processing logic may select a second view based on a corresponding tracking quality metric indicating a reduction in the uncertainty value exceeding a threshold. In some embodiments, the processing logic may select a second view based on other criteria. In one embodiment, the processing logic may automatically select a second view during a treatment phase of the radiation delivery system (e.g., without user intervention).

[0067] Figure 3A Illustration 300 is an example of an imaging system with multiple viewpoints having a region of interest according to embodiments of the present disclosure. Illustration 300 includes imaging system 302, which may be an X-ray (kV or MV) imaging system, an MR imaging system, or any other type of imaging system. In embodiments, imaging system 302 may correspond to Figure 1A kV imaging source 810 or Figure 1B The imaging system 1210. In some embodiments, the imaging system 302 may include an imaging detector (not shown), as previously described in Figure 1A As described in -C.

[0068] In one embodiment, the imaging system 302 may be coupled to a ring-shaped gantry, for example... Figure 1A The ring-shaped frame 820. In one embodiment, the imaging system 302 may be coupled to a robotic arm, for example... Figure 1B The robotic arm 1202. In some embodiments, the imaging system 302 may be coupled to a C-arm frame, for example... Figure 1C C-arm frame 1410.

[0069] Figure 300 also includes a region of interest 312. In an embodiment, the region of interest 312 may correspond to a patient's region of interest. The region of interest 312 may include one or more target and / or tracking structures.

[0070] Figure 300 includes a first viewpoint 304, a second viewpoint 306, a third viewpoint 308, and a fourth viewpoint 310, which may correspond to different viewpoints from which the region of interest 312 can be captured. In an embodiment, the first viewpoint 304 may correspond to the viewpoint from which a first image is captured, such as... Figure 2 As described in box 210. In some embodiments, the second viewpoint 306, the third viewpoint 308, and the fourth viewpoint 310 may correspond to subsets of multiple viewpoints considered for capturing subsequent images of the region of interest 312, as shown in Figure 2 The descriptions at boxes 230 and 240.

[0071] In an embodiment where the imaging system 302 is coupled to a gantry, as the imaging system 302 rotates around the region of interest 312 via the gantry, the first viewing angle 304, the second viewing angle 306, the third viewing angle 308, and the fourth viewing angle 310 can correspond to different angles of the imaging system 302 relative to the region of interest 312. In an embodiment where the imaging system is coupled to a robotic arm, as the imaging system 302 is positioned along a path around the region of interest 312 via the robotic arm, the first viewing angle 304, the second viewing angle 306, the third viewing angle 308, and the fourth viewing angle 310 can correspond to different orientations relative to the region of interest 312.

[0072] Figure 3B This is illustration 350, an example of selecting a viewpoint for capturing subsequent images based on a tracking quality metric according to embodiments of the present disclosure. In illustration 350, a first viewpoint 304 corresponds to the viewpoint from which the first image is captured. As previously described, the image may have associated uncertainty values ​​(e.g., uncertainty 352).

[0073] The second viewpoint 306, the third viewpoint 308, and the fourth viewpoint 310 can each have corresponding tracking quality metrics, metric 354, metric 356, and metric 358, respectively. As previously mentioned, the tracking quality metrics can indicate the reduction in the uncertainty 352 value of the image captured from the first viewpoint 304. After determining the metrics 354, 356, and 358, the processing logic of the processing device can select one of the second viewpoint 306, the third viewpoint 308, or the fourth viewpoint 310 to capture subsequent images of the region of interest 312.

[0074] refer to Figure 3B The values ​​for metric 354, metric 356, and metric 358 are 2.4, 8.2, and 6.3, respectively. Because metric 356 has the highest value, it indicates the maximum reduction in the uncertainty 352 value of the first image that a subsequent image captured from the third viewpoint 308 can have relative to the second and fourth viewpoints 310, which have lower corresponding metrics (e.g., metric 354 and metric 358). Therefore, the processing logic can select the third viewpoint 308 for capturing subsequent images.

[0075] In an embodiment, the processing logic can select the timing for capturing subsequent images from the third-view perspective 308. For example, if the region of interest 312 is a location that may be affected by the patient's respiratory movements, the specific timing for capturing subsequent images from the third-view perspective can be aligned with a specific respiratory phase of the patient to maximize the reduction of location uncertainty.

[0076] although Figure 3A and 3BFour perspectives are shown for capturing subsequent images, but embodiments of this disclosure may include any number of perspectives that can be considered for capturing subsequent images. Furthermore, although... Figure 3B The selection of a third-viewpoint 308 based on the maximum reduction of uncertainty is described, but embodiments of this disclosure may utilize other criteria to select the viewpoint for capturing subsequent images.

[0077] Figure 4 A flowchart depicts a method 400 for updating a model based on receiving subsequent images of a region of interest, according to embodiments of the present disclosure. Method 400 can be executed by processing logic, which may include hardware (e.g., circuitry, dedicated logic, programmable logic, processor, processing device, central processing unit (CPU), system-on-a-chip (SoC), etc.), software (e.g., instructions that run / execute on the processing device), firmware (e.g., microcode), or a combination thereof. In embodiments, various portions of method 400 may be generated as previously described in... Figure 1A The processing logic of the processing device of the radiation delivery system described in -1C is executed.

[0078] refer to Figure 4 Method 400 illustrates example functionality used in various embodiments. Although specific functional blocks (“blocks”) are disclosed in Method 400, these blocks are examples. That is, the embodiments are well-suited for performing the various other blocks or variations thereof described in Method 400. It should be understood that the blocks in Method 400 may be performed in a different order than presented, and not all blocks in Method 400 may be performed.

[0079] Method 400 begins at box 410, where processing logic causes the imaging system to capture subsequent images of the ROI. The subsequent images can be captured from a viewpoint selected based on a reduction in the uncertainty value associated with the first image, as previously described in... Figure 2 As described herein. In an embodiment, subsequent images may be captured during the treatment phase performed by the radiation delivery system.

[0080] At box 420, the processing logic receives the subsequent image from the imaging system.

[0081] At box 430, the processing logic updates a model based on the subsequent image update, including one or more parameters associated with the ROI. In an embodiment, the model may correspond to... Figure 2 The model generated at box 220. In some embodiments, updating the model may include modifying, adding, and / or removing parameters of the model. For example, upon receiving a subsequent image including additional information associated with the location of a radiosensitive structure within the ROI, the processing logic may modify the model's parameters corresponding to the location of the radiosensitive structure within the ROI.

[0082] Figure 5 This is a block diagram of an example computing device 500 capable of performing one or more operations described herein, according to some embodiments. The computing device 500 can be connected to other computing devices in a LAN, intranet, extranet, and / or the Internet. The computing device can operate as a server machine in a client-server network environment or as a client in a peer-to-peer network environment. The computing device can be provided by a personal computer (PC), set-top box (STB), server, network router, switch, or bridge, or any machine capable of executing a set of instructions (sequentially or otherwise) specifying the actions to be taken by that machine. Furthermore, although only a single computing device is shown, the term "computing device" should also be understood to include any collection of computing devices that individually or jointly execute a set (or more) of instructions to perform the methods discussed herein.

[0083] Example computing device 500 may include processing devices (e.g., general-purpose processors, PLDs, etc.) 502, main memory 504 (e.g., synchronous dynamic random access memory (DRAM), read-only memory (ROM)), static memory 506 (e.g., flash memory and data storage devices 518), which can communicate with each other via bus 530.

[0084] Processing device 502 may be provided by one or more general-purpose processing devices such as microprocessors, central processing units, etc. In illustrative examples, processing device 502 may include a Complex Instruction Set Computing (CISC) microprocessor, a Reduced Instruction Set Computing (RISC) microprocessor, a Very Long Instruction Word (VLIW) microprocessor, or a processor implementing other instruction sets or combinations of instruction sets. Processing device 502 may also include one or more special-purpose processing devices, such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), network processors, etc. According to one or more aspects of this disclosure, processing device 502 may be configured to perform the operations described herein, for performing the operations and steps discussed herein.

[0085] The computing device 500 may also include a network interface device 508 capable of communicating with the network 520. The computing device 500 may also include a video display unit 510 (e.g., a liquid crystal display (LCD) or a cathode ray tube (CRT)), an alphanumeric input device 512 (e.g., a keyboard), a cursor control device 514 (e.g., a mouse), and a sound signal generation device 516 (e.g., a speaker). In one embodiment, the video display unit 510, the alphanumeric input device 512, and the cursor control device 514 may be combined into a single component or device (e.g., an LCD touchscreen).

[0086] According to one or more aspects of this disclosure, data storage device 518 may include computer-readable storage medium 528 on which one or more instruction sets may be stored, the instruction sets including view selection instructions 525 for performing the operations described herein. During execution of instructions by computing device 500, the instructions may also reside wholly or at least partially in main memory 504 and / or processing device 502, which also constitute computer-readable media. Instructions may also be transmitted or received on network 520 via network interface device 508.

[0087] Although computer-readable storage medium 528 is shown as a single medium in the illustrative example, the term "computer-readable storage medium" should be understood to include a single medium or multiple media (e.g., a centralized or distributed database and / or associated caches and servers) that store one or more sets of instructions. The term "computer-readable storage medium" should also be understood to include any medium capable of storing, encoding, or carrying a set of instructions for execution by a machine and for causing the machine to perform the methods described herein. Therefore, the term "computer-readable storage medium" should be understood to include, but is not limited to, solid-state memory, optical media, and magnetic media.

[0088] It should be noted that the methods and apparatus described herein are not limited to use only in medical diagnostic imaging and treatment. In alternative embodiments, the methods and apparatus herein can be used for applications outside the field of medical technology, such as industrial imaging and non-destructive testing of materials. In such applications, for example, "treatment" can generally refer to the implementation of an operation controlled by a treatment planning system, such as the application of beams (e.g., radiation, sound, etc.), and "target" can refer to a non-anatomical object or area.

[0089] The foregoing description sets forth numerous specific details, such as examples of specific systems, components, methods, etc., to provide a good understanding of several embodiments of this disclosure. However, those skilled in the art will appreciate that at least some embodiments of the invention can be practiced without these specific details. In other instances, well-known components or methods have not been described in detail or presented in simple block diagram form to avoid unnecessarily obscuring this disclosure. Therefore, the specific details set forth are merely exemplary. Specific embodiments may differ from these exemplary details and are still considered to be within the scope of this disclosure.

[0090] Throughout this specification, references to "an embodiment" or "an embodiment" mean the specific feature, structure, or characteristic described in connection with an embodiment included in at least one embodiment. Therefore, the appearance of the phrase "in an embodiment" or "in one embodiment" in different places throughout the specification does not necessarily refer to the same embodiment.

[0091] Although the operations of the methods are shown and described herein in a specific order, the order of operations of each method can be changed so that some operations can be performed in reverse order, or so that some operations can be performed at least partially concurrently with other operations. In another embodiment, the instructions or sub-operations of different operations can be intermittent or alternating.

[0092] The above description of illustrative implementations of the invention (including those described in the abstract) is not intended to be exhaustive or to limit the invention to the precise forms disclosed. While specific implementations and examples of the invention have been described herein for illustrative purposes, those skilled in the art will recognize that various equivalent modifications are possible within the scope of the invention. The terms “example” or “exemplary” are used herein to mean used as an example, instance, or illustration. Any aspect or design described herein as “example” or “exemplary” is not necessarily to be construed as preferred or advantageous over other aspects or designs. Rather, the use of the terms “example” or “exemplary” is intended to present the concept in a specific manner. As used in this application, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise stated or clearly apparent from the context, “X includes A or B” is intended to represent any natural inclusion substitution. That is, “X includes A or B” is satisfied in any of the foregoing cases if X includes A; X includes B; or X includes both A and B. Furthermore, the articles “a” and “an” used in this application and the appended claims should generally be interpreted as meaning “one or more” unless otherwise stated or clearly indicated from the context. Additionally, the use of the terms “an embodiment”, “one embodiment”, “one implementation”, or “one implementation” throughout the document is not intended to refer to the same embodiment or implementation unless so described. Furthermore, the terms “first,” “second,” “third,” “fourth,” etc., as used herein are designations to distinguish different elements and their numerical designation does not necessarily have ordinal meaning.

Claims

1. A device for selecting an imaging viewing angle, comprising: Imaging system; and A processing device, operatively coupled to the imaging system, to: Images from a first-person perspective that identify regions of interest (ROIs), the images having uncertainty values ​​associated with the ROIs; Generate a model based on the image, including one or more parameters associated with the ROI; The model is used to determine multiple tracking quality metrics for multiple viewpoints, and the imaging system can capture subsequent images of the ROI from the multiple viewpoints, the multiple tracking quality metrics indicating the reduction of uncertainty values ​​associated with the ROI in the subsequent images; as well as Based on the corresponding tracking quality index among the plurality of tracking quality indices that indicate a reduction in the uncertainty value for the second viewpoint, a second viewpoint is selected for the subsequent image from the plurality of viewpoints.

2. The apparatus of claim 1, wherein, The device includes a radiation delivery system.

3. The apparatus of claim 2, wherein, The second perspective is automatically selected via the processing device during the treatment process performed through the radiation delivery system.

4. The apparatus of claim 1, wherein, The imaging system includes an X-ray imaging system.

5. The apparatus of claim 1, wherein, The imaging system includes a magnetic resonance (MR) imaging system.

6. The apparatus of claim 1, wherein, The processing equipment is also used for: The imaging source captures subsequent images of the ROI from the second viewpoint; Receive subsequent images from the imaging source; and The model is updated based on the subsequent images.

7. The apparatus of claim 1, wherein, The imaging system is operatively coupled to a ring gantry, and the first viewing angle corresponds to a first angle of the imaging system relative to the ROI, and the second viewing angle corresponds to a second angle of the imaging system relative to the ROI.

8. The apparatus of claim 1, wherein, The model includes respiratory motion trajectories associated with the ROI.

9. The apparatus of claim 1, wherein, The one or more parameters include one or more radiation-sensitive structures within the ROI.

10. The apparatus of claim 1, wherein, The one or more parameters include the visibility of the ROI.

11. A method for selecting an imaging viewpoint, comprising: Images from a first-person perspective that identify regions of interest (ROIs), the images having uncertainty values ​​associated with the ROIs; Generate a model based on the image, including one or more parameters associated with the ROI; The model is used to determine multiple tracking quality metrics for multiple viewpoints, and subsequent images of the ROI can be captured from the multiple viewpoints by an imaging system. The multiple tracking quality metrics indicate the reduction of uncertainty values ​​associated with the ROI in the subsequent images. as well as The processing device selects a second viewpoint from the plurality of viewpoints for the subsequent image based on a corresponding tracking quality index among the plurality of tracking quality indices that indicate a reduction in the uncertainty value for the second viewpoint.

12. The method of claim 11, wherein, The subsequent images will be captured by the imaging system of the radiation delivery system.

13. The method of claim 12, wherein, The second perspective is automatically selected via the processing device during the treatment process performed through the radiation delivery system.

14. The method of claim 11, wherein, The imaging system includes an X-ray imaging system.

15. The method of claim 11, wherein, The imaging system includes a magnetic resonance (MR) imaging system.

16. The method of claim 11, further comprising: The imaging system captures subsequent images of the ROI from the second viewpoint; Receive subsequent images from the imaging system; as well as The model is updated based on the subsequent images.

17. The method of claim 11, wherein, The imaging system is operatively coupled to a ring gantry, and the first viewing angle corresponds to a first angle of the imaging system relative to the ROI, and the second viewing angle corresponds to a second angle of the imaging system relative to the ROI.

18. The method of claim 11, wherein, The model includes respiratory motion trajectories associated with the ROI.

19. The method of claim 11, wherein, The one or more parameters include one or more radiation-sensitive structures within the ROI.

20. The method of claim 11, wherein, The one or more parameters include the visibility of the ROI.

21. A non-transitory computer-readable storage medium comprising instructions that, when executed by a processing device, cause the processing device to: Images from a first-person perspective that identify regions of interest (ROIs), the images having uncertainty values ​​associated with the ROIs; Generate a model based on the image, including one or more parameters associated with the ROI; The model is used to determine multiple tracking quality metrics for multiple viewpoints, and subsequent images of the ROI can be captured from the multiple viewpoints by an imaging system. The multiple tracking quality metrics indicate the reduction of uncertainty values ​​associated with the ROI in the subsequent images. as well as The processing device selects a second viewpoint from the plurality of viewpoints for the subsequent image based on a corresponding tracking quality index among the plurality of tracking quality indices that indicate a reduction in the uncertainty value for the second viewpoint.

22. The non-transitory computer-readable storage medium of claim 21, wherein, The subsequent images will be captured by the imaging system of the radiation delivery system.

23. The non-transitory computer-readable storage medium of claim 22, wherein, The second perspective is automatically selected via the processing device during the treatment process performed through the radiation delivery system.

24. The non-transitory computer-readable storage medium of claim 21, wherein, The imaging system includes an X-ray imaging system.

Images