Field-to-target assignment in radiation therapy planning

Abstract

The invention relates to field-to-target assignment in radiotherapy planning. A method (900) for generating and optimizing a radiotherapy plan for treating a plurality of targets (822-826) using a plurality of fields (802-806), each field (802-806) comprising a set of controllable points and each controllable point comprising a set of beam geometry parameters; wherein: the method (900) comprises: determining (908) a plurality of field-target assignments (812-816) for the plan by assigning each field (802-806) of the plurality of fields (802-806) to a respective non-empty subset of the plurality of targets (822-826); at least one field (802-806) is assigned to each target; and at least one field (802-806) is assigned to a non-empty proper subset of the plurality of targets.

Description

Technical Field

[0001] This invention relates to a radiotherapy plan, and more particularly to a method for generating or optimizing a radiotherapy plan that reduces or eliminates dose bridging or irradiation of healthy tissue located between multiple tumors. Background Technology

[0002] The use of energy to treat medical conditions encompasses known areas of current technological efforts. For example, radiation therapy is an important component of many treatment programs aimed at reducing or eliminating unwanted tumors. Unfortunately, the applied energy cannot inherently distinguish between unwanted material and desired or even vital adjacent tissues, organs, etc. (which may be referred to as "organs at risk," OARs) that are essential for the patient's continued survival. Therefore, energy, such as radiation, is typically applied in a carefully administered manner to at least attempt to confine the energy within a given target volume. So-called radiation therapy programs are commonly used in the aforementioned contexts.

[0003] Radiotherapy plans typically include specific values ​​for each of the various treatment platform "machine" parameters across multiple continuous fields. For example, machine parameters of a radiotherapy system may include the irradiation angle of the treatment beam or the configuration of the blades of a multi-leaf collimator.

[0004] Treatment plans for radiotherapy procedures are typically generated automatically through a process known as optimization. As used herein, “optimization” will be understood as improving candidate treatment plans, without necessarily ensuring that the result of optimization is actually a single optimal solution. Such optimization typically involves automatically adjusting one or more physical therapy machine parameters (usually while adhering to one or more corresponding constraints in these areas) and mathematically calculating possible corresponding treatment outcomes (such as dose levels) to identify a given set of treatment parameters representing a good trade-off between desired treatment outcomes and the avoidance of undesirable side effects.

[0005] In some cases, optimization is performed based on multiple different criteria or clinical goals. This may involve the use of a cost function, which comprises a mathematical model that attempts to balance conflicting clinical objectives or goals to produce a treatment plan with the lowest “cost.” Clinical objectives often take the form of requiring a specific metric to be greater than or less than a threshold. For example, a clinical objective or goal might require a specific percentage of the OAR volume to receive less than a certain number of doses. As another example, a clinical objective or goal might require a specific percentage of the planned target volume (PTV) to receive at least a certain number of doses. The cost function can be the sum of terms related to deviations from the clinical objective.

[0006] For cancer patients with multiple tumors, a challenge in planning optimization is the occurrence of "dose bridging." Dose bridging refers to radiation therapy that results in healthy tissue located between multiple targets receiving a higher dose than expected.

[0007] For example, dose bridging may occur in volumetric intensity-modulated arc therapy (VMAT) or intensity-modulated radiotherapy (IMRT).

[0008] In VMAT, radiation is delivered to the patient as the radiation source rotates around the patient in one or more arcs. An “arc” can be a “full” arc rotating 360º or a “partial” arc rotating less than 360º. During continuous rotation, the dose rate, beam shape, and gantry speed are dynamically adjusted so that the radiation is optimally aligned with one or more targets while avoiding exposure to surrounding healthy tissue (“organ at risk”, OAR).

[0009] In IMRT, radiation is delivered from multiple static angles. Radiation is delivered in discrete steps, where the radiotherapy machine rotates to different fixed angles and delivers radiation at each angle.

[0010] In VMAT and IMRT, the concept of a "field" refers to a set of optimizable control points, each associated with a set of beam geometry parameters. Beam geometry parameters can include treatment isocenter, bed position, or collimator angle. Another beam geometry can be a gantry angle (in IMRT) or a gantry start-stop angle range (in VMAT). In VMAT, a field comprises a single arc with an optimizable set of control points. In IMRT, a field refers to a single radiation beam directed at the patient from a specific angle or control point. Dose bridging may occur in VMAT or IMRT planning, where a field irradiates multiple physically separated targets, resulting in radiation being delivered to healthy tissue between the physically separated targets.

[0011] Existing methods exist aimed at reducing dose bridging. Some known techniques in VMAT therapy involve using a five-part VMAT arc at a predetermined location and a cost function that includes a dedicated normal tissue target specifically tailored to reduce dose bridging. The principle behind these existing techniques is that by providing more directions into which treatment doses can be directed, the planning optimizer can satisfy the targets of both the target and normal tissue, thereby effectively reducing dose bridging.

[0012] However, in existing VMAT and IMRT programs, the optimization process employs a method where all fields are designed to treat all targets equally. This leads to a reduction in suboptimal dose bridging. Summary of the Invention

[0013] According to a first aspect of the invention, a method is provided for generating and optimizing a radiotherapy plan for treating multiple targets using multiple fields.

[0014] According to a second aspect of the invention, a radiotherapy planning computer system according to the invention is provided.

[0015] According to a third aspect of the invention, a computer program product according to the invention is provided, comprising a non-transient computer-readable storage medium encoded with instructions operable to be executed by a processor.

[0016] Optional features are specified in the instruction manual.

[0017] A first aspect of the invention includes a method for generating and optimizing a radiotherapy plan for treating multiple targets using multiple fields, each field comprising an optimizable set of control points and each control point comprising a set of beam geometry parameters; wherein: the method includes: determining a plurality of field-target assignments of the plan by assigning each of the plurality of fields to a corresponding non-empty subset of the plurality of targets; at least one field being assigned to each target; and at least one field being assigned to a non-empty true subset of the plurality of targets.

[0018] In this disclosure, the term "field-target" assignment refers to the assignment of a field to a non-empty subset of multiple targets.

[0019] It is important to note that some fields in the field can be assigned to non-proper subsets of the target; that is, some fields can be assigned to all targets. However, at least one field in the field is assigned to a proper subset of the target.

[0020] In one example, multiple fields can be individually assigned to the corresponding proper subsets of the target. In another example, the majority of fields (e.g., >50%) can be individually assigned to the corresponding proper subsets of the target. As yet another example, the vast majority of fields (e.g., >90%) can be individually assigned to the corresponding proper subsets of the target.

[0021] A proper subset can include a subset that has any number of fewer elements than the full target set. In one example, a proper subset may include one fewer element than the full target set. In another example, a proper subset, as used by this method, may include significantly fewer elements (e.g., >30%) than the full target set. In yet another example, a proper subset, as used by this method, may include more than 50% fewer elements than the full target set. In still another example, a proper subset, as used by this method, may include more than 90% fewer elements than the full target set.

[0022] Advantageously, by allocating multiple fields (instead of all fields treating all targets as in the prior art) to a true subset of the target, there is greater flexibility regarding the tissue portion irradiated by each field, making it possible to reduce the irradiation incident on healthy tissue, thereby obtaining a total dose distribution with reduced dose bridging.

[0023] Field-target allocation can be followed by further (routine) dose-based optimization.

[0024] In some cases, the number of fields (e.g., VMAT arcs) can be predetermined before the treatment planning process. The number of fields can be optimized in a separate process prior to this planning process.

[0025] Optionally, each of the planned field-target assignments is non-exclusive, such that during the optimization of the plan, targets from unassigned fields can be planned for use in treatment through that field.

[0026] In other words, during optimization, if irradiating the target with a field that was never assigned to the target proves to be advantageous or more efficient, then the optimizer will allow it to do so. This could occur, for example, when treating a target with an unassigned field for the overall plan results in dose bridging or irradiation of healthy tissue that would lead to a reduction in dose delivery by all fields.

[0027] Determine the cost of field-target allocation

[0028] Optionally, the method may further include calculating the cost associated with each possible allocation of each field to a plurality of non-empty subsets of targets in a plurality of fields, wherein the plurality of field-target allocations of the scheme comprise a set of allocations having the minimum sum of associated costs.

[0029] The cost associated with allocating a field to a subset of targets can be viewed as a measure of the suitability of that field for treating that subset of targets.

[0030] Alternatively, the cost associated with assigning a field to a non-empty subset of a target can be calculated by one or more of the following:

[0031] (i) The cost of a plan generated by using a plan target to execute a plan optimizer while restricting the plan to a non-empty subset of the fields and targets;

[0032] (ii) Quantification of the overlap between the target and the organ at risk under allocation;

[0033] (iii) The number of collimator blades involved in the allocation;

[0034] (iv) The amount of healthy tissue exposed in the collimator apertures of all targets in the non-empty subset covering the target;

[0035] (v) The distance from the radiation source to a non-empty subset of the target; and

[0036] (vi) The sum of blade opening distances in the collimator apertures of the non-empty subset that conforms to the target.

[0037] The methods listed above for calculating the cost associated with the allocation of a non-empty subset of the field to the target are exemplary and not exhaustive, and variations and combinations of the methods (i) to (vi) listed above are within the scope of this disclosure.

[0038] The method (i) described above can involve conventional objective function-based optimization, where machine parameters are iteratively changed, the resulting dose distribution is calculated, and then the cost of a treatment plan as defined by the machine parameters is calculated based on a cost function of dose. Iterative modification of the machine parameters continues until the cost function based on dose is minimized. For example, the cost function can penalize deviations from a dose-based objective (such as a minimum dose to the target or a maximum dose to an organ at risk). The method (i) described above for determining the cost associated with a field-target allocation can include using a cost function to determine the cost of a treatment plan that includes only the field and target of that allocation.

[0039] The method described above (ii) takes into account the overlap between healthy tissue (OAR) and the target in the beam field of view, such as as seen through a multi-leaf collimator aperture. The greater the overlap between healthy tissue and the target, the higher the cost associated with the corresponding field-target assignment.

[0040] The method described above (iii) takes into account the range of movement required by the collimator blades to achieve a specific field-target allocation. Here, the greater the range of movement of the collimator blades, the greater the use of time and machine resources and the higher the cost of field-target allocation.

[0041] The method described above (iv) takes into account the amount of healthy tissue exposed due to the collimator aperture of the field attempting to irradiate the target to which the field is being assigned. The greater the amount of healthy tissue irradiated in this manner, the higher the cost of a particular field-target assignment.

[0042] The above method (v) takes into account that, under a specific field-target allocation, the greater the distance between the target subset and the radiation source, the higher the cost of the field-target allocation.

[0043] Method (vi) takes into account that the greater the distance the blade needs to move to cover the non-empty subset of the target, the higher the cost of the corresponding field assignment.

[0044] Beam geometry parameters

[0045] As mentioned above, the field comprises a set of control points, each characterized by beam geometry parameters.

[0046] Optionally, the beam geometry parameters include one or more of the following:

[0047] The position of the blades in the multi-leaf collimator;

[0048] Collimator angle;

[0049] Treatment centers, etc.

[0050] Bed location; and

[0051] Frame start-stop angle range.

[0052] Optionally, the beam geometry is optimized to be static during rack rotation; or: the beam geometry is optimized to be dynamic during rack rotation.

[0053] When determining the cost of a specific field-target allocation, the optimal set of beam geometry parameters for the field used to irradiate the subset of targets in that allocation is determined. Therefore, the cost is not determined based on arbitrary fields, but rather on the cost of fields with the optimal geometry for the subset of treatment targets.

[0054] In VMAT, the rack start-stop angle range is one of the beam geometry parameters. When using IMRT, the static rack angle is used instead of the rack start-stop angle as the beam geometry parameter.

[0055] In some implementations, one or more of the beam geometry parameters can be fixed, while others can be optimized or varied. For example, all beam geometry parameters except for the collimator aperture can be predetermined or fixed. Any combination of beam geometry parameters can be fixed, while the remaining beam geometry parameters can be optimized or varied.

[0056] For example, the collimator angle can be fixed or predetermined to avoid dose bridging. This can be achieved by selecting the collimator angle so that there are no gaps between multiple targets when viewed from the field of view of the beam and in the direction of movement of the collimator blades.

[0057] The extent to which dose bridging occurs in a given field depends on the collimator aperture; however, the remaining beam geometry parameters, such as bed position or gantry start-stop angle range (or any of the parameters listed above), can be selected optimally so that the collimator aperture can reduce dose bridging more effectively or most efficiently.

[0058] Binary matching

[0059] Optionally, field-target assignment includes minimum-cost binary matching.

[0060] Typically, minimum-cost bipartite matching divides the set into two subsets, where every element (or "vertices") from each subset is connected to every element in the other subset. The "cost" of each connection (or "edge") is determined, and then the set of edges that produce the minimum total cost of the matching is output.

[0061] therefore:

[0062] Bipartite matching uses the complete bipartite graph of vertices and edges;

[0063] The set of vertices comprises the union of (i) the set of fields and (ii) the set of all non-empty subsets of targets, such that each vertex represents a field or a subset of targets;

[0064] Each field vertex is connected to all target subset vertices via an edge representing one of the assignments;

[0065] Each edge has an edge cost associated with its corresponding assignment;

[0066] and

[0067] The method also includes finding the set of edges with the minimum total edge cost under the constraint that at least one field is assigned to each target.

[0068] Bipartite matching is a mathematical algorithm that finds the set of edges with the minimum total edge cost under the constraint that at least one field is assigned to each target. This problem can be solved in several ways. Some methods for solving the bipartite matching problem include: the Kuhn-Munkres algorithm; the continuous shortest path algorithm; the cost scaling algorithm; the primal-dual algorithm; the auction algorithm; and the minimum cost maximum flow algorithm.

[0069] Initial collimator blade aperture

[0070] Optionally, the optimization includes determining the initial collimator blade aperture of the field by adjusting a first set of collimator blades to shape the collimator aperture to cover the primary target of the assigned field; and / or further determining the initial collimator blade aperture of the field by adjusting a second set of collimator blades that do not intersect with the first set of collimator blades to shape the aperture to cover one or more secondary targets of the unassigned field.

[0071] Here, the second group of collimator blades, which "does not intersect" with the first group of collimator blades, means that no blade in the first group is a member of the second group.

[0072] Optionally, the method further includes determining the initial collimator blade aperture of the field by adjusting the first and / or second set of collimator blades so that the aperture covers one or more targets that have not yet been assigned a field but at least partially overlap with the primary and / or secondary targets in the beam field of view of the collimator.

[0073] In other words, when determining the initial collimator blade apertures, the blade positions are first adjusted to fit only the primary target (i.e., the target in the assigned field). Then, if possible, the remaining blades are adjusted to fit the secondary target. By using only the remaining blades for the secondary target, no gap is exposed between the primary and secondary targets. Conversely, if blades fitted around the primary target are moved to fit the secondary target, undesirable gaps may be exposed between the primary and secondary targets.

[0074] As described above, after the blades have been adapted around the primary and secondary targets, the blade position can be further adjusted to include any target tissue that may overlap with the primary or secondary target in the collimator's beam field of view. That is, the blade position is adjusted to include any target or target portion that can be adapted without causing dose bridging or irradiation of healthy tissue.

[0075] Treatment plan

[0076] Optionally, multiple fields correspond to multiple VMAT treatment arcs.

[0077] Typically, the examples provided in this disclosure relate to VMAT treatment. However, alternatively, multiple fields may correspond to multiple intensity-modulated radiotherapy (IMRT) fields or VMAT-IMRT hybrid fields. Other types of radiotherapy may also be used according to the principles of this disclosure, wherein the radiotherapy type uses multiple fields to irradiate multiple targets.

[0078] Optionally, the method further includes: dividing one or more VMAT treatment arcs into subfields; and assigning each subfield to a different subset or a different true subset of the target.

[0079] Advantageously, this provides greater flexibility in field allocation to the target by allowing subfields to be assigned to the target. This, in turn, allows for greater control over the final dose distribution in order to avoid dose bridging.

[0080] Optionally, the method also includes generating and optimizing a radiotherapy plan based on multiple field-target assignments determined for the plan.

[0081] For example, after the field-target allocation has been determined, a final radiotherapy plan can be generated using standard dose-based objective function optimization. Standard optimization may be subject to limitations imposed by the field-target allocation determined according to the principles of this disclosure.

[0082] Other aspects of the invention

[0083] A second aspect of the invention includes a radiotherapy planning computer system comprising a memory and a processor configured to perform the methods as defined above.

[0084] A third aspect of the invention includes a computer program product comprising a non-transient computer-readable storage medium encoded with instructions operable for execution by a processor to perform the methods as defined above.

[0085] According to a fourth aspect of the invention, an apparatus is provided for generating and optimizing a radiotherapy plan for treating multiple targets using multiple fields, each field comprising a set of optimizable control points and each control point comprising a set of beam geometry parameters, the apparatus comprising:

[0086] A device for determining planned field-target assignments by assigning each of the multiple fields to a corresponding non-empty subset of the multiple targets;

[0087] At least one field is assigned to each target; and

[0088] At least one of the fields is assigned to a non-empty true subset of multiple targets.

[0089] Devices for generating or optimizing treatment plans and devices for determining multiple field-target assignments may include the computer systems described below. Attached Figure Description

[0090] To better understand this disclosure, some embodiments will now be described by way of example with reference to the accompanying drawings.

[0091] Figure 1 A block diagram of an example computing system is shown, on which some of the embodiments described herein can be implemented.

[0092] Figure 2 A block diagram of selected components of a radiotherapy system is shown, on which some embodiments of the invention can be implemented.

[0093] Figure 3A Elements of an exemplary radiotherapy system are shown, on which a radiotherapy plan generated according to some embodiments of the invention can be implemented.

[0094] Figure 3B The beam field of view of the multi-leaf collimator is shown.

[0095] Figure 4 This illustrates dose distribution involving dose bridging and irradiation of healthy tissue between targets, achieved in the prior art.

[0096] Figures 5A to 5C Three consecutive (but not coherent) control points of an exemplary first treatment field or arc according to the prior art are shown.

[0097] Figures 5D to 5FThree consecutive (but not coherent) control points of an exemplary second treatment field or arc according to the prior art are shown.

[0098] Figure 6 The present invention illustrates a dose distribution that can be achieved according to the principles of this disclosure, which results in reduced dose bridging or reduced irradiation of healthy tissue between targets.

[0099] Figures 7A to 7C Three consecutive (but not coherent) control points of an exemplary first treatment field or arc according to the principles of this disclosure are shown.

[0100] Figures 7D to 7F Three consecutive (but not coherent) control points of an exemplary second treatment field or arc are shown in accordance with the principles of this disclosure.

[0101] Figure 8 The distribution of a field to a subset of multiple targets according to the principles of this disclosure is shown.

[0102] Figure 9 This is a flowchart for generating a radiotherapy plan using field-target allocation, based on the principles of this disclosure.

[0103] Figure 10 The field-to-target binary matching according to the principles of this disclosure is shown.

[0104] Figure 11 It is a flowchart used to determine the set of field-target assignments associated with minimum cost. Detailed Implementation

[0105] Reference will now be made in detail to several embodiments. While the subject matter will be described in conjunction with alternative embodiments, it will be understood that they are not intended to limit the claimed subject matter to these embodiments. Rather, the claimed subject matter is intended to cover alternatives, modifications, and equivalents, all of which may be included within the scope of the claimed subject matter as defined by the appended claims.

[0106] Furthermore, numerous specific details are set forth in the following detailed description to provide a thorough understanding of the claimed subject matter. However, those skilled in the art will recognize that embodiments may be practiced without these specific details or using their equivalents. In other instances, well-known methods, processes, components, and circuits have not been described in detail so as not to unnecessarily obscure aspects and characteristics of the subject matter.

[0107] The following description is intended to enable those skilled in the art to make and use embodiments of the invention; it is presented in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be apparent to those skilled in the art, and the general principles defined herein can be applied to other embodiments and applications without departing from the scope of this disclosure. Therefore, the invention is not limited to the embodiments shown, but should be accorded the widest scope consistent with the principles and features disclosed herein.

[0108] Computer System

[0109] Figure 1 A block diagram of an example computing system 100 on which embodiments described herein may be implemented is shown. In a basic configuration, system 100 includes at least one processing unit 102 and a memory 104. Figure 1 This most basic configuration is shown by dashed line 106. System 100 may also have additional optional features and / or functions. For example, system 100 may also include additional storage devices (removable and / or non-removable), including but not limited to solid-state, magnetic, or optical disks or tapes. Figure 1 Such additional storage devices are illustrated by removable storage device 108 and non-removable storage device 120. System 100 may also include communication connections(s)122, which allow devices to communicate with other devices, for example, in a network environment using logical connections to one or more remote computers.

[0110] System 100 also includes multiple input devices 124, such as keyboards, mice, pens, voice input devices, touch input devices, etc. It also includes multiple output devices 126, such as display devices, speakers, printers, etc.

[0111] exist Figure 1 In the example, memory 104 includes computer-readable instructions, data structures, program modules, etc. Depending on how it will be used, system 100 (by executing appropriate instructions, etc.) can be used to implement a planning system that generates a radiotherapy plan using the field-target allocation method disclosed herein.

[0112] Radiotherapy system

[0113] Figure 2 This is a block diagram on which selected components of a radiotherapy system 200 according to an embodiment of the present invention can be implemented. Figure 2In the example, system 200 includes an accelerator and a beam delivery system 204, which is operable to generate and / or accelerate a beam 201. The accelerator and beam delivery system can generate and deliver various types of beams, including, for example, X-ray (photon) beams. The operation and parameters of the accelerator and beam delivery system 204 are controlled such that the intensity, energy, size, and / or shape of the beam are dynamically adjusted or controlled during patient treatment according to an optimized radiotherapy plan generated by system 100 and stored within system 100 as discussed above.

[0114] Nozzle 206 is used to aim the beam at various locations within the patient's body supported on a patient support device 208 (e.g., a chair, bed, or table) within the treatment room (e.g., various locations of a target). For example, the target may be an organ, a portion of an organ (e.g., a volume or region within an organ), a tumor, diseased tissue, or the patient's outline.

[0115] Nozzle 206 can be mounted on the frame structure or can be part of the frame structure. Figure 3A The gantry structure can be movable relative to the patient support device 208, which can also be movable. The accelerator and beam delivery system 204 can be mounted on or part of the gantry structure; alternatively, the accelerator and beam delivery system can be separate from (but in communication with) the gantry structure.

[0116] Figure 2 The control system 210 receives and executes a prescribed treatment plan generated and / or optimized according to embodiments of the present invention. The control system 210 includes a computing system having a processor, memory, input devices (e.g., a keyboard), and an optional display. Figure 1 System 100 is an example of such a platform used for control system 210. Control system 210 can receive data about the operation of system 200. Control system 210 can control parameters of the accelerator and beam delivery system 204, nozzle 206, and patient support device 208 based on the data received by control system 210 and according to the radiotherapy plan. These parameters include parameters such as beam energy, intensity, size and / or shape, nozzle orientation, and the position of patient support device (and patient) relative to the nozzle.

[0117] Figure 3A Exemplary elements of a radiotherapy system 300 for treating patient 304 are shown. System 300 is Figure 2 An example of how the radiotherapy system 200 is implemented. The gantry 302 can rotate about an axis; for example, the gantry can rotate about the patient on the bed. The patient can be moved by moving the bed. Although in Figure 3AIn the example, patient 304 is supine, but the invention is not limited thereto. Alternatively, patient 304 may be seated in a chair or positioned in any orientation. The gantry 302 can be controlled by a treatment system using an optimized treatment plan generated according to embodiments of the invention.

[0118] Figure 3B A multi-leaf collimator (MLC) 350 is shown as part of a radiotherapy system 300. The MLC is used to shape a radiation beam into, for example, a three-dimensional shape conforming to one or more targets to be treated. The MLC includes multiple movable blades 352, which are typically made of a high-density material such as tungsten and configured to block radiation. The MLC blades 352 are arranged in two opposing rows, wherein each blade can move independently relative to the other blades. These two rows of blades can form an “aperture” 356 through which radiation can pass. By adjusting the position of each blade, the aperture 356 formed by the blades can shape the beam into a shape 354 or contour conforming to the target in the beam's field of view.

[0119] Dose bridging and collimator orifice

[0120] Figure 4 The dose distribution dd achieved by existing technology is shown. pa In this example, there are eight different tumors or targets, namely targets T1-T8, located in different parts of the patient's body. It can be seen that some targets are close to each other, while others are at a certain distance from each other. Here, targets T1, T7, and T8 are located close to each other in the first group, targets T3 and T5 are located close to each other in the second group, and targets T2, T4, and T6 are located close to each other in the third group. However, the first, second, and third groups of targets are located at a certain distance from each other. A radiation beam attempting to simultaneously irradiate all three groups of targets would typically also irradiate healthy tissue found between these target groups. This is due to dd pa As shown, healthy tissue (shown by shaded areas) was also irradiated while radiation was being delivered to targets T1-T8 simultaneously.

[0121] A key part of the optimization plan involves optimizing the position of the MLC blades at each control point. Specifically, from the beam field of view, the MLC blades should form apertures 356 that conform as closely as possible to the target profile. Iterative optimization algorithms are typically used to further refine the blade apertures within the target profile to best meet clinical objectives. This involves determining the initial blade apertures 356 to use at the start of optimization for each control point.

[0122] The existing strategy for determining the initial blade aperture is to position the blades such that the radiation beam conformally fits all targets (with some margin and / or correction factor). However, in the presence of multiple spatially separated targets, attempting to conformally fit the beam to all targets by selecting appropriate apertures may be disadvantageous, as this could lead the optimizer to provide technical solutions involving more dose bridging.

[0123] Figures 5A to 5F Two targets, 502 and 504, are shown from the perspective of the radiation beam. Figures 5A to 5C The first field (or arc) is shown, while Figures 5D to 5F The second session (or arc) is shown, and the first and second sessions are related to the same treatment plan.

[0124] Figures 5A to 5C Two targets, 502 and 504, are shown from the perspective of the radiation beam (i.e., the "beam field of view") at three consecutive (but not coherent) control points in the first field or arc of a treatment plan, according to the prior art. Figures 5A to 5C Also shown is a blade 352 for a multi-leaf collimator (MLC) used to shape the beam.

[0125] Figure 5A The initial blade aperture 550 is shown, which is likely a typical feature of the prior art. It can be seen that the aperture 550 formed by the MLC blade 352 will allow the radiation beam to irradiate two targets 502 and 504 in a single (initial) control point 500. However, since the two targets are spatially separated, the radiation beam also irradiates healthy tissue located between targets 502 and 504. This occurs because the optimizer of the prior art generates a plan attempting to treat two spatially separated targets from a single control point.

[0126] Figure 5B and Figure 5C Examples of consecutive (but not coherent) control points 500' and 500'' following an initial control point 500 as determined by a prior art planning optimizer are shown. It can be seen that significant dose bridging and irradiation of healthy tissue occurs at control point 500' because orifice 552 covers the gap between targets 502 and 504. At control point 500'', no dose bridging is observed, illustrating that prior art planning optimizers can sometimes generate control points that do not lead to dose bridging (this disclosure attempts to generate control points that do not lead to dose bridging more consistently).

[0127] Figures 5D to 5F The diagram shows three consecutive (but not coherent) control points for the second field or arc of the same treatment plan according to existing technology. Figures 5D to 5F Also shown is the blade 352 of the MLC that enables beamforming. Figures 5D to 5F and Figures 5A to 5C Similarly, and as can be seen, the first two control points shown, 590 and 590', both result in significant dose bridging and irradiation of healthy tissue due to orifices 560 and 562. This occurs because the planning optimizer attempts to treat two spatially separated targets 502 and 504 at the same control point and under the same MLC orifice. Figure 5C Like in the middle, Figure 5F The example illustrates an MLC orifice that does not produce any dose bridging or irradiation of healthy tissue. This demonstrates that some control points in the prior art do achieve reduced dose bridging, but the purpose of this disclosure is to reduce dose bridging more consistently and effectively.

[0128] Figure 6 An exemplary dose distribution is shown to avoid dose bridging or irradiation of healthy tissue. Figure 6 Eight exemplary targets T1-T8 that require treatment according to a treatment plan are shown. It can be seen that there are three groups of targets, including a first group of targets T1, T7, T8, a second group of targets T3, T5, and a third group of targets T2, T4, T6. Each group of targets includes targets that are spatially close to or spatially continuous with each other. Therefore, each group of targets can be treated by aligning the radiation beam with a single collimator aperture of the corresponding target group. However, if treatment is performed through a single collimator aperture, multiple groups of targets would result in the irradiation of healthy tissue between multiple groups of targets. This would be the case in the prior art.

[0129] Through this disclosure, it is clarified that not all fields attempt to treat all targets. This means that some fields or arcs treat only a proper subset of the targets. As will be explained in more detail below, this will result in a reduction in dose bridging, such as... Figure 6 As shown. In Figure 6 In the example, there is a reduced dose delivered to healthy tissue between the first group of targets (T1, T7, T8), the second group of targets (T3, T5), and the third group of targets (T2, T4, T6). Reduced dose bridging is one of the objectives of this disclosure.

[0130] Figures 7A to 7F Two targets, 702 and 704, are shown from the perspective of the radiation beam. Figures 7A to 7C The first field (or arc) is shown, while Figures 7D to 7F The second session (or arc) is shown; both the first and second sessions are related to the same treatment plan.

[0131] Figures 7A to 7C The two targets, 702 and 704, are shown from the perspective of the radiating beam (i.e., the "beam field of view"), similar to... Figures 5A to 5C .However, Figures 5A to 5C The blade aperture implemented using existing technology is shown, while Figures 7A to 7CAn exemplary blade aperture implemented by this disclosure is shown. Three consecutive (but not coherent) control points for a single arc are shown. Figures 7A to 7C Also shown is a blade 352 for the MLC used to shape the beam.

[0132] As can be seen, in Figures 7A to 7C In the example, dose bridging was avoided because no healthy tissue was irradiated between targets 702 and 704. This was achieved by assigning each field to a true subset of the targets.

[0133] The targets to which the field is assigned are called "primary targets" and include targets that the field must illuminate. Fields can also be assigned to other "secondary" targets, which may, but are not required to, be illuminated by the corresponding field. Figures 7A to 7C In the example, the field or arc is assigned to the first target 704. Therefore, target 704 is Figures 7A to 7C The field or arc's "primary target," and Figures 7A to 7C The target 704 must be treated during the field or arc period.

[0134] In this example, target 702 is a "minor" target, and target 702 can... Figures 7A to 7C Treatment may be administered during the field or arc period, but it is not mandatory to have it performed by [the relevant authority / organization]. Figures 7A to 7C Treatment is performed on the field or arc. Figures 7A to 7C In an exemplary field, the field is assigned to target 704 as the primary target, but to target 702 as the secondary target.

[0135] At the (initial) control point 700, it can be seen that the initial collimator aperture treats only the primary target 704, but not the secondary target 702. That is, the collimator blades are positioned to allow irradiation of the primary target 704, but block irradiation of the secondary target 702. This is because, according to the principles of this disclosure, the optimizer finds that it is impossible to irradiate both targets 702 and 704 from control point 700, nor to irradiate the healthy tissue between targets 702 and 704.

[0136] At the next exemplary control point 700', the collimator aperture is configured such that the radiation beam treats a portion of the primary target 704 and the secondary target 702. Thus, while the purpose of this disclosure is generally to treat only the target to which the field has been assigned, the field may also be assigned to the secondary target that is irradiated along with the primary target, if the position of the secondary target relative to the primary target allows for this.

[0137] This is Figure 7CAs shown, both the primary target 704 and the secondary target 702 are irradiated from the same control point. Here, due to the relative positions of the primary target 704 and the secondary target 702, both the primary and secondary targets can be irradiated from the same control point without any dose bridging or irradiation of healthy tissue.

[0138] Figures 7D to 7F It shows the relationship with Figures 7A to 7C The first field or arc shown represents three consecutive (but not coherent) control points related to the same treatment plan as the second field or arc. In this example, the field is assigned to target 702 as the primary target, not target 704 as the primary target. The field is assigned to target 704 as the secondary target.

[0139] At control point 790, only the primary target 702 is treated. This is because, according to the principles of this disclosure, the planning optimizer finds that it is impossible to simultaneously avoid irradiating healthy tissue located between the primary target 702 and the secondary target 704 from the same control point.

[0140] At control point 790', the primary target 702 is treated, and a portion of the secondary target 704 is also treated. This is because, according to the principles of this disclosure, the planning optimizer finds that the primary target 702 and a portion of the secondary target 704 can be treated from the same control point without dose bridging.

[0141] At control point 790'', both primary target 702 and secondary target 704 are treated. This is because, according to the principles of this disclosure, the planning optimizer finds that both primary target 702 and secondary target 704 can be treated from the same control point without dose bridging.

[0142] Field-target allocation

[0143] As noted above, in the prior art, a planning optimizer typically attempts to create a radiotherapy plan in which all fields are designed to treat all targets equally. In the prior art, when spatially separated targets exist and the plan is developed such that all fields attempt to treat all targets, the resulting plan leads to dose bridging. That is, when attempting to irradiate spatially separated targets from the same control point, healthy tissue between the targets also receives (harmful) radiation.

[0144] In contrast, this disclosure proposes a "multi-field-multi-target" technique aimed at generating better plans for spatially separated targets. That is, each field is assigned to a subset of the targets. At least one field is assigned to a proper subset of the targets.

[0145] In one example, multiple fields can be assigned to a proper subset of the target. In another example, many fields (e.g., >30%) can be assigned to a proper subset of the target. In yet another example, the majority of fields (e.g., >50%) can be assigned to a proper subset of the target. In yet another example, the vast majority of fields (e.g., >90%) can be assigned to a proper subset of the target.

[0146] In one example, a proper subset of the target may include one fewer element than the complete target set. In another example, a proper subset of the target may include several fewer elements than the complete target set. In yet another example, a proper subset of the target may include significantly fewer elements than the complete target set (e.g., >30%). In yet another example, a proper subset of the target may include 50% fewer elements than the complete target set. In yet another example, a proper subset of the target may include 90% fewer elements than the complete target set.

[0147] As noted above, a "field" refers to a set of optimizable control points, each associated with a set of beam geometry parameters. By assigning each field to a subset of targets, each field can seek the most favorable beam geometry, thereby avoiding dose bridging resulting from treating each corresponding subset of targets through the respective field. In other words, the field-target allocation of this disclosure allows the final dose distribution provided by all fields to conform to the shape of multiple targets by allowing each field to treat a different subset of targets, while avoiding healthy tissue between multiple targets. This is impossible in scenarios where the optimizer attempts to provide a plan where all fields cover all targets. Therefore, this technique is relevant to any plan with at least two spatially separated targets and at least two fields or arcs.

[0148] For ease of explanation, consider a treatment, denoted as... F={f 1 ,f 2 ,…,f n } n fields or arcs, and m targets T={t 1 ,t 2 ,…,t m } Where n≥2 and m≥2. According to this disclosure, the treatment planning optimizer allocates between fields and targets such that each field treats only the target it is assigned to. The targets of the assigned fields can be "primary" or "secondary," where a "primary" target is a target that the field must irradiate, while a secondary target is a target that the field can irradiate, provided that this does not result in, for example, excessive dose bridging.

[0149] According to the principles of this disclosure, field-target allocation is non-exclusive, meaning that if the field... f i Assigned for treatment of primary targets t j The plan optimizer can still generate plans, where if such a plan is excellent in achieving the plan's objectives, then the field... f i Treatment targets other than the primary target (e.g., the program may irradiate secondary or other targets).

[0150] Figure 8 The field-target assignment of this disclosure is shown. Figure 8 This relates to VMAT treatment, and multiple fields, i.e., field 1 to field n, are shown on the left. As explained above, in VMAT, the concept of a "field" refers to a set of optimizable control points, each associated with a set of beam geometry parameters, such as treatment isocenter, bed position, collimator angle, and gantry start / stop angle ranges. In VMAT, a field comprises a single arc with the set of optimizable control points. Therefore, in Figure 8 In the diagram, each partial circle with an arrow at one end represents a field comprising a single VMAT arc. This can be a full arc with a rack start / stop angle range of 360º, or a partial arc with a rack start / stop angle range of less than 360º.

[0151] Figure 8 Multiple target subsets, namely subset 1 to subset n, are shown on the right. Each target is represented by a circle. Targets assigned to a field are represented by shaded circles, while targets with corresponding unassigned fields are represented by hollow circles. Target subset 1 has five shaded circles and four hollow circles. Field 1 is assigned to target subset 1; that is, in this example, field 1 is assigned to a target subset containing five targets. Target subset 2 has three shaded circles and six hollow circles. Field 2 is assigned to target subset 2; that is, in this example, field 2 is assigned to a target subset containing three targets. Field n is assigned to target subset n; that is, in this example, field n is assigned to a target subset containing four targets.

[0152] As shown in the figure, each field is assigned to a subset of targets. At least one of these subsets is a "true" subset, because a true subset cannot include all targets and must be a subset smaller than the complete set of targets. By assigning each field to a subset of targets (where at least one field is assigned to a true subset of the targets), the resulting dose distribution can be controlled, thereby avoiding dose bridging.

[0153] This method can also use subfields. That is, in some cases of VMAT treatment, the arc field can be divided into subfields before the target allocation step. For example, the first half of the arc (the first subfield) can be allocated to the first group of targets, while the second half of the arc field (the second subfield) can be allocated to another group of targets.

[0154] Figure 9 This is a flowchart 900 related to field-target allocation according to an embodiment of the present invention.

[0155] In step 902, information relating to the number of fields used in radiotherapy is received. This number can be predefined, input by a human user, or determined by a planning optimizer. In this method, there are at least two fields in the treatment plan; in VMAT, this means there are at least two arcs in the treatment plan.

[0156] In step 904, information related to a set of multiple target volumes is received. That is, this method is applicable to scenarios where there are at least two target volumes. The received information may include the location and number of target volumes.

[0157] In step 906, the cost of the hypothetical assignment from each field to each subset of the target set is calculated. For example, if there are n fields, there are n field-target assignments. Furthermore, if there are m targets, there are 2 m There are possible target subsets. The constraint is that each field must be assigned to at least one target, and the empty set is excluded from the possible target subsets. Therefore, when excluding the empty set, there exist 2 m -1 possible target subsets. In n fields and 2 m In the case of -1 target subset, there exists a range from n fields to 2 m -1 target subset n×(2 m -1) possible allocations, and each possible allocation has an associated cost. The method for calculating the cost of each hypothetical allocation will be described below.

[0158] In step 908, under the constraint that each target is subject to at least one field therapy, a set of field-target assignments with the minimum total cost is identified.

[0159] In step 910, a radiotherapy plan is generated and optimized using the determined set of field-target assignments.

[0160] Advantageously, this method assigns each field or arc to a subset of multiple targets, such that the total dose distribution obtained through multiple fields or arcs avoids dose bridging by positioning the total dose distribution at spatially continuous targets rather than at healthy tissue between spatially separated targets.

[0161] Calculate the cost of a single field-target assignment

[0162] General Field f i Assigned to target The cost of a set can be mathematically written as Here, C can be referred to as the field used for application. f i Treat k targets The cost of geometric fit is a metric. This cost can be calculated in several ways, and some examples include:

[0163] i) The same objective as the original plan can be used, but only the field can be used. f i and target To execute a standard program optimizer. For example, a standard program optimizer can use an objective function with conflicting cost terms, where the total cost of the objective function should be minimized.

[0164] If the objective function is used to calculate only the field f i and target The cost of the optimization plan can be viewed as the cost of field-target allocation. .

[0165] ii) When using the field f i Therapeutic targets The geometry of time can be quantized in f i The overlap between the target and the organ at risk, as observed from the beam field of view at each control point, is used for assessment. These quantities can be combined (e.g., summed) to obtain the cost of field-target allocation. .

[0166] iii) For f i Each control point can be identified with a quantity associated with the number of blades participating in achieving the set of conformal orifices targeting the target. This quantity across all control points can be combined to obtain the cost of field-target allocation. .

[0167] iv) In f i At each control point in the data, a target can be defined. A conformal leaf opening is generated, and a measure of the amount of healthy tissue exposed within the opening that does not overlap with any target can be quantified. This measure across all control points can be combined (e.g., summed), and the combined measure can be viewed as the cost of field-target allocation. .

[0168] v) in f i At each control point, the distance from the target to the radiation source can be determined. These distances can be combined (e.g., summed) across control points to determine the cost of field-target allocation. .

[0169] vi) in f i At each control point in the data, a target can be defined. The system generates conformal blade openings and quantifies the distance the blades need to travel to achieve these openings. These metrics across all control points can be combined (e.g., summed), and the combined metrics can be viewed as the cost of field-target allocation. .

[0170] The cost or suitability of assigning any particular field to any particular subset of targets can be determined using any of the methods described above, and this can be used to determine the optimal set of field-target assignments as discussed below.

[0171] Describe the assignment using set theory.

[0172] The cost associated with each possible hypothesis assignment from each field to each possible subset of targets can be expressed by the following formula:

[0173]

[0174] Formula 1

[0175]

[0176] Formula 2

[0177] Here:

[0178] C = Cost

[0179] F = the set of fields

[0180] x =product

[0181] T = Target set

[0182] The power set of P(T)=T, that is, the set of all target subsets.

[0183] += nonempty set

[0184] \0 = Exclude empty set

[0185] P + (T) = the power set of all non-empty subsets of the target.

[0186] =Set of real numbers

[0187] The concept of the set of all subsets is called the power set and is denoted as P(T) according to the notation used in traditional set theory. The set of all subsets includes all possible combinations of subsets (including the empty set). For example, the set of 3 targets E1, E2, and E3 will have a power set that includes subsets of the following target sets (the empty subset is not listed): {E1}; {E2}; {E3}; {E1, E2}; {E1, E3}; {E2, E3}; {E1, E2, E3}.

[0188] According to traditional set theory, a nonempty set is represented by the symbol +. Therefore, the formula... In mathematics, the power set of the set of targets T that have no empty subsets is described (see Equation 2).

[0189] Each field to the power set P + Each assignment of each element in (T) can be mathematically written as two sets F and P. + The product of (T), i.e. Each field to the power set P + Each allocation of each element in (T) has an associated cost, which is represented by C in Formula 1.

[0190] Bipartite graph

[0191] The problem of finding the optimal set of assignments from each field to the corresponding target subset can be formulated as a combinatorial problem in a bipartite graph. Figure 10 The field f1-f is shown n The set of and the non-empty subsets of the target s1-s r A bipartite graph between them. As can be seen, in Figure 10 In a bipartite graph, each field f i Connect to all possible target subsets s r Each of these connections has an associated cost—for example, a connection that leads to excessive dose bridging can have a higher cost. By considering the cost of each possible allocation, a final set of allocations is determined such that the overall cost is minimized.

[0192] The above steps can be described using minimum-cost bisection matching, such as in... Figure 11 Chinese combination Figure 10 As shown. First, construct the complete bipartite graph 1102. The complete bipartite graph can be written as:

[0193] G={ V , E}

[0194] Formula 3

[0195] In other words, a bipartite graph G consists of vertices V and edges E, with each edge connecting two vertices.

[0196] Then:

[0197]

[0198] Formula 4

[0199] In other words, the set of all vertices V includes (i) the set of all fields F (i.e., f1-f). n (ii) All non-empty subsets P of the target set T + The set (i.e., s1-s) r The union of ), see Figure 11 Step 1104.

[0200] Then:

[0201]

[0202] Formula 5

[0203] Here:

[0204] u is a field and an element of the set of fields F.

[0205] v is a subset of the target set and an element of the set of all non-empty subsets of the target set T.

[0206] (u, v) is an edge between field u and target subset v, and is an element of the set of edges.

[0207] This means that nodes u and v are connected by an edge if and only if u represents a field and v represents a subset of targets.

[0208] In other words, the field u assigned to the target subset v is an element of the set of edges E. This is in Figure 11 Step 1106 is shown.

[0209] So:

[0210]

[0211] Formula 6

[0212] Therefore, each edge e (which represents the allocation from field u to the target subset v) has an associated cost. This is in Figure 11 Step 1108 is shown.

[0213] Then, the optimization process attempts to find the set of edges:

[0214]

[0215] Formula 7

[0216] Make:

[0217]

[0218] Make:

[0219]

[0220] Minimized.

[0221] In the above formula:

[0222] e = edge

[0223] n = number of edges / number of fields

[0224] s = a subset of all target sets

[0225] U n k=1 s k = The union of all subsets used in the mapping

[0226] T = the set of all targets

[0227] k = mapping index

[0228] C(f k , s k = The cost of allocating field k to a subset of target k

[0229] In other words, when field k is assigned to a subset of target k, this is represented as edge k and has an associated cost C(f). k , s k Then, the sum of the costs of all such allocations needs to be minimized. This is in Figure 11 Step 1110 is shown. In other words, the problem will be simplified to finding the allocation from each field to the corresponding subset of targets such that all targets are treated by at least one field, and the total cost is minimized. This can be viewed as a variation of the concept of “minimum-cost bipartite matching,” which can be solved in polynomial time using algorithms such as the Kuhn-Munkres algorithm, the continuous shortest path algorithm, the cost scaling algorithm, the primal dual algorithm, the auction algorithm, and the minimum-cost maximum flow algorithm.

[0230] Non-exclusivity

[0231] The method described above can be used to assign each field to a subset of targets, where the subset of targets to which a field is assigned can be referred to as the "primary" target. However, during optimization, the field is not necessarily limited to irradiating the primary target, but can also irradiate other targets referred to as "secondary" targets. This can occur when it is possible to irradiate the primary target and one or more secondary targets, while still avoiding or reducing dose bridging between the primary and secondary targets.

[0232] Similarly, the program can be optimized so that the field can irradiate a target that is neither a primary nor a secondary target, but the target can still be irradiated without causing dose bridging.

[0233] Beam geometry assigned to each field-target

[0234] Before calculating the cost of field-target allocation, the beam geometry of the corresponding field is optimized, as if the field only treats its primary target. For example, the isocenter, bed angle, gantry start / stop range, or collimator angle of the field can be optimized for the primary target of the allocated field. Then, the cost of field-target allocation is calculated based on the optimal beam geometry for a specific field and target subset. In the binary matching described above, the cost of each field-target allocation calculated under the optimal beam geometry can be associated with each edge between each field and the target subset of the allocated field.

[0235] When determining the optimal beam geometry for each field-target assignment, beam geometry parameters including collimator angle, bed angle, and isocenter can be optimized such that they have the same value for a given arc, or alternatively, that their values ​​vary as the arc progresses.

[0236] Once the problem of matching each field to a subset of targets has been solved, the optimal beam geometry used to calculate the cost of each field-target allocation can be used to treat the patient. Alternatively, after the optimal (minimum cost) field-target allocation has been determined, conventional dose-based optimization can be performed to generate a final radiotherapy plan. In this case, the dose-based optimization will be limited by the field-target allocation generated according to the principles of this disclosure.

[0237] Determine the initial collimator aperture

[0238] The collimator aperture for a specific field or arc can be initialized using the following steps. First, the position of a first set of collimator blades is adjusted so that the collimator aperture shapes to cover only the primary target of the assigned field. Then, the position of a second set of collimator blades, which does not intersect with the first set of collimator blades, is adjusted so that the aperture shapes to cover secondary targets that have not yet been assigned a field. Finally, the positions of the first and / or second sets of collimator blades can be further adjusted to cover one or more targets that have not yet been assigned a field but at least partially overlap with the primary and / or secondary targets. This method allows the blade positions to expand to cover as many primary, secondary, and other targets as possible, but reduces the exposed area between targets.

[0239] Therefore, embodiments of the invention have been described. Although the invention has been described in particular embodiments, it should be understood that the invention should not be construed as limited to such embodiments, but rather as interpreted in accordance with the appended claims.

Claims

1. A method for generating and optimizing a radiotherapy plan for treating multiple targets using multiple fields, each field comprising an optimizable set of control points and each control point comprising a set of beam geometry parameters; in: The method includes: determining the planned field-target allocation by assigning each of the plurality of fields to a corresponding non-empty subset of the plurality of targets; At least one field is assigned to each target; and At least one field is assigned to a non-empty true subset of the plurality of targets.

2. The method of claim 1, further comprising calculating a cost associated with each possible allocation of each of the plurality of fields to a plurality of non-empty subsets of the plurality of targets, and wherein, The planned multiple field-target allocations comprise a set of allocations with the minimum total associated costs.

3. The method according to claim 1 or claim 2, wherein, The multiple field-target assignments in the plan are non-exclusive, such that during the optimization of the plan, targets in unassigned fields can be planned for use in treatment through the fields.

4. The method of claim 2, wherein, The cost associated with the assignment of a non-empty subset of the field to the target can be calculated using one or more of the following: (i) The cost of a plan generated by using a plan target to execute a plan optimizer while restricting the plan to a non-empty subset of the field and the target; (ii) Quantification of the overlap between the target and the organ at risk under the aforementioned allocation; (iii) The number of collimator blades involved in the allocation; (iv) The amount of healthy tissue exposed in the collimator apertures of all targets in the non-empty subset covering the target; (v) The distance from the radiation source to a non-empty subset of the target; as well as (vi) The distance from the non-empty subset of the target to the center of the isocentric plane on an axis parallel to the blade direction.

5. The method of any preceding claim, wherein, The beam geometry parameters include one or more of the following: The position of the blades in the multi-leaf collimator; Collimator angle; Treatment centers, etc. Bed location; and Rack start / stop angle range.

6. The method of any preceding claim, wherein, The beam geometry is optimized to be static during rack rotation; or the beam geometry is optimized to be dynamic during rack rotation.

7. The method of any preceding claim, wherein, The allocation includes minimum-cost binary matching.

8. The method according to claim 7, wherein: The binary matching uses a complete bipartite graph of vertices and edges; The set of vertices comprises the union of (i) the set of fields and (ii) the set of all non-empty subsets of targets, such that each vertex represents a field or a subset of targets; Each field vertex is connected to all target subset vertices via an edge representing one of the assignments; Each edge has an edge cost associated with its corresponding assignment; and The method also includes finding the set of edges with the minimum total edge cost under the constraint that at least one field is assigned to each target.

9. The method of any preceding claim, wherein, The optimization includes determining the initial collimator blade aperture of the field through the following operations: Adjust the first set of collimator blades to shape the collimator apertures into a primary target covering the assigned field; And / or: The initial collimator blade aperture of the field is also determined by the following operation: Adjust the second set of collimator blades, which do not intersect with the first set of collimator blades, so that the aperture is shaped to cover one or more secondary targets that have not yet been assigned the field.

10. The method of claim 9, further comprising determining the initial collimator blade aperture of the field by: Adjust the first set of collimator blades and / or the second set of collimator blades so that the aperture covers one or more targets that have not yet been assigned the field but at least partially overlap with the primary target and / or the secondary target.

11. The method of any preceding claim, wherein, The multiple fields correspond to multiple VMAT treatment arcs.

12. The method of claim 11, further comprising: Divide one or more VMAT treatment arcs into subfields; as well as Each subfield is assigned to a different subset of targets.

13. The method according to any of the preceding claims further comprises generating and optimizing a radiotherapy plan based on the plurality of field-target assignments determined for the plan.

14. A radiotherapy planning computer system, comprising a memory and a processor configured to perform the method according to any one of claims 1 to 13.

15. A computer program product comprising a non-transient computer-readable storage medium encoded with instructions executable by a processor to perform the method according to any one of claims 1 to 13.

Images