Devices and methods for radiotherapy system alignment and calibration

Abstract

Disclosed herein are devices and methods for determining the position and alignment of radiotherapy system components. One variation of a device that is used to determine the positions of a therapeutic radiation source and radiation imager and their relative alignment is a phantom device comprising a substrate and a plurality of radiopaque elements attached to the substrate. Measurements by the radiation imager of the phantom device may be used to determine the physical positions of the therapeutic radiation source, beam-shaping components, and the radiation imager. In some variations, this position information is included in a radiation beam model. The position and / or alignment information of system components may be used to steer the therapeutic radiation beams to compensate for any offset positioning or misalignments. Also disclosed herein are steering coils for a linear accelerator and jaws having movable radiation-blocking blocks that adjust the size and / or position of the jaw aperture.

Description

DEVICES AND METHODS FOR RADIOTHERAPY SYSTEM ALIGNMENT AND CALIBRATIONCROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Patent Application Serial No.63 / 558,233 filed February 27, 2024, and U.S. Provisional Patent Application Serial No.63 / 744,621 filed January 13, 2025, which are hereby incorporated by reference in their entirety.BACKGROUND

[0002] External beam radiotherapy is a cancer treatment modality that precisely directs radiation beamlets to a cancerous tumor with the goal of delivering an elevated dose of radiation to the tumor while limiting the radiation exposure of surrounding healthy tissue. An external beam radiotherapy system includes a therapeutic radiation source, such as a linear accelerator (linac) or a radioactive compound (e.g., cobalt-60), a radiation imager to measure the radiation emitted by the therapeutic radiation source, and one or more beam-shaping devices to help shape and / or focus the therapeutic radiation beam so that it irradiates the tumor and avoids irradiating non-cancerous tissue. Some radiotherapy systems also include an imaging system, such as a kV CT imaging system, which is used to take images of the patient positioned within the radiotherapy system to ensure that the patient is in a desired orientation and position to receive radiotherapy. A radiotherapy treatment plan is generated by the radiation oncologist that specifies the prescribed radiation dose to the tumor, the radiation dose limits of surrounding tissue (include organs-at-risk or OARs), and details of the radiation fluence that needs to be emitted by the therapeutic radiation source in order to attain the prescribed dose distribution. The treatment plan assumes that the patient will be in a particular position and orientation on the day of treatment in order for the prescribed radiation to be delivered. If the patient is in a different position and / or orientation on the day of treatment, then the radiation delivery will not be accurate. The onboard imaging system (e.g., kV CT imaging system) is often used to check the patient’s position and orientation, and to allow the clinician to adjust their position to best match the planning position.

[0003] In addition to precise patient positioning on the day of treatment, it is also important that the components and sub-systems of the radiotherapy system are precisely aligned to each other and referenced to a common coordinate system. Even when the utmost care is taken duringthe assembly of a radiotherapy system, there may be slight misalignments between components that may cumulatively sum to a systemic misalignment that impacts the precision of radiation dose delivery. For example, the alignment of the beam-shaping components such as the jaw(s) and / or multi-leaf collimator(s) with the therapeutic radiation source needs to be accurate so that the radiation beamlets directed to the patient precisely irradiate the target tumor, and not nontarget tissue. Moreover, the alignment and / or position of the components and / or sub-systems of a radiotherapy system may change as they are moved around the patient treatment area, which may also impact radiation delivery. Accordingly, improved methods for measuring and characterizing the relative position(s) and alignment of different components of a radiotherapy system are desirable.SUMMARY

[0004] Disclosed herein are devices and methods for determining the position and alignment of radiotherapy system components. One variation of a device that may be used to determine the positions of a therapeutic radiation source and radiation imager and their relative alignment may be a phantom comprising a substrate and a plurality of radiopaque elements attached to the substrate at different locations. The relative positions of the radiopaque elements are predetermined. The phantom may be placed on a patient platform of a radiotherapy system and advanced into the patient treatment area for irradiation by the therapeutic radiation source. Measurements by the radiation imager of the emitted radiation may be used in the methods described herein to determine the physical positions of the therapeutic radiation source and the radiation imager. Some methods may comprise generating output transforms that map a pixel on the radiation imager relative to a center location of the radiotherapy system (e.g., isocenter). Also described herein are methods of using the radiation imager data of the phantom to determine the physical positions of one or more beam-shaping components, such as the jaw(s) and dynamic multi-leaf collimator.

[0005] Also disclosed herein are methods for characterizing the alignment of the therapeutic radiation source, the beam-shaping components, and the radiation imager. Previous methods of characterizing such alignment required the use of radiation film (e.g., Gafchromic film) to generate a “starshof ’ image of radiation beamlets emitted by the therapeutic radiation source at a plurality of firing angles. In an idealized arrangement, the radiation beamlets all intersect at a point. However, in reality, the radiation beamlets do not all intersect at a point. A minimumtangent circle (i.e., a circle with the smallest radius where all beamlet paths are tangent or intersect the circle) may be used to approximate the location of the radiotherapy system center. For example, the center of the best-fit circle may approximate the system isocenter and its radius (i.e., size) may be an approximation of the error of the radiotherapy system center (e.g., isocenter). However, the methods described herein do not require the use of radiation film, and instead, use imaging data acquired by the radiation imager, with and without the phantom, to generate a similar or equivalent “starshof ’ image that may be used to approximate the location of the radiotherapy system center and also to calculate an approximation of the error of the radiotherapy system center. Some variations of the methods described herein may also be used to determine a 3-D radiotherapy system isocenter, which cannot be readily determined from a film-based “starshof ’ image and analysis.

[0006] Disclosed herein are methods for determining a center of a radiotherapy system. In one variation, the method may comprise acquiring imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of a radiotherapy system and acquiring the imaging data using a radiation imager, iteratively adjusting parameters of a virtual model (e.g., beam model) of the radiotherapy system to generate simulated radiation imager data until the simulated radiation imager data with the acquired radiation imager data until an error value is below a threshold, and determining a center point of the radiotherapy system based on the adjusted parameters of the virtual model of the radiotherapy system. Determining the center point may comprise determining positions of the therapeutic radiation source and the radiation imager from the adjusted parameters of the virtual model. Optionally, some variations may comprise generating an output transform function that maps the radiation imager data relative to the center point of the radiotherapy system.

[0007] Also disclosed herein are methods for generating a plot for characterizing a radiotherapy system center. In one variation, the method may comprise acquiring imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of a radiotherapy system and acquiring the imaging data using a radiation imager. The radiotherapy system may further comprise jaws and a multi-leaf collimator disposed in a radiation beam path of the therapeutic radiation source. The method may comprise generating radiation imager output transforms from the acquired imaging data of the phantom, acquiring air scan imaging data of even and odd leaves of the multi-leaf collimator from multiple firing angles using the radiation imager to determine a location of edges of the even and oddleaves, applying the radiation imager output transforms to the location of the edges of the even and odd leaves to determine a center of two central leaves of the multi-leaf collimator from multiple firing angles, generating a plot of the center location for each firing angle relative to a stationary coordinate system, and fitting a minimum circle to points on the plot.

[0008] Disclosed herein is a radiotherapy system comprising a therapeutic radiation source and methods for determining structural deflections (e.g., due to gravity and / or motion and / or wear and tear) and compensating for the deflections to help the radiation maintain alignment with the system isocenter at all gantry angles. The therapeutic radiation source may comprise a linear accelerator and steering coils mounted on the linear accelerator to adjust the trajectory or path of the radiation beam (or beamlet). The electric current through the steering coils may be adjusted so that the radiation beam is better aligned with system isocenter across multiple (e.g., all) firing positions and / or gantry angles. This is in contrast to other radiotherapy systems where the electric current through steering coils is / are kept constant and / or are set based on ion chamber output. Optionally, some radiotherapy systems may comprise jaws having two opposing jaw blocks movable on a rail, and the position of the jaws and / or jaw blocks on the rail may be adjusted so that the radiation beam aligns with the system isocenter.

[0009] Also disclosed herein is a dose calculation method that utilizes a beam model that incorporates system alignment data for each firing position and / or gantry angle. The beam model may be used for dose calculations in treatment planning and / or quality assurance (QA) procedures.

[0010] Also disclosed herein are methods for compensation for radiation beamlet deviations in a radiotherapy system. One variation may comprise determining a deviation from a radiotherapy system isocenter of a radiation beamlet generated by a therapeutic radiation source for each firing positions of the therapeutic radiation source, determining a radiation beamlet shift for each firing position that compensates for the deviation at that firing position, and compensating for the deviation of the radiation beamlet at each firing position by adjusting radiotherapy system components to move the radiation beamlet according to the radiation beamlet shift for that firing position. Determining the deviation from the radiotherapy system isocenter may comprise acquiring imaging data of a phantom, where the phantom may comprise a plurality of radiopaque beads. The imaging data may be acquired by a radiation imager from multiple firing angles of the therapeutic radiation source. The method may optionally comprise generating a look-up table that comprises the radiation beamlet shift for each firing position ofthe therapeutic radiation source and storing the look-up table in a controller memory of the radiotherapy system. Compensating for the deviation of the radiation beamlet may comprise accessing the look-up table based on firing position to determine the radiation beamlet shift for that firing position. Methods may comprise generating radiotherapy system component instructions for each radiation beamlet shift at each firing position, where the look-up table may further comprise the radiotherapy system component instructions. Compensating for the deviation of the radiation beamlet may comprise executing the radiotherapy system component instructions for that firing position. The radiotherapy system components may include one or more of the therapeutic radiation source, a multi -leaf collimator, and jaws comprising a pair of jaw blocks.

[0011] In some variations, the therapeutic radiation source may comprise a linac and a plurality of steering coils attached to the linac. Compensating for the deviation may comprise controlling an electric current through the steering coils such that the radiation beamlet is moved according to the radiation beamlet shift for that firing position. For example, compensating for the deviation may comprise controlling an electric current through the steering coils to move the radiation beamlet according to the radiation beamlet shift in the look-up table for that firing position. Alternatively, or additionally, the radiotherapy system component instructions may comprise steering coil current parameters, and compensating for the deviation may comprise controlling an electric current through the steering coils according to the steering coil current parameters for that firing position. Steering coil current parameters may include one or more of magnitude, direction, and frequency.

[0012] Additionally, or alternatively, a radiotherapy system may comprise jaws located in a path of the radiation beamlet generated by the therapeutic radiation source. The jaws may comprise a pair of movable jaw blocks mounted on a rail. Compensating for the deviation may comprise moving the jaw blocks on the rail such that the radiation beamlet is moved according to the radiation beamlet shift for that firing position. For example, compensating for the deviation may comprise moving the jaw blocks on the rail according to the radiation beamlet shift in the look-up table for that firing position. Alternatively, or additionally, the radiotherapy system component instructions may comprise jaw block shift values, and compensating for the deviation may comprise moving the jaw blocks on the rail according to the jaw block shift values in the look-up table for that firing position. Jaw block shift values may include one or more of a shift amount and a shift direction along the rails. In some variations, the jaw blockshift values may comprise a first jaw block shift value for a first jaw block, and a second jaw block shift value for a second jaw block. The first jaw block shift value and the second jaw block shift value may not be the same or may be the same. In some variations, a jaw aperture between the jaw blocks may remain constant for each firing position, and compensating for the deviation may comprise moving both jaw blocks according to the jaw block shift values in the look-up table.

[0013] In some variations, the deviation of the radiation beamlet may comprise a deviation along a first axis and a deviation along a second axis, and compensating for the deviation of the radiation beamlet may comprise executing instructions for the therapeutic radiation source to correct the deviation along the first axis and executing instructions for the jaws to correct the deviation along the second axis. In some variations, compensating for the deviation of the radiation beamlet comprises executing instructions for the therapeutic radiation source to correct the deviation along the first axis and to correct the deviation along the second axis.

[0014] In some variations, determining the deviation from the radiotherapy system isocenter may further comprise generating radiation imager output transforms from the acquired imaging data of the phantom calculating a center location of the radiation beamlet using the radiation imager output transforms and the acquired imaging data at each firing position, and calculating the deviation of the center location of the radiation beamlet from the radiotherapy system isocenter for each firing position. Determining the radiation beamlet shift for each firing position that compensates for the deviation at that firing position may comprise calculating a first difference between the center location of the radiation beamlet along a first axis, calculating a second difference between the center location of the radiation beamlet along a second axis, and generating a beamlet shift vector comprising a first shift that compensates for the first difference along the first axis and a second shift that compensates for the second difference along the second axis. Optionally, determining the radiation beamlet shift for each firing position that compensates for the deviation at that firing position may further comprise calculating a third difference between the center location of the radiation beamlet along a third axis, and the beamlet shift vector may include a third shift that compensates for the third difference along the third axis.

[0015] Also disclosed herein are methods for characterizing radiation beamlet deviations. One example of a method may comprise acquiring imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of aradiotherapy system and acquiring the imaging data using a radiation imager, where the radiotherapy system further comprises jaws and a multi -leaf collimator disposed in a radiation beam path of the therapeutic radiation source, calculating a center location for the multi-leaf collimator for each firing angle using the acquired imaging data, calculating a center location for the jaws for each firing angle using the acquired imaging data, calculating deviations of the center locations for the multi-leaf collimator and the jaws from the radiotherapy system isocenter for each firing angle, and calculating radiation beamlet deviations from the radiotherapy system isocenter for each firing angle by combining the deviations of the center locations for the multileaf collimator and the jaws. Optionally, the method may also comprise incorporating the deviations of the center locations for the multi-leaf collimator and the jaws into a radiation beam model.

[0016] Any of the methods disclosed herein for characterizing displacements of beam-shaping components of a radiotherapy components may optionally be used for simulating radiation dose delivery. One example of a method may comprise acquiring imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of a radiotherapy system and acquiring the imaging data using a radiation imager, where the radiotherapy system further comprises jaws and a multi-leaf collimator disposed in a radiation beam path of the therapeutic radiation source, generating radiation imager output transforms from the acquired imaging data of the phantom, calculating a center location for the multi-leaf collimator for each firing angle using the radiation imager output transforms and the acquired imaging data, calculating a center location for the jaws for each firing angle using the radiation imager output transforms and the acquired imaging data, calculating deviations of the center locations for the multi-leaf collimator and the jaws from the radiotherapy system isocenter for each firing angle, and incorporating the multi -leaf collimator center location and the jaw center location deviations from system isocenter into a radiation beam model used to simulate a radiation dose delivery.BRIEF DESCRIPTION OF THE DRAWINGS

[0017] FIG. 1 A depicts a conceptual schematic of one variation of a radiotherapy system.

[0018] FIG. IB depicts a front (or top) view of one variation of a phantom device.

[0019] FIG. 1C depicts a perspective cross-sectional view of the phantom device of FIG. IB.

[0020] FIG. ID depicts a conceptual arrangement of how a phantom device may be placed on the couch of one variation of a radiotherapy system.

[0021] FIG. IE depicts a perspective view of one variation of a phantom device mounted on a couch of a radiotherapy system.

[0022] FIG. IF depicts a side view of the phantom device of FIG. IE mounted on the couch of a radiotherapy system.

[0023] FIG. 1G depicts a photograph of the front view (from inside the bore of a radiotherapy system) of the phantom device attached to the couch.

[0024] FIG. 1H depicts a photograph of the front view (from inside the bore of a radiotherapy system) of another variation of the phantom device attached to the couch.

[0025] FIG. II depicts a perspective view of one variation of a phantom device.

[0026] FIG. 2A depicts a flowchart representation of one variation of a method for generating MVD output transforms.

[0027] FIG. 2B depicts a plot of the locations of the phantom’s radiopaque elements at a plurality of firing angles ranging from 0° to 360°.

[0028] FIG. 2C depicts a flowchart representation of one variation of a method for determining physical locations of the linac and MV detector (MVD).

[0029] FIG. 3 A depicts a flowchart representation of one variation of a method for generating a starshot image based on acquired MVD data and applying the MVD output transforms derived from the MVD imaging data of a phantom comprising radiopaque beads or balls.

[0030] FIG. 3B depicts a plot of beam centerlines for 5 firing angles for a radiotherapy system, generated using the methods disclosed herein that utilize MVD imaging data of a phantom comprising radiopaque beads or balls, acquired during an experiment.

[0031] FIG. 3C depicts a plot generated from a film-based starshot plot of beam centerlines for 5 firing angles, acquired during an experiment, for the same radiotherapy system that was characterized in FIG. 3B.

[0032] FIG. 3D depicts a plot beam centerlines for 50 firing angles for the radiotherapy system of FIG. 3B, generated using the methods disclosed herein that utilize MVD imaging data of a phantom comprising radiopaque beads or balls, acquired during an experiment.

[0033] FIG. 4A depicts a film image acquired during an experiment of one of the starshot experiments for one of the radiotherapy systems.

[0034] FIG. 4B depicts a plot generated from the film image of FIG. 4 A.

[0035] FIG. 4C depicts a plot of the beam centerlines for 5 starshot angles generated using the methods disclosed herein that utilize MVD imaging data of a phantom comprising radiopaque beads or balls, acquired during an experiment.

[0036] FIG. 4D depicts a plot of the beam centerlines for 50 starshot angles generated using the methods disclosed herein that utilize MVD imaging data of a phantom comprising radiopaque beads or balls, acquired during an experiment.

[0037] FIG. 4E depicts a plot of the beam centerlines for 5 starshot angles and its associated minimum tangent circle (in the blue dotted line) as determined from the methods disclosed herein, overlaid with the beam centerlines for 50 starshot angles and the associated minimum tangent circle (in the red solid line).

[0038] FIG. 4F depicts a table that summarizes the results of the film-based starshot experiments and the phantom-based starshot experiments conducted over 6 radiotherapy systems A-F.

[0039] FIG. 5 A depicts a flowchart representation of one variation of a method for calculating and characterizing a 3-D isosphere for a radiotherapy system.

[0040] FIG. 5B depicts a plot of an isosphere for a radiotherapy system that was generated using the method described in FIG. 5 A, based on data acquired during an experiment.

[0041] FIG. 6 depicts a flowchart representation of one variation of a method for characterizing deviations / offsets of a multi-leaf collimator (MLC) and jaws of a radiotherapy system.

[0042] FIG. 7 A depicts a plot of MLC leaf deviations over multiple firing / gantry angles, based on data acquired during an experiment.

[0043] FIG. 7B depicts a plot of jaw block edge deviations over multiple firing / gantry angles, based on data acquired during an experiment.

[0044] FIG. 7C depicts a plot of jaw center deviations over multiple firing / gantry angles, based on data acquired during an experiment.

[0045] FIG. 7D depicts a plot of an axial slice through the dose distribution to a target using a beam model without MLC and jaw corrections.

[0046] FIG. 7E depicts a plot of an axial slice through the dose distribution to a target using a beam model with MLC and jaw corrections.

[0047] FIG. 7F depicts a plot of the difference between the dose distributions of FIG. 7D and FIG. 7E.

[0048] FIG. 7G depicts the dose profile along the X-axis as measured on film and the dose profile as calculated from a simulation using a beam model without the corrections for MLC and jaw deviations.

[0049] FIG. 7H depicts the dose profile along the X-axis as measured on film and the dose profile as calculated from a simulation using a beam model having the corrections for MLC and jaw deviations.

[0050] FIG. 71 depicts the simulated film results (e.g., dose distribution) derived from a simulation using a beam model without the corrections for MLC and jaw deviations.

[0051] FIG. 7J depicts the simulated film results (e.g., dose distribution) derived from a simulation using a beam model having the corrections for MLC and jaw deviations.

[0052] FIG. 8 A depicts a photograph of one variation of a steering coil.

[0053] FIG. 8B depicts a schematic of one variation of a therapeutic radiation source having one or more steering coils.

[0054] FIG. 8C depicts a front elevational view of another variation of a therapeutic radiation source having one or more steering coils.

[0055] FIG. 8D depicts a back elevational view of the therapeutic radiation source of FIG. 8C.

[0056] FIG. 9 depicts a side view of one variation of a jaws assembly comprising two jaw blocks.

[0057] FIG. 10 depicts a flowchart representation of one variation of a method for compensating for radiation beamlet deviations in a radiotherapy system.DETAILED DESCRIPTION

[0058] Disclosed herein are devices and methods for determining the position and alignment of radiotherapy system components. One variation of a device that is used to determine the positions of a therapeutic radiation source and radiation imager and their relative alignment is a phantom device comprising a substrate and a plurality of radiopaque elements attached to the substrate at different locations. Measurements by the radiation imager of the phantom device may be used to determine the physical positions of the therapeutic radiation source, beamshaping components, and the radiation imager and incorporating the position information in a radiation beam model. In some variations, the position and / or alignment information of radiotherapy system components are used to steer the therapeutic radiation beams to compensate for any offset positioning or misalignments. Also disclosed herein are steering coils for a linearaccelerator which may be used to compensate for beamlet offsets or misalignments. Optionally, radiotherapy systems may comprise jaws having movable radiati on -blocking blocks that are configured to adjust the size and / or position of the jaw aperture to compensate for beamlet deviations. Radiotherapy system components that are moved or rotated quickly may be subject to dynamic acceleration forces, which may cause changes in the relative alignment and position of different components to each other. This may introduce imprecisions and / or errors in the radiation delivered to a patient. Moreover, these dynamic loading forces may impact the alignment and / or position of components differently for different firing positions of the therapeutic radiation source. For example, the relative alignment and / or position between components may be impacted by gravity, and as these components are moved through different firing positions and / or gantry angles, their alignment and / or positions may vary depending on their orientation in the gravitational field. The devices and methods described herein may characterize these deviations for each firing position and / or gantry angle, so that they may be accounted for in dosimetry calculations, treatment planning and optimization, and / or used to control radiotherapy components to compensate for the deviations during radiation delivery.Radiotherapy Systems and Devices

[0059] The devices and methods described herein may be used with any external beam radiotherapy system. A radiotherapy system may comprise a rotatable gantry, a therapeutic radiation source mounted on the gantry, one or more beam-shaping components disposed within the radiation beam generated by the therapeutic radiation source, a radiation imager also mounted on the gantry at a location that is across from the therapeutic radiation source, and a couch or patient platform that is configured to move within the patient treatment area of the radiotherapy system. FIG. 1 A depicts a conceptual schematic of one variation of a radiotherapy system 100. The radiotherapy system may comprise a rotatable gantry (e.g., a circular gantry) 102, a therapeutic radiation source comprising a linear accelerator (linac) 104, beam-shaping components 106 including jaws 108 and a dynamic multi-leaf collimator (MLC) 110, a radiation imager comprising an MV detector 112, and a patient platform or couch 114 movable within the patient treatment area 116. The gantry 102 may be rotatable about an axis of rotation that extends within a bore of the radiotherapy system 100. The axis of rotation may intersect with a center or isocenter of the radiotherapy system. In this example, the center 118 of the radiotherapy system may be the center of the circular gantry. The center 118 may, in somevariations, also be the system isocenter, which may be a defined reference point or origin point (0,0,0) in the stationary reference frame, such as the reference frame of the treatment room. In some variations, the center 118 may be located midway (e.g., halfway) between the linac 104 and the MV detector 112. The X-axis, Y-axis, and Z-axis depicted in FIG. 1 A may be the IEC-X axis, IEC-Y axis, and IEC-Z axis respectively.

[0060] The gantry 102 may be continuously rotatable 360° about its axis of rotation around the patient treatment area 116. In some variations, the gantry 102 may be configured to rotate about 50 RPM or more, e.g., about 60 RPM or more, about 70 RPM or more. Rotating the gantry 102 allows the linac 104 to emit radiation toward the patient treatment area 116 from a plurality of firing positions or firing angles, which may be referenced by their angular position from 0°- 359° (which spans 360 degrees). For radiotherapy systems comprising a continuously rotatable gantry, linac firing angles correspond to gantry angles, and may be referred to interchangeably. The jaws 108 may comprise high-density, radiation-blocking material such as tungsten that are movable so that they block radiation from the linac 104 along a certain extent (e.g., x-axis or y- axis). An aperture through the jaws between the radiation-blocking materials may be referred to as the jaw opening or aperture. In some variations, the size of the jaw opening may be fixed and not adjustable, while in other variations, the size of the jaw opening may be adjusted by moving the radiation-blocking material. For example, jaws may comprise two or more movable blocks comprising high-density materials (e.g., tungsten), and the jaw aperture or opening refers to the space between the blocks that allow radiation to pass through. This creates a radiation field that has the size and shape of the space between the blocks. In the examples described herein, the jaws 108 may comprise a pair of opposing jaw blocks that may be movable toward or away from each other to define the size of the jaw aperture or jaw opening. The motion of the pair of blocks may be in synchrony and / or symmetry with each other. Some radiotherapy systems may comprise two sets of jaws that each comprise a pair of jaw blocks, where the first set of jaws may be located above the MLC and the second set of jaws may be located below the MLC. The MLC 110 may be a dynamic MLC comprising a plurality of movable high-density, radiationblocking leaves that block radiation along a different extent from the jaws (e.g., y-axis or x- axis). The MLC and the one or more sets of jaws may be mechanically arranged such that their central axes are aligned. Operating together, the jaws and the MLC shape the radiation generated by the linac 104, shaping the wide radiation beam from the linac into one or more radiation beamlets. Selectively adjusting the aperture of the jaws and selectively opening and closingindividual MLC leaves shapes the radiation that is delivered to the patient area, which may be in accordance with a prescribed or desired dose distribution as part of a treatment session (i.e., with a patient in the patient treatment area) and / or a quality assurance procedure (i.e., without a patient in the patient treatment area). The MV detector 112 located across from the linac 104 is configured to measure the emitted radiation after it has passed through the patient treatment area 116. The data acquired by the MV detector may be referred to as MVD imaging data or MVD data. In some situations, such as during a treatment session, the radiation will have interacted with a patient. In other situations, such as during a quality assurance (QA) procedure, the radiation may have interacted with a phantom and / or measurement device on the couch 114 in the patient treatment area 116 before it is measured by the MV detector 112. The data measured by the MV detector may be used to generate an MV detector image of the of the phantom. The shape, size, location, and / or orientation of the radiopaque portions of a phantom device may be readily determined from the MVD imaging data. In some QA procedures, there is nothing located in the patient treatment area, and the radiation emitted from the linac 104 will not have interacted with anything (other than the bore cover) before it is measured by the MV detector 112. This may be referred to as an air scan. The MV detector 112 may be any suitable radiation imager, including for example, an electronic portal imaging device (EPID). The operation of the components of the radiotherapy system 100 may be coordinated by one or more controllers having one or more processors (not shown). For example, the radiotherapy system 100 may comprise a first controller that is in electrical communication with, and operates, the components on the rotatable gantry and a second controller that is in electrical communication with, and operates, the components that are not on the rotatable gantry (e.g., couch). The first and second controller may communicate and synchronize with each other to coordinate radiation delivery, transmit machine instructions, and acquire sensor or imaging data. The one or more controllers may comprise one or more machine-readable non-transitory memories to store machine instructions, commands, and data.Phantom Device with Radiopaque Elements

[0061] A phantom device (also referred to as a phantom) comprising a plurality of radiopaque elements may be used in the methods disclosed herein for determining the position and alignment of radiotherapy system components. One variation of a phantom device may comprise a substrate and a plurality of radiopaque elements with known shape(s), size(s), and position(s)mounted to the substrate using a radiotransparent mounting mechanism. Examples of radiopaque elements may comprise beads, balls, and / or blocks of a high-density (i.e., radiopaque) material. The size, shape, and location of the radiopaque elements may be predetermined and known. In some variations, the substrate may define a plane and the mounting mechanisms may comprise structures that extend from the plane of the substrate such that the radiopaque elements are not located on the same plane as the substrate. This may allow the radiopaque elements to be imaged using the MV detector with little or no interference from the substrate. Examples of mounting mechanisms may include stand-offs, screws, rods, cylinders, dowels, posts, tubes, and / or pins. The mounting mechanisms may be made of a lower-density, non-radiopaque (i.e., radiotransparent) material that, when imaged using an MV detector, is not visible or conspicuous as compared to the radiopaque elements. For example, mounting mechanisms may comprise various types of plastic such as polyether ether ketone (PEEK) and / or carbon fiber such as graphite fiber. In some variations, the substrate may comprise similarly lower-density, non- radiopaque (i.e., radiotransparent) material so that even if the substrate is within the irradiation field of the linac and imaged by the MV detector, there is little or no interference with the images of the radiopaque elements. The substrate may have any shape or size that is sufficient to securely retain the radiopaque elements and to be placed or attached to the radiotherapy system couch. For example, the substrate may have a square shape, rectangular shape, triangular shape, circular shape, oval shape, or any desired shape that is structurally stable and allows the phantom to be coupled to the couch and remain stationary. Some variations of a phantom device may optionally comprise a couch mounting assembly that couples the substrate to the couch.

[0062] The phantom device may have any number of radiopaque elements, for example, 2 or more elements, 3 or more elements, 4 or more elements, 5 or more elements, 10 or more elements, etc. The radiopaque elements may be arranged in any symmetric or asymmetric pattern. They may be coplanar or occupy more than one plane in space. Three or more of the plurality of radiopaque elements may be collinear. In one variation, the radiopaque elements are coupled to the substrate, for example, by mounting mechanisms or retention mechanism. Radiopaque elements may be located at or near the edges of the substrate, for example, at the corners of a square or rectangular shaped substrate. The arrangement pattern of the radiopaque elements may be bilaterally or radially symmetric, while in other variations, the arrangement pattern of the radiopaque elements may not have any axis of symmetry. While the examples described and depicted herein are of phantom devices with particular geometries, a particularnumber of radiopaque elements in certain arrangements, it should be understood that other variations of phantom devices according to the description above, having a different geometry, different number of radiopaque elements in different arrangements, may be used in conjunction with any of the methods disclosed herein. One or more such phantoms may be used to determine the position and / or offset of a kV imaging system isocenter and / or the position and / or offset of a PET imaging isocenter relative to the MV (i.e., therapeutic radiation source or delivery system) isocenter.

[0063] While the variations of the methods and devices described use a phantom device with an MV detector, it should be understood that similar methods may be used with other types of imaging detectors or imaging modalities. For example, any of the phantom devices described herein may be used with a kV imaging system (e.g., comprising a kV radiation source and a kV detector) for determining the position and alignment of the kV imaging system components and / or radiotherapy system components in accordance with the methods disclosed herein. In some variations, the radiopaque beads may be sized and shaped such that they are tuned for kV imaging. Alternatively, or additionally, the phantom device may comprise PET-avid beads or seeds, or any suitable PET sources. A phantom having one or more PET-avid beads may be used to determine the position and alignment of the PET imaging system components and / or radiotherapy system components, in accordance with the methods disclosed herein. Some phantoms may comprise beads suitable for multiple imaging modalities. For example, a phantom may comprise radiopaque beads for X-ray based imaging modalities (e.g., kV CT and / or MV imaging) and PET-avid beads for PET imaging.

[0064] FIG. IB depicts a front view of one variation of a phantom device 120 and FIG. 1C depicts a perspective cross-sectional view of the phantom device 120, where the cross-section is taken along line 1C-1C. The phantom device 120 may comprise a square-shaped substrate 122, 4 radiopaque elements, here shown as beads 124, and mounting mechanisms comprising 4 posts 126. The posts 126 retain the beads 124 at the corners of the substrate 122 such that the beads 124 are coplanar to each other, in a separate plane from the substrate 122. The posts 126 may be made of a plastic material, such as any of the materials described above, and the beads 124 may be made of tungsten or any high-density, radiopaque material. Optionally, the phantom device 120 may comprise polymeric covers 128 sized and shaped to be disposed over the posts 126, which may help to further secure the beads 124 on the posts, or may be removed before imaging. Optional components such as caps or rings may be used to protect the beads and / or posts fromimpact. These may be left in place during imaging or removed prior to imaging. The size, shape, and number of beads 124, as well as their arrangement on the substrate may be predetermined and this information may be stored in a machine-readable memory of a controller or processor, and may be used in one or more of the methods described below. FIG. ID depicts one variation of how the phantom device 126 may be placed on the couch 114 of a radiotherapy system, which may be used to advance the phantom device 120 into the patient treatment area 116 or bore of the rotatable gantry 102. The phantom device may be placed at a non-perpendicular angle to the couch, or the phantom device may be placed such that it is perpendicular to the couch as depicted in FIG. ID. In some of the methods described herein, the couch 114 may be used to advance the phantom device 120 into the patient treatment area, where the linac may be activated from multiple firing angles to image the phantom device 120. The MV detector may then acquire the imaging data, which may be processed or analyzed to determine the position and alignment of the linac and MV detector, and / or the position and alignment of the jaws and MLC.

[0065] FIGS. 1E-1G depict another variation of a phantom device 130, where the phantom device may further comprise a couch mounting assembly. FIG. IE depicts a perspective view of a phantom device 130 comprising a couch mounting assembly 131 configured to attach the phantom device 130 to the couch 114. FIG. IF depicts a side view of the phantom device 130 attached to the couch 114. Phantom device 130 may comprise a circular substrate 132, 4 radiopaque elements, e.g., beads 134, and mounting mechanisms comprising 4 tubes 136. The tubes 136 may retain the beads 134 near the edges of the substrate 132 such that the beads 134 form a square shape and are coplanar to each other, in a separate plane from the substrate 132. While tubes are shown here to retain the beads, any suitable mounting mechanism may be used. Similarly, when tubes are shown as the mounting mechanism, they may be attached to the substrate using any suitable coupling mechanism, for example, by friction fit or screw fit. As conceptually depicted in FIG. IF, the field of the radiation from the linac (irradiation field 133) irradiates the beads 134 but not the substrate 132. The tubes 136 may be made of a carbon fiber material and the beads 134 may be made of tungsten. FIG. II depicts another variation of a phantom device having a rectangular frame substrate, 4 radiopaque elements or beads 134, and 4 mounting mechanisms comprising tubes 136. The tubes 136 may be thin carbon tubes that retain the beads within the tube lumen and may provide sufficient structural support and protection while being made of a material with little or no impact to the imaging of the beads 134.Optionally, the phantom device 130 may comprise polymeric covers sized and shaped to be disposed over the tubes 136, which may help to further secure the beads 134 on the posts or may be removed before imaging. The size, shape, and number of beads 134, as well as their arrangement on the substrate may be predetermined and this information may be stored in a machine-readable memory of a controller or processor, which may be used with one or more of the methods described herein.

[0066] The couch mounting assembly 131 may comprise a clamp 137 and a bracket 135 that attaches the substrate 132 to the clamp 137. The bracket 135 may span a substantial length of the substrate 132 so that the substrate may be stabilized when the phantom device 130 is secured to the couch. The bracket 135 may be attached to the substrate using any suitable attachment mechanism, for example, using one or more screws, dowels, by friction fit, adhesives, and any interlocking mechanism, alone or in combination with any of these attachment mechanisms. In this variation, the bracket 135 may be attached to the substrate 132 using screws 138. For example, a bracket may have an L-shape comprising a vertical plate and a horizonal plate that meet at a junction. The phantom substrate may be attached to the vertical plate, and the horizontal plate may be connected to the clamp to engage the couch. The clamp 137 may have a clip that is configured to engage with the edge of the couch 114. While a clamp is described here, it should be understood that the couch mounting assembly may comprise alternative or additional mechanisms for attaching the phantom device to the couch. For example, a couch mounting assembly may comprise a bracket with grooves or slots that allow the phantom device to be slid over the end of the couch and a locking mechanism to secure the bracket to the couch. Providing a couch mounting assembly may help facilitate the placement of the phantom device in generally the same location, in a repeatable fashion, to promote consistency across multiple quality assurance (QA) procedures.

[0067] FIG. 1G is a photograph of the front view (from inside the bore of the radiotherapy system) of the phantom device 130 attached to a couch and advanced into the bore or patient treatment area. While the substrate 132 in the examples described and depicted above may have a circular or a rectangular shape, it should be understood that the substrate 132 may have any desirable shape, size or geometry. FIG. 1H depicts one variation of a phantom device having a substrate 132 comprising a rectangular shape and optional extension elements 140. Extension elements 140 may be frames, handles, brackets, grooves, or any structure that helps a user to position and install the phantom device as described above. While the extension elements maybe integral or otherwise fixably attached to the substrate, in some variations, the extension elements may be detachable from the substrate.Methods

[0068] One or more of the phantom devices described above may be used in any of the methods described below to determine the physical position and alignment of radiotherapy system components, as well as to determine a center location of the radiotherapy system (e.g., isocenter). The methods below comprise the acquisition and use of MVD imaging data of the phantom device, as well as MVD imaging data of air scans of the beam-shaping components in various configurations, in combination with a simulation (e.g., computer-generated) model of the radiotherapy system to determine the physical positions of the linac and MV detector and a center of the radiotherapy system within a margin of error comparable to film-based methods of determining a center of the radiotherapy system. Also disclosed herein are methods for using MVD imaging data of a phantom device and air scans to generate an MVD output transform that maps pixels on the MV detector to a stationary reference frame (e.g., a reference frame or coordinate system that has an origin at the center of the radiotherapy system). That is, an MVD output transform may map the locations of the MVD pixels to a stationary coordinate system so that the imaging data at a particular MVD pixel is mapped to a location on a plane that intersects the radiotherapy system center (e.g., the couch plane). In some variations, an MVD output transform may be used to reconstruct MVD images and / or reference the location of rotating components (e.g., linac, MV detector, jaws, MLC) to a stationary coordinate system (e.g., couch, patient).

[0069] In one variation, the phantom devices may be used to determine the physical position and / or alignment of the multi-leaf collimator (MLC) and / or jaw(s). A method may comprise acquiring MVD imaging data of the phantom device and analyzing the MVD imaging data to characterize gantry angle-specific radiation beamlet alignments, including but not limited to offsets and / or deviations of the MLC and / or jaw(s) relative to system isocenter. The method may comprise determining MLC leaf displacements (e.g., in IEC-X) or tilts and / or determining jaw deflections or displacements (e.g., in IEC-Y) for multiple (e.g., every) gantry angles. The MLC leaf displacements or tilts, and / or the jaw deflections or displacements for multiple gantry angles may be incorporated into a radiation beam model. A radiation beam model is a simulation of radiation behavior or characteristics (e.g., transmission, scattering, absorption, reflection, etc. ofX-rays or photons) within a radiotherapy system. The radiation beam model may include data about the arrangement of the radiotherapy system components that may impact radiation beamlets, including, but not limited to, the alignment and position of one or more jaws, multileaf collimator (MLC) leaves, dose monitoring chamber(s), and linac characteristics, including radiation output and energy. A radiation beam model that incorporates more data about the physical radiotherapy system may provide a more accurate simulation of radiation beam data than a beam model that does not include such data and instead uses theoretical approximations. In particular, radiotherapy systems comprising beam-shaping or collimating systems and a therapeutic radiation source moving at high speeds (e.g., rotating at about 40-60 RPM) may be subject to high-order oscillations or deviations that may impact the precision of dose delivery to small target regions. While relatively small oscillations may not have a large effect on dose delivery to larger target regions (e.g., > 2 cm), such oscillations may have a substantial impact dose delivery to small target regions. This impact may be pronounced for radiation delivery that relies on the intersection of radiation beamlets fired from a confluence of angles in order to irradiate a target region located near the system isocenter, e.g., for a radiotherapy system that moves a therapeutic radiation source rapidly about the patient area during radiation delivery. Data about MLC leaf displacements or tilts, and / or the jaw deflections or displacements for multiple gantry angles derived from the methods described herein may be used to improve the accuracy of a radiation beam model, which may improve the accuracy of radiation dose calculations.

[0070] The methods described herein may be implemented in a program (e.g., software) stored in a computer readable storage medium and executed by a processor.

[0071] FIG. 2A is a flowchart representation of one variation of a method for generating MVD output transforms. Method 200 may comprise acquiring 202 MVD imaging data of the phantom from a plurality of firing angles by rotating the linac and MVD around the phantom, calculating 204 the locations of the phantom beads from the MVD images for the plurality of firing angles, determining 208 the physical positions of the linac and MVD, and a center point of the radiotherapy system, using the calculated locations of the phantom beads, and generating 210 MVD output transforms that map MVD imaging data relative to the center point location of the radiotherapy system. Acquiring 202 MVD imaging data of the phantom device may comprise placing or attaching the phantom device (e.g., any of the phantom devices described above) onto the couch of a radiotherapy system, advancing the couch into the radiation field (e.g., radiationbeam, radiation beamlet(s)) of the linac such that the beads are in the radiation field, and firing radiation from the linac from multiple firing angles (i.e., gantry angles) while measuring MVD imaging data of the phantom device at the different firing angles. In some variations, the beamshaping components (e.g., jaw and the MLC) may be completely open, allowing for a full-field irradiation of the phantom device. MVD imaging data acquired at a particular firing angle may comprise pixels with varying intensities that represent the location of each of the beads. For example, a location of a dark spot or group of pixels may represent the location of a bead. The centroid of a group of dark pixels may be calculated and the centroid may be a location of a bead as measured in the plane of the MV detector. The method may comprise calculating a number of centroids of dark pixels or spots, up to the number of beads on the phantom device. In some cases, the MVD imaging data may reveal fewer dark pixels than the number of beads because two or more beads may be collinear at a particular firing angle. In some variations, those MVD imaging data may be excluded from further analyses (i.e., discarded).

[0072] Once the locations of the beads as determined from the acquired MVD imaging data are calculated, optionally, method 200 may comprise generating 206 a plot of the phantom bead locations for the plurality of firing angles. One example of an optional plot is depicted in FIG. 2B. In this example, the phantom device has 4 beads and for MVD imaging data acquired at each firing angle (i.e., gantry angle), the x-axis location of each of the 4 beads is determined. The optional plot(s) represent the x-axis location on the MVD image for each of the 4 beads, across firing angles ranging from 0°-360°. Each color represents a different radiopaque bead of the phantom device. For example, referring to the left plot, at firing angle 157° (about halfway between 135° and 180°), a first bead is at x-position -175, a second bead is at x-position -90, a third bead is at x-position +60, and a fourth bead is at x-position +180.

[0073] Determining 208 the physical locations of the linac and MVD may comprise iteratively adjusting one or more parameters (e.g., degrees of freedom) of a virtual (i.e., simulation, computer-generated, software) model of the radiotherapy system, simulating the acquisition of MVD imaging data of the phantom device on the virtual model, determining whether the simulated MVD imaging data approximates the empirically-acquired MVD imaging data within a specified error tolerance, and iterating on the one or more parameters of the virtual model until the simulated MVD imaging data converges to the empirically-acquired MVD imaging data. FIG. 2C depicts a flowchart representation of one variation of this method. Method 250 may comprise defining 212 an initial set of positions of the linac, MVD, and phantom for a computer-generated (e.g., virtual) model of the radiotherapy system, simulating 214 projected bead locations at the MVD, based on the defined set of positions and gantry / firing angles for each, and comparing 218 the simulated locations of the phantom beads with the measured (i.e., empirically determined) locations of the phantom beads to determine an error value. If the error value meets or exceeds a threshold, method 250 may comprise defining 220 different positions for the linac and MVD and repeating the simulation 214 and comparison 218. If the error value is below a threshold, method 250 may comprise assigning 222 the current set of simulation positions of the linac and MVD as their physical positions, defining 224 a gantry rotational axis and a radial axis based on the physical positions of the linac and MVD, and determining 226 a center point of the radiotherapy system based on the rotational axis and the radial axis. In some variations, determining the center point location of the radiotherapy system may comprise finding the intersection of the rotational axis and the radial axis. The computer-generated model of the radiation system may have an initial linac position and orientation, and an initial MVD position and orientation. These initial positions and orientations may be the intended locations and positions of the linac and the MVD according to a CAD schematic, or may be a nominal set of locations and positions. The relative positions of the phantom device beads are known and fixed, so they act as a stationary reference for each iteration. In each iteration, the positions and / or orientations of the linac and MVD positions and / or orientations may be adjusted. For example, in one variation, the parameters that may be iteratively adjusted may comprise a z-axis location of the linac, a z-axis location of the MV detector plane, an x-axis location of the MV detector (or two x-axis locations for a 2-panel MV detector), and a y-axis location of the MV detector (or two y-axis locations for a 2-panel MV detector). In some variations, there may be other parameters of the linac and MVD that are adjustable, in addition to their position and orientation. One or more of these parameters may be adjusted in each iteration until the error value is below the acceptable error threshold. The predetermined or known positions and sizes of the radiopaque beads on the phantom stored in a computer-readable medium of a controller processor may be used to simulate 214 the projected bead locations at the (virtual) MVD.

[0074] As described above, method 200 may comprise generating 210 an output transform for the MVD. In some variations, generating an MVD output transform may occur after the physical locations of the linac and MV detector are determined, along with a center point of the radiotherapy system. An MV detector may comprise multiple detector panels and an output transform may be generated for each of the detector panels. For example, a radiotherapy systemmay have an MVD comprising two detector panels, and in this system, method 200 may comprise generating two MVD output transforms, one output transform for the first MV detector panel and a second output transform for the second MV detector panel. An MVD output transform may be a projection matrix or mapping that translates MVD imaging data of a pixel (which is acquired on the plane of the MV detector) to a corresponding pixel that is coplanar with the center of the radiotherapy system. The output transform may account for the vertical (e.g., z-axis), planar (e.g., x-, y-axis), and angle (e.g., y-axis roll) shifts of the MV detector, as well as vertical shifts (e.g., z-axis) of the linac. In some variations, the MVD output transform maps MVD imaging data from the plane of the MV detector to the radiotherapy system isoplane, which includes the isocenter and may include the couch plane. MVD output transforms may be used to move the imaging data acquired on the MVD plane to a region on the couch plane incorporating orthogonal and angle correction, where the region is referenced to the stationary coordinate system at least partially defined by the calculated radiotherapy system center (e.g., isocenter). In some variations, MVD output transforms map MVD imaging data to a corresponding region in the couch plane. This may help the MV detector imaging data more accurately represent the object (e.g., patient, phantom-device) on the couch.

[0075] In some variations, an MVD output transform may be a homogeneous transformation matrix that represents a perspective projection of points at the MV detector plane onto the rotating plane (e.g., the plane of the rotating gantry) along lines to the center of the linac. This may facilitate imaging processing of MVD imaging data to be referenced to the rotating plane. An MVD output transform may be generated for each firing angle and applied to imaging data acquired at that firing angle. In one variation, points on the MV detector pixels may be referenced from the origin at the center of the MVD field and represented as columns of a matrix in units of (mm), e.g., [x;y;0;l]. The MVD output transform multiplied by the column vectors may be normalized so that the homogeneous coordinate is one; that is, divide each resulting column by its fourth element. In some variations, multiplying the MV detector pixel data (e.g., which may be in the form of a matrix) may project the MV detector pixels or points to the x-y plane of the rotating gantry.

[0076] Some methods may use the physical positions and orientations of the linac and MV detector and the one or more MVD output transforms to characterize (e.g., QA) the alignment of the beam-shaping components relative to the linac and MV detector. FIG. 3 A is a flowchart representation of one variation of a method for generating a starshot image based on acquiredMVD imaging data and applying the MVD output transforms derived from the MVD imaging data of the beaded phantom. Method 300 may comprise acquiring MVD imaging data of air scans with the jaws and MLC in different configurations. Method 300 may comprise acquiring 302 MVD imaging data of the even and odd leaves of the MLC from a plurality of firing angles by rotating the linac and MVD in the absence of a phantom, with the jaw completely open, identifying 304 the edges of the even and odd MLC leaves from the MVD imaging data, determining 306 the location of the edges of the even and odd MLC leaves for a plurality of firing angles and applying the MVD output transforms to the acquired MVD images of the even and odd leaves of the MLC, determining 308 the center (x, z coordinates) of the two central leaves of the MLC for a plurality of firing angles, and generating 310 e.g., either empirically or by calculation based on 306, generating 310 a 2-D (x, z) plot of the MLC center location and / or center beamline for each firing angle (e.g., relative to the stationary coordinate system) and fitting 312 a minimum circle to the points and / or center beamline on the 2-D (x, z) plots. In some variations, the 2-D (x,z) plot of the MLC center location may be defined relative to the stationary coordinate system using the MVD output transforms (i.e., when determining 306 the location of the edges of the even and odd MLC leaves), or using an additional transform that maps points on the rotating plane (e.g., the plane of the rotating gantry) to the stationary coordinate system (e.g., the coordinate system of the couch). Acquiring 302 MVD imaging data of the even and odd leaves of the MLC from a plurality of firing angles may comprise acquiring a first set of imaging data of the even MLC leaves by opening the even numbered MLC leaves and closing the odd numbered MLC leaves, and acquiring a second set of imaging data of the odd MLC leaves by opening the odd numbered MLC leaves while closing the even numbered MLC leaves. Determining 308 the center of the two central leaves of the MLC may comprise calculating the center based on the location of the edges of the even and odd MLC leaves determined in step 306, and / or may comprise empirically measuring the center by emitting radiation from the linac with the two central MLC leaves open. For example, in some variations, an MLC may comprise 64 leaves so to determine the center of the MLC, radiation may be emitted from the linac while leaf #32 and leaf #33 are opened (and at least leaf #31 and leaf #34 are closed). The center of the MLC may be determined for multiple firing angles, ranging from about 5 firing angles (e.g., similar to the number of beams used in traditional film-based starshot measurements) to about 50 firing angles or more (e.g., with a rapidly rotating gantry that may acquire images from many firing / gantry angles in a few seconds). The firing angles may beevenly spaced or unevenly spaced. Fitting 312 a minimum circle to the points may comprise adjusting each point in the 2-D (x, z) plot along a line from initial coordinates to the linac position until a minimum fit is achieved. In one variation, a minimum tangent circle may be defined by finding the minimum size of a circle that would be tangential to the center radiation beamlets. The size of the minimum tangent circle represents the magnitude of the error or imprecision of the location of the radiotherapy system center. For example, one variation of a method for fitting 312 a minimum circle to the points may include creating beam centerlines from center of the MLC to the linac and fitting a minimum tangent circle to those lines. The endpoints of the beam centerlines may be adjusted until a minimum fit is attained.

[0077] FIG. 3B is a plot generated using the methods described herein (which may use MVD output transforms derived from MVD imaging data of a beaded phantom; also referred to as phantom-based methods) with MVD imaging data acquired from 5 firing angles. The MVD imaging data was acquired in an experiment on a radiotherapy system having a rotatable gantry and by rotating the linac and MVD around the patient area. The center radiation beamlets (i.e., with the center MLC leaves open, or calculated from MLC leaf data) were plotted for 5 firing angles. The center radiation beamlets (322a, 322b, 322c, 322d, 322e) are plotted as dotted lines, which intersect to form a star-like shape. The minimum tangent circle 320 was defined or fitted by finding the minimum size of a circle that would be tangential to the center radiation beamlets. The size of the minimum tangent circle represents the magnitude of the error or imprecision of the location of the radiotherapy system center. In this experiment, the minimum tangent circle was calculated to have a radius of 0.532 mm. FIG. 3C is a plot generated from a film-based starshot plot that measured the center radiation beamlets emitted from 5 firing angles, for the same radiotherapy system that was characterized in FIG. 3B. In this film-based method, the central leaf or leaves of the MLC were opened to shape the radiation emitted by the linac into a central beamlet. In this experiment, the minimum tangent circle was measured to have a radius of 0.48 mm. FIG. 3D depicts a plot generated using the phantom-based methods described herein, but instead of displaying the data for 5 firing angles as in FIG. 3B, this plot displays the MVD imaging data acquired from 50 firing angles. The minimum tangent circle based on the MVD imaging data from 50 firing angles was calculated to be 0.53 mm.

[0078] FIGS. 4A-4F depict the results of an experiment where the isocenter for multiple radiotherapy systems was characterized (i.e., in a QA procedure) using the traditional film-based starshot analysis and also using the phantom-based methods described herein. During theexperiment, a phantom device having 4 radiopaque beads (similar to the ones described herein) was advanced into a radiotherapy machine and images were captured of the phantom device using an electronic portal imaging device (EPID). The EPID had a first high-resolution acquisition mode (2560 x 384) and a mid-resolution acquisition mode (1280 x 192) which were used to capture images of the phantom device at 50 unique gantry positions and analyzed by software having instructions to execute one or more of the methods described herein to fit a curve to data relative to gantry angle. A minimum tangent radius to the 5 lines was found using software executing the instructions to perform the methods described herein. Additionally, traditional film-based starshot measurements were taken at 5 angles and analyzed by software provided by RIT. The film-based starshot analysis was performed according to AAPM Task Group 142, and utilized analysis software provided by RADIOLOGICAL IMAGING TECHNOLOGY, INC. (RIT). For comparison, values were exported from the fitted curve at the 5 angles specified by TG-142 guidelines to generate beamline center offsets. These two sets of tests were performed on 6 different machines over 6 months and the phantom-based starshot results were analyzed and compared with the film-based starshot results.

[0079] The results of the experiment are depicted in FIGS. 4A-4F. Overall, both phantombased starshot measurements and film-based starshot measurements were within TG-142 tolerances. FIG. 4A depicts a film image of one of the starshot experiments for one of the radiotherapy systems. FIG. 4B depicts a plot generated from the film image of FIG. 4A, where it was determined that the minimum tangent circle had a radius of 0.22 mm. FIGS. 4C-4D depict plots generated from software analyzing the phantom-based data using the methods described herein, acquired on the same radiotherapy system as for FIGS. 4A-4B. FIG. 4C depicts a plot of the beam centerlines for the 5 starshot angles. Based on this plot, the minimum tangent circle had a radius of -0.352. FIG. 4D depicts a plot of beam centerlines for 50 gantry angles. Based on this plot, the minimum tangent circle has a radius of -0.352. Both of the plots depicted in FIGS. 4C-4D are at IEC XZ plane with gantry axis set to (0,0).

[0080] Phantom measurements were taken and analyzed according to the phantom-based methods described herein for another radiotherapy system. FIG. 4E depicts a plot of the beam centerlines for 5 starshot angles and its associated minimum tangent circle (in the blue dotted line). This data is overlaid with the beam centerlines for 50 starshot angles and its associated minimum tangent circle (in the red solid line). Based on the 50-angle data, the minimum tangent circle fit -0.388 mm and based on the 5-angle data, the minimum tangent circle fit is -0.358 mm.

[0081] FIG. 4F depicts a table that summarizes the results of the film-based starshot experiments and the phantom-based starshot experiments conducted over 6 radiotherapy systems A-F. The table includes information pertaining to the systems, film measurements, phantom measurements, image resolution acquisition, statistical analysis method, and corresponding values.

[0082] These experimental results indicate that the devices and methods described herein may be used to in a QA procedure to characterize the radiation isocenter of a radiotherapy system and have a comparable level of precision and accuracy as film-based starshot analysis.

[0083] The MVD imaging data acquired of the beaded phantom and processed according to the methods described herein may be used to calculate a 3-D center of a radiotherapy system. A film-based starshot plot may be used to find and characterize the center of the radiotherapy system at a particular plane (e.g., an x-z plane at a particular y-axis location) in the patient treatment area. Such a center is a 2-D center because it is located on the plane where the film is installed. With the MVD imaging data acquired of the beaded phantom and additional air scans, a 3-D center of the radiotherapy system may be found. The error or imprecision of the 3-D center may be represented by a sphere (which may be referred to as the isosphere). FIG. 5A is a flowchart representation of one variation of a method for calculating and characterizing a 3-D isosphere. Method 500 may comprise acquiring 502 MVD imaging data of the even and odd leaves of the MLC from a plurality of firing angles by rotating the linac and MVD in the absence of a phantom, with the jaws completely open, acquiring 504 open field MVD imaging data of all MLC leaves open, jaws open, identifying 506 the edges of the jaws from the MVD imaging data, determining 508 the location of the edges of the jaws for a plurality of firing angles and applying the MVD output transforms to the open field images, determining 510 the center (e.g., y- coordinate) of a jaw field for a plurality of firing angles using the transformed MVD open field images, identifying 512 the edges of the even and odd MLC leaves, determining 514 the location of the edges of the even and odd MLC leaves for a plurality of firing angles and applying the MVD output transforms to the acquired MVD images of the even and odd leaves of the MLC, determining 516 the center (x, z coordinates) of the two central leaves of the MLC for a plurality of firing angles relative to stationary coordinate system using the transformed MVD images of the even and odd leaves of the MLC, generating 518 a 3-D plot (x, y, z coordinates) of the center location and / or center beamline of a plurality of firing angles, and fitting 520 a sphere of minimum radius (i.e., isosphere) such that each (x,y,z) coordinate for each firing angle isapproximately on the surface or inside of the sphere. Fitting 520 a minimum sphere to the points may comprise adjusting each point in the 3-D (x, y, z) plot along line from initial coordinates to the linac position until a minimum fit is achieved. For example, one variation of a method for fitting 520 a minimum sphere to the points may include creating beam centerlines from center of the MLC to the linac and fitting a minimum tangent sphere to those lines. The endpoints of the beam centerlines may be adjusted until a minimum fit is attained. FIG. 5B is an example of a plot of an isosphere for a radiotherapy system that was generated using the method described in FIG. 5 A.

[0084] Some methods may comprise acquiring MVD imaging data of the phantom device and analyzing the MVD imaging data to characterize gantry angle-specific radiation beamlet alignments, including but not limited to offsets and / or deviations of the MLC and / or jaw(s) relative to system isocenter. The method may comprise determining MLC leaf displacements (e.g., in IEC-X) or tilts and / or determining jaw deflections or displacements (e.g., in IEC-Y) for multiple (e.g., every) gantry angles. This information may be incorporated into a radiation beam model, which may be used for dose calculations as part of treatment planning and / or quality assurance evaluations. This information may be incorporated into methods that convert a radiation fluence map into radiotherapy system machine instructions or commands (e.g., fluence map segmentation methods), so that executing the machine instructions will result in the delivery of radiation in accordance with the radiation fluence map. One variation of a method may comprise using the phantom to generate MVD output transforms (e.g., using methods 200, 250 described above and depicted in FIGS. 2A and 2C), acquiring MVD data of the MLC and using the MVD output transforms to determine the MLC center location for a plurality of gantry angles (i.e., firing angles) (e.g., using methods 300 and 500 described above and depicted in FIGS. 3 A and 5 A), acquiring MVD data of the jaw and using the MVD output transforms to determine the jaw center location for a plurality of gantry angles (e.g., using method 500 described above and depicted in FIG. 5 A), calculating a deviation of the MLC and the jaw from the system isocenter for each gantry angle by comparing the MLC center location and the jaw center location for each gantry angle with the system isocenter (e.g., mechanical isocenter), and generating a radiation beam model that simulates the effects of the MLC and / or jaw deviations (at each gantry angle) on emitted radiation beamlets and / or delivered radiation dose.

[0085] FIG. 6 is a flowchart representation of one variation of a method for characterizing gantry angle-specific radiation beamlet deviations (e.g., deflections, offsets, misalignments, etc.)relative to the radiotherapy system isocenter. These gantry angle-specific radiation beamlet deviations may include one or more of MLC leaf center location deviations, MLC tilt, and / or jaw center location deviations for each of a plurality of gantry angles. These component deviations may be integrated into a beam model of the radiotherapy system. Incorporating information about the collimation components into the beam model may help improve dosimetric agreement between the measured delivered dose and the simulated / calculated dose, especially for treatment plans and / or dose delivery to small target regions located near system isocenter. Method 600 may comprise acquiring 602 MVD imaging data of the phantom from a plurality of firing angles by rotating the linac and MVD around the phantom, generating 604 MVD output transforms that map MVD imaging data relative to the center point location of the radiotherapy system, acquiring 606 MVD imaging data of the even and odd leaves of the MLC from a plurality of firing angles by rotating the linac and MVD in the absence of a phantom, with the jaws completely open, calculating 608 the MLC center location (e.g., x-coordinate) for each firing angle by applying the MVD transforms on the acquired MVD imaging data to determine the location of MLC leaf edges, acquiring 612 open field MVD imaging data with all MLC leaves open, jaws open, calculating 614 the jaw center location (e.g., y-coordinate) for each firing angle by applying the MVD output transforms to the acquired MVD imaging data to determine the location of the edges of the jaws, and incorporating 618 the MLC center locations and the jaw center locations into a radiation beam model. Alternatively, or additionally, method 600 may comprise calculating 610 a deviation of the MLC center location from system isocenter for each firing angle, and / or calculating 616 a deviation of the jaw center location from system isocenter for each firing angle. One or both of the deviations from steps 610, 616 may optionally be incorporated 620 into the radiation beam model. MLC or MLC leaf center deviations may include displacements or offsets along the X-axis (e.g., IEC-X) direction and / or Y-axis (e.g., IEC-Y) direction. In some variations, MLC deviations may include MLC tilt angles for each of multiple gantry angles. Jaw center deviations may include displacements or deflections along the Y-axis (e.g., IEC-Y) direction and / or the X-axis (e.g., IEC-X) direction. The MLC center location may include an X-coordinate or Y-coordinate, while the jaw center location may include a Y-coordinate or X-coordinate, respectively. In one variation, generating 604 MVD output transforms may comprise one or more of the steps described in methods 200, 250. Calculating 608 the MLC center location for each firing angle (or gantry angle) may comprise one or more of the steps described in methods 300, 500. In some variations, calculating 614 thejaw center location for each firing angle may comprise one or more of the steps described in method 500.

[0086] These component deviations may optionally be included in segmentation methods used to transform a radiation fluence map (which specifies the desired amount of radiation per volume or voxel of a target region or area to be treated) to radiotherapy system machine instructions or commands such that by executing these instructions or commands, the delivered radiation fluence would match the radiation distribution according to the radiation fluence map. Segmentation methods may use similar information (e.g., arrangement of beam generation and beam-shaping components, any deviations, offsets, or dynamic artifacts during motion or rotation) as may be contained within a beam model in order to generate machine instructions or commands to deliver radiation with a desired distribution profile. In one variation, machine instructions may comprise a discretized fluence sinogram. The discretized fluence sinogram may be a 3-D matrix where the matrix dimensions are the number of MLC leaves, number of firing positions, and number of beam stations, and each matrix entry is the radiation intensity (e.g., the number of linac pulses). As an example, the 3-D matrix of a discretized fluence sinogram for a radiotherapy system having 64 MLC leaves, 50 firing positions, and N beam stations may comprise a 64 x 50 x N matrix. Component deviations that impact the opening of MLC leaves (e.g., adjacent MLC leaves) may be accounted for during segmentation by adjusting the radiation intensity for one or more entries in the discretized fluence sinogram such that the resultant delivered dose approximates the desired radiation fluence map. This may include, for example, shifting radiation intensities between two adjacent MLC leaves to enact a “feathering” effect, changing the combinations of open MLC leaves, and / or adjusting radiation intensity values (e.g., increasing or decreasing the number of pulses) in the discretized fluence sinogram. Optionally, component deviations may be considered during radiation beamlet or fluence map optimization during radiotherapy treatment planning. The optimizer may account for such deviations when generating radiation beamlets for the planned fluence map by adjusting beamlet intensity values to emulate MLC misalignments and / or any oscillations of the system components that give rise to beamlet deviations. Accurate information about the precise arrangement of the radiotherapy components may help improve the quality of a segmentation method so that the delivered radiation more closely matches the fluence map.

[0087] FIGS. 7A-7J depict the results of an experiment performed using method 600 of FIG. 6. A phantom device comprising 4 radiopaque elements or beads (referred herein as the “4-ballphantom”) was imaged at 60 RPM by a megavoltage (MV) detector, and the MV detector imaging data was analyzed and processed using the methods described herein to characterize gantry angle-specific beamlet offsets relative to system (e.g., mechanical) isocenter. The deviation or displacement of MLC leaves in the X-axis direction and displacement of the jaws in the Y-axis direction were calculated for each gantry (firing) angle. The amount of displacement of the MLC leaves and the jaws was incorporated into a beam model as gantry angle-specific leaf filters and jaw profiles. A dose calculation for a treatment plan for a 1.5 cm central small target was calculated using the beam model that incorporated the MLC and jaw displacement data (with the appropriate corrections). A dose calculation for the same treatment plan was calculated using a beam model without the MLC and jaw displacement data (without any corrections). These two dose calculations were compared to the ion chamber point dose measurement of the dose delivered to a 1.5 cm central small target.

[0088] FIG. 7A depicts a plot of the MLC center location variation over multiple firing / gantry angles as calculated using the acquired MV detector data in conjunction with the 4-ball phantom. FIG 7B depicts the jaw center location variation over multiple firing / gantry angles as calculated using the acquired MV detector data in conjunction with the 4-ball phantom. As shown in FIG. 7 A, the MLC X-shift for central radiation beamlets (e.g., beamlets that cross through the radiotherapy system isocenter) varies with angle as a sinusoid of 0.2mm amplitude and 0.3mm offset. FIG. 7B depicts a plot of the locations / positions of edges the jaws (i.e., jaw blocks) over multiple firing / gantry angles as calculated using the acquired MV detector data in conjunction with the 4-ball phantom. As described above, jaws comprise a pair of jaw blocks, labeled “Jaw 2” and “Jaw 1” in the upper and lower plots, respectively. The positions of the center of the jaw block edges were determined as described above for method 500 (e.g., steps 502, 504, 506, 508, 510). A Fourier fitting method was applied to the acquired MVD imaging data of the center of the jaw edges to calculate a jaw center for each gantry angle. The Fourier fitting method (or any desired fitting method or algorithm) may help account for known offsets and / or alignments of the jaws, MLC, and / or 4-ball phantom radiopaque beads to calculate the jaw center at each gantry angle. FIG. 7C are plots derived from the data in FIG. 7B, which show that jaw displacement followed an irregular trajectory totaling 0.4mm amplitude, with negative displacement peaking at 150°, positive displacement peaking at 315°, and a plateau near 220°. FIG. 7D is an axial slice through the dose distribution to a small target as calculated using a beam model without the corrections for MLC and jaw deviations, while FIG. 7E is an axial slicethrough the dose distribution to a small target as calculated using a beam model with the corrections for MLC and jaw deviations. FIG. 7F depicts the difference between the dose distribution of FIG. 7D and the dose distribution of FIG. 7E.

[0089] The plots in FIGS. 7G-7J demonstrate that incorporating gantry angle-specific alignment offsets for collimation components improved dosimetric agreement between the actual delivered radiation dose (e.g., to film) and a dose calculation based on a beam model. FIG. 7G depicts the dose profile along the X-axis as measured on film (blue line) after delivering radiation according to the treatment plan, and the dose profile as calculated from a simulation using a beam model without the corrections for MLC and jaw deviations. FIG. 7H depicts the dose profile along the X-axis as measured on film (blue line) after delivering radiation according to the treatment plan, and the dose profiled as calculated from a simulation using a beam model with the corrections for MLC and jaw deviations. The two dose profiles in FIG. 7G differed by 3.77% while the two dose profiles in FIG. 7H differed by 0.56%. Visually, a profile across the target evolved from a sharp monotonic peak to shallower bimodal peaks. FIG. 71 depicts the simulated film results (e.g., dose distribution) derived from a simulation using a beam model without the corrections for MLC and jaw deviations. FIG. 7H depicts the simulated film results derived from a simulation using a beam model with the corrections for MLC and jaw deviations. The film QA passing rate (2% / 2mm) for the FIG. 71 dose distribution was 84.2% while the film QA passing rate (2% / 2mm) for the FIG. 7J dose distribution was 93.6%. Dose calculations for larger targets were negligibly affected.Correcting Radiation Beamlet DeviationsSteering Coil Control based on Firing Position

[0090] Radiation beamlet deviations (e.g., deflections, offsets, misalignments, etc.) from the radiotherapy system isocenter which may arise due to the cumulative deviations in mechanical components, such as the beam-generating and / or beam-shaping components. In some variations, these deviations may vary for each firing position of the therapeutic radiation source. The methods described above calculate these deviations by measuring imaging data acquired of a phantom device. There may be a different beamlet deviation for each firing position of the therapeutic radiation source. The plots depicted in FIGS. 3B, 3D, 4C, 4D, and 5B are examples of variable beamlet deviations for each firing position or gantry angle. In variations where the therapeutic radiation source comprises a linac, beamlet deviations may be corrected usingsteering coils attached to the linac. In some variations, steering coils for the linac may be configured to compensate for beamlet deviations in the IEC-X and IEC-Y directions. In some variations, a therapeutic radiation source may comprise a source of electrons, a linac through which the electrons are accelerated as they move from a proximal end of the linac to a distal end of the linac, and a target material located at the distal end of the linac. A therapeutic radiation beam is generated when the accelerated electrons strike the target material and cause a release of high-energy photons or radiation that are directed toward the beam-shaping components (e.g., jaws, MLC). The steering coils adjust the X-ray radiation beam position by steering the trajectory of the electron beam as it is being accelerated through the linac. Controlling the trajectory of the electron beam causes it to strike the target material at different locations, thereby changing the location from which X-ray radiation beams are generated and emanate from the linac. Steering coils affect the trajectory of the electron beam via magnetic fields generated by electric currents flowing through the coils. The magnitude and direction of the currents in the steering coils generate magnetic fields that affect the directionality of the electron beam, which results in corresponding changes in the directionality of the radiation beams. In some variations, the steering coils may comprise two set of dipole bending coils mounted at the proximal end of the linac, where the electron beam has the lowest energy (because the electrons have not yet been accelerated through the linac). Each set of steering coils (e.g., dipole bending coils) may be controlled with separate currents to provide independent electron beam steering (e.g., bending) in the X direction and Y direction. In some variations, a radiotherapy system may comprise a power supply that is electrically connected to the coils to drive current into the coils based on system controller commands.

[0091] FIG. 8A depicts one variation of a steering coil 800. Steering coil 800 may be a magnetic coil comprising electrical wires that are wound together in a closed loop. The shape of the loop may be circular, rectangular, or any desirable shape for attached to a linear accelerator. FIG. 8B is a schematic depiction of one variation of a therapeutic radiation source comprising an electron gun 801, a linear accelerator or linac 803, and steering coils 802, 804, and 806. In this variation, there is another steering coil 808 located opposite steering coil 802, but is not visible in FIG. 8B. The linac 803 may comprise a plurality of stacked cylindrical cavities which facilitate the acceleration of electrons within the linac. The electron gun 801 emits a beam of electrons traveling down the inside of the linac 803 in the direction of arrow 805. The steering coils 804 and 806 are mounted on the linac 803 at locations that are opposite each other (i.e.,180° from each other), and are configured to adjust the electron beam trajectory along a first axis (e.g., X-axis). The pair of steering coils 804, 806 are operated together to adjust the electron beam trajectory along the first axis. The steering coils 802 and 808 are mounted on the linac 803 at locations that are opposite each other (i.e., 180° from each other), and are configured to adjust the electron beam trajectory along a second axis (e.g., Y-axis). The pair of steering coils 802, 808 are operated together to adjust the electron beam trajectory along the second axis. The steering coils 802 and 808 are at locations on the linac 803 that are 90° offset from the steering coils 804 and 806. In this variation, steering coil 806 is located higher up on the linac 803 than the steering coil 804, however, in other variations, pairs of steering coils may be located directly across from each other at the same height of the linac. FIG. 8C depicts a front elevational view of another variation of a therapeutic radiation source comprising an electron gun 801, a linear accelerator or linac 803, and steering coils 812, 814, 816, and 818, and FIG. 8D depicts a back elevational view of the therapeutic radiation source of FIG. 8C. The steering coils are all located at the same height along the length of the linac 803. The pair of steering coils 814, 816 are operated together to adjust the electron beam trajectory along the first axis (e.g., Y-axis), and the pair of steering coils 812, 818 are operated together to adjust the electron beam trajectory along the second axis (e.g., X-axis). Mounting the steering coils at the same height or location along the length of the linac may help to steer the electron beam more evenly.

[0092] In one variation, a look-up table (LUT) comprising commands for electrical currents for each firing position and / or gantry angle may be stored in the memory of the system controller. One method for generating a LUT for controlling the steering coil(s) in a therapeutic radiation source may comprise determining a radiation beamlet deviation for each firing position and / or gantry angle, determining the beamlet shift vectors for correcting the radiation beamlet deviation, calculating the current(s) for each of the steering coils to attain the beamlet shift vectors, and storing the current(s) in a LUT, indexed by firing position and / or gantry angle. Any of the methods described herein may be used to characterize the beamlet deviation for each firing position and / or gantry angle. Alternatively, or additionally, in some variations, the radiation beamlet deviation may be similar across all firing positions and / or gantry angles, in which case, the electrical currents through the steering coils may be the same (or unchanged) for each firing position and / or gantry angle.Jaw Position Control based on Firing Position

[0093] Alternatively, or additionally, beamlet deviations may be corrected by moving the jaws, such that the location of the jaw aperture compensates for beamlet shifts and / or offsets. In some variations, the jaw aperture size may not change for different firing positions and / or gantry angles, but may be shifted to provide the same radiation beamlet field size at a different location. Adjusting the position of the jaws so that the jaw aperture is in a different location may help compensate for beamlet deviations, which may include deviations in the IEC-Y direction, IEC-Y direction, and / or deviations around the radiation beam axis (e.g., rotational deviations around the beam axis, from about 1° to about 90°, from about 1° to about 359°, etc.). In one variation, jaws may comprise two opposing jaw blocks that are movable along a rail. The rail may be curved or straight. The jaw blocks may be independently movable, and may each have their own motors to adjust their location along the rail. Radiation beamlet deviations along a particular axis, e.g., Y- axis (also referred to as the IEC-Y axis) may be compensated for by moving one or both of the jaw blocks. The jaw blocks may be moved such that the radiation beamlet field size is kept relatively constant while adjusting the location of the beamlet. The position of the first and second jaw blocks along a curved rail may be adjusted by the same shift amount or may be adjusted by a different shift amount, depending on their location on the curved rail and / or the firing position, in order to help maintain a constant radiation beamlet field size at system isocenter. In some variations, both jaw blocks may move together in the same direction along the rail, but may be moved by different shift amounts. In contrast to other systems where the position of the jaws and / or the position of the individual jaw blocks may be adjusted solely based on the jaw block’s local position sensor value (e.g., encoder count value) and desired jaw aperture size, the methods described herein may also adjust the position of the jaws and / or the position of the individual jaw blocks to compensate for radiation beamlet deviations. The position of the jaws and / or individual jaw blocks may be changed on a per-firing position and / or per-gantry angle basis to compensate for beamlet deviation(s). Alternatively, or additionally, the position of the jaws and / or individual jaw blocks may be changed along a smooth motion profile across firing positions and / or gantry angles. For example, the desired locations or positions of the jaws and / or jaw blocks for multiple firing positions and / or gantry angles may be used to define a motion trajectory for the jaws and / or jaw blocks. The jaws and / or jaw blocks may be gradually moved along the defined motion trajectory as the gantry moves through different angles and / or positions, such that the jaws and / or jaw blocks are at the desired position at thefiring positions and / or gantry angles to compensate for the radiation beamlet deviations at those firing positions and / or gantry angles. While some variations may adjust the position(s) of the jaws for every firing position and / or gantry angle, in other variations, the radiation beamlet deviation may be similar across all firing positions and / or gantry angles, in which case, the position(s) of the jaws may be the same (or unchanged) for each firing position and / or gantry angle.

[0094] FIG. 9 depicts a side view of one variation of a jaws assembly. Jaws assembly 900 may comprise a first jaw block 902 mounted on a first jaw carriage 903, a second jaw block 904 opposite the first jaw block and mounted on a second jaw carriage 905, a jaw arc rail 906 upon which the first and second jaw carriages may be positioned and / or moved, a first jaw motion assembly 908 configured to move the first jaw carriage 903, and a second jaw motion assembly 910 configured to move the second jaw carriage 905. A jaw motion assembly may comprise a motor, a jaw motor linkage that couples the motor to the jaw carriage and a jaw block position encoder. The jaw opening or jaw aperture 912 may be the space between the first jaw block 902 and the second jaw block 904. Movement of the jaw carriages 903, 905 on the rail 906 also moves jaw blocks 902, 904 respectively, which in turn may adjust the location of the radiation beamlet through the jaw aperture 912 along an axis, e.g., the IEC-Y axis. The shift values for each jaw block 902, 904 may be different for each firing position and / or gantry angle, in order to compensate for the radiation beamlet deviation at each firing position and / or gantry angle. Alternatively, or additionally, the radiation beamlet deviation may be similar across all firing positions and / or gantry angles, in which case, the position(s) of the jaws may be the same (or unchanged) for each firing position and / or gantry angle.

[0095] Radiation beamlet deviations may be corrected and / or compensated for by moving the beam using steering coils and / or moving the jaws, alone or in combination. Each firing position of the therapeutic radiation source may have specific steering coil current(s) and jaw position(s) to help align the generated radiation beamlet as closely as possible to the system isocenter. In some variations, a look-up table (LUT), indexed by firing position and / or gantry angle, may be saved in the memory of a radiotherapy system controller that contains steering coil current and jaw positions for each linac firing position and / or gantry angle. The configuration of these beamsteering and beam-shaping components may be changed for each firing position and / or gantry angle based on the data in the LUT. Alternatively, or additionally, the radiation beamlet deviation may be similar across all firing positions and / or gantry angles, in which case, theposition(s) of the jaws and / or the electrical current(s) through the steering coils may be the same (or unchanged) for each firing position and / or gantry angle.

[0096] FIG. 10 depicts a flowchart representation of one variation of a method for compensating for radiation beamlet deviations in a radiotherapy system. Method 1000 may comprise determining 1002 a deviation from a radiotherapy system isocenter of a radiation beamlet generated by a therapeutic radiation source for each firing positions of the therapeutic radiation source, determining 1004 a radiation beamlet shift for each firing position that compensates for the deviation at that firing position, and compensating 1006 for the deviation of the radiation beamlet at each firing position by adjusting radiotherapy system components to move the radiation beamlet according to the radiation beamlet shift for that firing position. Determining 1002 a radiation beamlet deviation at a firing position and / or gantry angle may comprise any of the methods described herein, for example, the methods described and depicted in FIGS. 3A and 5A. Determining 1002 radiation beamlet deviations for multiple firing position and / or gantry angles may comprise acquiring imaging data of a beaded phantom using a radiation imager, from multiple firing angles of the therapeutic radiation source. These imaging data of the beaded phantom may be used to generate radiation imager output transforms, calculating a center location of the radiation beamlet using the radiation imager output transforms and the acquired imaging data at each firing position, and calculating the deviation of the center location of the radiation beamlet from the radiotherapy system isocenter for each firing position. The radiation beamlet deviation at each firing position and / or gantry angle may be the cumulative effect of deviations in the mechanical components of the radiotherapy system, such as the MLC and / or the jaws and / or the therapeutic radiation source. In some variations, determining 1004 the radiation beamlet shift for each firing position to compensate for the deviation at that firing position may comprise calculating a first difference between the center location of the radiation beamlet along a first axis (e.g., IEC-X), calculating a second difference between the center location of the radiation beamlet along a second axis (e.g., IEC-Y), and generating a beamlet shift vector that has a first shift that compensates for the first difference along the first axis and a second shift that compensates for the second difference along the second axis. In some variations, the beamlet shift may be implemented by adjusting linac steering coil currents to correct for shifts along the first axis, and adjusting the position of the jaws to correct for shifts along the second axis. Some variations may further comprise calculating a third difference between the center location of the radiation beamlet along a thirdaxis (e.g., IEC-Z), and generating a 3-D vector that includes the shifts along three axes (e.g., IEC-X, IEC-Y, IEC-Z).

[0097] Method 1000 may optionally comprise generating a look-up table that comprises the radiation beamlet deviation, and / or radiation beamlet shift, and / or instructions for adjusting radiotherapy system components, indexed by firing position and / or gantry angle. The look-up table may be stored in a controller memory of the radiotherapy system and then accessed during radiation delivery to control the operation of the system components such that they compensate for radiation beamlet deviations. Instructions for the electric current(s) for the steering coil(s) may comprise current magnitude, direction, and / or frequency (e.g., duty cycle, pulse width modulation, etc.). Instructions for the jaws may comprise the jaw shift value(s) and direction along the rail for each of the jaw blocks and / or jaw aperture size, etc. The look-up table and the information contained therein may be generated during radiotherapy system installation and / or calibration procedures and referenced during radiation delivery. The look-up table may be periodically updated, as desired.

[0098] Compensating 1006 for, or correcting, radiation beamlet deviations may comprise adjusting a variety of components of the radiotherapy system. In one variation, radiation beamlet deviations may be corrected by moving the therapeutic radiation source (e.g., adjusting the linac position), moving the MLC, and / or moving the jaws. Alternatively, or additionally, radiation beamlet deviations may be corrected using the steering coils of the therapeutic radiation source. For example, beamlet deviations may be corrected solely using the steering coils, or beamlet deviations may be corrected by adjusting the linac position so that it is better aligned with system isocenter and also using the steering coils to further tune the beamlet trajectory so that it further aligns with system isocenter. In some variations, the beamlet deviations may be corrected or compensated for by simultaneously adjusting the steering coil currents of the linac and the position of the jaws and / or jaw blocks. For example, the steering coil currents may adjust for beamlet deviations along IEC-X and / or IEC-Y, and the jaws may adjust for beamlet deviations along IEC-Y. In some variations, the compensation for beamlet deviations in IEC-Y may involve adjusting both the steering coil current and position of the jaws, so that in combination, the beamlet deviation along IEC-Y may be corrected (via the linac steering coils) while also tuning the penumbra of the radiation beamlet (via the jaws), which may help maintain a consistent radiation beam profile. The adjustments to the linac, MLC, and / or jaws may be performed in discrete steps, for example, adjusting one or more of those components at eachfiring position and / or gantry angle, and / or adjusting one or more components continuously as the gantry rotates so that as the linac moves through each firing position, the beam-steering and / or beam-shaping components are in the configurations and / or positions that provide the desired compensation for beamlet deviations at that firing position. In some variations, compensating for the radiation beamlet deviations may comprise continuous adjustments to the position of the jaws through multiple firing positions (e.g., continuously and / or gradually along a motion trajectory) and discrete adjustments to the steering coil currents at each firing position (e.g., in discrete increments or values at each firing position).

[0099] The above-described systems and methods can be implemented in any number of ways, alone or in combination. For example, at least some methods of the present technology may be implemented using hardware, firmware, software, or a combination thereof. When implemented in firmware and / or software, the firmware and / or software code can be executed on any suitable processor or collection of logic components, whether provided in a single device or distributed among multiple devices. In some variations, beamlet deviations and / or component misalignments may be measured and / or characterized, but not corrected by adjusting component operation or alignments, and / or may or may not be incorporated in a radiation beam model. Such data may be included in a graphical user interface that informs the user of beamlet deviations and / or component misalignments. The user may incorporate this data to adjust or generate a radiotherapy treatment plan, and / or in any manner to adapt the delivery of radiation to a patient or phantom.

[0100] In this respect, various aspects described herein may be embodied as a computer readable storage medium (or multiple computer readable storage media) (e.g., a computer memory, one or more floppy discs, compact discs, optical discs, magnetic tapes, flash memories, circuit configurations in Field Programmable Gate Arrays or other semiconductor devices, or other non-transitory medium or tangible computer storage medium) encoded with one or more programs that, when executed on one or more computers or other processors, perform methods that implement the various embodiments of the invention discussed above. The computer readable medium or media can be transportable, such that the program or programs stored thereon can be loaded onto one or more different computers or other processors to implement various aspects of the present invention as discussed above.

[0101] The terms “program” or “software” are used herein in a generic sense to refer to any type of computer code or set of computer-executable instructions that can be employed toprogram a computer or other processor to implement various aspects of embodiments as discussed above. Additionally, it should be appreciated that according to one aspect, one or more computer programs that when executed perform methods disclosed herein need not reside on a single computer or processor but may be distributed in a modular fashion amongst a number of different computers or processors to implement various aspects of the inventions disclosed herein.

[0102] Computer-executable instructions may be in many forms, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. Typically, the functionality of the program modules may be combined or distributed as desired in different variations.

[0103] Also, data structures may be stored in computer-readable media in any suitable form. For simplicity of illustration, data structures may be shown to have fields that are related through location in the data structure. Such relationships may likewise be achieved by assigning storage for the fields with locations in a computer-readable medium that convey relationship between the fields. However, any suitable mechanism may be used to establish a relationship between information in fields of a data structure, including through the use of pointers, tags or other mechanisms that establish relationship between data elements.

[0104] Also, the acts performed as part of the method may be ordered in any suitable way. Accordingly, various methods may be constructed in which acts are performed in an order different than illustrated, which may include performing some acts simultaneously, even though shown as sequential acts in illustrative examples.

Claims

CLAIMS1. A method for compensating for radiation beamlet deviations in a radiotherapy system, the method comprising: determining a deviation from a radiotherapy system isocenter of a radiation beamlet generated by a therapeutic radiation source for each firing position of the therapeutic radiation source; determining a radiation beamlet shift for each firing position that compensates for the deviation at that firing position; and compensating for the deviation of the radiation beamlet at each firing position by adjusting radiotherapy system components to move the radiation beamlet according to the radiation beamlet shift for that firing position.

2. The method of claim 1, wherein determining the deviation from the radiotherapy system isocenter comprises acquiring imaging data of a phantom comprising a plurality of radiopaque beads, wherein the imaging data is acquired by a radiation imager from multiple firing angles of the therapeutic radiation source.

3. The method of claim 1, further comprising generating a look-up table that comprises the radiation beamlet shift for each firing position of the therapeutic radiation source and storing the look-up table in a controller memory of the radiotherapy system, and wherein compensating for the deviation of the radiation beamlet comprises accessing the look-up table based on firing position to determine the radiation beamlet shift for that firing position.

4. The method of claim 3, further comprising generating radiotherapy system component instructions for each radiation beamlet shift at each firing position, wherein the lookup table further comprises the radiotherapy system component instructions, and wherein compensating for the deviation of the radiation beamlet comprises executing the radiotherapy system component instructions for that firing position.

5. The method of claim 4, wherein the radiotherapy system components include one or more of the therapeutic radiation source, a multi -leaf collimator, and jaws comprising a pair of jaw blocks.

6. The method of claim 1, wherein the therapeutic radiation source comprises a linac and a plurality of steering coils attached to the linac, and wherein compensating for the deviation comprises controlling an electric current through the steering coils such that the radiation beamlet is moved according to the radiation beamlet shift for that firing position.

7. The method of claim 3, wherein the therapeutic radiation source comprises a linac and a plurality of steering coils attached to the linac, and wherein compensating for the deviation comprises controlling an electric current through the steering coils to move the radiation beamlet according to the radiation beamlet shift in the look-up table for that firing position.

8. The method of claim 4, wherein the therapeutic radiation source comprises a linac and a plurality of steering coils attached to the linac, wherein the radiotherapy system component instructions comprise steering coil current parameters, and wherein compensating for the deviation comprises controlling an electric current through the steering coils according to the steering coil current parameters for that firing position.

9. The method of claim 8, wherein steering coil current parameters include one or more of magnitude, direction, and frequency.

10. The method of claim 1, wherein the radiotherapy system comprises jaws located in a path of the radiation beamlet generated by the therapeutic radiation source, wherein the jaws comprise a pair of movable jaw blocks mounted on a rail, and wherein compensating for the deviation comprises moving the jaw blocks on the rail such that the radiation beamlet is moved according to the radiation beamlet shift for that firing position.

11. The method of claim 3, wherein the radiotherapy system comprises jaws located in a path of the radiation beamlet generated by the therapeutic radiation source, wherein the jaws comprise a pair of movable jaw blocks mounted on a rail, and wherein compensating for the deviation comprises moving the jaw blocks on the rail according to the radiation beamlet shift in the look-up table for that firing position.

12. The method of claim 4, wherein the radiotherapy system comprises jaws located in a path of the radiation beamlet generated by the therapeutic radiation source, wherein the jaws comprise a pair of movable jaw blocks mounted on a rail, wherein the radiotherapy system component instructions comprises jaw block shift values, and wherein compensating for thedeviation comprises moving the jaw blocks on the rail according to the jaw block shift values in the look-up table for that firing position.

13. The method of claim 12, wherein jaw block shift values include one or more of shift amount and shift direction along the rails.

14. The method of claim 13, wherein the jaw block shift values comprise a first jaw block shift value for a first jaw block, and a second jaw block shift value for a second jaw block.

15. The method of claim 14, wherein the first jaw block shift value and the second jaw block shift value are not the same.

16. The method of claim 12, wherein a jaw aperture between the jaw blocks remains constant for each firing position, and compensating for the deviation comprises moving both jaw blocks according to the jaw block shift values in the look-up table.

17. The method of claim 5, wherein the deviation of the radiation beamlet comprises a deviation along a first axis and a deviation along a second axis, and wherein compensating for the deviation of the radiation beamlet comprises executing instructions for the therapeutic radiation source to correct the deviation along the first axis and executing instructions for the jaws to correct the deviation along the second axis.

18. The method of claim 5, wherein the deviation of the radiation beamlet comprises a deviation along a first axis and a deviation along a second axis, and wherein compensating for the deviation of the radiation beamlet comprises executing instructions for the therapeutic radiation source to correct the deviation along the first axis and to correct the deviation along the second axis.

19. The method of claim 2, wherein determining the deviation from the radiotherapy system isocenter further comprises: generating radiation imager output transforms from the acquired imaging data of the phantom, calculating a center location of the radiation beamlet using the radiation imager output transforms and the acquired imaging data at each firing position, andcalculating the deviation of the center location of the radiation beamlet from the radiotherapy system isocenter for each firing position.

20. The method of claim 19, wherein determining the radiation beamlet shift for each firing position that compensates for the deviation at that firing position comprises calculating a first difference between the center location of the radiation beamlet along a first axis, calculating a second difference between the center location of the radiation beamlet along a second axis, and generating a beamlet shift vector comprising a first shift that compensates for the first difference along the first axis and a second shift that compensates for the second difference along the second axis.

21. The method of claim 20, wherein determining the radiation beamlet shift for each firing position that compensates for the deviation at that firing position further comprises calculating a third difference between the center location of the radiation beamlet along a third axis, and wherein the beamlet shift vector includes a third shift that compensates for the third difference along the third axis.

22. A method for characterizing radiation beamlet deviations, the method comprising: acquiring imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of a radiotherapy system and acquiring the imaging data using a radiation imager, wherein the radiotherapy system further comprises jaws and a multi -leaf collimator disposed in a radiation beam path of the therapeutic radiation source; calculating a center location for the multi-leaf collimator for each firing angle using the acquired imaging data; calculating a center location for the jaws for each firing angle using the acquired imaging data; calculating deviations of the center locations for the multi-leaf collimator and the jaws from the radiotherapy system isocenter for each firing angle; andcalculating radiation beamlet deviations from the radiotherapy system isocenter for each firing angle by combining the deviations of the center locations for the multi-leaf collimator and the jaws.

23. The method of claim 22, further comprising incorporating the deviations of the center locations for the multi-leaf collimator and the jaws into a radiation beam model.

24. A method for characterizing displacements of beam-shaping components of a radiotherapy component for simulating radiation dose delivery, the method comprising: acquiring imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of a radiotherapy system and acquiring the imaging data using a radiation imager, wherein the radiotherapy system further comprises jaws and a multi -leaf collimator disposed in a radiation beam path of the therapeutic radiation source; generating radiation imager output transforms from the acquired imaging data of the phantom; calculating a center location for the multi-leaf collimator for each firing angle using the radiation imager output transforms and the acquired imaging data; calculating a center location for the jaws for each firing angle using the radiation imager output transforms and the acquired imaging data; calculating deviations of the center locations for the multi-leaf collimator and the jaws from the radiotherapy system isocenter for each firing angle; and incorporating the multi-leaf collimator center location and the jaw center location deviations from system isocenter into a radiation beam model used to simulate a radiation dose delivery.

25. A method for determining a center of a radiotherapy system, the method comprising:acquiring, using a radiation imager, imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of a radiotherapy system; iteratively adjusting parameters of a virtual model of the radiotherapy system to generate simulated radiation imager data until the simulated radiation imager data with the acquired radiation imager data until an error value is below a threshold; and determining a center point of the radiotherapy system based on the adjusted parameters of the virtual model of the radiotherapy system.

26. The method of claim 23, wherein determining the center point comprises determining positions of the therapeutic radiation source and the radiation imager from the adjusted parameters of the virtual model.

27. The method of claim 24, further comprising generating an output transform function that maps the radiation imager data relative to the center point of the radiotherapy system.

28. A method for generating a plot for characterizing a radiotherapy system center, the method comprising: acquiring imaging data of a phantom comprising a plurality of radiopaque beads from multiple firing angles of a therapeutic radiation source of a radiotherapy system and acquiring the imaging data using a radiation imager, wherein the radiotherapy system further comprises jaws and a multi -leaf collimator disposed in a radiation beam path of the therapeutic radiation source; generating radiation imager output transforms from the acquired imaging data of the phantom; acquiring air scan imaging data of even and odd leaves of the multi-leaf collimator from multiple firing angles using the radiation imager to determine a location of edges of the even and odd leaves;applying the radiation imager output transforms to the location of the edges of the even and odd leaves to determine a center of two central leaves of the multi-leaf collimator from multiple firing angles; generating a plot of the center location for each firing angle relative to a stationary coordinate system; and fitting a minimum circle to points on the plot.

Images