Radiotherapy planning optimization method and apparatus

Abstract

Embodiments of the present disclosure relate to radiation therapy plan optimization methods and apparatus. To facilitate optimizing a radiation therapy plan for a particular patient using a particular radiation therapy device, by one method, a control circuit iteratively accesses a fluence space optimization solution for the radiation therapy plan for the particular patient, generates control point parameters for the particular radiation therapy device from the fluence space optimization solution, determines a treatment delivery time using the control point parameters for the particular radiation therapy device, and modifies the fluence space optimization solution when the treatment delivery time is suboptimal.

Description

Technical Field

[0001] These teachings as a whole involve the planned target volume of energy therapy for patients according to an energy-based treatment plan, and more specifically involve optimizing an energy-based treatment plan. Background Technology

[0002] The use of energy to treat medical conditions encompasses areas known in the prior art. For example, radiation therapy is a crucial component of many treatment programs aimed at reducing or eliminating unwanted tumors. Unfortunately, applied energy cannot inherently distinguish between unwanted material and adjacent tissues, organs, etc., that are expected to be preserved or even are essential for the patient's continued survival. Therefore, energy, such as radiation, is typically applied in a carefully managed manner to at least attempt to confine the energy to a given target volume. So-called radiation therapy programs are commonly used for the aforementioned purposes.

[0003] Radiation therapy plans typically include specified values ​​for each of the various treatment platform parameters during each of multiple consecutive radiation fields. Treatment plans for radiation therapy courses are often automatically generated through a process known as optimization. As used herein, "optimization" should be understood as improving candidate treatment plans without ensuring that the optimized result is actually the only optimal solution. This optimization typically involves automatically adjusting one or more physical therapy parameters (usually while observing one or more corresponding limitations in these aspects) and mathematically calculating possible corresponding treatment outcomes (such as dose levels) to identify a given set of treatment parameters that represents a good trade-off between the desired treatment outcome and the avoidance of undesirable side effects.

[0004] Radiation therapy planning optimization typically involves addressing two sub-problems. First, fluence space optimization focuses on finding the dosimetrically optimal radiation distribution geometry. Second, control point optimization focuses on finding an efficient way to deliver this distribution (or a near-optimal distribution) using the intended treatment machine. Summary of the Invention

[0005] In one or more embodiments, a method for optimizing a radiotherapy plan for a specific patient using a specific radiotherapy device is disclosed, the method comprising: by a control circuit: iteratively: accessing a fluence space optimization solution for the radiotherapy plan for the specific patient; generating control point parameters for the specific radiotherapy device based on the fluence space optimization solution; determining a treatment delivery time using the control point parameters for the specific radiotherapy device; and modifying the fluence space optimization solution when the treatment delivery time is suboptimal.

[0006] In one or more embodiments, an apparatus for optimizing a radiotherapy plan for a specific patient using a specific radiotherapy device is disclosed. The apparatus includes: control circuitry configured to iteratively: access a fluence space optimization solution for the radiotherapy plan for the specific patient; generate control point parameters for the specific radiotherapy device based on the fluence space optimization solution; determine a treatment delivery time using the control point parameters for the specific radiotherapy device; and modify the fluence space optimization solution when the treatment delivery time is suboptimal.

[0007] In one or more embodiments, a nontransitory computer-readable medium is disclosed for optimizing a radiotherapy plan for a specific patient using a specific radiotherapy apparatus. The nontransitory computer-readable medium includes instructions stored on the nontransitory computer-readable medium that, when executed on a processor, perform the following steps: iteratively: accessing a fluence-space optimized solution for the radiotherapy plan for the specific patient; generating control point parameters for the specific radiotherapy apparatus based on the fluence-space optimized solution; determining a treatment delivery time using the control point parameters for the specific radiotherapy apparatus; and modifying the fluence-space optimized solution when the treatment delivery time is suboptimal. Attached Figure Description

[0008] By providing a radiotherapy planning optimization method and apparatus as described in the following specific embodiments, various needs are at least partially met, particularly when studied in conjunction with the accompanying drawings, wherein:

[0009] Figure 1 Including block diagrams configured according to various embodiments of these teachings;

[0010] Figure 2 Includes flowcharts of various embodiments configured according to these teachings;

[0011] Figure 3 Including flowcharts of various embodiments configured according to these teachings; and

[0012] Figure 4 This includes schematic diagrams of various embodiments configured according to these teachings.

[0013] The elements in the figures are shown for simplicity and clarity and are not necessarily drawn to scale. For example, the size and / or relative positioning of some elements in the figures may be exaggerated relative to other elements to aid in understanding the various embodiments of this teaching. Furthermore, common but well-known elements that are useful or necessary in commercially viable embodiments are generally not depicted to facilitate less obstructive observation of the various embodiments of this teaching. Certain actions and / or steps may be described or depicted in a particular order, and those skilled in the art will understand that such specificity regarding the order is not actually necessary. The terms and expressions used herein have the ordinary technical meaning as attributed to those skilled in the art, unless otherwise stated herein with a different specific meaning. Unless otherwise specifically stated, the word "or" as used herein should be interpreted as having a disjunctive structure rather than a conjunctive structure. Detailed Implementation

[0014] Generally, these various embodiments can provide a method for optimizing a radiotherapy plan for a specific patient using a specific radiotherapy device. One method involves control circuitry iteratively accessing a fluence-space optimized solution for the patient-specific radiotherapy plan, generating control point parameters for the specific radiotherapy device based on the fluence-space optimized solution, determining the treatment delivery time using the control point parameters for the specific radiotherapy device, and modifying the fluence-space optimized solution when the treatment delivery time is suboptimal.

[0015] The generation of the aforementioned control point parameters may include, through a method, generating a lossless transformation from input fluence information to control point parameters, enabling a particular radiotherapy apparatus to (at least theoretically) deliver a fluence at least nearly equivalent to the fluence space optimization solution. The generation of control point parameters may initially be performed without any limitation on the number of automatically assigned control points. Alternatively, or in combination with the foregoing, the generation of control point parameters may initially be performed without any limitation on the time required for a particular radiotherapy apparatus to complete the input fluence information.

[0016] By means of a method, determining treatment delivery time using control point parameters for a specific radiotherapy device may include determining the time requirement for each of a plurality of different radiotherapy device motion axes.

[0017] By means of a method, the aforementioned modified fluence space optimization solution may include modifying the fluence space optimization solution according to the individual time requirements of at least some of the multiple different radiotherapy device motion axes.

[0018] If necessary, these teachings will also allow the fluence space optimization solution to be updated based on an increased level of compression for the identified fluence details. By means of a method, to provide the aforementioned increased level of compression for the identified fluence details, these teachings will optionally provide: generating a time-demand ranking list of different radiotherapy device motion axes for at least some control points; identifying the fluence details corresponding to slower radiotherapy device motion axes to provide the identified fluence details; and applying the increased level of compression to the identified fluence details. As an illustrative example, the aforementioned fluence details may include fluence lines.

[0019] If needed, these teachings will further allow for responding to user requests by presenting information including a calculated dose distribution based on increased compression levels for the identified infusion details.

[0020] These teachings will also allow for the application of therapeutic radiation to specific patients using specific radiotherapy devices, in accordance with a radiotherapy plan.

[0021] In one approach, these teachings may include a non-transitory computer-readable medium having a computer program that itself includes instructions, when executed by a computer, to cause the computer to perform any or more of the steps, functions, and / or activities described herein, including, for example: iteratively accessing a fluence-space optimized solution for a particular patient's radiotherapy plan, generating control point parameters for a particular radiotherapy apparatus based on the fluence-space optimized solution, determining a treatment delivery time using the control point parameters for the particular radiotherapy apparatus, and modifying the fluence-space optimized solution when the treatment delivery time is suboptimal.

[0022] With this configuration, these teachings can provide a method for optimizing radiotherapy plans, in which fluence space optimization and control point optimization are iteratively solved starting from the complete fluence space optimization solution and gradually reducing the fidelity requirement of control point optimization until a solution with acceptable treatment delivery time is reached.

[0023] These and other benefits will become clearer after a careful review and study of the following detailed description. Refer now to the accompanying drawings, especially... Figure 1 The first thing to be presented is an illustrative device 100 that is compatible with many of the teachings in these teachings.

[0024] In this particular example, the enabling device 100 includes a control circuit 101. As a "circuit", the control circuit 101 therefore includes a structure comprising at least one (and typically many) conductive paths (e.g., paths made of conductive metals such as copper or silver) that carry current in an ordered manner, and which will typically also include corresponding electrical components (passive components such as resistors and capacitors and active components such as any of a variety of semiconductor-based devices, depending on the case) to allow the circuit to implement the control aspects of these teachings.

[0025] This control circuit 101 may include a fixed-purpose hardwired hardware platform (including, but not limited to, an application-specific integrated circuit (ASIC, an integrated circuit designed for a specific purpose rather than for general purposes), a field-programmable gate array (FPGA), etc.) or may include a partially or fully programmable hardware platform (including, but not limited to, a microcontroller, a microprocessor, etc.). Architectural options for these structures are well known and understood in the art and need not be further described herein. The control circuit 101 is configured (e.g., by using corresponding programming that will be fully understood by those skilled in the art) to perform one or more of the steps, actions, and / or functions described herein.

[0026] It should be understood that the control circuit 101 may include a single integrated platform, or may include multiple such circuits that work together with each other.

[0027] Control circuitry 101 is operatively coupled to memory 102. Memory 102 may be integrated into control circuitry 101 or physically separated ( wholly or partially) from control circuitry 101 as needed. Memory 102 may also be local to control circuitry 101 (e.g., both sharing a common circuit board, chassis, power supply, and / or enclosure), or may be partially or completely remote to control circuitry 101 (e.g., memory 102 is physically located in another facility, metropolitan area, or even country compared to control circuitry 101). Like control circuitry 101, memory 102 may comprise a single structure or may comprise multiple memory platforms that collectively comprise the “memory” of this device 100.

[0028] In addition to information such as optimization information for a specific patient and information about a specific radiotherapy platform, the memory 102 can be used, for example, to non-transitory store computer instructions that, when executed by the control circuitry 101, cause the control circuitry 101 to operate as described herein. (As used herein, the reference to "non-transitory" should be understood to mean the non-transitory state of the stored content (and therefore excludes the case where the stored content constitutes only a signal or wave), rather than the volatility of the storage medium itself, and therefore includes non-volatile memory (such as read-only memory (ROM)) as well as volatile memory (such as dynamic random access memory (DRAM)).

[0029] Alternatively, the control circuitry 101 may also be operatively coupled to the user interface 103. The user interface 103 may include any of a variety of user input mechanisms (such as, but not limited to, keyboards and keypads, cursor control devices, touch-sensitive displays, voice recognition interfaces, gesture recognition interfaces, etc.) and / or user output mechanisms (such as, but not limited to, visual displays, audio transducers, printers, etc.) to facilitate receiving information and / or instructions from the user and / or providing information to the user.

[0030] If needed, the control circuitry 101 can also be operatively coupled to a network interface (not shown). The control circuitry 101 configured in this way can communicate with other components (both within and outside the device 100) via the network interface. Network interfaces, including both wireless and non-wireless platforms, are well known in the art and will not be described in detail here.

[0031] By means of a method, computed tomography apparatus 106 and / or other imaging apparatus 107 known in the art can provide some or all of any desired patient-related imaging information.

[0032] In this illustrative example, control circuitry 101 is configured to ultimately output an optimized energy-based treatment plan (e.g., an optimized radiotherapy plan 113). This energy-based treatment plan typically includes specified values ​​for each of various treatment platform parameters during each of multiple consecutive radiation fields. In this case, the energy-based treatment plan is generated through an optimization process, an example of which is further provided herein.

[0033] In one method, control circuitry 101 is operatively coupled to an energy-based treatment platform 114, which is configured to direct treatment to a patient having at least one treatment volume 105 and one or more organs at risk, according to an optimized energy-based treatment plan 113. Figure 1Therapeutic energy 112 is delivered to the corresponding patient 104 (represented by the first organ at risk 108 to the Nth organ at risk 109). These teachings are generally applicable to use with any of a variety of energy-based therapeutic platforms / devices. In a typical application setting, the energy-based therapeutic platform 114 will include an energy source, such as a radiation source 115 of ionizing radiation 116.

[0034] In one method, the radiation source 115 can be selectively moved along an arcuate pathway via a gantry (where the pathway is at least partially contained within the patient themselves during treatment administration). The arcuate pathway may, as needed, comprise a complete or nearly complete circle. In another method, control circuitry 101 controls the movement of the radiation source 115 along the arcuate pathway and can accordingly control when the radiation source 115 begins to move, stops moving, accelerates, decelerates, and / or the speed at which the radiation source 115 travels along the arcuate pathway.

[0035] As an illustrative example, radiation source 115 may include, for example, an X-ray source based on a radio frequency (RF) linear particle accelerator (a linear accelerator). A linear accelerator is a particle accelerator that greatly increases the kinetic energy of charged subatomic particles or ions by subjecting charged particles to a series of oscillating potentials along a linear beamline. This particle accelerator can be used to generate ionizing radiation (e.g., X-rays) 116 and high-energy electrons.

[0036] A typical energy-based treatment platform 114 may also include one or more support devices 110 (such as a treatment bed) to support the patient 104 during treatment, one or more patient fixation devices 111, a gantry or other movable mechanism that allows selective movement of the radiation source 115, and one or more energy shaping devices (e.g., beam shaping devices 117, such as jaws, multi-leaf collimators, etc.) to provide selective energy shaping and / or energy modulation as needed.

[0037] In a typical application setting, this document assumes that the patient support device 110 is selectively controllable to move in any direction (i.e., any X, Y, or Z direction) via control circuitry 101 during an energy-based therapeutic session. Since the foregoing components and systems are well known in the art, no further detailed description of these aspects is provided herein except where particularly relevant to the description.

[0038] Now for reference Figure 2 The process 200 will be described, which can be executed, for example, in conjunction with the application settings described above (and more specifically via the control circuit 101 described above). Generally, the process 200 is used to facilitate the generation of an optimized radiotherapy plan 113, thereby facilitating the treatment of a specific patient with therapeutic radiation using a specific radiotherapy platform according to the optimized radiotherapy plan.

[0039] As indicated by reference numeral 201 in the attached diagram, many steps of process 200 occur iteratively, and particularly during iterations of the radiotherapy planning optimization process. In one approach, the described steps are performed during each optimization iteration. In another approach, the described steps are performed only during certain iterations (e.g., a first number of iterations may occur without process 200 being performed).

[0040] At box 202, control circuit 101 accesses the fluence-space optimized solution for a radiotherapy plan for a specific patient. (Fluorescence represents the radiation flux integrated over time and includes fundamental metrics in dosimetry (i.e., the measurement and calculation of the absorbed dose of ionizing radiation in matter and tissue). Thus, by a method, radiotherapy plan optimization has begun and has progressed to the point where a fluence-space optimized solution is available.

[0041] At box 203, control circuitry 101 generates control point parameters for a specific radiotherapy apparatus based on the accessed fluence space optimization solution. (A control point is a specific location or angle at which the radiation beam is turned on or off during treatment. Control points are typically used to help shape the radiation beam and optimize dose distribution to target the tumor while protecting surrounding healthy tissue.) Generating these control point parameters may include generating a lossless transformation from input fluence information to control point parameters, enabling a specific radiotherapy apparatus to deliver (at least theoretically) a fluence at least nearly equivalent to the fluence space optimization solution. (As used herein, it should be understood that the term "lossless" includes "nearly lossless," as a completely lossless scenario may be practically difficult or even impossible to achieve. By one method, and by way of example, "nearly lossless" may be specified when the sum of the absolute differences between the optimal and reproduced fluence does not exceed 5% of the total fluence intensity.)

[0042] One method allows the generation of the aforementioned control point parameters to initially be performed without any limitation on the number of automatically assigned control points. Alternatively, or in combination with the foregoing, these teachings would also allow the generation of the aforementioned control point parameters to initially be performed without any limitation on the time required to complete the input dose information for a specific radiotherapy device.

[0043] At block 204, control circuitry 101 uses the generated control point parameters for a specific radiotherapy apparatus to determine the treatment delivery time. In one approach, this treatment delivery time represents the delivery time from start to finish. Alternatively, the treatment delivery time can be specified in some other way, if desired. In one approach, these teachings would allow for determining the timing requirements for each of several different radiotherapy apparatus motion axes (such as different delivery positions of the radiation source, different positions of a particular patient during treatment, etc.).

[0044] At box 205, control circuitry 101 then determines whether the aforementioned treatment delivery time is suboptimal. In a typical application setting, this determination will identify whether the treatment delivery time exceeds a predetermined maximum time. This maximum time can be specified by the user, can be a maximum time stipulated by the treatment facility, can be a default value adopted by control circuitry 101, or can be determined in other ways as needed.

[0045] When the determination concludes that the treatment delivery time is not suboptimal, the process can tolerate any of the various responses. By a particular method, the optimization of the radiotherapy plan can now conclude because an acceptable sequence of control points has been determined. If necessary, the control points can be further pruned, for example, by removing very small apertures, or the control points can be reordered to improve delivery time. After these control points have been defined, the optimization process can also proceed in different ways depending on the optimization scheme. This might mean, for example, terminating the optimization process and accepting the control points as is, or there might be further optimization steps (e.g., a step based on a new dose calculation with improved accuracy using the control points defined above).

[0046] However, when the determination concludes that the treatment delivery time is suboptimal, at block 206, control circuitry 101 modifies the fluence space optimization solution for use in subsequent iterative loops of reusing the process 200. Modifying the fluence space optimization solution may include adapting it to the individual time requirements of at least some of the multiple different radiotherapy device motion axes.

[0047] Alternatively, modifying the injection space optimization solution may also include updating the injection space optimization solution based on an increased level of compression for the identified injection detail 207. (For now, see reference only.) Figure 3 Now, we will describe an illustrative method for providing such content.

[0048] At block 301, control circuitry 101 generates a sorted list of time requirements for different radiotherapy device motion axes for at least some control points. At block 302, control circuitry 101 then identifies the infusion details corresponding to the slower radiotherapy device motion axes to provide the identified infusion details. And at block 303, the process 300 provides an increased level of compression applied to the identified infusion details.

[0049] One method allows the aforementioned dose details to at least partially comprise dose rows. Diffusion can be represented using a dose map, which is a grid-like representation of dose distribution. Such a map can present intensity levels at different points across the treatment field. This map is typically divided into small regions called pixels or cells, and the intensity level in each cell represents the amount of radiation to be delivered to that particular region. In this context, a "dose row" refers to a single row of pixels or cells within that grid. Thus, a dose row is essentially a one-dimensional slice or portion of a dose map spanning the treatment field along a specified direction. Each pixel in the row indicates the intensity level of the beam to be delivered to the corresponding portion of the treatment area.

[0050] By an optional method, and as indicated at optional box 304, the process 300 may monitor user requests. In response to detecting the user request, and as indicated at optional box 305, the control circuit 101 may present information including a calculated dose distribution based on the now increased compression level for the identified dosing details.

[0051] Refer again Figure 2 And as indicated by reference numeral 208 in the attached diagram, the above iterative process ends. After iteration, an optimized radiotherapy plan is now available, and at option 209, the process will allow therapeutic radiation to be delivered to a specific patient using a specific radiotherapy platform according to this optimized radiotherapy plan.

[0052] Further details consistent with these teachings will now be presented. It should be understood that the specific details of these examples are for illustrative purposes and are not intended to imply any particular limitation on these teachings.

[0053] First, for some readers, it may be helpful to know additional information about current optimization techniques.

[0054] Flux space optimization typically involves finding a set of flux maps whose dose distribution minimizes a given cost function based on planning objectives. This is generally a relatively well-performing problem and can be solved efficiently using standard gradient-based methods. To deliver flux space optimization solutions using a radiotherapy platform, the flux maps need to be converted into control point parameters, such as multi-leaf collimator blade positions and monitoring units.

[0055] In control point optimization, a predefined number of control points are typically optimized to deliver a dose distribution close to the fluence space optimization solution, and may include cost terms related to treatment time (directly or indirectly) and / or other parameters associated with treatment complexity, such as aperture shape constraints. In practice, control point optimization typically aims to maximize the fidelity of dose delivery (e.g., by minimizing conversion losses) because it is constrained by speed limits applicable to the machine's motion axes.

[0056] Control point optimization is typically non-convex and has high degrees of freedom. Variations of gradient methods used to solve this problem are sensitive to initialization, and the applicant has determined that such methods can easily get trapped in suboptimal solutions. Furthermore, due to performance reasons, control point optimization methods often use coarse approximations of first and second gradients, thus casting doubt on the validity of the solutions. Therefore, the applicant has determined that machine-generated infusions may deviate significantly from the optimal infusion space solution. This can lead to suboptimal dose distributions and degraded plan quality.

[0057] For a given set of planning objectives, the optimal achievable plan quality is typically not revealed to the planner because the fluence space optimization solution is usually only within the radiotherapy planning optimization algorithm. Even if such a baseline is provided, improving the performance of control point optimization methods for each specific case requires complex additional tools and interfaces that allow planners to manually allocate more control points in certain directions or relax speed constraints by extending delivery times. Instead of the foregoing or in combination, planners may use case-sensitive initialization methods or choose from a variety of control point optimization solver algorithms. The applicant has determined that handling such tools makes the plan generation process more cumbersome and time-consuming.

[0058] Given the above background, typical existing methods find suboptimal solutions to the control point optimization problem simply by using approximate gradient methods or stochastic search strategies with a large number of trials. This means that the transformation from fluence space optimization of the fluence map to control point parameters can be very costly. Nevertheless, the distortion of the transformation is usually simply ignored. Instead, radiotherapy planning optimization continues without knowing how much the planning quality could be improved if the transformation loss were adjusted.

[0059] To address at least some of the problems mentioned above, these teachings can provide methods for optimizing radiotherapy planning, where the transformations described are not formulated as optimization problems. Therefore, the approach of solving control point optimization as a non-convex high-dimensional problem is avoided. Instead, the process begins with an optimal solution in the fluence space, generates an optimal quality plan through lossless control point creation, and then gradually increases fluence compression until the desired level of planning complexity (or treatment delivery time) is reached.

[0060] Figure 4 Illustrative examples of these aspects are presented. It should be understood that the details of these examples are for illustrative purposes and are not intended to imply any limitation on these teachings.

[0061] In addition to the cost function 401 based on the planned objectives and the injection optimizer 402 that operates on the cost function 401 and the corresponding application field geometry, the illustrative device 400 also includes a high-fidelity converter 403, an aperture sorter 404, a delivery time calculator 405, and an image compressor 406.

[0062] The high-fidelity converter 403 performs a non-destructive conversion from the input fluence map to control point parameters, enabling the treatment machine to deliver a fluence at least nearly identical to the input. It should be noted that there is no limit to the time required for the machine to generate those fluences. Furthermore, the number of control points required is neither set nor limited. Instead, the high-fidelity converter 403 can allocate as many control points as needed to facilitate non-destructive conversion. The number of control points required may depend on, for example, the complexity of the input fluence. Such converters sometimes use variations of the sliding window technique, and these teachings will allow for the use of such converters.

[0063] Aperture sorter 404 finds a reasonable sorting order for the apertures of the multi-leaf collimator so that the blades in adjacent control points do not waste excessive delivery time due to unnecessary back-and-forth movements. (If needed, these teachings will support including this sorting activity in the high-fidelity converter 403).

[0064] The delivery time calculator 405 is configured to calculate the time required for the therapeutic machine to treat a given sequence of control points. As part of the total calculation time, the delivery time calculator 405 can also be configured to provide detailed timing information for each axis of motion of the therapeutic machine.

[0065] Furthermore, image compressor 406 may include a conventional adjustable compressor used in image processing to reduce the size of the input letter line to a desired level. These teachings will allow for a variety of approaches in these areas, ranging from methods based on traditional wavelet decomposition to more advanced methods utilizing neural networks or other forms of artificial intelligence.

[0066] Assuming the aforementioned components are used, illustrative use cases for optimizing radiotherapy plans will now be provided.

[0067] At the outset, the user / planner can set the desired delivery time T0409. The injection space optimization solution 411 is then obtained by running the injection optimizer 402 against a given set of planning objectives 407 and a field geometry 408 associated with the treatment setup. The image compressor 406 can be set to an initial compression level of zero.

[0068] The high-fidelity converter 403 outputs a defined number of control points 410 and control point parameters required for non-destructive delivery of the injection volume to optimize the injection space. The aperture sequencer 404 outputs the aperture sorting order 412, and the delivery time calculator 405 outputs the total delivery time T of the baseline plan. t 413.

[0069] Assuming total delivery time T tIf 413 is greater than the expected delivery time T0409, then the injection row (or the entire graph) corresponding to the slowest axis can be identified, and for the injection row (or the entire graph) that causes the deceleration (or is most responsible for causing the deceleration), the compression level in the image compressor 406 is increased by a step size ε, which can then be applied as the compression level in the image compressor 406 to the injection.

[0070] The compressed fluence 414 is provided to the fluence optimizer, which updates the previous fluence space optimization solution for the uncompressed fluence. The high-fidelity converter 403 updates the control points 410 and the parameters required for lossless delivery of the new solution, and the aperture sorter 404 outputs a new sorting order of these output apertures.

[0071] Delivery Time Calculator 405 recalculates total delivery time T t 413, and as previously described, compress the appropriate injection volume and provide a new injection space optimization solution.

[0072] This iterative method can continue in a loop as described (as indicated by the arrow in the attached diagram, number 415) until the total delivery time is equal to or less than the expected delivery time. (These teachings will also allow the process to be stopped, for example, after a given number of iterations.)

[0073] If needed, and upon detecting a user / planner's request for intermediate planning information, these instructions will allow the calculation of the dose distribution corresponding to the compressed injection volume and the presentation of that information via the user interface.

[0074] If needed, this training will allow the reinforcement learning agent to be trained using control point-specific times calculated by the delivery time calculator 405 to apply a specific level of compression. Using this method, it is possible to minimize the impact of compression on program quality.

[0075] Similarly, if needed, these instructions can be configured to provide users / planners with a desired or a given number of plan alternatives, ranging from the highest quality but slowest delivery plans to lower quality but fastest delivery plans. Users / planners can then choose the plan best suited to the needs of a given application setup.

[0076] These teachings can help minimize the time the optimizer spends searching the large non-convex solution space of the control point optimization problem. They also help minimize the time users / planners spend searching the unproductive parts of the solution space. Unlike typical existing methods, these teachings provide the optimal fluence space optimization solution and its corresponding delivery time as a benchmark for a given cost function. Furthermore, these teachings enable users / planners to make informed decisions regarding the trade-offs between plan quality and delivery time. Those skilled in the art will also understand that these teachings are GPU-friendly, computationally fast, and easily interactive with artificial intelligence techniques to offload plan generation tasks from human users / planners.

[0077] Further aspects of these teachings are provided by the subject matter of the following clauses (it should be understood that any one of these clauses may be combined with one or more other clauses in the other clauses as needed).

[0078] Clause 1. A method for using a patient-specific optimized radiotherapy plan that uses a specific radiotherapy device, the method comprising: by control circuitry: iteratively: accessing a fluence space optimization solution of the patient-specific radiotherapy plan; generating control point parameters for the specific radiotherapy device based on the fluence space optimization solution; determining a treatment delivery time using the control point parameters for the specific radiotherapy device; and modifying the fluence space optimization solution when the treatment delivery time is suboptimal.

[0079] Clause 2. The method according to Clause 1, wherein generating control point parameters includes generating a lossless transformation from input fluence information to control point parameters, such that a particular radiotherapy device is at least theoretically capable of delivering a fluence at least nearly equivalent to the fluence space optimization solution.

[0080] Clause 3. The method described in Clause 2, wherein the generation of control point parameters is initially performed without any limitation on the number of automatically assigned control points.

[0081] Clause 4. The method according to Clause 2, wherein the generation of control point parameters is initially performed without any time limit required to complete the input of injection volume information for a particular radiotherapy device.

[0082] Clause 5. The method according to Clause 1, wherein determining the treatment delivery time using control point parameters for a particular radiotherapy device includes: determining the time requirement for each of a plurality of different radiotherapy device motion axes.

[0083] Clause 6. The method according to Clause 5, wherein modifying the fluence space optimization solution includes modifying the fluence space optimization solution based on the individual time requirements of at least some of the multiple different radiotherapy device motion axes.

[0084] Clause 7. The method according to Clause 6 further comprises: generating a time requirement sorting list for different radiotherapy device motion axes for at least some control points; identifying the injection details corresponding to slower radiotherapy device motion axes to provide the identified injection details; applying an increased compression level to the identified injection details; and wherein modifying the injection space optimization solution comprises updating the injection space optimization solution according to the increased compression level applied to the identified injection details.

[0085] Clause 8. The method described in Clause 7, wherein the volume details include the volume line.

[0086] Clause 9. The method of Clause 7 further comprises: in response to a user request, presenting information including a calculated dose distribution based on an increased level of compression for the identified dose details.

[0087] Clause 10. The method described in Clause 1 further includes: administering therapeutic radiation to a specific patient using a specific radiotherapy device in accordance with a radiotherapy plan.

[0088] Clause 11. An apparatus for optimizing a radiotherapy plan for a specific patient using a specific radiotherapy apparatus, the apparatus comprising: control circuitry configured to: iteratively: access a fluence space optimization solution for the radiotherapy plan for the specific patient; generate control point parameters for the specific radiotherapy apparatus based on the fluence space optimization solution; determine a treatment delivery time using the control point parameters for the specific radiotherapy apparatus; and modify the fluence space optimization solution when the treatment delivery time is suboptimal.

[0089] Clause 12. The apparatus of Clause 11, wherein the control circuitry is configured to generate control point parameters by generating a lossless transformation from input fluence information to control point parameters, such that a particular radiotherapy apparatus is at least theoretically capable of delivering a fluence at least nearly equivalent to the fluence space optimization solution.

[0090] Clause 13. The apparatus according to Clause 12, wherein the control circuitry is configured to generate control point parameters by initially generating control point parameters without any limitation on the number of automatically assigned control points.

[0091] Clause 14. The apparatus of Clause 12, wherein the control circuitry is configured to generate control point parameters by generating control point parameters initially without any limitation on the time required to complete input dose information for a particular radiotherapy device.

[0092] Clause 15. The apparatus of Clause 11, wherein the control circuitry is configured to determine the treatment delivery time using control point parameters of a particular radiotherapy device by determining the time requirement for each of a plurality of different radiotherapy device motion axes.

[0093] Clause 16. The apparatus according to Clause 15, wherein the control circuitry is configured to modify the fluence space optimization solution by modifying the fluence space optimization solution according to the individual time requirements of at least some of the multiple different radiotherapy device motion axes.

[0094] Clause 17. The apparatus according to Clause 16, wherein the control circuitry is further configured to: generate a time requirement sorting list for different radiotherapy device motion axes for at least some control points; identify the flux details corresponding to slower radiotherapy device motion axes to provide the identified flux details; apply an increased compression level to the identified flux details; and such that modifying the flux space optimization solution includes updating the flux space optimization solution according to the increased compression level applied to the identified flux details.

[0095] Clause 18. The apparatus as described in Clause 17, wherein the volume details include a volume line.

[0096] Clause 19. The apparatus of Clause 17, wherein the control circuitry is further configured to: in response to a user request, present information including a calculated dose distribution based on an increased level of compression for the identified dose details.

[0097] Clause 20. A nontransitory computer-readable medium for optimizing a radiotherapy plan for a specific patient using a specific radiotherapy apparatus, the nontransitory computer-readable medium comprising instructions stored on the nontransitory computer-readable medium that, when executed on a processor, perform the following steps: iteratively: accessing a fluence-space optimized solution for the radiotherapy plan for the specific patient; generating control point parameters for the specific radiotherapy apparatus based on the fluence-space optimized solution; determining a treatment delivery time using the control point parameters for the specific radiotherapy apparatus; and modifying the fluence-space optimized solution when the treatment delivery time is suboptimal.

[0098] Those skilled in the art will recognize that various modifications, alterations, and combinations can be made to the above embodiments without departing from the scope of the invention, and such modifications, alterations, and combinations should be considered within the scope of the inventive concept.

Claims

1. A method for optimizing a radiotherapy plan for a specific patient, the radiotherapy plan using a specific radiotherapy device, the method comprising: Control circuit: Iteratively: Access the fluence-space optimization solution for the radiotherapy plan for the specific patient; Control point parameters for the specific radiotherapy device are generated based on the fluence space optimization solution. The treatment delivery time is determined using the control point parameters for the specific radiotherapy device. When the treatment delivery time is suboptimal, the injection space optimization solution is modified.

2. The method according to claim 1, wherein generating the control point parameters comprises: A lossless transformation from input fluence information to the control point parameters is generated, such that the specific radiotherapy device is at least theoretically capable of delivering a fluence at least nearly equivalent to the fluence space optimization solution.

3. The method of claim 2, wherein the generation of the control point parameters is initially performed without any limitation on the number of automatically assigned control points.

4. The method of claim 2, wherein the generation of the control point parameters is initially performed without any limitation on the time required for the specific radiotherapy device to complete the input injection information.

5. The method of claim 1, wherein determining the treatment delivery time using the control point parameters for the specific radiotherapy device comprises: Determine the time requirements for each of the multiple different radiotherapy device motion axes.

6. The method according to claim 5, wherein modifying the injection space optimization solution includes: The infusion space optimization solution is modified based on the individual time requirements of at least some of the multiple different radiotherapy device motion axes.

7. The method according to claim 6, further comprising: Generate a sorted list of time requirements for the different motion axes of the radiotherapy device for at least some control points; The injection details are indicated corresponding to the motion axis of the slower radiotherapy device to provide the indicated injection details; An increased level of compression is applied to the identified volume details; Furthermore, modifying the injection space optimization solution includes updating the injection space optimization solution based on the increased compression level for the identified injection details.

8. The method of claim 7, wherein the volume details include a volume line.

9. The method according to claim 7, further comprising: In response to a user request, information including a calculated dose distribution is presented, the calculated dose distribution being based on the increased compression level for the identified dosage details.

10. The method according to claim 1, further comprising: According to the radiotherapy plan, therapeutic radiation is administered to the specific patient using the specific radiotherapy device.

11. An apparatus for optimizing a radiotherapy plan for a specific patient, the radiotherapy plan using a specific radiotherapy apparatus, the apparatus comprising: The control circuit is configured iteratively: Access the fluence-space optimization solution for the radiotherapy plan for the specific patient; Control point parameters for the specific radiotherapy device are generated based on the fluence space optimization solution. The treatment delivery time is determined using the control point parameters for the specific radiotherapy device. When the treatment delivery time is suboptimal, the injection space optimization solution is modified.

12. The apparatus of claim 11, wherein the control circuitry is configured to generate the control point parameters by generating a lossless transformation from input fluence information to the control point parameters, such that the particular radiotherapy apparatus is at least theoretically capable of delivering a fluence at least nearly equivalent to the fluence space optimization solution.

13. The apparatus of claim 12, wherein the control circuitry is configured to generate the control point parameters by initially generating the control point parameters without any limitation on the number of automatically assigned control points.

14. The apparatus of claim 12, wherein the control circuitry is configured to generate the control point parameters by initially generating the control point parameters without any limitation on the time required to complete the input injection information for the particular radiotherapy device.

15. The apparatus of claim 11, wherein the control circuitry is configured to determine the treatment delivery time using the control point parameters of the particular radiotherapy device by determining the time requirement for each of a plurality of different radiotherapy device motion axes.

16. The apparatus of claim 15, wherein the control circuitry is configured to modify the fluence space optimization solution by modifying the fluence space optimization solution according to the individual time requirements of at least some of the plurality of different radiotherapy device motion axes.

17. The apparatus of claim 16, wherein the control circuit is further configured to: Generate a sorted list of time requirements for the different motion axes of the radiotherapy device for at least some control points; The injection details are indicated corresponding to the motion axis of the slower radiotherapy device to provide the indicated injection details; An increased level of compression is applied to the identified volume details; Modifying the injection space optimization solution includes updating the injection space optimization solution based on the increased compression level for the identified injection details.

18. The apparatus of claim 17, wherein the dosing details include dosing lines.

19. The apparatus of claim 17, wherein the control circuit is further configured to: In response to a user request, information including a calculated dose distribution is presented, the calculated dose distribution being based on the increased compression level for the identified dosage details.

20. A non-transitory computer-readable medium for optimizing a radiotherapy plan for a specific patient, the radiotherapy plan using a specific radiotherapy apparatus, the non-transitory computer-readable medium including instructions stored on the non-transitory computer-readable medium, which, when executed on a processor, perform the following steps: Iteratively: Access the fluence-space optimization solution for the radiotherapy plan for the specific patient; Control point parameters for the specific radiotherapy device are generated based on the fluence space optimization solution. The treatment delivery time is determined using the control point parameters for the specific radiotherapy device. When the treatment delivery time is suboptimal, the injection space optimization solution is modified.

Images