X-ray transmission image analysis for evaluation of linac isocenter quality

Abstract

To provide systems and methods for determining a radiation isocenter of a linear accelerator (LINAC).SOLUTION: Determining the radiation isocenter may include determining a set of three-dimensional (3D) radiation beam axes 908 of the LINAC from two-dimensional (2D) radiation transmission images. The radiation isocenter 904 may be determined based on at least the set of 3D radiation beam axes. Determining the set of 3D radiation beam axes may include constructing a 3D radiation beam axis based on a determined location of a beam axis of a radiation beam generated with a gantry of the LINAC at an angle relative to a reference gantry angle, a determined center of a shadow of a radiation opaque marker in the radiation field of the radiation beam, and the gantry angle.SELECTED DRAWING: Figure 10

Description

Cross - reference to related applications

[0001]

[0001] This application claims the benefit of priority of U.S. Provisional Patent Application No. 63 / 176,494, filed on April 19, 2021, which is hereby incorporated by reference in its entirety.

Technical Field

[0002]

[0002] Technical Field

[0003]

[0003] The present invention generally relates to radiation therapy in which a medical linear accelerator (LINAC) delivers a radiation beam to an exact point within a patient. In particular, the present invention relates to analyzing and displaying two - dimensional X - ray transmission images for evaluating the quality of the radiation isocenter of a LINAC.

Background Art

[0004]

[0004] Background Art

[0005]

[0005] 1.1 Overview of LINAC

[0006]

[0006] Radiation therapy is a type of cancer treatment that uses a beam of powerful high - energy radiation to kill cancer cells. FIGS. 1 - 3 illustrate a medical linear accelerator (LINAC) 100 including a gantry 102 and a couch 106. As shown in FIG. 4, the gantry 102 includes a collimator head 104 that defines the field of the radiation beam 412 generated by the LINAC 100. During treatment, the gantry 102 of the LINAC 100 delivers the radiation beam to an exact location within a patient (not shown) supported on the couch 106. To customize the radiation treatment plan, the gantry 102 can rotate around the patient, and the couch 106 can change the orientation of the patient relative to the rotation plane of the gantry 102.

[0007]

[0007] As shown in Figure 2, the gantry 102 may be capable of rotating a full 360° around the patient to optimize the entry of the radiation beam into the patient. As shown in Figure 2, the gantry 102 may rotate about the gantry rotation axis 208. As shown in Figure 3, the couch 106 may rotate about the couch rotation axis 310. This allows the patient's orientation to be adjusted relative to the rotation plane of the gantry 102 in order to further optimize the entry of the radiation beam 412 into the patient.

[0008]

[0008] In an ideal LINAC 100, the radiation beam would remain focused on a fixed point in space (i.e., on the tumor in the patient) as the gantry 102 rotates. Similarly, an ideal LINAC 100 would keep the tumor fixed within the radiation beam as the patient moves due to the rotation of the couch 106.

[0009]

[0009] As shown in Figure 4, the LINAC100 may have an imaging device 416 (e.g., an electronic portal imaging device (EPID)) which is used to confirm the patient's position before treatment by measuring the intensity of X-rays transmitted from the radiation port through the patient during the treatment session.

[0010]

[0010] 1.2 Radiation Isocenter

[0011]

[0011] The objective of radiotherapy is to deliver the highest possible dose of radiation to the tumor while minimizing the dose to surrounding healthy cells. To do so, radiotherapy includes the objectives of (1) keeping the radiation beam precisely focused on the patient's tumor as the gantry 102 rotates, and (2) keeping the patient's tumor fixed within the radiation beam as the couch 106 rotates. These two objectives, however, are rarely fully achieved. The gantry 102 generally flexes under its own weight, substantially blurring the radiation beam. The rotation of the couch 106 is also often not perfectly circular, and / or the axis of rotation 310 of the couch 106 generally does not intersect the axis of rotation 208 of the gantry 102.

[0012]

[0012] To minimize the effects of these common LINAC imperfections, the tumor should be positioned at a single point in space that minimizes the maximum beam-to-target error. This location is called the isocenter.

[0013]

[0013] 1.3 Isocenter size

[0014]

[0014] The maximum beam axis miss distance for all combinations of gantry and couch rotation is the isocenter size. A smaller isocenter will result in more precise treatment.

[0015]

[0015] 1.4 Existing Methods

[0016]

[0016] Existing methods for analyzing isocenters include two main steps: (1) measuring EPID X-ray transmission images of radiopaque markers positioned at (or near) the isocenter with different gantry and couch rotations, and (2) analyzing the images to determine the size of the isocenter.

[0017]

[0017] The setup shown in Figure 4 is typically used to accomplish the first step. In Figure 4, the LINAC 100 delivers a radiation beam 412 through a radiopaque marker 414 located at the isocenter. An imaging device 416 acquires a two-dimensional transmission image of the radiation beam 412 as it passes through the marker 414.

[0018]

[0018] A detailed view of the marker assembly 500, including the radiopaque marker 414, is shown in Figure 5. In Figure 5, the high-density spherical marker 414 is connected to a low-density support rod 502, which is then connected to a base 504. The base 504 is set on top of the couch 106 so that the marker 414 can be positioned at (or near) the isocenter.

[0019]

[0019] Figure 6 shows a typical image 600 (e.g., an EPID image) that has passed through the marker 414. In Figure 6, the dark square region 602 is created by the square radiation field of the radiation beam 412 that exposes the imaging device 410, and the brighter inner circular shape 604 is created by the shadow of the radiopaque marker 414 located within the radiation field.

[0020]

[0020] Existing methods for measuring isocenters involve rotating the gantry 102 through a set of fixed gantry angles while the couch 106 remains at 0°, and acquiring an EPID image for each gantry position. Additional EPID images are acquired while the gantry 102 remains fixed at 0 and the couch 106 rotates through a set of fixed couch angles.

[0021]

[0021] The second step of the existing method includes analyzing the images described above to determine the size of the isocenter. This step can be further broken down into two subprocesses: (2a) using image processing techniques to find the center of the field and the center of the marker 414 in each image (an example of an image 700 with the determined field and marker locations is shown in Figure 7), and (2b) calculating the largest marker center to field center distance for all acquired images and assigning this maximum value to the isocenter size. [Overview of the project]

[0022]

[0022] The existing use of field-to-marker distances for all EPID images acquired to analyze isocenters models a complex three-dimensional process using a two-dimensional coordinate system (CS). Existing methods maintain data acquired from EPID images in a two-dimensional image CS. In doing so, it is often difficult to intuitively understand the nature of the problem when a LINAC isocenter problem arises. Aspects of the present invention can overcome one or more of the problems of existing methods (for example, by replacing and improving subprocess step 2b of the existing method).

[0023]

[0023] Aspects of the present invention may include (1) using a linear accelerator (LINAC) to acquire a radiographic image of a radiopaque marker positioned at (or near) an isocenter using different gantry and / or couch rotations, and (2) analyzing the image to determine the size of the isocenter, the marker placement error (i.e., the isocenter position relative to the current marker position), and / or couch walkout. Analyzing the image may include finding the center of the radiation field and the center of the marker in the image.

[0024]

[0024] With regard to determining the isocenter size, instead of comparing the center of the field and the center of the marker to find the maximum difference and assigning the maximum distance as the isocenter size, the center of the field and the center of the marker can be transformed from a two-dimensional (2D) image coordinate system (CS) to a three-dimensional (3D) (e.g., real-world) CS. Transitioning from a 2D image CS to a 3D image CS may offer the advantage of enabling more flexible analysis and intuitive display in 3D that more accurately reflects the underlying process.

[0025]

[0025] In some embodiments, when analyzing images to depict the isocenter of LINAC, gantry and couch rotations may be treated differently. When the gantry rotates, the field moves, but the markers do not. Similarly, when the couch rotates, the markers move, but the field remains stationary. In some embodiments, images may therefore be grouped into two categories: (1) images for isocenter determination and / or marker placement error determination that can be obtained using gantry rotation while the couch remains at a fixed couch angle (e.g., 0°), and (2) images for couch walkout determination that can be obtained using couch rotation while the gantry remains at a fixed gantry angle (e.g., 0°).

[0026]

[0026] One aspect of the present invention may provide a method for determining a set of three-dimensional (3D) radiation beam axes of a linear accelerator (LINAC) from a two-dimensional (2D) radiographic image, wherein the set of 3D radiation beam axes includes at least first and second 3D radiation beam axes. The method may also include determining the radiation isocenter of the LINAC based on at least the set of 3D radiation beam axes.

[0027]

[0027] In some embodiments, determining the first 3D radiation beam axis may include positioning the LINAC gantry at a first gantry angle relative to a reference gantry angle, and positioning the gantry may include rotating the gantry about the gantry rotation axis. In some embodiments, determining the first 3D radiation beam axis may include using the gantry positioned at the first gantry angle to operate the LINAC to generate a first radiation beam. In some embodiments, determining the first 3D radiation beam axis may include using the gantry positioned at the first gantry angle to operate the LINAC imaging device to obtain a first two-dimensional (2D) radiation transmission image indicative of the radiation field of the first radiation beam after passing near a radiation-opaque marker. In some embodiments, determining the first 3D radiation beam axis may include determining, based on the first 2D radiation transmission image, the location of the beam axis of the first radiation beam and the center of the shadow of the radiation-opaque marker in the radiation field of the first radiation beam. In some embodiments, determining the first 3D radiation beam axis may include constructing the first 3D radiation beam axis based on the determined location of the beam axis of the first radiation beam, the determined center of the shadow of the radiation-opaque marker in the radiation field of the first radiation beam, and the first gantry angle.

[0028]

[0028] In some embodiments, constructing the first 3D radiation beam axis may include rotating the first 2D radiation transmission image about a first image rotation axis that coincides with the determined center of the shadow of the radiation-opaque marker in the radiation field of the first radiation beam and is parallel to the gantry rotation axis. In some embodiments, the first 2D radiation transmission image may be rotated by an amount equal to the first gantry angle. In some embodiments, the first 3D radiation beam axis may coincide with the determined location of the beam axis of the first radiation beam in the rotated first 2D radiation transmission image and be perpendicular to the plane of the rotated first 2D radiation transmission image.

[0029]

[0029] In some alternative embodiments, constructing the first 3D radiation beam axis may include generating an initial 3D radiation beam axis that coincides with the determined location of the beam axis of the first radiation beam in the first 2D radiation transmission image and that is perpendicular to the plane of the first 2D radiation transmission image, and rotating the initial 3D radiation beam axis about a first image rotation axis that coincides with the determined center of the shadow of the radiation-opaque marker in the radiation field of the first radiation beam and that is parallel to the gantry rotation axis, where the initial 3D radiation beam axis may be rotated by an amount equal to the first gantry angle.

[0030]

[0030] In some embodiments, determining the second 3D radiation beam axis may include positioning the gantry of the LINAC at a second gantry angle relative to a reference gantry angle. In some embodiments, determining the second 3D radiation beam axis may include using the gantry positioned at the second gantry angle to operate the LINAC to generate a second radiation beam. In some embodiments, determining the second 3D radiation beam axis may include using the gantry positioned at the second gantry angle to operate the imaging device of the LINAC to acquire a second 2D radiation transmission image showing the radiation field of the second radiation beam after passing near the radiation-opaque marker. In some embodiments, determining the second 3D radiation beam axis may include determining, based on the second 2D radiation transmission image, the location of the beam axis of the second radiation beam and the center of the shadow of the radiation-opaque marker in the radiation field of the second radiation beam. In some embodiments, determining the second 3D radiation beam axis may include constructing the second 3D radiation beam axis based on the determined location of the beam axis of the second radiation beam, the determined center of the shadow of the radiation-opaque marker in the radiation field of the second radiation beam, and the second gantry angle.

[0031]

[0031] In some embodiments, constructing a second 3D radiation beam axis may involve rotating a second 2D radiographic image about a second image rotation axis that coincides with a determined center of the shadow of the radiopaque marker in the radiation field of the second radiation beam and is parallel to the gantry rotation axis. In some embodiments, the second 2D radiographic image may be rotated by an amount equal to the second gantry angle. In some embodiments, the second 3D radiation beam axis may coincide with a determined location of the beam axis of the second radiation beam in the rotated second 2D radiographic image and be perpendicular to the plane of the rotated second 2D radiographic image.

[0032]

[0032] In some alternative embodiments, constructing a second 3D radiation beam axis may include generating a first second 3D radiation beam axis that coincides with a determined location of the beam axis of the second radiation beam in the second 2D radiographic image and is perpendicular to the plane of the second 2D radiographic image, and rotating the first second 3D radiation beam axis about a second image rotation axis that coincides with a determined center of the shadow of the radiopaque marker in the radiation field of the second radiation beam and is parallel to the gantry rotation axis, the first second 3D radiation beam axis may be rotated by an amount equal to the second gantry angle.

[0033]

[0033] In some embodiments, the couch may be positioned at a fixed couch angle while the LINAC is used to generate first and second radiation beams and the imaging device of the LINAC is used to acquire first and second 2D radiographic images.

[0034]

[0034] In some embodiments, determining a radiation isocenter may involve determining the off-beam distance between a location in 3D space and a 3D radiation beam axis for each 3D radiation beam axis in a set of 3D radiation beam axes. In some embodiments, determining a radiation isocenter may involve determining which of the determined off-beam distances for a 3D radiation beam axis in a set of 3D radiation beam axes is the largest, and the largest of the determined off-beam distances for a 3D radiation beam axis in a set of 3D radiation beam axes may be the maximum off-beam distance for the location. In some embodiments, determining a radiation isocenter may involve finding a location in 3D space that has the smallest maximum off-beam distance. In some embodiments, the off-beam distance between a location in 3D space and a 3D radiation beam axis may be the shortest distance between the location and the 3D radiation beam axis. In some embodiments, the size of the radiation isocenter may be the maximum off-beam distance for the location.

[0035]

[0035] In some embodiments, the method may further include positioning the tumor at a determined radiation isocenter.

[0036]

[0036] In some embodiments, the method may further include determining a marker movement vector for each couch angle in a set of couch angles, the set of couch angles may include at least a first couch angle relative to a reference couch angle and a second couch angle relative to a reference couch angle. In some embodiments, the method may further include determining, based on the determined marker movement vector, the distance from the marker to the 3D radiation beam axis error for each combination of a couch angle in the set of couch angles and a 3D radiation beam axis in a set of 3D radiation beam axes, the distance from the marker to the 3D radiation beam axis error may represent the shortest distance between the location of the radiopaque marker and the 3D radiation beam axis error.

[0037]

[0037] In some embodiments, determining the marker movement vector for a first couch angle may involve using LINAC to generate a first reference radiation beam using a LINAC couch positioned at a reference couch angle and a LINAC gantry positioned at a reference gantry angle. In some embodiments, determining the marker movement vector for a first couch angle may involve using LINAC's imaging device to acquire a reference couch angle two-dimensional (2D) radiographic image showing the radiation field of the first reference radiation beam after it has passed near a radiopaque marker, using a couch positioned at a reference couch angle and a gantry positioned at a reference gantry angle. In some embodiments, determining the marker movement vector for a first couch angle may involve determining the center of the shadow of the radiopaque marker in the radiation field of the first reference radiation beam based on the reference couch angle 2D radiographic image. In some embodiments, determining the marker movement vector for a first couch angle may include positioning the couch at the first couch angle. In some embodiments, determining the marker movement vector for a first couch angle may include using LINAC to generate a second reference radiation beam using a couch positioned at the first couch angle and a gantry positioned at a reference gantry angle. In some embodiments, determining the marker movement vector for a first couch angle may include using the imaging device of LINAC to acquire a 2D radiographic image of the first couch angle showing the radiation field of the second reference radiation beam after it has passed near the radiopaque marker, using a couch positioned at the first couch angle and a gantry positioned at a reference gantry angle. In some embodiments, determining the marker movement vector for a first couch angle may include determining the center of the shadow of the radiopaque marker in the radiation field of the second reference radiation beam, based on the 2D radiographic image of the first couch angle.In some embodiments, determining the marker movement vector for a first couch angle may involve determining the marker movement vector for a first couch angle based on a comparison between the center of the shadow of the radiopaque marker in the radiation field of a first reference radiation beam and the center of the shadow of the radiopaque marker in the radiation field of a second reference radiation beam.

[0038]

[0038] In some embodiments, determining the marker movement vector for a second couch angle may include positioning the couch at the second couch angle. In some embodiments, determining the marker movement vector for a second couch angle may include using LINAC to generate a third reference radiation beam using a couch positioned at the second couch angle and a gantry positioned at a reference gantry angle. In some embodiments, determining the marker movement vector for a second couch angle may include using the imaging device of LINAC to acquire a second couch angle 2D radiographic image showing the radiation field of the third reference radiation beam after it has passed near a radiopaque marker, using a couch positioned at the second couch angle and a gantry positioned at a reference gantry angle. In some embodiments, determining the marker movement vector for a second couch angle may include determining the center of the shadow of the radiopaque marker in the radiation field of a third reference radiation beam based on a 2D radiographic image of the second couch angle. In some embodiments, determining the marker movement vector for a second couch angle may include determining the marker movement vector for a second couch angle based on a comparison between the center of the shadow of the radiopaque marker in the radiation field of a first reference radiation beam and the center of the shadow of the radiopaque marker in the radiation field of a third reference radiation beam.

[0039]

[0039] In some embodiments, the method may further include determining a clinical isocenter, which comprises determining a spatial location that minimizes the distance from the largest marker to the beam axis error. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the 3D radiation beam axis error may include predicting a couch rotation axis based on the shadows of radiopaque markers for different couch angles. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the 3D radiation beam axis error may include predicting a set of positions for radiopaque markers at a new reference position displaced by couch rotation around the predicted couch rotation axis. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the beam axis error may include determining the predicted distance from the marker to the 3D beam axis error for each combination of couch angles from a set of couch angles and 3D radiation beam axes from a set of 3D radiation beam axes, based on the determined marker movement vector, where the predicted distance from the marker to the 3D radiation beam axis error may represent the shortest distance between the location of the radiopaque marker and the 3D radiation beam axis error. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the beam axis error may include determining a reference marker position that minimizes the distance from the largest marker to the beam axis error. In some embodiments, the method may further include positioning the tumor at the determined clinical isocenter.

[0040]

[0040] Another aspect of the present invention may provide an apparatus configured to determine a set of three-dimensional (3D) radiation beam axes of a linear accelerator (LINAC) from a two-dimensional (2D) radiographic image, the set of 3D radiation beam axes may include at least first and second 3D radiation beam axes. The apparatus may be configured to determine the radiation isocenter of the LINAC based on at least the set of 3D radiation beam axes.

[0041]

[0041] Another aspect of the present invention may provide a method for determining a marker movement vector for each couch angle in a set of couch angles of a linear accelerator (LINAC), the set of couch angles may include at least a first couch angle relative to a reference couch angle and a second couch angle relative to a reference couch angle. Based on the determined marker movement vector, the method may include determining the distance from the marker to the 3D radiation beam axis error for each combination of a couch angle in the set of couch angles and a 3D radiation beam axis in a set of 3D radiation beam axes, the distance from the marker to the 3D radiation beam axis error may represent the shortest distance between the location of the radiopaque marker and the 3D radiation beam axis error.

[0042]

[0042] A further aspect of the present invention may provide a device configured to determine a marker movement vector for each couch angle in a set of couch angles of a linear accelerator (LINAC), the set of couch angles may include at least a first couch angle relative to a reference couch angle and a second couch angle relative to a reference couch angle. The device may be configured to determine, based on the determined marker movement vector, the distance from the marker to the 3D radiation beam axis error for each combination of a couch angle in the set of couch angles and a 3D radiation beam axis in a set of 3D radiation beam axes, the distance from the marker to the 3D radiation beam axis error may represent the shortest distance between the location of the radiopaque marker and the 3D radiation beam axis error.

[0043]

[0043] Another aspect of the present invention may provide a computer program that includes instructions for adapting an apparatus to perform any of the methods described above. Yet another aspect of the present invention may provide a carrier that contains a computer program, the carrier may be an electrical signal, an optical signal, a radio signal, or a computer-readable storage medium.

[0044]

[0044] Another aspect of the present invention may provide an apparatus including a processing circuit and a memory. The memory may contain instructions that can be executed by the processing circuit, thereby enabling the apparatus to operate to perform any of the methods described above.

[0045]

[0045] Yet another aspect of the present invention may provide an apparatus adapted to any of the methods described above.

[0046]

[0046] Yet another aspect of the present invention may provide any combination of the aspects described above.

[0047]

[0047] Further variations of the system and method are described below in the embodiments for carrying out the invention of the present invention.

[0048]

[0048] The accompanying drawings incorporated herein and forming part thereof illustrate various non-limiting embodiments of the invention. In the drawings, similar reference numerals indicate identical or functionally similar elements. [Brief explanation of the drawing]

[0049] [Figure 1]

[0049] A medical linear accelerator (LINAC) is given as an example. [Figure 2]

[0050] This illustrates the rotation of the LINAC gantry. [Figure 3]

[0051] This illustrates the rotation of the LINAC couch. [Figure 4]

[0052] Examples of LINAC's radiation beam and imaging panel are shown. [Figure 5]

[0053] Several embodiments of marker assemblies are illustrated below. [Figure 6]

[0054] Examples of unprocessed radiographic images in several configurations are provided. [Figure 7]

[0055] Examples of processed radiographic images in several configurations are provided. [Figure 8]

[0056] The difference between the position of a radiopaque marker and the center of the radiation field is illustrated in several embodiments. [Figure 9]

[0057] Processed 2D images showing the discovered radiation field, the center of the radiation field, the shadow of the discovered marker, and the central axis of the radiation field are illustrated in several embodiments. [Figure 10]

[0058] Several embodiments illustrate the rotation of the original 2D image to a new location, centered on the center of the marker's shadow, by an amount and direction corresponding only to the amount and direction of gantry rotation during image acquisition. [Figure 11]

[0059] Several embodiments illustrate the transformation from 2D image coordinates to 3D real-world coordinates, and the superposition of beam axes defining isocenters. [Figure 12]

[0060] This document illustrates the calculation of tumor beam off-axis distance for different couch marker walkout positions with varying couch rotation angles and combinations of gantry and couch angles, in several different configurations. [Figure 13]

[0061] Several examples of the process are given. [Figure 14]

[0062] Several embodiments of the process for determining the first 3D radiation beam axis are illustrated. [Figure 15]

[0063] Several approaches to determining a radiation isocenter are illustrated. [Figure 16]

[0064] This document illustrates several approaches to determining a couch walkout. [Figure 17]

[0065] Several methods for determining the marker movement vector with respect to the couch angle are illustrated. [Figure 18]

[0066] Several embodiments of the LINAC controller are illustrated below. [Figure 19]

[0067] Several embodiments of the apparatus are illustrated below. [Modes for carrying out the invention]

[0050]

[0068] In this application, the term "beam axis" may refer to a three-dimensional vector representing the center of a radiation beam.

[0051]

[0069] In this application, the term "beam-axis off-axis distance" may refer to the shortest distance between the target (e.g., the center of a tumor) and any point along the beam axis.

[0052]

[0070] In this application, the terms "CS" or "coordinate system" may refer to a two-dimensional (2D) CS, such as the CS of an image (e.g., an EPID image), or a three-dimensional (3D) CS, such as the CS of a linear accelerator (LINAC) (e.g., LINAC100).

[0053]

[0071] In this application, the term “couch walkout” may refer to the amount by which a patient’s tumor moves as the couch 106 rotates through its full range of motion, when positioned at (or near) the isocenter.

[0054]

[0072] In this application, the term "couch" may refer to a component (e.g., couch 106) of a LINAC (e.g., LINAC 100) that supports a patient.

[0055]

[0073] In this application, the term “imaging device” (e.g., electronic portal imaging device (EPID)) may refer to a component of a LINAC (e.g., LINAC 100) (e.g., imaging device 416) that measures the intensity of X-rays transmitted from the radiation port through the patient (e.g., during a treatment session). The imaging device may electronically convert the radiation signal into a two-dimensional (2D) digital X-ray image, for example, to verify the correct beam placement relative to the patient’s anatomical structures.

[0056]

[0074] In this application, the term “gantry” may refer to a component (e.g., gantry 102) of the LINAC (e.g., LINAC 100) that rotates around the patient while delivering a radiation beam.

[0057]

[0075] In this application, the term "isocenter" may refer to a location in space that minimizes the off-axis radiation beam distance for all gantry angles.

[0058]

[0076] In this application, the term "marker placement error" may refer to a three-dimensional vector between the actual LINAC isocenter location and the marker placement location when acquiring EPID images for isocenter analysis.

[0059]

[0077] In some embodiments, the LINAC100 may (1) acquire a radiographic image of a radiopaque marker 414 positioned at (or near) the isocenter using different gantry and / or couch rotations, and (2) analyze the image to determine the size of the isocenter, the marker placement error (i.e., the isocenter position relative to the current marker position), and / or the couch walkout. Analyzing the image may include finding the center of the radiation field and the center of the marker in the image.

[0060]

[0078] 2.1 Isocenter

[0061]

[0079] In some embodiments, determining the isocenter may involve analyzing images (e.g., EPID images) in which the couch 106 remains at a fixed couch angle (e.g., 0°) while the gantry angle changes (e.g., through its entire range of motion). In some embodiments, determining the isocenter may involve analyzing only images in which the couch 106 remains at a fixed couch angle while the gantry angle changes. In some embodiments, this may involve determining the location of the beam axis of the radiation beam in the image and finding the center of the marker 414 in the image (see, for example, Figures 6-8). As shown in Figures 9 and 10, in some embodiments, this may involve transforming the beam axis location and the center of the marker from a two-dimensional (2D) image coordinate system (CS) to a three-dimensional (3D) CS. In some embodiments, as shown in Figures 9 and 10, the gantry 102 rotates while the marker 414 remains fixed in space, so the conversion from a 2D image CS to a 3D (e.g., real-world) CS can use the shadow 906 of the marker 414 as a fixed reference point (for example, the determination of all beam axes would be relative to the location of the marker, which is a fixed point in space).

[0062]

[0080] In some embodiments, as shown in Figure 9, depicting the isocenter may involve using image processing techniques for each image acquired by the imaging device 416 (e.g., EPID) of the LINAC 100 while the gantry 102 is rotating and the couch angle remains fixed, to (i) determine the location of the beam axis 908 of the radiation beam 412 generated using the gantry 102 of the LINAC 100, and (ii) find the center of the shadow 906 of the marker 414 in the radiation field 902 of the radiation beam 412. In some embodiments, determining the location of the beam axis 908 may involve finding the center 904 of the radiation field 902 of the radiation beam 412 generated using the gantry 102 of the LINAC 100. In some alternative embodiments, determining the location of the beam axis 908 may involve finding the average of the center 904 of the radiation field 902 generated using the gantry 102 of the LINAC 100 with (i) a first collimator angle and (ii) a second collimator angle that is 180 degrees different from the first collimator angle, (for example, to account for errors when the radiation field is not well calibrated).

[0063]

[0081] In some embodiments, as shown in Figure 10, depicting the isocenter may involve constructing a beam axis 908 that coincides with a determined location of the beam axis 908 in the image and is perpendicular to the imaging plane for each image acquired by the imaging device 416 while the gantry 102 is rotating and the couch angle remains fixed. In some embodiments, as shown in Figure 10, for each image acquired by the imaging device 416 while the gantry 102 is rotating and the couch angle remains fixed (except for images acquired at a gantry angle of 0° where rotation would not be necessary), the constructed beam axis 908 may be rotated to a new position around the center 1010 of the shadow 906 of the marker 414 by an amount and direction equal to the gantry rotation used when the image was acquired. Several alternative embodiments, as shown in Figure 10, may include rotating a first 2D radiographic image about a first image rotation axis that coincides with a determined center of the shadow of the radiopaque marker in the radiation field of the first radiation beam and is parallel to the gantry rotation axis, for each image acquired by the imaging device 416 while the gantry 102 is rotating and the couch angle remains fixed (except for images acquired at a gantry angle of 0° where rotation would not be required), the first 2D radiographic image may be rotated by an amount equal to the gantry rotation angle, and the first 3D radiation beam axis may coincide with a determined location of the beam axis of the first radiation beam in the rotated first 2D radiographic image and be perpendicular to the plane of the rotated first 2D radiographic image. Figure 10 shows the rotated radiation field 1002, the center 1004 of the rotated radiation field 1002, the shadow 1006 of the marker 414 in the rotated radiation field 1002, and the rotated beam axis 1008 in several embodiments.

[0064]

[0082] In some embodiments, as shown in Figure 11, rotating the constructed beam axis by an amount equal to the gantry rotation angle (or rotating the image by an amount equal to the gantry rotation angle and using the rotated image to construct the beam axis) can create a set of three-dimensional beam axes (e.g., including rotated beam axes 1008, 1110, and 1112, as well as the constructed beam axis 1108 with a gantry rotation angle of 0°) having a common shadow 1106 of the marker 414 in the radiation field 1102 of the radiation beam 412 at gantry 102 (and couch 106 with a fixed couch angle) as a reference point (e.g., for the calculation of the isocenter 1114). In some embodiments, it may be necessary to use the three-dimensional beam axes (e.g., including 3D beam axes 1008, 1108, 1110, and 1112) to calculate the isocenter 1114. Some embodiments may involve calculating the isocenter 1114 by finding a point in space that minimizes the maximum beam-axis-off distance between that point and all of the three-dimensional beam axes. Some embodiments may involve determining the size of the isocenter as the maximum off-axis distance between the center of the isocenter 1114 and all of the three-dimensional beam axes under consideration.

[0065]

[0083] 2.2 Marker Placement Error

[0066]

[0084] In some embodiments, the three-dimensional isocenter analysis described in Section 2.1 above may have the additional benefit of separating any marker placement errors from the resulting isocenter size. For example, as shown in Figure 11, the marker positioning error may be determined as a vector 1116 from (a) the center of the shadow 1106 of the marker 414 in the radiation field 1102 of the radiation beam 412 at a gantry rotation angle of 0° (and couch 106 at a fixed couch angle) to (b) the location of the isocenter 1114 found.

[0067]

[0085] 2.3 Couch Walkout

[0068]

[0086] In some embodiments, this may include performing a couch walkout determination to determine how much a tumor positioned at the isocenter 1114 will move as the couch 106 rotates. In some embodiments, in contrast to an isocenter determination in which the marker 414 remains fixed and the radiation field changes as the gantry 102 rotates, the LINAC 100 may perform a couch walkout measurement using a fixed radiation field (e.g., the gantry 102 positioned at a fixed gantry angle) and a marker 414 that (potentially) moves as the couch 106 rotates.

[0069]

[0087] In some embodiments, measuring a couch walkout may involve using image processing techniques to find the center of the shadow of marker 414 in the radiation field of a radiation beam generated using gantry 102 for each image acquired by the imaging device 416 (e.g., EPID) of LINAC 100 while couch 106 rotates and the gantry angle remains fixed. For example, as shown in Figure 12, the center of the shadow 1106 of marker 414 at couch 106 with a couch angle of 0° and the centers of the shadows of marker 414 (e.g., shadows 1202, 1204, 1206, and 1208) can be found using couch 106 at different non-zero couch angles, with gantry 102 at a fixed gantry angle (e.g., 0°).

[0070]

[0088] Some embodiments may include using the position of the shadow 1106 of marker 414 in the radiation field at couch 106 with a couch angle of 0° as the baseline, and calculating the movement vector of the shadow of marker 414 from each image using non-zero couch rotation. Figure 12 shows the movement vectors by dotted lines from the center of the shadow 1106 of marker 414 at couch 106 with a couch angle of 0° to the respective centers of the shadows 1202, 1204, 1206, and 1208 of marker 414 at couch 106 with a non-zero couch angle.

[0071]

[0089] Some embodiments may involve calculating the error distance between the couch angle marker 414 and the 3D beam axis of the gantry angle for different combinations of gantry rotation and couch rotation. For example, as shown in Figure 12, the marker-to-beam axis distance 1210 can be calculated between the shadow 1208 of the marker 414 for a particular non-zero couch angle and the 3D beam axis 1112 for a particular non-zero gantry angle.

[0072]

[0090] In some embodiments, calculating the distance from the marker to the beam axis for all combinations of gantry rotation and couch rotation may allow us to understand how much the radiation beam will miss the tumor for all combinations of gantry angle and couch angle (even though the images for couch rotation may have all been acquired using gantry 102 fixed at one angle).

[0073]

[0091] 2.4 Flowchart

[0074]

[0092] Figure 13 illustrates process 1300 in several embodiments. In some embodiments, one or more steps of process 1300 may be performed by LINAC 100 (e.g., the controller of LINAC 100). In some embodiments, one or more steps of process 1300 may be performed by an apparatus (e.g., apparatus 1900 in Figure 19) in addition or alternatively. In some embodiments, process 1300 may include step 1302 of determining a set of three-dimensional (3D) radiation beam axes (e.g., radiation beam axes 1008, 1108, 1112, and 1114). In some embodiments, the set of 3D radiation beam axes may include at least the first and second 3D radiation beam axes 1008 and 1110. In some embodiments, process 1300 may include step 1304 of determining a radiation isocenter 1114 based on at least the set of 3D radiation beam axes.

[0075]

[0093] Figure 14 illustrates a process 1400 for determining a first 3D radiation beam axis 1008 in several embodiments. In some embodiments, one or more steps of process 1400 may be performed by LINAC 100 (e.g., the controller of LINAC 100). In some embodiments, one or more steps of process 1400 may be performed by an apparatus (e.g., the apparatus 1900 in Figure 19) in addition or alternatively. In some embodiments, process 1400 may be performed in step 1302 of process 1300 shown in Figure 13.

[0076]

[0094] In some embodiments, process 1400 may include step 1402 of positioning the gantry 102 of LINAC 100 at a first gantry angle (e.g., 90°) with respect to a reference gantry angle (e.g., 0°). In some embodiments, positioning the gantry 102 may include rotating the gantry 102 about the gantry rotation axis 208.

[0077]

[0095] In some embodiments, process 1400 may include step 1404 of using LINAC 100 to generate a first radiation beam using gantry 102 positioned at a first gantry angle. In some embodiments, process 1400 may include step 1406 of using imaging device 416 of LINAC 100 to acquire a first two-dimensional (2D) radiographic image showing the radiation field 902 of the first radiation beam after it has passed near the radiopaque marker 414, using gantry 102 positioned at the first gantry angle. In some embodiments, as shown in Figure 9, process 1400 may include step 1408 of determining, based on the first 2D radiographic image, the location of the beam axis 908 of the first radiation beam and the center of the shadow 906 of the radiopaque marker 414 in the radiation field 902 of the first radiation beam.

[0078]

[0096] In some embodiments, process 1400 may include step 1410 of constructing a first 3D radiation beam axis 1008 based on a determined location of the beam axis 908 of a first radiation beam, a determined center of the shadow 906 of the radiopaque marker 414 in the radiation field 902 of the first radiation beam, and a first gantry angle. In some embodiments, as shown in Figure 10, step 1410 of constructing the first 3D radiation beam axis 1008 may include rotating a first 2D radiographic image about a first image rotation axis that coincides with the determined center of the shadow 906 of the radiopaque marker 414 in the radiation field 902 of the first radiation beam and is parallel to the gantry rotation axis 208. In some embodiments, the first 2D radiographic image may be rotated by an amount equal to the first gantry angle. In some embodiments, as shown in Figure 10, the first 3D radiation beam axis 1008 may coincide with a determined location of the beam axis of the first radiation beam in the rotated first 2D radiographic image and be perpendicular to the plane of the rotated first 2D radiographic image.

[0079]

[0097] In some alternative embodiments, as shown in Figure 10, step 1410 for constructing the first 3D radiation beam axis 1008 may include (i) generating an initial 3D radiation beam axis 908 that coincides with a determined location of the beam axis 908 of the first radiation beam in the first 2D radiographic image and is perpendicular to the plane of the first 2D radiographic image, and (ii) rotating the initial 3D radiation beam axis about a first image rotation axis that coincides with a determined center of the shadow 906 of the radiopaque marker 414 in the radiation field of the first radiation beam and is parallel to the gantry rotation axis. In some embodiments, the initial 3D radiation beam axis may be rotated by an amount equal to the first gantry angle.

[0080]

[0098] In some embodiments, determining the second 3D radiation beam axis 1110 (for example, in step 1302 of process 1300) may include positioning the gantry 102 of LINAC 100 at a second gantry angle (e.g., 180°) relative to a reference gantry angle (e.g., 0°). In some embodiments, determining the second 3D radiation beam axis 1110 may include using LINAC 100 to generate a second radiation beam using the gantry 102 positioned at the second gantry angle. In some embodiments, determining the second 3D radiation beam axis 1110 may include using the imaging device 416 of LINAC 100 to acquire a second 2D radiographic image showing the radiation field of the second radiation beam after it has passed near the radiopaque marker 414, using the gantry 102 positioned at the second gantry angle. In some embodiments, determining the second 3D radiation beam axis 1110 may include determining the location of the beam axis of the second radiation beam and the center of the shadow of the radiopaque marker 414 in the radiation field of the second radiation beam, based on a second 2D radiographic image. In some embodiments, determining the second 3D radiation beam axis 1110 may include constructing the second 3D radiation beam axis 1110 based on the determined location of the beam axis of the second radiation beam, the determined center of the shadow of the radiopaque marker 414 in the radiation field of the second radiation beam, and the second gantry angle.

[0081]

[0099] In some embodiments, constructing a second 3D radiation beam axis 1110 (for example, in step 1302 of process 1300) may involve rotating a second 2D radiographic image about a second image rotation axis that coincides with a determined center of the shadow of the radiopaque marker 414 in the radiation field of the second radiation beam and is parallel to the gantry rotation axis 208. In some embodiments, the second 2D radiographic image may be rotated by an amount equal to the second gantry angle. In some embodiments, the second 3D radiation beam axis may coincide with a determined location of the beam axis of the second radiation beam in the rotated second 2D radiographic image and be perpendicular to the plane of the rotated second 2D radiographic image.

[0082]

[0100] In some embodiments, constructing a second 3D radiation beam axis 1110 (for example, in step 1302 of process 1300) may include (i) generating a first second 3D radiation beam axis that coincides with a determined location of the beam axis of the second radiation beam in the second 2D radiographic image and is perpendicular to the plane of the second 2D radiographic image, and (ii) rotating the first second 3D radiation beam axis about a second image rotation axis that coincides with a determined center of the shadow of the radiopaque marker in the radiation field of the second radiation beam and is parallel to the gantry rotation axis. In some embodiments, the first second 3D radiation beam axis may be rotated by an amount equal to the second gantry angle.

[0083]

[0101] In some embodiments, the couch 106 may be positioned at a fixed couch angle (e.g., 0°) while the LINAC 100 is used to generate first and second radiation beams (for example, in step 1302 of process 1300, and / or in at least steps 1402, 1404, and 1406 of process 1400) and the imaging device of the LINAC is used to acquire first and second 2D radiographic images.

[0084]

[0102] Figure 15 illustrates a process 1500 for determining a radiation isocenter 1114 in several embodiments. In some embodiments, one or more steps of process 1500 may be performed by LINAC 100 (e.g., the controller of LINAC 100). In some embodiments, one or more steps of process 1500 may be performed by an apparatus (e.g., the apparatus 1900 in Figure 19) in addition or alternatively. In some embodiments, process 1500 may be performed in step 1304 of process 1300 shown in Figure 13.

[0085]

[0103] In some embodiments, process 1500 may include step 1502 determining the off-beam distance between a location in 3D space and a 3D radiation beam axis for each 3D radiation beam axis in a set of 3D radiation beam axes (e.g., including 3D radiation beam axes 1008, 1108, 1112, and 1114). In some embodiments, the off-beam distance between a location in 3D space and a 3D radiation beam axis may be the shortest distance between the location and the 3D radiation beam axis. In some embodiments, process 1500 may include step 1504 determining which of the determined off-beam distances for a 3D radiation beam axis in a set of 3D radiation beam axes is the largest. In some embodiments, the largest of the determined off-beam distances for a 3D radiation beam axis in a set of 3D radiation beam axes may be the maximum off-beam distance for a location. In some embodiments, process 1500 may include step 1506 finding the location in 3D space that has the smallest maximum off-beam distance. In some embodiments, the size of the radiation isocenter 1114 may be the maximum off-axis beam distance for a given location. In some alternative embodiments, instead of finding a location that minimizes the maximum off-axis beam distance, other metrics can be used (e.g., minimizing the average off-axis beam distance, minimizing the root mean square error of the off-axis beam distance, or a combination thereof).

[0086]

[0104] In some embodiments, as shown in Figure 13, process 1300 may include an optional step 1306 of determining the marker placement error. In some embodiments, the marker placement error may be determined as a vector from the center of the shadow of the radiopaque marker 414 in the radiation field to a determined radiation isocenter 1114. In some embodiments, process 1300 may include using the determined marker placement error to move the radiopaque marker 414 to the determined radiation isocenter 1114.

[0087]

[0105] In some embodiments, as shown in Figure 13, process 1300 may include an optional step 1308 in which a couch walkout is determined.

[0088]

[0106] In some embodiments, as shown in Figure 13, the process 1300 may include an optional step 1310 of positioning the tumor at a determined radiation isocenter 1114.

[0089]

[0107] Figure 16 illustrates a process 1600 for determining a couch walkout in several embodiments. In some embodiments, one or more steps of process 1600 may be performed by LINAC 100 (e.g., the controller of LINAC 100). In some embodiments, one or more steps of process 1600 may be performed by an apparatus (e.g., apparatus 1900 in Figure 19) in addition or alternatively. In some embodiments, process 1600 may be performed in step 1308 of process 1300 shown in Figure 13.

[0090]

[0108] In some embodiments, process 1600 may include step 1602 of determining a marker movement vector for each couch angle in a set of couch angles. In some embodiments, the set of couch angles may include a first couch angle relative to a reference couch angle (e.g., 0°) and a second couch angle relative to the reference couch angle. In some embodiments, process 1600 may include step 1604 of determining the distance from the marker to the 3D radiation beam axis error (e.g., distance 1210 from the marker to the 3D radiation beam axis error) for each combination of a couch angle in the set of couch angles and a 3D radiation beam axis in the set of 3D radiation beam axes, based on the determined marker movement vector. In some embodiments, the distance from the marker to the 3D radiation beam axis error may represent the shortest distance between the location of the radiopaque marker and the 3D radiation beam axis error.

[0091]

[0109] Figure 17 illustrates a process 1700 for determining a marker movement vector for a couch angle (e.g., a first couch angle) in several embodiments. In some embodiments, one or more steps of process 1700 may be performed by LINAC 100 (e.g., a controller of LINAC 100). In some embodiments, one or more steps of process 1700 may be performed by an apparatus (e.g., apparatus 1900 in Figure 19) in addition or alternatively. In some embodiments, process 1700 may be performed in step 1602 of process 1600 and / or step 1308 of process 1300 shown in Figure 13.

[0092]

[0110] In some embodiments, process 1700 may include step 1702 of using LINAC 100 to generate a first reference radiation beam using the couch 106 of LINAC 100 positioned at a reference couch angle and the gantry 102 of LINAC 100 positioned at a reference gantry angle (e.g., 0°). In some embodiments, process 1700 may include step 1704 of using the imaging device 416 of LINAC 100 to obtain a reference couch angle two-dimensional (2D) radiographic image showing the radiation field 1102 of the first reference radiation beam after it has passed near a radiopaque marker 414, using the couch 106 positioned at a reference couch angle and the gantry 102 positioned at a reference gantry angle. In some embodiments, process 1700 may include step 1706 of determining the center of the shadow 1106 of the radiopaque marker 414 in the radiation field 1102 of a first reference radiation beam, based on a reference couch angle 2D radiographic image.

[0093]

[0111] In some embodiments, process 1700 may include step 1708 of positioning the couch at a first couch angle. In some embodiments, process 1700 may include step 1710 of using LINAC 100 to generate a second reference radiation beam using the couch 106 positioned at the first couch angle and the gantry 102 positioned at a reference gantry angle. In some embodiments, process 1700 may include step 1712 of using the imaging device 416 of LINAC 100 to obtain a first couch angle 2D radiographic image showing the radiation field of the second reference radiation beam after it has passed near a radiopaque marker, using the couch 106 positioned at the first couch angle and the gantry 102 positioned at a reference gantry angle. In some embodiments, process 1700 may include step 1714 of determining the center of the shadow 1208 of the radiopaque marker 414 in the radiation field of a second reference radiation beam, based on a 2D radiographic image of a first couch angle. In some embodiments, process 1700 may include step 1716 of determining a marker movement vector for the first couch angle, based on a comparison between the center of the shadow 1106 of the radiopaque marker 414 in the radiation field 1102 of the first reference radiation beam and the center of the shadow 1208 of the radiopaque marker in the radiation field of the second reference radiation beam.

[0094]

[0112] In some embodiments, determining the marker movement vector for a second couch angle in step 1602 may include positioning the couch 106 at the second couch angle. In some embodiments, determining the marker movement vector for a second couch angle in step 1602 may include using the LINAC 100 to generate a third reference radiation beam using the couch 106 positioned at the second couch angle and the gantry 102 positioned at the reference gantry angle. In some embodiments, determining the marker movement vector for a second couch angle in step 1602 may include using the imaging device 416 of the LINAC 100 to acquire a second couch angle 2D radiographic image showing the radiation field of the third reference radiation beam after it has passed near the radiopaque marker 414, using the couch 106 positioned at the second couch angle and the gantry 102 positioned at the reference gantry angle. In some embodiments, determining the marker movement vector for a second couch angle in step 1602 may include determining the center of the shadow 1206 of the radiopaque marker 414 in the radiation field of a third reference radiation beam based on a 2D radiographic image of the second couch angle. In some embodiments, determining the marker movement vector for a second couch angle in step 1602 may include determining the marker movement vector for a second couch angle based on a comparison between the center of the shadow 1106 of the radiopaque marker 414 in the radiation field 1102 of a first reference radiation beam and the center of the shadow 1206 of the radiopaque marker in the radiation field of a third reference radiation beam.

[0095]

[0113] In some embodiments, process 1600 may further include determining a clinical isocenter. In some embodiments, determining a clinical isocenter may include determining a spatial location that minimizes the distance from the largest marker to the beam axis error. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the 3D radiation beam axis error may include predicting a couch rotation axis based on the shadows of the radiopaque markers for different couch angles. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the 3D radiation beam axis error may include predicting a set of positions for the radiopaque markers 414 at a new reference position displaced by couch rotation around the predicted couch rotation axis. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the beam axis error may include determining the predicted distance from the marker to the 3D radiation beam axis error for each combination of couch angles from a set of couch angles and 3D radiation beam axes from a set of 3D radiation beam axes, based on the determined marker movement vector, where the predicted distance from the marker to the 3D radiation beam axis error may represent the shortest distance between the location of the radiopaque marker and the 3D radiation beam axis error. In some embodiments, determining a spatial location that minimizes the distance from the largest marker to the beam axis error may include determining a reference marker position that minimizes the distance from the largest marker to the beam axis error. In some embodiments, process 1600 may further include the step of positioning the tumor at the determined clinical isocenter.

[0096]

[0114] 2.5 LINAC Controller

[0097]

[0115] Figure 18 is a block diagram of the controller 1800 of LINAC100 in several embodiments. As shown in Figure 18, the controller 1800 is a processing circuit (PC) 1802 which may include one or more processors (P) 1855 (e.g., one or more general-purpose microprocessors and / or one or more other processors such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), and the like), and the processors may be colocated in a single housing or a single data center or may be geographically distributed (i.e., the system may be a distributed computing device), and the processing circuit (PC) 1802 and the controller 1800 are connected to a network 1810 (e.g., an Internet Protocol (IP) network) to which a network interface 1868 is connected. The network interface 1868 includes a transmitter (Tx) 1865 and a receiver (Rx) 1867 for enabling the transmission of data to and reception of data from other nodes; a gantry rotating device 1850 configured to rotate the gantry 102 about a gantry rotation axis 208; a couch rotating device 1852 configured to rotate the couch 106 about a couch rotation axis 310; a radiation beam generator 1854 configured to generate an electron beam in the waveguide of the LINAC 100; and a local memory unit (also known as a "data storage system") 1808, which may include one or more non-volatile storage devices and / or one or more volatile storage devices. In embodiments where the PC 1802 includes a programmable processor, a computer program product (CPP) 1841 may be provided. In some embodiments, the CPP 1841 may include a computer-readable medium (CRM) 1842 that stores a computer program (CP) 1843 comprising computer-readable instructions (CRI) 1844. In some embodiments, the CRM 1842 may be a non-temporary computer-readable medium such as a magnetic medium (e.g., a hard disk), an optical medium, a memory device (e.g., random-access memory, flash memory), and the like.In some embodiments, the CRI 1844 of the computer program 1843 may be configured, when executed by the PC 1802, to cause the CRI to cause the LINAC 100 to perform the steps described herein (for example, one or more steps described herein with reference to the flowchart herein). In other embodiments, the controller 1800 may be configured to perform the steps described herein without requiring code. That is, for example, the PC 1802 may consist of simply one or more ASICs. Thus, the features of the embodiments described herein may be implemented in hardware and / or software.

[0098]

[0116] 2.6 Equipment

[0099]

[0117] Figure 19 is a block diagram of the apparatus 1900 in several embodiments. As shown in Figure 19, the apparatus 1900 may comprise a processing circuit (PC) 1902 which may include one or more processors (P) 1955 (e.g., one or more general-purpose microprocessors and / or one or more other processors such as application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), and the like), the processors which may be colocated in a single housing or in a single data center or geographically distributed (i.e., the system may be a distributed computing device), a network interface 1968 which includes a transmitter (Tx) 1965 and a receiver (Rx) 1967 to enable the apparatus 1900 to send data to and receive data from other nodes connected to a network 1910 (e.g., an Internet Protocol (IP) network) to which the network interface 1968 is connected, and a local storage unit (also known as a "data storage system") 1908 which may include one or more non-volatile storage devices and / or one or more volatile storage devices. In embodiments where the PC1902 includes a programmable processor, a computer program product (CPP) 1941 may be provided. In some embodiments, the CPP 1941 may include a computer-readable medium (CRM) 1942 that stores a computer program (CP) 1943 comprising computer-readable instructions (CRI) 1944. In some embodiments, the CRM 1942 may be a non-temporary computer-readable medium such as a magnetic medium (e.g., a hard disk), an optical medium, a memory device (e.g., random access memory, flash memory), and the like. In some embodiments, the CRI 1944 of the computer program 1943 may be configured such that, when executed by the PC1902, the CRI causes the device 1900 to perform the steps described herein (e.g., one or more steps described herein with reference to the flowchart herein). In other embodiments, the device 1900 may be configured to perform the steps described herein without requiring code.That is, for example, PC1902 may consist of simply one or more ASICs. Therefore, the features of the embodiments described herein may be implemented in hardware and / or software.

[0100]

[0118] While various embodiments have been described herein, it should be understood that they are presented only as examples and not as limitations. Therefore, the breadth and scope of this disclosure should not be limited by any of the illustrative embodiments described above. Furthermore, any combination of the elements described above in all possible variations thereof is encompassed by this disclosure unless otherwise indicated herein or unless it is clearly inconsistent with the context.

[0101]

[0119] In addition, the process described above and illustrated in the diagrams is shown as a series of steps, but this is for illustrative purposes only. Therefore, it is intended that some steps may be added, some steps may be omitted, the order of the steps may be rearranged, and some steps may be performed in parallel.

Claims

1. determining a set of three-dimensional (3D) radiation beam axes of a linear accelerator (LINAC) from the two-dimensional (2D) radiographic images, where the set of 3D radiation beam axes includes at least a first 3D radiation beam axis and a second 3D radiation beam axis; determining a radiation isocenter of the LINAC based on at least the set of 3D radiation beam axes; A method comprising:

2. Determining the first 3D radiation beam axis comprises: positioning a gantry of the LINAC at a first gantry angle relative to a reference gantry angle, where positioning the gantry comprises rotating the gantry about a gantry axis of rotation; using the LINAC to generate a first radiation beam with the gantry positioned at the first gantry angle; using an imaging device of the LINAC to obtain a first two-dimensional (2D) radiographic image indicative of a radiation field of the first radiation beam after passing near a radiopaque marker with the gantry positioned at the first gantry angle; determining a location of a beam axis of the first radiation beam and a center of a shadow of the radiopaque marker in a radiation field of the first radiation beam based on the first 2D radiographic image; constructing the first 3D radiation beam axis based on the determined location of a beam axis of the first radiation beam, the determined center of a shadow of the radiopaque marker in a radiation field of the first radiation beam, and the first gantry angle; The method of claim 1 , comprising:

3. constructing the first 3D radiation beam axis comprises rotating the first 2D radiographic image about a first image rotation axis that coincides with a determined center of a shadow of the radiopaque marker in the radiation field of the first radiation beam and is parallel to the gantry rotation axis; the first 2D radiographic image is rotated an amount equal to the first gantry angle; the first 3D radiation beam axis coincides with the determined location of a beam axis of the first radiation beam in the rotated first 2D radiographic image and is perpendicular to a plane of the rotated first 2D radiographic image. The method of claim 2.

4. Constructing the first 3D radiation beam axis comprises: generating an initial 3D radiation beam axis coincident with the determined location of the beam axis of the first radiation beam in the first 2D radiographic image and perpendicular to a plane of the first 2D radiographic image; rotating the initial 3D radiation beam axis about a first image rotation axis that coincides with a determined center of a shadow of the radiopaque marker in the radiation field of the first radiation beam and is parallel to the gantry rotation axis; The method of claim 2 , wherein the initial 3D radiation beam axis is rotated an amount equal to the first gantry angle.

5. Determining the second 3D radiation beam axis comprises: positioning the gantry of the LINAC at a second gantry angle relative to the reference gantry angle; using the LINAC to generate a second radiation beam with the gantry positioned at the second gantry angle; using the imaging device of the LINAC to acquire a second 2D radiographic image indicative of a radiation field of the second radiation beam after passing near the radiopaque marker with the gantry positioned at the second gantry angle; determining a location of a beam axis of the second radiation beam and a center of a shadow of the radiopaque marker in the radiation field of the second radiation beam based on the second 2D radiographic image; constructing the second 3D radiation beam axis based on the determined location of a beam axis of the second radiation beam, the determined center of a shadow of the radiopaque marker in a radiation field of the second radiation beam, and the second gantry angle; The method according to any one of claims 2 to 4, comprising:

6. constructing the second 3D radiation beam axis comprises rotating the second 2D radiographic image about a second image rotation axis that coincides with a determined center of a shadow of the radiopaque marker in the radiation field of the second radiation beam and is parallel to the gantry rotation axis; the second 2D radiographic image is rotated an amount equal to the second gantry angle; the second 3D radiation beam axis coincides with the determined location of the beam axis of the second radiation beam in the rotated second 2D radiographic image and is perpendicular to the plane of the rotated second 2D radiographic image. The method according to claim 5.

7. Constructing the second 3D radiation beam axis comprises: generating an initial second 3D radiation beam axis coincident with the determined location of the beam axis of the second radiation beam in the second 2D radiographic image and perpendicular to a plane of the second 2D radiographic image; rotating the initial second 3D radiation beam axis about a second image rotation axis that coincides with the determined center of the shadow of the radiopaque marker in the radiation field of the second radiation beam and is parallel to the gantry rotation axis; The method of claim 5 , wherein the initial second 3D radiation beam axis is rotated an amount equal to the second gantry angle.

8. 6. The method of claim 5, wherein a couch is positioned at a fixed couch angle while the LINAC is used to generate the first and second radiation beams and the imaging device of the LINAC is used to acquire the first and second 2D radiographic images.

9. Determining the radiation isocenter comprises: determining, for each 3D radiation beam axis of the set of 3D radiation beam axes, an off-axis beam distance between a location in 3D space and the 3D radiation beam axis; determining which of the determined off-axis beam distances for the 3D radiation beam axes of the set of 3D radiation beam axes is largest; and finding a location in 3D space that has the smallest maximum beam off-axis distance; 5. The method of claim 1, wherein the largest of the determined off-axis distances for the 3D radiation beam axes of the set of 3D radiation beam axes is a maximum off-axis distance for the location.

10. 10. The method of claim 9, wherein the size of the radiation isocenter is the maximum beam off-axis distance for the location.

11. The method according to any one of claims 1 to 4, further comprising positioning the tumor at the determined radiation isocenter.

12. determining a marker movement vector for each couch angle of a set of couch angles, where the set of couch angles includes at least a first couch angle relative to a reference couch angle and a second couch angle relative to the reference couch angle; determining a distance from a marker to a 3D radiation beam axis error for each combination of a couch angle from the set of couch angles and a 3D radiation beam axis from the set of 3D radiation beam axes based on the determined marker movement vector, wherein the distance from the marker to the 3D radiation beam axis error represents a shortest distance between a location of a radiopaque marker and the 3D radiation beam axis error. The method of any one of claims 1 to 4, further comprising:

13. Determining the marker movement vector for the first couch angle includes: using the LINAC to generate a first reference radiation beam with a couch of the LINAC positioned at the reference couch angle and a gantry of the LINAC positioned at a reference gantry angle; using an imaging device of the LINAC to acquire a reference couch angle two-dimensional (2D) radiographic image indicative of a radiation field of the first reference radiation beam after passing near the radiopaque marker with the couch positioned at the reference couch angle and the gantry positioned at the reference gantry angle; determining a center of a shadow of the radiopaque marker in a radiation field of the first reference radiation beam based on the reference couch angle 2D radiographic image; positioning the couch at the first couch angle; using the LINAC to generate a second reference radiation beam with the couch positioned at the first couch angle and the gantry positioned at the reference gantry angle; using the imaging device of the LINAC to acquire a first couch angle 2D radiographic image indicative of a radiation field of the second reference radiation beam after passing near the radiopaque marker, with the couch positioned at the first couch angle and the gantry positioned at the reference gantry angle; determining a center of a shadow of the radiopaque marker in the radiation field of the second reference radiation beam based on the first couch angle 2D radiographic image; determining the marker movement vector for the first couch angle based on a comparison of a center of a shadow of the radiopaque marker in the radiation field of the first reference radiation beam and a center of a shadow of the radiopaque marker in the radiation field of the second reference radiation beam; The method of claim 12 comprising:

14. Determining the marker movement vector for the second couch angle includes: positioning the couch at the second couch angle; using the LINAC to generate a third reference radiation beam with the couch positioned at the second couch angle and the gantry positioned at the reference gantry angle; using the imaging device of the LINAC to acquire a second couch angle 2D radiographic image indicative of a radiation field of the third reference radiation beam after passing near the radiopaque marker, with the couch positioned at the second couch angle and the gantry positioned at the reference gantry angle; determining a center of a shadow of the radiopaque marker in the radiation field of the third reference radiation beam based on the second couch angle 2D radiographic image; determining the marker movement vector for the second couch angle based on a comparison of a center of a shadow of the radiopaque marker in the radiation field of the first reference radiation beam and a center of a shadow of the radiopaque marker in the radiation field of the third reference radiation beam; The method of claim 13 comprising:

15. 13. The method of claim 12, further comprising determining a clinical isocenter, wherein determining the clinical isocenter comprises determining a location in space that minimizes a distance from a maximum marker to a 3D radiation beam axis error.

16. Determining a location in space that minimizes a distance from the maximum marker to a 3D radiation beam axis error comprises: predicting a couch rotation axis based on the shadow of the radiopaque marker for different couch angles; predicting a set of positions of the radiopaque markers at new reference positions displaced by a couch rotation about the predicted couch rotation axis; determining a predicted marker to 3D radiation beam axis error distance for each combination of a couch angle from the set of couch angles and a 3D radiation beam axis from the set of 3D radiation beam axes based on the determined marker movement vector, wherein the predicted marker to 3D radiation beam axis error distance represents a shortest distance between a location of a radiopaque marker and the 3D radiation beam axis error; determining a reference marker position that minimizes a distance from the maximum marker to a 3D radiation beam axis error; The method of claim 15 comprising:

17. The method of claim 15 , further comprising positioning a tumor at the determined clinical isocenter.

18. determining a set of three-dimensional (3D) radiation beam axes of a linear accelerator (LINAC) from the two-dimensional (2D) radiographic images, where the set of 3D radiation beam axes includes at least first and second 3D radiation beam axes; determining a radiation isocenter of the LINAC based on at least the set of 3D radiation beam axes; The apparatus is configured to:

19. determining a marker movement vector for each couch angle of a set of couch angles of a linear accelerator (LINAC), where the set of couch angles includes at least a first couch angle relative to a reference couch angle and a second couch angle relative to the reference couch angle; determining a distance from a marker to a 3D radiation beam axis error for each combination of a couch angle from the set of couch angles and a 3D radiation beam axis from the set of 3D radiation beam axes based on the determined marker movement vector, wherein the distance from the marker to the 3D radiation beam axis error represents a shortest distance between a location of a radiopaque marker and the 3D radiation beam axis error. A method comprising:

Images