Method for quantifying patient setup errors in radiation therapy - Patents.com

Abstract

A method and system for determining changes between planning and treatment images of a subject is described, the method comprising the steps of: defining one or more clinical volumes on a planning image of a subject and defining a planning envelope volume around the clinical volumes for the planning image, acquiring treatment images from the subject for locations corresponding to locations of the planning image, the treatment images having the same planning envelope volume as the planning image, determining locations of the one or more clinical volumes on the treatment images relative to the planning envelope volume, and determining an inclusion metric for one or more of the one or more clinical volumes defining a degree of inclusion of the clinical volume on the treatment image within the planning envelope volume on the treatment image.

Description

[Technical field]

[0001] The present invention relates to the field of medical imaging and medical image processing, and in particular to identifying changes in medical images due to changes in the anatomical position of the patient when the scan is acquired, and particularly to the field of radiation therapy. [Background technology]

[0002] During radiation therapy, the patient undergoes planning medical images, most commonly computed tomography (CT) but increasingly magnetic resonance (MR). From these images, a plan is created that is used for multiple radiation treatment sessions. These treatment sessions are known as fractions. One of several problems with this approach is that the planning images are not an accurate representation of the patient's anatomy at each treatment session. The position of the patient's anatomy varies randomly from session to session relative to its appearance on the planning images. There are also systematic changes in the patient's anatomy that can occur over time due to weight loss, tumor shrinkage, or other such processes. Image Guided Radiotherapy (IGRT) and Adaptive Radiotherapy (ART) are techniques that use imaging over the course of treatment to monitor anatomical changes and optimize the patient's position in the case of IGRT, or modify the treatment plan in the case of ART.

[0003] To generate a radiation treatment that is robust to variations in patient position, margins are applied to specific regions identified on the planning images. The oncologist identifies the Clinical Target Volume (CTV) as the region that may contain disease to be treated with radiation. Margins are applied to the CTV to create the Planning Target Volume (PTV). The PTV is the region to which radiation is applied to ensure that the CTV receives the required radiation dose. The PTV is a spatial envelope that includes all possible locations of the CTV due to variations in anatomical location. A similar definition exists for sensitive healthy organs that should be avoided. Organs at Risk (OARs) have margins applied to generate the Planning Risk Volume (PRV). These concepts were formalized in ICRU Reports 50, 62. Since the present invention applies equally to both target and healthy regions, we group these definitions together, with the CTV and OAR grouped together as Clinical Volumes (CV), and the PTV and PRV grouped together as Planning Envelope Volumes (PEV). The PEV is most often explicitly defined during radiation treatment planning by a contour drawn during planning, but can also be defined implicitly, such as the CV plus a 10 mm margin.

[0004] In IGRT and ART, a key decision is whether the CV is still contained by the respective PEV. CTVs that move outside the PTV have an increased risk of being missed by radiation therapy. Similarly, OARs that move outside the PRV have an increased risk of moving into high dose regions. If the CV extends outside the PEV, the original planning assumptions are invalid and the patient position must be shifted to bring the CV back within the PEV or a new plan should be considered. In scenarios where the user wants to perform position correction, the majority of radiation therapy machines allow position correction by automated linear movement of the patient couch. More advanced systems allow automatic correction of the patient angle and linear shift.

[0005] A flow chart of an exemplary current process 100 for IGRT / ART is shown in FIG. 1. The process has two stages. A planning stage 101 is performed before the first radiation treatment. In step 103, single or multiple clinical volumes (CVs) are identified on a planning 3D image, such as a CT scan. Some CVs correlate accurately with organs visible in the scan, while others, such as clinical target volumes that identify areas of risk for cancer disease, may not. In step 104, PEVs are generated around the single or multiple CVs to account for possible movement of the CVs during treatment. The number of PEVs may differ from the number of CVs, for example, because PEVs are not generated for all CVs or because multiple CVs may be combined into a single PEV. Then, in step 105, a radiation treatment plan is created using the PEVs from step 104 to optimize the dose delivered to the treatment target while keeping the dose to the risk organs below a set threshold.

[0006] The treatment phase 102 is performed over one or more treatment phases. Step 106 is performed for each or selected treatment fractions, where an imaging system integrated with the treatment machine acquires an image before performing the treatment, typically up to 15 minutes before the start of the treatment, while the patient remains in the treatment position during this time. In step 107, the user visually compares this image with the planning image to verify that each critical CV from step 104 is still contained within the associated PEV. Typically, this is done on a dedicated imaging workstation that allows a variety of different methods of image comparison. Based on this evaluation, the user decides whether to proceed with the treatment in step 120, possibly with corrections for patient position.

[0007] In the prior art, most IGRT and ART systems only provide tools for manual subjective assessment of whether the CV is contained within the PEV. The few systems that provide automated assessments designed for real-time assessment of the patient's anatomy provide, for example, a Boolean yes / no indicator that shows green if the CV is within the PEV and red if it is not (Non-Patent Document 3). The problem with such indicators is that they do not provide a means to quantify change and track possible development of a problem over time.

[0008] Some systems have attempted to use quantitative measures to track changes over time. Wang et al. (Non-Patent Document 4) used Varian's Velocity "Adaptive Monitoring Navigator" to track the volumetric changes of structures and any shifts in the structure center position. These metrics are useful when the CV being tracked is a rigid body, but in modern radiotherapy, some CVs are very complex and do not simply move. When tracking the movement of the center of the CV, parts of the CV that move outside of their PEV may be missed.

[0009] A survey of academic studies has found many other papers reporting on frameworks and metrics for monitoring patient changes during radiation therapy. In particular, Brouwer et al. (Non-Patent Document 5) conducted a systematic review of multiple papers on techniques for ART in head and neck cancer. The metrics of anatomical changes they identified from 51 studies were body weight, body thickness, CV volume, CV density, CV position, CV angle, and variations of these (e.g., changes, speed of changes, or changes occurring at specific anatomical points). As mentioned above, in complex CVs, these metrics may miss multiple parts of the CV that move outside of their PEV.

[0010] Several studies (6-8) have reported more complex metrics of anatomical change based either on direct changes in patient images or on changes in structures identified on the patient images.

[0011] Schaly et al. (Non-Patent Document 6) investigated the difference in gamma index between multiple CT scans. The gamma index is a method developed for the comparison of two 3D radiation dose distributions (Non-Patent Document 9). It is sensitive to both changes in voxel intensity and the match distance, which is the distance between a voxel in the reference image and a voxel of the same intensity in the test image. The gamma index is certainly sensitive to changes in the location of the CV, but as a global image-based metric, it does not explicitly measure the adequacy of the PEV encompassing the CV, and a low gamma index score is sensitive to anatomical changes anywhere in the image.

[0012] Fiorino et al. (Non-Patent Document 7) reported the use of Jacobian Volume Histograms to quantify changes in deforming organs. In this method, a deformable registration is applied between the reference and test images. The Jacobian matrix of the deformable vector field is then calculated and multiple values ​​in the region of the CV are plotted in a histogram. This provides a measure of the expansion and contraction of voxels throughout the CV. This has the advantage over Schaly's method of focusing on a specific CV rather than the entire volume. However, this does not directly address the issue of the validity of the PEV encompassing the CV. Changes occurring within the CV that do not extend beyond the PEV will still produce large changes in the Jacobian.

[0013] Hargrave et al. (Non-Patent Document 8) reported the closest approach to the present invention. This measurement uses the Hausdorff distance between the original CV and the treated CV as a metric. The Hausdorff distance is a measurement of the worst-case difference between two sets of contours. This will find significant changes in the complex CV, but is not explicitly linked to the question of clinical interest of whether the CV moves outside the PEV. Further contraction and movement of the CV within the PEV is recorded as a change, not a clinical issue. This lack of directionality in the metric makes its use in tracking changes more difficult. [Prior art documents] [Non-patent literature]

[0014] [Non-Patent Document 1] ICRU Report 50: Prescribing, Recording and Reporting Photon Beam Therapy: Journal of the ICRU, Vol. 26, No. 1, September 1993 [Non-Patent Document 2] ICRU Report 62: Prescribing, Recording and Reporting Photon Beam Therapy (Supplement to ICRU Report 50): Journal of the ICRU, Vol. 32, No. 1, November 1999 [Non-Patent Document 3] https: / / viewray.com / mri-guided-smart / [Non-Patent Document 4] Wang et al., Adaptive radiotherapy based on statistical process control for oropharyngeal cancer: J Appl Clin Med Phys 2020; 21:9:171-177 [Non-Patent Document 5] Brouwer et al. Identifying patients who may benefit from adaptive radiotherapy: Does the literature on anatomic and dosimetric changes in head and neck organs at risk during radiotherapy provide information to help? Radiotherapy and Oncology 115 (2015) 285-294 [Non-Patent Document 6] Schaly et al., Alert system for monitoring changes in patient anatomy during radiation therapy of head and neck cancer. J Appl Clin Med Phys. 2021; 22: 168-174. [Non-Patent Document 7] Fiorino et al. Introducing the Jacobian-volume-histogram of deforming organs: application to parotid shrinkage evaluation, Phys. Med. Biol. 56 (2011) 3301-3312 [Non-Patent Document 8] Hargrave et al. A feature alignment score for online cone‐beam CT‐based image‐guided radiotherapy for prostate cancer. Med Phys. 45 (7), July 2018 2898-2911 [Non-Patent Document 9] Low et al., A technique for the quantitative evaluation of dose distributions. Med Phys. 1998; 25: 656-661. Summary of the Invention [Problem to be solved by the invention]

[0015] That is, the present invention solves one or more of the following problems. To help identify changes in a patient's anatomical location, a simple, quantifiable metric is needed that easily correlates changes between images acquired over a period of time. Preferably, this information can be used for treatment planning in the presence of patient geometric changes and can be presented in a meaningful way to a clinical user. [Means for solving the problem]

[0016] According to the invention, there is provided a method for determining changes between planning and treatment images of a subject, the method comprising the steps of: defining one or more clinical volumes on a planning image of a subject and defining a planning envelope volume around the clinical volumes for the planning image, acquiring treatment images from the subject for positions corresponding to positions of the planning image, the treatment images having the same planning envelope volume as the planning image, determining positions of the one or more clinical volumes on the treatment images relative to the planning envelope volume, and determining an inclusion metric for one or more of the one or more clinical volumes defining a degree of inclusion of the clinical volume on the treatment image within the planning envelope volume on the treatment image.

[0017] In one embodiment of the invention, determining the containment metric comprises the steps of designating one or more representative points on the surface of the clinical volume and determining a shortest distance for the one or more representative points of the clinical volume to the planning envelope volume, where if the representative point on the clinical volume is within the planning envelope volume, the shortest distance is classified as either a positive or negative internal distance, and if the representative point on the clinical volume is outside the planning envelope volume, the shortest distance is an external distance classified with an opposite sign to the internal distance, and the value of the containment metric is determined as either the minimum of the signed values ​​of the shortest distance when the internal distance is classified as positive, or the maximum of the signed values ​​of the shortest distance when the internal distance is classified as negative.

[0018] In a preferred embodiment of the invention, the planning image and the plurality of treatment images are 3D images, more preferably, said planning image and treatment images are CT, PET, SPECT or MRI images.

[0019] In a preferred embodiment of the present invention, after the planning envelope volume is defined, a distance transform is calculated around the planning envelope volume, which is used to determine a containment metric.

[0020] More preferably, the containment metric is determined by simulating a translation of one or more treatment clinical volumes relative to the planning envelope volume. In one embodiment of the present invention, the simulated translation is a linear translation.

[0021] In a preferred embodiment of the invention, determining the location of the one or more clinical volumes on the treatment images relative to the planning envelope volume includes one or more of: determining a geometric relationship between the planning image and the treatment image and mapping the location of the clinical volume across the treatment image; and identifying a plurality of anatomical features on the treatment image to locate the location of the clinical volume.

[0022] Preferably, the planning envelope volume is defined as the clinical volume plus a margin specified by a pre-established protocol. In one embodiment of the invention, the clinical volume is the clinical target volume or total tumor volume for a treatment plan. In an alternative embodiment of the invention, the clinical volume represents organs at risk that should be avoided in the treatment.

[0023] In a preferred embodiment, the steps are repeated over a set period of time to monitor changes in the containment metric on the treatment images, preferably the set period of time varying between 6 hours and 3 months.

[0024] In an embodiment of the present invention, the change in said containment metric over time is displayed to the user. In a preferred embodiment of the invention, the subject is positioned at a specified position within the scanner for the planning and treatment images, and a user of the system can adjust the subject's position according to one or more of the determined containment metrics.

[0025] Preferably, changes in the containment metric over time are displayed to a user, the display further indicating when the containment metric exceeds a pre-set threshold which indicates unsafe movement of the subject within the scanner.

[0026] Preferably, the method further comprises the step of displaying one or more of said determined inclusive metrics to a user of the system. In a preferred embodiment, the method further comprises using the containment metric to determine a 3D representation indicating movement margins indicating how the subject can be moved within the scanner while maintaining or improving the position of the clinical volume relative to the position of the planning envelope volume.

[0027] Preferably, the clinical volume is displayed as a 3D rendered surface and the containment metric for the surface is shown as a heatmap on the surface. In a further embodiment of the invention, there is provided a system for analyzing medical images to determine changes between planning and treatment images of a subject, the system comprising a processor configured to: determine one or more clinical volumes on a planning image of a subject and define a planning envelope volume around the clinical volumes for the planning image, analyze treatment images from the subject for locations corresponding to locations of the planning image, the treatment images having the same planning envelope volume as the planning image, determine locations of the one or more clinical volumes on the treatment images relative to the planning envelope volume, and determine an inclusion metric for one or more of the one or more clinical volumes defining a degree of inclusion of the clinical volume on the treatment image within the planning envelope volume on the treatment image.

[0028] Preferably, the processor is further configured to perform any of the steps of the method described above. In a preferred embodiment of the invention, the system further comprises a display for displaying at least one of said planning image, said treatment image, said one or more clinical volumes, and said planning envelope volume.

[0029] In a further embodiment of the invention, there is also provided a computer program product comprising a plurality of instructions, which when the program is executed by a computer, causes the computer to perform a plurality of steps of the method described above. [Brief description of the drawings]

[0030] Further details, aspects and embodiments of the present invention are now described, by way of example only, with reference to the drawings, in which like reference numbers are used to identify like or functionally similar elements, and in which elements are illustrated for simplicity and clarity and have not necessarily been drawn to scale. [Figure 1]FIG. 1 shows a flow chart of a known process for image-guided / adaptive radiation therapy. [Diagram 2] FIG. 2 shows a flow chart of a method according to a first embodiment of the invention. [Diagram 3] FIG. 3 shows a flow chart of a method according to a further embodiment of the invention. [Figure 4a] FIG. 4(a) shows the clinical volume and the planning envelope volume on the planning image. [Figure 4b] FIG. 4(b) shows the clinical volume and the planning envelope volume on multiple treatment images. [Figure 4c] FIG. 4(c) shows the clinical volume and planning envelope volume on multiple treatment images. [Figure 5a] FIG. 5(a) shows the determination of the measurement metric according to the location of the clinical volume relative to the location of the planning envelope volume. [Figure 5b] FIG. 5(b) illustrates the determination of the measurement metric according to the location of the clinical volume relative to the location of the planning envelope volume. [Figure 5c] FIG. 5(c) illustrates the determination of the measurement metric according to the location of the clinical volume relative to the location of the planning envelope volume. [Figure 6a-b] 6(a) and 6(b) illustrate a method for calculating a measurement metric according to one embodiment of the present invention. [Figure 6c] FIG. 6(c) illustrates a method for calculating the measurement metric according to one embodiment of the present invention. [Figure 7] FIG. 7 is a plot of measurement metrics versus time for an embodiment of the present invention. [Figure 8a] FIG. 8(a) is a rendered 3D view of a clinical volume according to one embodiment of the present invention. [Figure 8b] FIG. 8(b) is a rendered 3D view of a clinical volume according to one embodiment of the present invention. [Figure 9] FIG. 9 is a 3D view of information regarding measurement metrics for clinical volumes. [Figure 10]FIG. 10 is a simplified block diagram of a medical imaging system configured to display medical images to a user. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0031] The present invention is a system for monitoring and reporting changes between multiple medical images acquired at different times. For example, they may be multiple images acquired during radiation treatment and planning. A specific potential application is to support decision making in image guided radiotherapy (IGRT) and adaptive radiotherapy (ART), but the invention is not limited to this application. Other potential applications are in deformable registration quality control, disease progression tracking, or other image-guided medical interventions where movement of clinical volumes across successive images is required.

[0032] FIG. 10 shows a simplified block diagram of one example of a medical image processing system 1000 configured to enable display of medical images to a user. In the illustrated example, the medical image processing system 1000 includes one or more user terminals 1001 including, for example, workstations or the like configured to access medical images stored in, for example, a database 1002 or other data storage device. In the illustrated example, a single database 1002 is shown. However, it will be appreciated that the user terminal 1001 may be configured to access medical images from more than one data storage device. Furthermore, in the illustrated example, the database 1002 is shown as being external to the user terminal 1001. However, it will be appreciated that the user terminal 1001 may also be configured to access medical images stored locally in a local storage module indicated at 1010, on one or more internal storage elements such as a memory element indicated at 1003 or a disk element indicated at 1009. The user terminal 1001 further includes one or more signal processing modules, such as a signal processing module generally indicated at 1004. The one or more signal processing modules are configured to execute computer program code stored, for example, in the local storage module 1009. In the illustrated example, the one or more signal processing modules 1005 are configured to execute computer program code including one or more of the one or more patient change monitoring components, exemplified in this example as a metric calculation component 1005. In the illustrated example, the signal processing module 1004 is further configured to execute computer program code including one or more image display components 1006 configured to display any metrics of the patient changes as generated by the one or more metric calculation components 1005 in a manner meaningful to a user, such as on a display screen 1007.The medical image processing system 1000 may further include one or more user input devices, generally indicated at 1008, to enable a user to interact with computer program code or the like executing on the one or more signal processing modules 1004.

[0033] In FIG. 2, the process shown in FIG. 1 is shown incorporating the present invention. With respect to FIG. 1, the process has two stages: a planning stage 201, which takes place before the first radiation treatment, and a treatment stage 202. The treatment process is almost the same. Items 201-206 and 210 all correspond to the same items in FIG. 100. In step 203 of the planning stage, single or multiple clinical volumes are identified on a medical planning image of a subject. In one example of the present invention, the planning image and multiple treatment images are 3D images. Preferably, the planning image and the treatment images are CT, PET, SPECT or MRI images.

[0034] Some clinical volumes, such as clinical target volumes that identify areas of risk for cancer disease, may correlate accurately with organs visible in the scan, while others may not. In step 204, multiple planning envelope volumes are generated around a single or multiple clinical volumes to account for possible movement of the clinical volumes during treatment. The number of planning envelope volumes may differ from the number of clinical volumes, for example, because planning envelope volumes are not generated for all clinical volumes, or because multiple clinical volumes may be combined into a single planning envelope volume. Then, in step 205, a radiation treatment plan is created using the planning envelope volumes of step 204 to optimize the dose delivered to the treatment target while keeping the dose to the risk organs below a set threshold.

[0035] The treatment phase 202 is performed over one or more treatment phases. Step 206 is performed for each treatment fraction, or a selected number of treatment fractions, where an imaging system integrated with the treatment machine acquires a treatment image before administering the treatment, typically up to 15 minutes before the start of treatment, with the patient remaining in the treatment position during this time. Preferably, the treatment image for the subject is acquired for a position corresponding to the position of the planning image, such that the treatment image has the same planning envelope volume as the planning image.

[0036] The present invention supports user evaluation of the images such that step 107 of FIG. 1 is replaced by steps 207-209 of FIG. 2. In step 207, the location of one or more clinical volumes on the treatment image is identified, preferably the location is determined relative to the location of the planning envelope volume. There are various ways to perform the task of step 207, and the present invention does not rely on a particular method, but is performed in any technically feasible manner. For example, by determining a geometric relationship between the planning image and the treatment image and mapping the overall location of the volume. Another method is by manually or automatically identifying multiple volumes from the anatomical structure on the treatment image to locate the clinical volume. In step 208, an encapsulation metric, which will be described in more detail below, is calculated between one or more clinical volumes on the treatment image and the associated planning envelope volume from the planning image. Preferably, the encapsulation metric for one or more of the one or more clinical volumes defines the degree of encapsulation of the one or more clinical volumes on the treatment image within the planning envelope volume on the treatment image. This is shown to the user in step 209. This view can take several forms incorporating any combination of patient images, clinical and planning envelope volumes, metrics for this treatment session, trends in previous sessions, graphical representations of where problems were detected, metric values ​​organized by surface angle, or summary information in numeric or text format. Some of these elements are further illustrated in Figures 700, 800, and 900.

[0037] A further example of the method of the invention is shown in FIG. 3. Items 301-306 and 320 all correspond to the same items in FIG. 1. As in FIG. 2, in step 307 the location of one or more clinical volumes on the treatment image is determined, and in step 308 an inclusion metric between each of the one or more treatment clinical volumes and the associated planning envelope volume is calculated. In step 309, a correction to the location of the treatment image that best includes the one or more clinical volumes in the planning envelope volume and thus minimizes the inclusion metric is determined. If there are multiple inclusion metrics to be optimized, any standard method for managing multiple objectives in an optimization problem can be used, for example by setting multiple priority rules by volume type, or by weighting and summing the separate metrics to generate an overall score. In step 310, the individual inclusion metrics for this optimized treatment position are calculated. In step 311, the inclusion metric for the original patient position is reported to the user along with the value of the possible inclusion metric if an optimal correction is performed.

[0038] Diagram 400 shows several concepts of clinical volume and planning envelope volume in planning and treatment situations. Figure 4(a) shows the situation in a planning image. A coordinate system is defined with the origin at O ​​(item 402). The location of O is determined during the planning process and can be any point on the treatment image, but typically corresponds to the isocenter of the radiation therapy machine. The location of the clinical volume in the planning scan is shown as A0 (item 403). The planning envelope volume is shown as B (item 404) completely encompassing the clinical volume with margins. Figure 4(b) shows the situation in one of multiple treatment fractions (designated i). The time period between multiple treatment fractions typically varies between 6 hours and 1 week. Since the treatment is prescribed relative to the machine geometry, the origin (item 406) and the location of the planning envelope volume (item 408) are invariant for each treatment fraction. However, the patient will generally not sit or lie in the machine geometry in exactly the same position as during the planning phase, so the position of the clinical volume will change, as indicated by Ai (item 407). In this example of the invention, the clinical volume has moved to the upper right quadrant relative to the axis, but still remains within the boundaries of the planning envelope volume. Figure 4(c) shows the situation at different treatment fractions of multiple treatment fractions. Again, the origin (item 410) and the planning envelope volume (item 411) are not changed. The clinical volume (item 412) has not only moved in this example relative to the planning envelope volume, but has also been deformed compared to the original shape it had in the planning phase (item 403). The combination of movement and shape change has caused part of the clinical volume to move outside the space defined by the planning envelope volume, as indicated by item 413.

[0039] The containment metric used to quantify the extent to which a clinical volume lies within an associated planning envelope volume is defined as follows: For one or more points on the clinical volume, the shortest distance to the planning envelope volume is determined. In one example of the invention, if a point on the clinical volume is within the planning envelope volume, this distance is given a positive sign. If a point of the clinical volume is outside the planning envelope volume, this distance is given a negative sign. The overall containment metric is then the minimum of the one or more signed shortest distances.

[0040] An example of a containment metric is shown in Figure 5. Figure 5(a) shows a portion of a clinical volume (item 502) enclosed by a portion of a planning envelope volume (item 503). A representative point on the surface of the clinical volume is indicated by item 504. Successive distances from the representative point 504 on the clinical volume to the planning envelope volume 503 are shown as arrows of different lengths (item 505). The shortest distance from the representative point 504 on the clinical volume to the planning envelope volume 503 is indicated by a thick arrow (item 506).

[0041] FIG. 5(b) 507 illustrates the case where the clinical volume 509 is still completely enclosed within the planning envelope volume 508. One or more representative points are specified on the surface of the clinical volume. For each of the one or more representative points on the clinical volume, a shortest distance to the planning envelope volume is determined, indicated by thin arrows 510. In one example of the invention, if one or more representative points on the clinical volume are within the planning envelope volume, the distance is classified as a positive or negative internal distance, whereas if one or more representative points on the clinical volume are outside the planning envelope volume, the distance is an external distance classified with the opposite sign to the internal distance. The value of the containment metric is determined as either the minimum of the signed shortest distance values ​​for the one or more representative points where the internal distance is classified as positive, or the maximum of the signed shortest distance values ​​for the internal distance is classified as negative.

[0042] In this case, as shown in Figure 5(b), each distance from the clinical volume to the planning envelope volume has a positive value because the clinical volume is completely enclosed by the planning envelope volume. The smallest of these, which is the value reported as the containment metric, is indicated by the thick arrow 511. All of the other arrows in this figure have a length greater than the arrow indicated by 511.

[0043] FIG. 5(c) 512 illustrates the case where a portion of the clinical volume 513 has moved beyond the planning envelope volume 514. In this example of the invention, a portion of the clinical volume 520 is still within the planning envelope volume, and a portion of the clinical volume 521 is now outside the planning envelope volume. For each point on the clinical volume (both outside the planning envelope volume and inside the planning envelope volume), the shortest distance to the planning envelope volume is determined. For those points on the portion of the clinical volume 520 that is still within the planning envelope volume, this distance is given a positive value, indicated by the closed-tipped arrow 515. For those points on the region of the clinical volume 521 that is outside the planning envelope volume, this distance is given a negative value, indicated by the open-tipped arrow 516. The reported containment metric is the minimum of the multiple signed shortest distances, as indicated by the thick arrow 517. In this case, in this example of the invention, the containment metric also has a negative value, since the shortest distance is between the PEV and a portion of the CV that is outside the PEV.

[0044] In one example of the invention, after the planning envelope volume is defined, a distance transform is calculated around the planning envelope volume, which is used to determine a containment metric. Alternatively, the containment metric is determined by simulating a translation of one or more treatment clinical volumes relative to the planning envelope volume. In one embodiment of the invention, the simulated translation is a linear translation.

[0045] An example of how to efficiently calculate the containment metric is shown in FIG. 6. FIG. 6(a) shows the calculations during the radiation treatment planning stage 601. Once the PEV 602 (shown as a dashed line) is determined, a signed distance transform (603) is calculated and is shown in legend 604. The signed distance transform is a 3-dimensional matrix of signed distance values. In this example, the values ​​of the matrix are represented by contiguous radial sections 5 mm wide. The section at the center of the image is +25 to +30 mm from the PEV. This is surrounded by a radial section that is +20 to +25 mm. The image has contiguous concentric 5 mm sections that range from +5 mm to 0 mm at the inner boundary of the PEV, 0 to -5 mm outside the PEV, and -10 to -15 mm at the edge of the image furthest from the PEV.

[0046] The signed distance transform for determining the containment metric can be calculated by any method detailed in the scientific literature, such as the method by Borgefors [1]. This signed distance transform indicates the shortest distance from the clinical volume to the PEV for any point in the treatment coordinate system. In one example of the present invention, if the point is inside the PEV (as shown in FIG. 5(a)), the value is positive, and if the point is outside the PEV, the value is negative, as shown in FIG. 5(b).

[0047] The radiation treatment stage is shown in FIG. 6(b). The signed distance transform 603 from the planning stage is copied directly into the treatment space 606. That is, the three-dimensional matrix of signed distance values ​​is mapped into the treatment space represented in this figure by a series of radial sections with signed distance values ​​ranging from 25-30 mm to -15-10 mm. The surface of the clinical volume 607 is shown, and in this example of the invention, the clinical volume is completely contained within the planning envelope volume, so the containment metric has a positive value in this example of the invention. As shown, the clinical volume is offset in the upper right quadrant relative to the origin of the original planning envelope volume. Given this, and the location of the clinical volume 607 on the treatment image, the minimum of these signed minimum distance values ​​on the boundary of the clinical volume can be found by any means. A particular trivial method is by an exhaustive search along all boundary points of the clinical volume, but more efficient algorithms such as priority queuing or sorting algorithms are possible as well. Alternatively, the distance transform can be adapted with a convention where parts of the CV that are outside the PEV are assigned positive values ​​and parts of the CV that are inside the PEV are assigned negative values ​​(the opposite of the convention used in Figure 5). In this scenario, the containment measure is the maximum of the signed shortest distance values ​​on the boundary of the CV, rather than the minimum. This is shown in Figure 6(c).

[0048] The inclusion metric of the present invention can be calculated using 3D or 2D geometry. In the 3D case, the inclusion metric is sensitive to volume changes occurring in any direction. An example of the use of the 3D case is when the clinical volume being tracked is a clinical target volume, and movements in any direction are clinically important. In the 2D case, the inclusion metric is sensitive to volume changes occurring in a particular image axis. An example of the use of the 2D case is when the spinal cord is the clinical volume being tracked, but the clinical volume extends above and below the region of the therapeutic image. In this case, it may be better for the inclusion metric to be less sensitive to changes in the cranio-caudal direction, since those changes may not be clinically relevant. In the method described in figure 600, the difference between the 2D and 3D methods is in the calculation of the signed distance transform 603. To generate the 3D inclusion metric, the signed distance transform describes the distance from the clinical volume to the planning envelope volume in any direction. To generate a 2D inclusion metric, a signed distance transform describes the distance from the clinical volume to the planning envelope volume only in the 2D plane orthogonal to the excluded direction. A further extension of the invention is to clip the region over which the metric is calculated to avoid regions that are likely to produce inaccurate results, such as the top and bottom of a planning or treatment image sequence, or to remove regions far from the treatment area.

[0049] An alternative, but less efficient, method for calculating the minimum margin encapsulation metric is to simulate multiple linear translations of one or more treatment clinical volumes and identify the magnitude of linear translation required for each clinical volume to exceed its associated planning envelope volume.

[0050] In one example of the invention, the planning envelope volume is defined as the clinical volume plus a margin specified by a preset protocol. Preferably, the clinical volume is the clinical target volume or the total tumor volume for the treatment plan. Alternatively, the clinical volume represents organs at risk that should be avoided during treatment.

[0051] An important advantage of the minimum margin inclusion metric over others previously reported is that this inclusion metric is signed so that movement of clinical volumes outside the planning envelope volume (negative values ​​of the inclusion metric) appears differently than movement of clinical volumes inside the planning envelope volume (positive values ​​of the inclusion metric). In one example of the invention, the method also includes a step of displaying the determined one or more inclusion metrics to a system user. The principle of displaying the inclusion metric to the user can be extended to give greater insight and flexibility. Further embodiments of the invention use the inclusion metric to track changes in the patient. In one example of the invention, the steps of the method may be repeated over a set period of time to monitor changes in the inclusion metric on a number of treatment images. Preferably, the set period of time may vary between one day and three months. In some examples of the invention, the change in the inclusion metric is displayed to the user. To this end, a tracking view of the changes is shown in figure 700, which preferably plots the inclusion metric 701 for each day 703 of treatment, although other time intervals may be used. In this plot, multiple thresholds of interest 702 can be displayed, where values ​​of the containment metric above a pre-set threshold indicate unsafe movement of the subject within the scanner. The thresholds can be either global values ​​for each planning envelope volume, or individual different values ​​for each planning envelope volume. Each threshold can be set in a number of ways, such as as a user-entered value, as an organ-specific standard value, or algorithmically determined based on any combination of location, size, radiation dose tolerance or objective, treatment technique, and proximity to other clinical or planning envelope volumes. In this example, multiple points for this case 704 show a downward trend of clinical volumes moving outside the planning envelope volume on the sixth day of treatment.Other implementations of systems utilizing inclusion metrics for change tracking may include autonomously monitoring changes in the inclusion metrics against statistical criteria and using the inclusion metrics as input to decision support algorithms, such as those that advise when a statistically significant change in behavior has occurred.

[0052] While the global encapsulation metric is useful to aid in immediate decision making and error tracking, in one example of the invention, other views can also assist the user in understanding what is going on in a patient case. In a further embodiment of the invention, Fig. 8(a) and Fig. 8(b) show two alternative 3D surface rendered views of a clinical volume 801. In Fig. 8(a), each cube 801 represents a portion of the clinical volume surface. The surface can be partitioned algorithmically, preferably according to a grid or mesh set by the user, or based on any combination of the patient geometry, surface curvature, and dose grid or multiple image parameters. If the partitioning is fine enough, the surface appears smooth and continuous to the user (see Fig. 8(b)). Orientation labels 802 allow the user to relate the 3D rendered surface view to the patient's anatomy. In this example of the invention, the encapsulation metric is determined using the minimum signed distance, but also applies to the encapsulation metric determined as the maximum distance as described above. The representation of each element of that surface is modified by means such as color, transparency, texture, or label according to the shortest distance (or maximum distance if an alternative method of calculating the metric is used) to the planning envelope volume at that location on the surface of the clinical volume, as shown in legend 803. An example of this representation is a heat map, where lower values ​​of the metric are displayed hotter. The user can set any other color scheme that they find more useful. This allows the user to quickly identify the largest regions of interest 804. Figure 8(b) shows a 3D surface image of the volume, rather than a cubic version.

[0053] The only means by which the user can immediately correct any patient position problems highlighted by the determination of the inclusion metric is to move the patient table, which may only allow linear movement. In one example of the invention, the subject may be positioned in a specified position within the scanner for planning and treatment images, and the system user can adjust the subject's position within the scanner according to one or more determined inclusion metrics.

[0054] Therefore, a further embodiment calculates a minimum margin metric by orientation and displays this information to the user. A particular way to achieve this is to assign each point on the surface of the clinical volume to an orientation based on the 3D angle of the structure surface at this point. The 3D angle of the structure surface can be easily determined from the gradient of the signed distance transform 603. Other standard methods exist in computer graphics to determine the 3D angle of the orientation of a surface. The continuous 3D angle values ​​for all points on the surface are then classified (binned) into discrete orientations based on a number of pre-set limits. In the calculation of the containment metric, the points on the CV surface that fall into each orientation bin are treated independently of each other, resulting in a separate containment metric value for each orientation.

[0055] Another method achieves a directional containment metric by simulating the patient's position shift in each direction and reporting the distance moved in each direction before the clinical volume exceeds the planning envelope volume. This advises the user if the couch can be moved in any direction without moving the clinical volume outside the planning envelope volume. This directional minimum margin metric can be displayed to the user in many forms such as a simple table, using multiple 2D or 3D directional arrows of length equal to the metric value, or using a 2D or 3D representation of the CV shape with multiple indicators on the edges corresponding to each direction. An example of how to display this information is shown in figure 900. The 3D view is shown with indicators 901 indicating the minimum margin metric in each direction. The multiple directions are provided by directional labels 902. As before, the color of the indicator indicates the minimum margin in each direction as explained by legend 903. In this case, the user sees that the margin metric is slightly worse in the backward direction 904 than in the forward direction 905, and this situation can be improved with a small forward movement of the couch.

[0056] [1] Borgefors, On Digital Distance Transforms in Three Dimensions, Computer Vision and Image Understanding, 1996, 64 (3); 368-376 Other image-guided interventional medical modalities involve tracking of a patient's anatomy over time.

[0057] Embodiments of the present invention may be applied to any or all of the following: picture archiving and communication systems (PACS); advanced visualization workstations; image acquisition workstations; web-based or cloud-based medical information and imaging systems; radiation treatment planning systems (TPS); radiation therapy linear accelerator consoles; and radiation therapy proton beam consoles.

[0058] The present invention has been described with reference to the accompanying drawings. However, it will be appreciated that the present invention is not limited to the specific examples described herein and illustrated in the accompanying drawings. Moreover, since the illustrated embodiments of the present invention can be implemented in large part using electronic components and circuits known to those skilled in the art, details will not be described beyond the extent deemed necessary as illustrated above for the understanding and appreciation of the underlying concepts of the present invention and so as not to obscure or confuse the teachings of the present invention.

[0059] The present invention may be implemented in a computer program for running on a computer system, comprising at least code portions for performing the steps of a method according to the present invention when executed on a programmable apparatus such as a computer system or for enabling a programmable apparatus to perform the functions of a device or system according to the present invention.

[0060] A computer program is a listing of instructions, such as a particular application program and / or an operating system. A computer program may include, for example, one or more of subroutines, functions, procedures, object methods, object implementations, executable applications, applets, servlets, source code, object code, shared libraries / dynamic load libraries, and / or other sequences of instructions designed to execute on a computer system. Thus, some examples describe a non-transitory computer program product having stored executable program code for automatic contouring of cone beam CT images.

[0061] The computer program may be stored internally in a tangible non-transitory computer readable storage medium or may be transmitted to the computer system via a computer readable transmission medium. All or a portion of the computer program may be provided in a computer readable medium that is permanently, removably or remotely coupled to an information processing system. The tangible non-transitory computer readable medium may include, by way of example and not limitation, any number of the following: magnetic storage media including disk and tape storage media; optical storage media such as compact disk media (e.g., CD ROM, CD-R, etc.) and digital video disk storage media; non-volatile memory storage media including semiconductor-based memory units such as FLASH memory, EEPROM, EPROM, ROM; ferromagnetic digital memory; MRAM; volatile storage media including registers, buffers or caches, main memory, RAM, etc.

[0062] A computer process typically includes an active (running) program or part of a program, current program values ​​and state information, and resources used by an operating system to manage the execution of the process. An operating system (OS) is software that manages the sharing of a computer's resources and provides programmers with the interface used to access those resources. An operating system processes system data and user input, and responds by allocating and managing tasks and internal system resources as services to users and programs of the system.

[0063] A computer system may include, for example, at least one processing unit, associated memory, and a number of input / output (I / O) devices. When executing a computer program, the computer system processes information in accordance with the computer program and generates resultant output information via the I / O devices.

[0064] The present invention has been described herein above with reference to specific examples of its embodiments, however it will be apparent that various modifications and changes may be made thereto without departing from the scope of the present invention as set forth in the appended claims, which are not limited to the specific examples described above.

[0065] Those skilled in the art will recognize that the boundaries between logic blocks are merely exemplary, and that alternative embodiments may merge logic blocks or circuit elements, or impose alternative decompositions of functionality on the various logic blocks or circuit elements. Thus, it should be understood that the architectures depicted herein are merely exemplary, and that in fact many other architectures may be implemented which achieve the same functionality.

[0066] Any arrangement of components to achieve the same functionality is effectively "associated" such that the desired functionality is achieved. Thus, any two components combined herein to achieve a particular functionality can be considered to be "associated" with one another such that the desired functionality is achieved, regardless of architecture or intermediate components. Similarly, any two components so associated can also be considered to be "operably connected" or "operably coupled" with one another to achieve the desired functionality.

[0067] Moreover, those skilled in the art will recognize that the boundaries between the operations described above are merely exemplary. Multiple operations may be combined into a single operation, a single operation may be distributed into additional operations, and operations may be performed with at least partial overlap in time. Additionally, alternative embodiments may further include multiple instances of a particular operation, and the order of operations may be changed in various other embodiments.

[0068] However, other modifications, variations and alternatives are possible, and the specification and drawings are accordingly to be regarded in an illustrative rather than a restrictive sense. In the claims, any reference signs in parentheses shall not be construed as limiting the claims. The term "comprises" does not exclude the presence of other elements or steps than those recited in the claim. Furthermore, the terms "a" or "an" as used herein are defined as one or more. Also, the use of introductory phrases such as "at least one" and "one or more" in the claims should not be construed as meaning that the introduction of an element of another claim by the indefinite article "a" or "an" limits a particular claim that includes such an introduced claim element to an invention that includes only one such element, even if the same claim includes the introductory phrase "one or more" or "at least one" and an indefinite article such as "a" or "an". The same applies to the use of definite articles. Unless otherwise stated, terms such as "first" and "second" are used to arbitrarily distinguish between the elements they describe. Thus, these terms are not necessarily intended to indicate a temporal or other priority of such elements. The mere fact that certain measures are recited in mutually different claims does not indicate that a combination of these measures cannot be used to advantage.

Claims

1. 1. A method for determining changes between planning and treatment images of a subject, comprising: defining one or more clinical volumes on the planning images of a subject and defining a planning envelope volume around the clinical volumes for the planning images; acquiring treatment images from the subject at locations corresponding to locations of the planning images, the treatment images having the same planning envelope volume as the planning images; determining a position of the one or more clinical volumes on the treatment image relative to the planning envelope volume; and determining an inclusion metric for one or more of the one or more clinical volumes defining a degree of inclusion of the clinical volume on the treatment image within the planning envelope volume on the treatment image.

2. Determining the inclusion metric comprises: designating one or more representative points on the surface of the clinical volume; determining a shortest distance for the one or more representative points of the clinical volume to the planning envelope volume; if the representative point on the clinical volume is within the planning envelope volume, the shortest distance is classified as either a positive or negative internal distance, and if the representative point on the clinical volume is outside the planning envelope volume, the shortest distance is an external distance classified with the opposite sign to the internal distance; 2. The method of claim 1, wherein the containment metric value is determined as either the minimum of the shortest signed distance values ​​of the distances for the one or more keypoints when the internal distance is classified as positive, or the maximum of the shortest signed distance values ​​of the distances for the one or more keypoints when the internal distance is classified as negative.

3. The method of claim 1 or 2, wherein the planning image and the plurality of treatment images are 3D images.

4. The method of claim 3 , wherein the planning image and the plurality of treatment images are CT, PET, SPECT or MRI images.

5. 5. The method of claim 1, wherein after the planning envelope volume is defined, a distance transform is calculated around the planning envelope volume, and said distance transform is used to determine the containment metric.

6. The method of any one of claims 1 to 4, wherein the containment metric is determined by simulation of the movement of the one or more treatment clinical volumes relative to the planning envelope volume.

7. The method of claim 6 , wherein the simulated movement is a linear movement.

8. Determining the location of the one or more clinical volumes on the treatment image relative to the planning envelope volume includes: determining a geometric relationship between the planning image and the treatment image and mapping a location of the clinical volume across the treatment image; and identifying a plurality of anatomical features on the therapeutic image to locate the clinical volume.

9. The method according to any one of claims 1 to 8, wherein the planning envelope volume is defined as the clinical volume plus a margin specified by a pre-set protocol.

10. The method of any one of claims 1 to 9, wherein the clinical volume is a clinical target volume for a treatment plan or a total tumor volume.

11. The method according to any one of claims 1 to 9, wherein said clinical volume represents an organ at risk to be avoided in the treatment.

12. The method of any one of claims 1 to 11, wherein the steps are repeated over a set period of time to monitor changes in the containment metric on the plurality of therapeutic images.

13. The method of claim 12, wherein the set period of time varies between six hours and three months.

14. The method of claim 12 or 13, wherein the change in the containment metric over time is displayed to a user.

15. 15. The method according to any one of claims 1 to 14, wherein the subject is positioned in a specified position in a scanner for the planning and treatment images, and a user of the system can adjust the position of the subject according to one or more determined containment metrics.

16. 16. The method of claim 15 when dependent on claim 14, wherein changes in the containment metric over time are displayed to a user, the display further indicating when the containment metric exceeds a pre-set threshold indicative of unsafe movement of the subject within the scanner.

17. A method according to any preceding claim, further comprising the step of displaying one or more of the determined inclusive metrics to a user of the system.

18. 18. The method of any one of claims 1 to 17, further comprising using the containment metric to determine a 3D representation indicative of movement margins indicating how a subject can be moved within a scanner while maintaining or improving the position of the clinical volume relative to the position of the planning envelope volume.

19. A method according to any preceding claim, wherein the clinical volume is displayed as a 3D rendered surface and the containment metric for the surface is shown as a heatmap on the surface.

20. 1. A system for analyzing medical images to determine changes between planning and treatment images of a subject, comprising: a processor, the processor comprising: determining one or more clinical volumes on a planning image of a subject and defining a planning envelope volume around the clinical volumes for said planning images; analyzing treatment images from the subject for locations corresponding to locations of the planning images, the treatment images having the same planning envelope volume as the planning images; determining a position of the one or more clinical volumes on the treatment image relative to the planning envelope volume; determining an inclusion metric for one or more of the one or more clinical volumes defining a degree of inclusion of the clinical volume on the treatment image within the planning envelope volume on the treatment image.

21. The system of claim 20, wherein the processor is further configured to perform a number of steps of the method of any one of claims 2 to 19.

22. 23. The system of claim 21 or 22, further comprising a display for displaying at least one of the planning image, the treatment image, the one or more clinical volumes, and the planning envelope volume.

23. A computer program product comprising a plurality of instructions, said plurality of instructions, when said program is executed by a computer, causing said computer to perform the method of any one of claims 1 to 19.

Images